Mary Kane Goldstein

All Publications

• Laboratory-wide association study of survival with prostate cancer. Cancer Sohlberg, E. M., Thomas, I., Yang, J., Kapphahn, K., Velaer, K. N., Goldstein, M. K., Wagner, T. H., Chertow, G. M., Brooks, J. D., Patel, C. J., Desai, M., Leppert, J. T. 2020

  Abstract

  BACKGROUND: Estimates of overall patient health are essential to inform treatment decisions for patients diagnosed with cancer. The authors applied XWAS methods, herein referred to as "laboratory-wide association study (LWAS)", to evaluate associations between routinely collected laboratory tests and survival in veterans with prostate cancer.METHODS: The authors identified 133,878 patients who were diagnosed with prostate cancer between 2000 and 2013 in the Veterans Health Administration using any laboratory tests collected within 6 months of diagnosis (3,345,083 results). Using the LWAS framework, the false-discovery rate was used to test the association between multiple laboratory tests and survival, and these results were validated using training, testing, and validation cohorts.RESULTS: A total of 31 laboratory tests associated with survival met stringent LWAS criteria. LWAS confirmed markers of prostate cancer biology (prostate-specific antigen: hazard ratio [HR], 1.07 [95% confidence interval (95% CI), 1.06-1.08]; and alkaline phosphatase: HR, 1.22 [95% CI, 1.20-1.24]) as well laboratory tests of general health (eg, serum albumin: HR, 0.78 [95% CI, 0.76-0.80]; and creatinine: HR, 1.05 [95% CI, 1.03-1.07]) and inflammation (leukocyte count: HR, 1.23 [95% CI, 1.98-1.26]; and erythrocyte sedimentation rate: HR, 1.33 [95% CI, 1.09-1.61]). In addition, the authors derived and validated separate models for patients with localized and advanced disease, identifying 28 laboratory markers and 15 laboratory markers, respectively, in each cohort.CONCLUSIONS: The authors identified routinely collected laboratory data associated with survival for patients with prostate cancer using LWAS methodologies, including markers of prostate cancer biology, overall health, and inflammation. Broadening consideration of determinants of survival beyond those related to cancer itself could help to inform the design of clinical trials and aid in shared decision making.LAY SUMMARY: This article examined routine laboratory tests associated with survival among veterans with prostate cancer. Using laboratory-wide association studies, the authors identified 31 laboratory tests associated with survival that can be used to inform the design of clinical trials and aid patients in shared decision making.

  View details for DOI 10.1002/cncr.33341

  View details for PubMedID 33237577

• OrderRex clinical user testing: a randomized trial of recommender system decision support on simulated cases. Journal of the American Medical Informatics Association : JAMIA Kumar, A., Aikens, R. C., Hom, J., Shieh, L., Chiang, J., Morales, D., Saini, D., Musen, M., Baiocchi, M., Altman, R., Goldstein, M. K., Asch, S., Chen, J. H. 2020

  Abstract

  OBJECTIVE: To assess usability and usefulness of a machine learning-based order recommender system applied to simulated clinical cases.MATERIALS AND METHODS: 43 physicians entered orders for 5 simulated clinical cases using a clinical order entry interface with or without access to a previously developed automated order recommender system. Cases were randomly allocated to the recommender system in a 3:2 ratio. A panel of clinicians scored whether the orders placed were clinically appropriate. Our primary outcome included the difference in clinical appropriateness scores. Secondary outcomes included total number of orders, case time, and survey responses.RESULTS: Clinical appropriateness scores per order were comparable for cases randomized to the order recommender system (mean difference -0.11 order per score, 95% CI: [-0.41, 0.20]). Physicians using the recommender placed more orders (median 16 vs 15 orders, incidence rate ratio 1.09, 95%CI: [1.01-1.17]). Case times were comparable with the recommender system. Order suggestions generated from the recommender system were more likely to match physician needs than standard manual search options. Physicians used recommender suggestions in 98% of available cases. Approximately 95% of participants agreed the system would be useful for their workflows.DISCUSSION: User testing with a simulated electronic medical record interface can assess the value of machine learning and clinical decision support tools for clinician usability and acceptance before live deployments.CONCLUSIONS: Clinicians can use and accept machine learned clinical order recommendations integrated into an electronic order entry interface in a simulated setting. The clinical appropriateness of orders entered was comparable even when supported by automated recommendations.

  View details for DOI 10.1093/jamia/ocaa190

  View details for PubMedID 33106874

• "It's a disease of families": Neurologists' insights on how to improve communication and quality of life for families of Parkinson's disease patients CHRONIC ILLNESS Schwartz, R., Zulman, D., Gray, C., Goldstein, M. K., Trivedi, R. 2020; 16 (3): 201–11

  View details for DOI 10.1177/1742395318799852

  View details for Web of Science ID 000552340000004

• Life expectancy estimates for patients diagnosed with prostate cancer in the Veterans Health Administration. Urologic oncology Sohlberg, E. M., Thomas, I., Yang, J., Kapphahn, K., Daskivich, T. J., Skolarus, T. A., Shelton, J. B., Makarov, D. V., Bergman, J., Bang, C. K., Goldstein, M. K., Wagner, T. H., Brooks, J. D., Desai, M., Leppert, J. T. 2020

  Abstract

  PURPOSE: Accurate life expectancy estimates are required to inform prostate cancer treatment decisions. However, few models are specific to the population served or easily implemented in a clinical setting. We sought to create life expectancy estimates specific to Veterans diagnosed with prostate cancer.MATERIALS AND METHODS: Using national Veterans Health Administration electronic health records, we identified Veterans diagnosed with prostate cancer between 2000 and 2015. We abstracted demographics, comorbidities, oncologic staging, and treatment information. We fit Cox Proportional Hazards models to determine the impact of age, comorbidity, cancer risk, and race on survival. We stratified life expectancy estimates by age, comorbidity and cancer stage.RESULTS: Our analytic cohort included 145,678 patients. Survival modeling demonstrated the importance of age and comorbidity across all cancer risk categories. Life expectancy estimates generated from age and comorbidity data were predictive of overall survival (C-index 0.676, 95% CI 0.674-0.679) and visualized using Kaplan-Meier plots and heatmaps stratified by age and comorbidity. Separate life expectancy estimates were generated for patients with localized or advanced disease. These life expectancy estimates calibrate well across prostate cancer risk categories.CONCLUSIONS: Life expectancy estimates are essential to providing patient-centered prostate cancer care. We developed accessible life expectancy estimation tools for Veterans diagnosed with prostate cancer that can be used in routine clinical practice to inform medical-decision making.

  View details for DOI 10.1016/j.urolonc.2020.05.015

  View details for PubMedID 32674954

• OUTPATIENT CARE FRAGMENTATION PATTERNS AND ASSOCIATION WITH HOSPITALIZATION IN HIGH-RISK VA PATIENTS Zulman, D. M., Greene, A. L., Wagner, T. H., Maciejewski, M., Singer, S., Slightam, C., Trivedi, R., Vanneman, M., Asch, S., Yoon, J., Goldstein, M. K., Boothroyd, D. SPRINGER. 2020: S216

  View details for Web of Science ID 000567143600487

• Guided self-management targeting anxiety and activity participation in older Veterans. Aging & mental health Gould, C. E., Carlson, C., Wetherell, J. L., O'Hara, R., Goldstein, M. K., Loup, J. R., Ma, F., Beaudreau, S. A. 2020: 1–10

  Abstract

  Objectives: This study examined the acceptance, feasibility, and preliminary effects of a guided self-management intervention using video delivery and a telephone coach on anxiety and activity engagement.Method: Ten Veterans aged 60years or older with anxiety disorders determined by Structured Clinical Interview for Diagnostic and Statistical Manual 5th edition (SCID-5) at baseline visit participated in this non-randomized study examining a 4-week guided self-management intervention for anxiety. Feasibility was examined using participation engagement with the intervention. Measures of anxiety (Geriatric Anxiety Scale, PROMIS Anxiety Scale, Anxiety Control Questionnaire), depression (Patient Health Questionnaire 9-item), and activity participation (modified Activity Card Sort) administered at baseline and final (week 8) visit provided estimates of preliminary intervention effects. The Geriatric Anxiety Scale also was administered by phone at week 4. Participants completed a semi-structured qualitative interview at the final visit, which provided information about the acceptability, benefits of intervention, and barriers to engagement.Results: All participants (N=10) reported that the intervention somewhat or completely met their expectations, demonstrating intervention acceptability. Intervention completers (n=9) experienced reduced anxiety over the first 4weeks, alongside significant improvements in anxiety control and personalized activity goals across 8weeks. However, anxiety symptoms tended to return to baseline at follow-up. Participants identified the relaxation videos and promotion of a daily relaxation routine as the most helpful intervention components.Conclusions: Findings indicate that the intervention may improve activity participation and reduce anxiety. Thus, guided self-management interventions show promise for reducing distress and maintaining engagement later in life.

  View details for DOI 10.1080/13607863.2020.1758905

  View details for PubMedID 32397822

• Stakeholder Engagement for a Planned Automated Quality Measurement System SAGE OPEN Kalsy, M., Kelly, N., Meystre, S. M., Kim, Y., Bray, B. E., Bolton, D., Goldstein, M. K., Garvin, J. H. 2020; 10 (2)

  View details for DOI 10.1177/2158244020919459

  View details for Web of Science ID 000537286800001

• Receipt of Eye Care Services among Medicare Beneficiaries with and without Dementia. Ophthalmology Pershing, S., Goldstein, M. K., Henderson, V. W., Bundorf, M. K., Lu, Y., Rahman, M., Stein, J. D. 2020

  Abstract

  PURPOSE: To examine the relationship between dementia status and receipt of eye care among US Medicare beneficiaries.DESIGN: Retrospective, claims-based analysis.PARTICIPANTS: A 20% representative sample of Medicare beneficiaries who received care between January 1, 2006, and December 31,2015.METHODS: Dementia was identified from diagnosis codes documented in a beneficiary's first 3 years of observed Medicare enrollment. Eye care visits were identified from provider specialty codes on each encounter claim. We used multivariable Cox proportional hazards regression models with time-varying covariates to compare the likelihood of receiving eye care between beneficiaries with and without dementia. All models were adjusted for potential confounders, including demographics, urban/rural residence, systemic health (Charlson Index), and ocular comorbidities.MAIN OUTCOME MEASURES: Hazard ratio (HR) and 95% confidence interval (CI) for (1) being seen by any eye care provider (ophthalmologist or optometrist); (2) being seen by an ophthalmologist specifically; and (3) receiving cataract surgery (among beneficiaries with ophthalmologist encounters).RESULTS: A total of 4451200 beneficiaries met inclusion criteria; 3805718 (85.5%) received eye care during the study period, and 391 556 (8.8%) had diagnosed dementia. Some 73.4% of beneficiaries diagnosed with dementia saw an eye care provider during the study period and 55.4% saw an ophthalmologist versus 86.7% and 74.0% of beneficiaries, respectively, without dementia diagnoses. Compared with those without dementia diagnoses, beneficiaries with diagnosed dementia had lower likelihood of seeing any eye care provider (adjusted HR, 0.69; 95% CI, 0.69-0.70) and were less likely to see an ophthalmologist (adjusted HR, 0.55; 95% CI, 0.55-0.55). Among the subset of beneficiaries who did see ophthalmologists, those with diagnosed dementia were also less likely to receive cataract surgery than beneficiaries without diagnosed dementia (HR, 0.62; 95% CI, 0.62-0.63) and less likely to receive a cataract diagnosis (18% vs. 82%).CONCLUSIONS: US Medicare beneficiaries diagnosed with dementia are less likely to receive eye care thanthose without diagnosed dementia. Depending on visual acuity and functional status, this may have implications for injury prevention, physical and cognitive function, and quality of life. Further work is needed to identify barriers to receiving eye care, determine eye care services and settings that provide greatest value topatients with dementia, and implement measures to improve access to appropriate eye care.

  View details for DOI 10.1016/j.ophtha.2020.02.022

  View details for PubMedID 32317179

• Cataract Surgery Complexity and Surgical Complication Rates among Medicare Beneficiaries with and without Dementia. American journal of ophthalmology Pershing, S. n., Henderson, V. W., Goldstein, M. K., Lu, Y. n., Bundorf, M. K., Rahman, M. n., Stein, J. D. 2020

  Abstract

  To evaluate cataract surgery complexity and complications among US Medicare beneficiaries with and without dementia.Retrospective claims-based cohort study PARTICIPANTS: A 20% representative sample of Medicare beneficiaries, 2006-2015.Dementia was identified from diagnosis codes on or prior to each beneficiary's first-eye cataract surgery. For each surgery, we identified setting, routine versus complex coding, anesthesia provider type, duration, and any postoperative hospitalization. We evaluated 30- and 90-day complication rates-return to operating room, endophthalmitis, suprachoroidal hemorrhage, retinal detachment, retinal tear, macular edema, glaucoma, or choroidal detachment-and used adjusted regression models to evaluate likelihood of surgical characteristics and complications. Complications analyses were stratified by second-eye cataract surgery within 90 days postoperatively.We identified 457,128 beneficiaries undergoing first-eye cataract surgery, 23,332 (5.1%) with dementia. None of the evaluated surgical complications were more likely in dementia-diagnosed beneficiaries. There was also no difference in likelihood of non-ambulatory surgery center setting, anesthesiologist provider, or postoperative hospitalization. Dementia-diagnosed beneficiaries were more likely to have surgeries coded as complex (15.6% of cases versus 8.8%, p<0.0001), and surgeries exceeding 30 minutes (OR=1.21, 95%CI=1.17-1.25).Among US Medicare beneficiaries undergoing cataract surgery, those with dementia are more likely to have "complex" surgery" lasting over 30 minutes. However, they do not have greater likelihood of surgical complications, higher-acuity setting, advanced anesthesia care, or postoperative hospitalization. This may be influenced by case selection and may suggest missed opportunities to improve vision. Future research is needed to identify dementia patients likely to benefit from cataract surgery.

  View details for DOI 10.1016/j.ajo.2020.08.025

  View details for PubMedID 32828874

• Physician Usage and Acceptance of a Machine Learning Recommender System for Simulated Clinical Order Entry. AMIA Joint Summits on Translational Science proceedings. AMIA Joint Summits on Translational Science Chiang, J., Kumar, A., Morales, D., Saini, D., Hom, J., Shieh, L., Musen, M., Goldstein, M. K., Chen, J. H. 2020; 2020: 89–97

  Abstract

  Clinical decision support tools that automatically disseminate patterns of clinical orders have the potential to improve patient care by reducing errors of omission and streamlining physician workflows. However, it is unknown if physicians will accept such tools or how their behavior will be affected. In this randomized controlled study, we exposed 34 licensed physicians to a clinical order entry interface and five simulated emergency cases, with randomized availability of a previously developed clinical order recommender system. With the recommender available, physicians spent similar time per case (6.7 minutes), but placed more total orders (17.1 vs. 15.8). The recommender demonstrated superior recall (59% vs 41%) and precision (25% vs 17%) compared to manual search results, and was positively received by physicians recognizing workflow benefits. Further studies must assess the potential clinical impact towards a future where electronic health records automatically anticipate clinical needs.

  View details for PubMedID 32477627

• Technology Use and Preferences for Mental Health Self-Management Interventions among Older Veterans. International journal of geriatric psychiatry Gould, C. E., Loup, J., Kuhn, E., Beaudreau, S. A., Ma, F., Goldstein, M. K., Wetherell, J. L., Zapata, A. M., Choe, P., O'Hara, R. 2019

  Abstract

  OBJECTIVES: The United States Department of Veterans Affairs offers numerous technology-delivered interventions to self-manage mental health problems. It is unknown, however, what barriers older military veterans face to using these technologies and how willing they would be to use technologies for mental health concerns.METHODS: Seventy-seven veterans (Mage =69.16years; SD=7.10) completed interviews in a concurrent mixed methods study. Interviewers asked about technology ownership and described four modalities of delivering self-management interventions: printed materials, DVDs, Internet, and mobile apps. Interviewers obtained feedback about each modality's benefits, barriers, and facilitators. Participants ranked their self-management modalities preferences alone and compared with counseling. Multi-variable adjusted logistic regression and qualitative analyses were conducted to investigate the reasons contributing to preferences.RESULTS: Most reported owning a computer (84.4%), having home Internet (80.5%), and a smartphone (70.1%). Participants preferred printed materials (35.1%) over mobile apps (28.6%), Internet (24.7%), and DVDs (13.0%). Lower computer proficiency was associated with preferring DVDs; higher proficiency was associated with Internet and mobile interventions. Residing in an urban area was associated with mobile apps. When counseling was an option, 66% identified this as their first preference. Qualitative findings showed veterans' desire for information, training, and provider support with technology.CONCLUSIONS: Older veterans reported high technology ownership rates, but varied preferences for self-management interventions. Notably, two-thirds preferred some form of technology, which points to the importance of ensuring that providers offer existing technology-delivered interventions to older veterans. Veterans' strong preference for counseling emphasizes the need for human support alongside self-management. This article is protected by copyright. All rights reserved.

  View details for DOI 10.1002/gps.5252

  View details for PubMedID 31854029

• Differences in Receiving Eye Care Based on Dementia Status Pershing, S., Goldstein, M. K., Henderson, V. W., Lu, Y., Bundorf, M., Rahman, M., Andrews, C., Stein, J. D. ASSOC RESEARCH VISION OPHTHALMOLOGY INC. 2019

  View details for Web of Science ID 000488800705001

• High-Need Patients' Goals and Goal Progress in a Veterans Affairs Intensive Outpatient Care Program. Journal of general internal medicine Hsu, K. Y., Slightam, C., Shaw, J. G., Tierney, A., Hummel, D. L., Goldstein, M. K., Chang, E. T., Boothroyd, D., Zulman, D. M. 2019

  Abstract

  BACKGROUND: Healthcare systems nationwide are implementing intensive outpatient care programs to optimize care for high-need patients; however, little is known about these patients' personal goals and factors associated with goal progress.OBJECTIVE: To describe high-need patients' goals, and to identify factors associated with their goal progress DESIGN: Retrospective cohort study PARTICIPANTS: A total of 113 high-need patients participated in a single-site Veterans Affairs intensive outpatient care program.MAIN MEASURES: Two independent reviewers examined patients' goals recorded in the electronic health record, categorized each goal into one of three domains (medical, behavioral, or social), and determined whether patients attained goal progress during program participation. Logistic regression was used to determine factors associated with goal progress.RESULTS: The majority (n=72, 64%) of the 113 patients attained goal progress. Among the 100 (88%) patients with at least one identified goal, 58 set goal(s) in the medical domain; 60 in the behavioral domain; and 52 in the social domain. Within each respective domain, 41 (71%) attained medical goal progress; 34 (57%) attained behavioral goal progress; and 32 (62%) attained social goal progress. Patients with mental health condition(s) (aOR 0.3; 95% CI 0.1-0.9; p=0.03) and those living alone (aOR 0.4; 95% CI 0.1-1.0; p=0.05) were less likely to attain goal progress. Those with mental health condition(s) and those who were living alone were least likely to attain goal progress (interaction aOR 0.1 compared to those with neither characteristic; 95% CI 0.0-0.7; p=0.02).CONCLUSIONS: Among high-need patients participating in an intensive outpatient care program, patient goals were fairly evenly distributed across medical, behavioral, and social domains. Notably, individuals living alone with mental health conditions were least likely to attain progress. Future care coordination interventions might incorporate strategies to address this gap, e.g., broader integration of behavioral and social service components.

  View details for DOI 10.1007/s11606-019-05010-w

  View details for PubMedID 31140094

• Differences in Cataract Surgery Rates Based on Dementia Status. Journal of Alzheimer's disease : JAD Pershing, S., Henderson, V. W., Bundorf, M. K., Lu, Y., Rahman, M., Andrews, C. A., Goldstein, M., Stein, J. D. 2019

  Abstract

  BACKGROUND: Cataract surgery substantially improves patient quality of life. Despite the rising prevalence of dementia in the US, little is known about use of cataract surgery among this group.OBJECTIVE: To evaluate the relationship between dementia status and cataract surgery.METHODS: Using administrative insurance claims for a representative sample of 1,125,387 US Medicare beneficiaries who received eye care between 2006 and 2015, we compared cataract surgery rates between patients with and without dementia via multivariable regression models to adjust for patient characteristics. Main outcome measures were annual rates of cataract surgery and hazard ratio and 95% confidence interval (CI) for receiving cataract surgery.RESULTS: Cataract surgery was performed in 457,128 patients, 23,331 with a prior diagnosis of dementia. 16.7% of dementia patients underwent cataract surgery, compared to 43.8% of patients without dementia. 59 cataract surgeries were performed per 1000 dementia patients annually, versus 105 surgeries per 1000 nondementia patients. After adjusting for patient characteristics, dementia patients were approximately half as likely to receive cataract surgery compared to nondementia patients (adjusted HR = 0.53, 95% CI 0.53-0.54). Among the subset of patients who received a first cataract surgery, those with dementia were also less likely to receive second-eye cataract surgery (adjusted HR = 0.87, 95% CI 0.86-0.88).CONCLUSION: US Medicare patients with dementia are less likely to undergo cataract surgery than those without dementia. This finding has implications for quality of care and dementia progression. More information is necessary to understand why rates of cataract surgery are lower for these patients, and to identify conditions where benefits of surgery may outweigh risks.

  View details for PubMedID 30958371

• Differences in Cataract Surgery Rates Based on Dementia Status JOURNAL OF ALZHEIMERS DISEASE Pershing, S., Henderson, V. W., Bundorf, M., Lu, Y., Rahman, M., Andrews, C. A., Goldstein, M., Stein, J. D. 2019; 69 (2): 423–32

  View details for DOI 10.3233/JAD-181292

  View details for Web of Science ID 000470879800009

• An evaluation of clinical order patterns machine-learned from clinician cohorts stratified by patient mortality outcomes JOURNAL OF BIOMEDICAL INFORMATICS Wang, J. K., Hom, J., Balasubramanian, S., Schuler, A., Shah, N. H., Goldstein, M. K., Baiocchi, M. M., Chen, J. H. 2018; 86: 109–19

  View details for DOI 10.1016/j.jbi.2018.09.005

  View details for Web of Science ID 000460600800011

• An Evaluation of Clinical Order Patterns Machine-Learned from Clinician Cohorts Stratified by Patient Mortality Outcomes. Journal of biomedical informatics Wang, J. K., Hom, J., Balasubramanian, S., Schuler, A., Shah, N. H., Goldstein, M. K., Baiocchi, M. T., Chen, J. H. 2018

  Abstract

  OBJECTIVE: Evaluate the quality of clinical order practice patterns machine-learned from clinician cohorts stratified by patient mortality outcomes.MATERIALS AND METHODS: Inpatient electronic health records from 2010-2013 were extracted from a tertiary academic hospital. Clinicians (n=1,822) were stratified into low-mortality (21.8%, n=397) and high-mortality (6.0%, n=110) extremes using a two-sided P-value score quantifying deviation of observed vs. expected 30-day patient mortality rates. Three patient cohorts were assembled: patients seen by low-mortality clinicians, high-mortality clinicians, and an unfiltered crowd of all clinicians (n=1,046, 1,046, and 5,230 post-propensity score matching, respectively). Predicted order lists were automatically generated from recommender system algorithms trained on each patient cohort and evaluated against i) real-world practice patterns reflected in patient cases with better-than-expected mortality outcomes and ii) reference standards derived from clinical practice guidelines.RESULTS: Across six common admission diagnoses, order lists learned from the crowd demonstrated the greatest alignment with guideline references (AUROC range=0.86-0.91), performing on par or better than those learned from low-mortality clinicians (0.79-0.84, P<10-5) or manually-authored hospital order sets (0.65-0.77, P<10-3). The same trend was observed in evaluating model predictions against better-than-expected patient cases, with the crowd model (AUROC mean=0.91) outperforming the low-mortality model (0.87, P<10-16) and order set benchmarks (0.78, P<10-35).DISCUSSION: Whether machine-learning models are trained on all clinicians or a subset of experts illustrates a bias-variance tradeoff in data usage. Defining robust metrics to assess quality based on internal (e.g. practice patterns from better-than-expected patient cases) or external reference standards (e.g. clinical practice guidelines) is critical to assess decision support content.CONCLUSION: Learning relevant decision support content from all clinicians is as, if not more, robust than learning from a select subgroup of clinicians favored by patient outcomes.

  View details for PubMedID 30195660

• Testing the usability of a portable DVD player and tailored photo instructions with older adult veterans EDUCATIONAL GERONTOLOGY Gould, C. E., Zapata, A. L., Shinsky, D. N., Goldstein, M. K. 2018; 44 (1): 64–73

  View details for DOI 10.1080/03601277.2017.1395986

  View details for Web of Science ID 000428733000006

• Selecting Test Cases from the Electronic Health Record for Software Testing of Knowledge-Based Clinical Decision Support Systems. AMIA ... Annual Symposium proceedings. AMIA Symposium Usman, O. A., Oshiro, C., Chambers, J. G., Tu, S. W., Martins, S., Robinson, A., Goldstein, M. K. 2018; 2018: 1046–55

  Abstract

  Software testing of knowledge-based clinical decision support systems is challenging, labor intensive, and expensive; yet, testing is necessary since clinical applications have heightened consequences. Thoughtful test case selection improves testing coverage while minimizing testing burden. ATHENA-CDS is a knowledge-based system that provides guideline-based recommendations for chronic medical conditions. Using the ATHENA-CDS diabetes knowledgebase, we demonstrate a generalizable approach for selecting test cases using rules/ filters to create a set of paths that mimics the system's logic. Test cases are allocated to paths using a proportion heuristic. Using data from the electronic health record, we found 1,086 cases with glycemic control above target goals. We created a total of 48 filters and 50 unique system paths, which were used to allocate 200 test cases. We show that our method generates a comprehensive set of test cases that provides adequate coverage for the testing of a knowledge-based CDS.

  View details for PubMedID 31019657

• "It's a disease of families": Neurologists' insights on how to improve communication and quality of life for families of Parkinson's disease patients. Chronic illness Schwartz, R. n., Zulman, D. n., Gray, C. n., Goldstein, M. K., Trivedi, R. n. 2018: 1742395318799852

  Abstract

  Objectives Parkinson's disease presents an evolving challenge for patients and families due to an unpredictable disease trajectory and symptoms that complicate social interactions. In this study, we explore neurologists' perspectives on the challenges Parkinson's disease presents for families and the strategies they use to improve communication and quality of life. Methods We conducted hour-long semi-structured interviews with 16 neurologists at 4 care delivery institutions in the San Francisco Bay Area, focusing on techniques neurologists use to support families through the Parkinson's disease journey. Results Neurologists identified strategies for addressing caregiver-patient disagreements around symptom accuracy and negotiating driving safety. Family education is needed to contextualize patient symptoms and to identify psychosocial support resources. Unmet caregiver needs remain, particularly in the form of psychosocial support, respite care and support for unequal gender dynamics in the Parkinson's disease caregiving experience. Discussion Family members of Parkinson's disease patients face unique caregiving and interpersonal challenges due to the nature of the disease. Targeted education and structural support are needed to alleviate current burdens and allow for improved patient- and family-centered care.

  View details for PubMedID 30208725

• Automating Quality Measures for Heart Failure Using Natural Language Processing: A Descriptive Study in the Department of Veterans Affairs JMIR MEDICAL INFORMATICS Garvin, J., Kim, Y., Gobbel, G., Matheny, M. E., Redd, A., Bray, B. E., Heidenreich, P., Bolton, D., Heavirland, J., Kelly, N., Reeves, R., Kalsy, M., Goldstein, M., Meystre, S. M. 2018; 6 (1): e5

  Abstract

  We developed an accurate, stakeholder-informed, automated, natural language processing (NLP) system to measure the quality of heart failure (HF) inpatient care, and explored the potential for adoption of this system within an integrated health care system.To accurately automate a United States Department of Veterans Affairs (VA) quality measure for inpatients with HF.We automated the HF quality measure Congestive Heart Failure Inpatient Measure 19 (CHI19) that identifies whether a given patient has left ventricular ejection fraction (LVEF) <40%, and if so, whether an angiotensin-converting enzyme inhibitor or angiotensin-receptor blocker was prescribed at discharge if there were no contraindications. We used documents from 1083 unique inpatients from eight VA medical centers to develop a reference standard (RS) to train (n=314) and test (n=769) the Congestive Heart Failure Information Extraction Framework (CHIEF). We also conducted semi-structured interviews (n=15) for stakeholder feedback on implementation of the CHIEF.The CHIEF classified each hospitalization in the test set with a sensitivity (SN) of 98.9% and positive predictive value of 98.7%, compared with an RS and SN of 98.5% for available External Peer Review Program assessments. Of the 1083 patients available for the NLP system, the CHIEF evaluated and classified 100% of cases. Stakeholders identified potential implementation facilitators and clinical uses of the CHIEF.The CHIEF provided complete data for all patients in the cohort and could potentially improve the efficiency, timeliness, and utility of HF quality measurements.

  View details for PubMedID 29335238

• Effect of the Tool to Reduce Inappropriate Medications on Medication Communication and Deprescribing JOURNAL OF THE AMERICAN GERIATRICS SOCIETY Fried, T. R., Niehoff, K. M., Street, R. L., Charpentier, P. A., Rajeevan, N., Miller, P. L., Goldstein, M. K., O'Leary, J. R., Fenton, B. T. 2017; 65 (10): 2265–71

  Abstract

  To examine the effect of the Tool to Reduce Inappropriate Medications (TRIM), a web tool linking an electronic health record (EHR) to a clinical decision support system, on medication communication and prescribing.Randomized clinical trial.Primary care clinics at a Veterans Affairs Medical Center.Veterans aged 65 and older prescribed seven or more medications randomized to receipt of TRIM or usual care (N = 128).TRIM extracts information on medications and chronic conditions from the EHR and contains data entry screens for information obtained from brief chart review and telephonic patient assessment. These data serve as input for automated algorithms identifying medication reconciliation discrepancies, potentially inappropriate medications (PIMs), and potentially inappropriate regimens. Clinician feedback reports summarize discrepancies and provide recommendations for deprescribing. Patient feedback reports summarize discrepancies and self-reported medication problems.Primary: subscales of the Patient Assessment of Care for Chronic Conditions (PACIC) related to shared decision-making; clinician and patient communication. Secondary: changes in medications.29.7% of TRIM participants and 15.6% of control participants provided the highest PACIC ratings; this difference was not significant. Adjusting for covariates and clustering of patients within clinicians, TRIM was associated with significantly more-active patient communication and facilitative clinician communication and with more medication-related communication among patients and clinicians. TRIM was significantly associated with correction of medication discrepancies but had no effect on number of medications or reduction in PIMs.TRIM improved communication about medications and accuracy of documentation. Although there was no association with prescribing, the small sample size provided limited power to examine medication-related outcomes.

  View details for PubMedID 28804870

  View details for PubMedCentralID PMC5641237

• Development of a video-delivered relaxation treatment of late-life anxiety for veterans. International psychogeriatrics Gould, C. E., Zapata, A. M., Bruce, J., Bereknyei Merrell, S., Wetherell, J. L., O'Hara, R., Kuhn, E., Goldstein, M. K., Beaudreau, S. A. 2017: 1-13

  Abstract

  Behavioral treatments reduce anxiety, yet many older adults may not have access to these efficacious treatments. To address this need, we developed and evaluated the feasibility and acceptability of a video-delivered anxiety treatment for older Veterans. This treatment program, BREATHE (Breathing, Relaxation, and Education for Anxiety Treatment in the Home Environment), combines psychoeducation, diaphragmatic breathing, and progressive muscle relaxation training with engagement in activities.A mixed methods concurrent study design was used to examine the clarity of the treatment videos. We conducted semi-structured interviews with 20 Veterans (M age = 69.5, SD = 7.3 years; 55% White, Non-Hispanic) and collected ratings of video clarity.Quantitative ratings revealed that 100% of participants generally or definitely could follow breathing and relaxation video instructions. Qualitative findings, however, demonstrated more variability in the extent to which each video segment was clear. Participants identified both immediate benefits and motivation challenges associated with a video-delivered treatment. Participants suggested that some patients may need encouragement, whereas others need face-to-face therapy.Quantitative ratings of video clarity and qualitative findings highlight the feasibility of a video-delivered treatment for older Veterans with anxiety. Our findings demonstrate the importance of ensuring patients can follow instructions provided in self-directed treatments and the role that an iterative testing process has in addressing these issues. Next steps include testing the treatment videos with older Veterans with anxiety disorders.

  View details for DOI 10.1017/S1041610217000928

  View details for PubMedID 28592349

• Decaying relevance of clinical data towards future decisions in data-driven inpatient clinical order sets. International journal of medical informatics Chen, J. H., Alagappan, M., Goldstein, M. K., Asch, S. M., Altman, R. B. 2017; 102: 71-79

  Abstract

  Determine how varying longitudinal historical training data can impact prediction of future clinical decisions. Estimate the "decay rate" of clinical data source relevance.We trained a clinical order recommender system, analogous to Netflix or Amazon's "Customers who bought A also bought B..." product recommenders, based on a tertiary academic hospital's structured electronic health record data. We used this system to predict future (2013) admission orders based on different subsets of historical training data (2009 through 2012), relative to existing human-authored order sets.Predicting future (2013) inpatient orders is more accurate with models trained on just one month of recent (2012) data than with 12 months of older (2009) data (ROC AUC 0.91 vs. 0.88, precision 27% vs. 22%, recall 52% vs. 43%, all P<10(-10)). Algorithmically learned models from even the older (2009) data was still more effective than existing human-authored order sets (ROC AUC 0.81, precision 16% recall 35%). Training with more longitudinal data (2009-2012) was no better than using only the most recent (2012) data, unless applying a decaying weighting scheme with a "half-life" of data relevance about 4 months.Clinical practice patterns (automatically) learned from electronic health record data can vary substantially across years. Gold standards for clinical decision support are elusive moving targets, reinforcing the need for automated methods that can adapt to evolving information.Prioritizing small amounts of recent data is more effective than using larger amounts of older data towards future clinical predictions.

  View details for DOI 10.1016/j.ijmedinf.2017.03.006

  View details for PubMedID 28495350

• Extraction of left ventricular ejection fraction information from various types of clinical reports JOURNAL OF BIOMEDICAL INFORMATICS Kim, Y., Garvin, J. H., Goldstein, M. K., Hwang, T. S., Redd, A., Bolton, D., Heidenreich, P. A., Meystre, S. M. 2017; 67: 42–48

  Abstract

  Efforts to improve the treatment of congestive heart failure, a common and serious medical condition, include the use of quality measures to assess guideline-concordant care. The goal of this study is to identify left ventricular ejection fraction (LVEF) information from various types of clinical notes, and to then use this information for heart failure quality measurement. We analyzed the annotation differences between a new corpus of clinical notes from the Echocardiography, Radiology, and Text Integrated Utility package and other corpora annotated for natural language processing (NLP) research in the Department of Veterans Affairs. These reports contain varying degrees of structure. To examine whether existing LVEF extraction modules we developed in prior research improve the accuracy of LVEF information extraction from the new corpus, we created two sequence-tagging NLP modules trained with a new data set, with or without predictions from the existing LVEF extraction modules. We also conducted a set of experiments to examine the impact of training data size on information extraction accuracy. We found that less training data is needed when reports are highly structured, and that combining predictions from existing LVEF extraction modules improves information extraction when reports have less structured formats and a rich set of vocabulary.

  View details for PubMedID 28163196

  View details for PubMedCentralID PMC5575914

• What is a good medical decision? A research agenda guided by perspectives from multiple stakeholders. Journal of behavioral medicine Hamilton, J. G., Lillie, S. E., Alden, D. L., Scherer, L., Oser, M., Rini, C., Tanaka, M., Baleix, J., Brewster, M., Craddock Lee, S., Goldstein, M. K., Jacobson, R. M., Myers, R. E., Zikmund-Fisher, B. J., Waters, E. A. 2017; 40 (1): 52-68

  Abstract

  Informed and shared decision making are critical aspects of patient-centered care, which has contributed to an emphasis on decision support interventions to promote good medical decision making. However, researchers and healthcare providers have not reached a consensus on what defines a good decision, nor how to evaluate it. This position paper, informed by conference sessions featuring diverse stakeholders held at the 2015 Society of Behavioral Medicine and Society for Medical Decision Making annual meetings, describes key concepts that influence the decision making process itself and that may change what it means to make a good decision: interpersonal factors, structural constraints, affective influences, and values clarification methods. This paper also proposes specific research questions within each of these priority areas, with the goal of moving medical decision making research to a more comprehensive definition of a good medical decision, and enhancing the ability to measure and improve the decision making process.

  View details for DOI 10.1007/s10865-016-9785-z

  View details for PubMedID 27566316

• What is a good medical decision? A research agenda guided by perspectives from multiple stakeholders JOURNAL OF BEHAVIORAL MEDICINE Hamilton, J. G., Lillie, S. E., Alden, D. L., Scherer, L., Oser, M., Rini, C., Tanaka, M., Baleix, J., Brewster, M., Lee, S. C., Goldstein, M. K., Jacobson, R. M., Myers, R. E., Zikmund-Fisher, B. J., Waters, E. A. 2017; 40 (1): 52-68

  Abstract

  Informed and shared decision making are critical aspects of patient-centered care, which has contributed to an emphasis on decision support interventions to promote good medical decision making. However, researchers and healthcare providers have not reached a consensus on what defines a good decision, nor how to evaluate it. This position paper, informed by conference sessions featuring diverse stakeholders held at the 2015 Society of Behavioral Medicine and Society for Medical Decision Making annual meetings, describes key concepts that influence the decision making process itself and that may change what it means to make a good decision: interpersonal factors, structural constraints, affective influences, and values clarification methods. This paper also proposes specific research questions within each of these priority areas, with the goal of moving medical decision making research to a more comprehensive definition of a good medical decision, and enhancing the ability to measure and improve the decision making process.

  View details for DOI 10.1007/s10865-016-9785-z

  View details for Web of Science ID 000394342300005

  View details for PubMedCentralID PMC5296255

• High-Risk Drug-Drug Interactions Between Clinical Practice Guidelines for Management of Chronic Conditions. AMIA Joint Summits on Translational Science proceedings. AMIA Joint Summits on Translational Science Tso, G. J., Tu, S. W., Musen, M. A., Goldstein, M. K. 2017; 2017: 531–39

  Abstract

  Clinicians and clinical decision-support systems often follow pharmacotherapy recommendations for patients based on clinical practice guidelines (CPGs). In multimorbid patients, these recommendations can potentially have clinically significant drug-drug interactions (DDIs). In this study, we describe and validate a method for programmatically detecting DDIs among CPG recommendations. The system extracts pharmacotherapy intervention recommendations from narrative CPGs, normalizes the terms, creates a mapping of drugs and drug classes, and then identifies occurrences of DDIs between CPG pairs. We used this system to analyze 75 CPGs written by authoring entities in the United States that discuss outpatient management of common chronic diseases. Using a reference list of high-risk DDIs, we identified 2198 of these DDIs in 638 CPG pairs (46 unique CPGs). Only 9 high-risk DDIs were discussed by both CPGs in a pairing. In 69 of the pairings, neither CPG had a pharmacologic reference or a warning of the possibility of a DDI.

  View details for PubMedID 28815153

• The New HIT: Human Health Information Technology Leung, T. I., Goldstein, M. K., Musen, M. A., Cronkite, R., Chen, J. H., Gottlieb, A., Leitersdorf, E., Gundlapalli, A. V., Jaulent, M. C., Zhao, D. IOS PRESS. 2017: 768–72

  Abstract

  Humanism in medicine is defined as health care providers' attitudes and actions that demonstrate respect for patients' values and concerns in relation to their social, psychological and spiritual life domains. Specifically, humanistic clinical medicine involves showing respect for the patient, building a personal connection, and eliciting and addressing a patient's emotional response to illness. Health information technology (IT) often interferes with humanistic clinical practice, potentially disabling these core aspects of the therapeutic patient-physician relationship. Health IT has evolved rapidly in recent years - and the imperative to maintain humanism in practice has never been greater. In this vision paper, we aim to discuss why preserving humanism is imperative in the design and implementation of health IT systems.

  View details for PubMedID 29295202

• A MIXED-METHODS EVALUATION OF A VIDEO-DELIVERED RELAXATION TREATMENT OF ANXIETY IN OLDER VETERANS Gould, C. E., O'Hara, R., Zapata, A., Bruce, J., Merrell, S., Wetherell, J., Goldstein, M., Beaudreau, S. A. OXFORD UNIV PRESS INC. 2016: 448

  View details for Web of Science ID 000388585002119

• TESTING THE USABILITY OF A PORTABLE DVD PLAYER AND PHOTO INSTRUCTIONS GUIDE WITH OLDER VETERANS Gould, C. E., Zapata, A., Merrell, S., Bruce, J., Beaudreau, S. A., Goldstein, M., Wetherell, J., O'Hara, R. OXFORD UNIV PRESS INC. 2016: 568

  View details for Web of Science ID 000388585002588

• Multimorbidity is associated with anxiety in older adults in the Health and Retirement Study. International journal of geriatric psychiatry Gould, C. E., O'Hara, R., Goldstein, M. K., Beaudreau, S. A. 2016; 31 (10): 1105-1115

  Abstract

  The present study determined whether the number of medical conditions was associated with increased occurrence of anxiety and whether triads of medical conditions were associated with anxiety in a nationally representative sample of older Americans. We determined whether multimorbidity findings were unique to anxiety as compared with depressive symptoms.A sample of 4219 participants (65 years or older) completed anxiety and depression measures in the Health and Retirement Study 2006 wave. The logistic regression models' outcome was elevated anxiety (≥12 on five-item Beck Anxiety Inventory) or depressive symptoms (≥12 on eight-item Center for Epidemiological Studies Depression Scale). The predictor variable was a tally of seven self-report of doctor-diagnosed conditions: arthritis, cancer, diabetes, heart conditions, high blood pressure, lung disease, and stroke. Analyses were adjusted for age, gender, and depressive or anxiety symptoms. Associations among elevated anxiety or depressive symptoms and 35 triads of medical conditions were examined using Bonferroni corrected chi-square analyses.Three or more medical conditions conferred a 2.30-fold increase in elevated anxiety (95% confidence interval: 1.44-4.01). Twenty triads were associated with elevated anxiety as compared with 13 associated with depressive symptoms. Six of seven medical conditions, with the exception being stroke, were present in the majority of triads.Number of medical conditions and specific conditions are associated with increased occurrence of elevated anxiety. Compared with elevated depressive symptoms, anxiety is associated with greater multimorbidity. As anxiety and depression cause significant morbidity, it may be beneficial to consider these mental health symptoms when evaluating older adults with multimorbidity. Published 2016. This article is a U.S. Government work and is in the public domain in the USA.

  View details for DOI 10.1002/gps.4532

  View details for PubMedID 27441851

• Predicting inpatient clinical order patterns with probabilistic topic models vs conventional order sets. Journal of the American Medical Informatics Association Chen, J. H., Goldstein, M. K., Asch, S. M., Mackey, L., Altman, R. B. 2016

  Abstract

  Build probabilistic topic model representations of hospital admissions processes and compare the ability of such models to predict clinical order patterns as compared to preconstructed order sets.The authors evaluated the first 24 hours of structured electronic health record data for > 10 K inpatients. Drawing an analogy between structured items (e.g., clinical orders) to words in a text document, the authors performed latent Dirichlet allocation probabilistic topic modeling. These topic models use initial clinical information to predict clinical orders for a separate validation set of > 4 K patients. The authors evaluated these topic model-based predictions vs existing human-authored order sets by area under the receiver operating characteristic curve, precision, and recall for subsequent clinical orders.Existing order sets predict clinical orders used within 24 hours with area under the receiver operating characteristic curve 0.81, precision 16%, and recall 35%. This can be improved to 0.90, 24%, and 47% ( P  < 10 -20 ) by using probabilistic topic models to summarize clinical data into up to 32 topics. Many of these latent topics yield natural clinical interpretations (e.g., "critical care," "pneumonia," "neurologic evaluation").Existing order sets tend to provide nonspecific, process-oriented aid, with usability limitations impairing more precise, patient-focused support. Algorithmic summarization has the potential to breach this usability barrier by automatically inferring patient context, but with potential tradeoffs in interpretability.Probabilistic topic modeling provides an automated approach to detect thematic trends in patient care and generate decision support content. A potential use case finds related clinical orders for decision support.

  View details for DOI 10.1093/jamia/ocw136

  View details for PubMedID 27655861

• Automated Heart Failure Quality Measurement with Natural Language Processing Garvin, J. H., Kim, Y., Gobbel, G. T., Matheny, M. E., Redd, A., Bray, B. E., Heidenreich, P., Kalsy, M., Goldstein, M. K., Bolton, D., Nativi-Nicolau, J., Meystre, S. M. CHURCHILL LIVINGSTONE INC MEDICAL PUBLISHERS. 2016: S92

  View details for DOI 10.1016/j.cardfail.2016.06.292

  View details for Web of Science ID 000381064700257

• A Delphi process to address medication appropriateness for older persons with multiple chronic conditions (vol 16, 67, 2016) BMC GERIATRICS Fried, T. R., Niehoff, K., Tjia, J., Redeker, N., Goldstein, M. K. 2016; 16: 123

  View details for PubMedID 27312251

• Development of the Tool to Reduce Inappropriate Medications (TRIM): A Clinical Decision Support System to Improve Medication Prescribing for Older Adults. Pharmacotherapy Niehoff, K. M., Rajeevan, N., Charpentier, P. A., Miller, P. L., Goldstein, M. K., Fried, T. R. 2016; 36 (6): 694-701

  Abstract

  To create a clinical decision support system (CDSS) for evaluating problems with medications among older outpatients based on a broad set of criteria.Web-based CDSS development.Primary care clinics at a Veterans Affairs medical center.Forty veterans 65 years and older who were prescribed seven or more medications that included those for treatment of diabetes mellitus and hypertension.The Tool to Reduce Inappropriate Medications (TRIM) uses a program to extract age, medications, and chronic conditions from the electronic health record to identify high-risk patients and as input for evaluating the medication regimen. Additional health variables obtained through chart review and direct patient assessment are entered into a Web-based program. Based on a series of algorithms, TRIM generates feedback reports for clinicians. TRIM identified medication reconciliation discrepancies in 98% (39/40) of veterans, potentially inappropriate medications in 58% (23/40), potential problems with feasibility (based on poor adherence and/or cognitive impairment) in 25% (10/40), potential overtreatment of hypertension in 50% (20/40), potential overtreatment of diabetes in 43% (17/40), inappropriate dosing of renally excreted medications in 5% (2/40), and patient-reported adverse reactions in 5% (2/40).This evaluation of TRIM demonstrated that data elements can be extracted from the electronic health record to identify older primary care patients at risk for potentially problematic medication regimens. Supplemented with chart review and direct patient assessment, these data can be processed through clinical algorithms that identify potential problems and generate patient-specific feedback reports. Additional work is necessary to assess the effects of TRIM on medication deprescribing.

  View details for DOI 10.1002/phar.1751

  View details for PubMedID 27041466

  View details for PubMedCentralID PMC4919149

• Longitudinal Cognitive Trajectories of Women Veterans from the Women's Health Initiative Memory Study. Gerontologist Padula, C. B., Weitlauf, J. C., Rosen, A. C., Reiber, G., Cochrane, B. B., Naughton, M. J., Li, W., Rissling, M., Yaffe, K., Hunt, J. R., Stefanick, M. L., Goldstein, M. K., Espeland, M. A. 2016; 56 (1): 115-125

  Abstract

  A comparison of longitudinal global cognitive functioning in women Veteran and non-Veteran participants in the Women's Health Initiative (WHI).We studied 7,330 women aged 65-79 at baseline who participated in the WHI Hormone Therapy Trial and its ancillary Memory Study (WHIMS). Global cognitive functioning (Modified Mini-Mental State Examination [3MSE]) in Veterans (n = 279) and non-Veterans (n = 7,051) was compared at baseline and annually for 8 years using generalized linear modeling methods.Compared with non-Veterans, Veteran women were older, more likely to be Caucasian, unmarried, and had higher rates of educational and occupational attainment. Results of unadjusted baseline analyses suggest 3MSE scores were similar between groups. Longitudinal analyses, adjusted for age, education, ethnicity, and WHI trial assignment revealed differences in the rate of cognitive decline between groups over time, such that scores decreased more in Veterans relative to non-Veterans. This relative difference was more pronounced among Veterans who were older, had higher educational/occupational attainment and greater baseline prevalence of cardiovascular risk factors (e.g., smoking) and cardiovascular disease (e.g., angina, stroke).Veteran status was associated with higher prevalence of protective factors that may have helped initially preserve cognitive functioning. However, findings ultimately revealed more pronounced cognitive decline among Veteran relative to non-Veteran participants, likely suggesting the presence of risks that may impact neuropathology and the effects of which were initially masked by Veterans' greater cognitive reserve.

  View details for DOI 10.1093/geront/gnv663

  View details for PubMedID 26615021

• A Delphi process to address medication appropriateness for older persons with multiple chronic conditions. BMC geriatrics Fried, T. R., Niehoff, K., Tjia, J., Redeker, N., Goldstein, M. K. 2016; 16 (1): 67-?

  Abstract

  Frameworks exist to evaluate the appropriateness of medication regimens for older patients with multiple medical conditions (MCCs). Less is known about how to translate the concepts of the frameworks into specific strategies to identify and remediate inappropriate regimens.Modified Delphi method involving iterative rounds of input from panel members. Panelists (n = 9) represented the disciplines of nursing, medicine and pharmacy. Included among the physicians were two geriatricians, one general internist, one family practitioner, one cardiologist and two nephrologists. They participated in 3 rounds of web-based anonymous surveys.The panel reached consensus on a set of markers to identify problems with medication regimens, including patient/caregiver report of non-adherence, medication complexity, cognitive impairment, medications identified by expert opinion as inappropriate for older persons, excessively tight blood sugar and blood pressure control among persons with diabetes mellitus, patient/caregiver report of adverse medication effects or medications not achieving desired outcomes, and total number of medications. The panel also reached consensus on approaches to address these problems, including endorsement of strategies to discontinue medications with known benefit if necessary because of problems with feasibility or lack of alignment with patient goals.The results of the Delphi process provide the basis for an algorithm to improve medication regimens among older persons with MCCs. The algorithm will require assessment not only of medications and diagnoses but also cognition and social support, and it will support discontinuation of medications both when risks outweigh benefits and when regimens are not feasible or do not align with goals.

  View details for DOI 10.1186/s12877-016-0240-3

  View details for PubMedID 26979576

  View details for PubMedCentralID PMC4791884

• DYNAMICALLY EVOLVING CLINICAL PRACTICES AND IMPLICATIONS FOR PREDICTING MEDICAL DECISIONS Chen, J. H., Goldstein, M. K., Asch, S. M., Altman, R. B., Altman, R. B., Dunker, A. K., Hunter, L., Ritchie, M. D., Murray, T., Klein, T. E. WORLD SCIENTIFIC PUBL CO PTE LTD. 2016: 195–206

  View details for Web of Science ID 000386326200019

• Automating Guidelines for Clinical Decision Support: Knowledge Engineering and Implementation. AMIA ... Annual Symposium proceedings. AMIA Symposium Tso, G. J., Tu, S. W., Oshiro, C., Martins, S., Ashcraft, M., Yuen, K. W., Wang, D., Robinson, A., Heidenreich, P. A., Goldstein, M. K. 2016; 2016: 1189-1198

  Abstract

  As utilization of clinical decision support (CDS) increases, it is important to continue the development and refinement of methods to accurately translate the intention of clinical practice guidelines (CPG) into a computable form. In this study, we validate and extend the 13 steps that Shiffman et al.(5) identified for translating CPG knowledge for use in CDS. During an implementation project of ATHENA-CDS, we encoded complex CPG recommendations for five common chronic conditions for integration into an existing clinical dashboard. Major decisions made during the implementation process were recorded and categorized according to the 13 steps. During the implementation period, we categorized 119 decisions and identified 8 new categories required to complete the project. We provide details on an updated model that outlines all of the steps used to translate CPG knowledge into a CDS integrated with existing health information technology.

  View details for PubMedID 28269916

• Automating Performance Measures and Clinical Practice Guidelines: Differences and Complementarities. AMIA ... Annual Symposium proceedings. AMIA Symposium Tu, S. W., Martins, S., Oshiro, C., Yuen, K., Wang, D., Robinson, A., Ashcraft, M., Heidenreich, P. A., Goldstein, M. K. 2016; 2016: 1199-1208

  Abstract

  Through close analysis of two pairs of systems that implement the automated evaluation of performance measures (PMs) and guideline-based clinical decision support (CDS), we contrast differences in their knowledge encoding and necessary changes to a CDS system that provides management recommendations for patients failing performance measures. We trace the sources of differences to the implementation environments and goals of PMs and CDS.

  View details for PubMedID 28269917

• Test Case Selection in Pre-Deployment Testing of Complex Clinical Decision Support Systems. AMIA Joint Summits on Translational Science proceedings. AMIA Joint Summits on Translational Science Tso, G. J., Yuen, K., Martins, S., Tu, S. W., Ashcraft, M., Heidenreich, P., Hoffman, B. B., Goldstein, M. K. 2016; 2016: 240-249

  Abstract

  Clinical decision support (CDS) systems with complex logic are being developed. Ensuring the quality of CDS is imperative, but there is no consensus on testing standards. We tested ATHENA-HTN CDS after encoding updated hypertension guidelines into the system. A logic flow and a complexity analysis of the encoding were performed to guide testing. 100 test cases were selected to test the major pathways in the CDS logic flow, and the effectiveness of the testing was analyzed. The encoding contained 26 decision points and 3120 possible output combinations. The 100 cases selected tested all of the major pathways in the logic, but only 1% of the possible output combinations. Test case selection is one of the most challenging aspects in CDS testing and has a major impact on testing coverage. A test selection strategy should take into account the complexity of the system, identification of major logic pathways, and available resources.

  View details for PubMedID 27570678

• DYNAMICALLY EVOLVING CLINICAL PRACTICES AND IMPLICATIONS FOR PREDICTING MEDICAL DECISIONS. Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing Chen, J. H., Goldstein, M. K., Asch, S. M., Altman, R. B. 2016; 21: 195-206

  Abstract

  Automatically data-mining clinical practice patterns from electronic health records (EHR) can enable prediction of future practices as a form of clinical decision support (CDS). Our objective is to determine the stability of learned clinical practice patterns over time and what implication this has when using varying longitudinal historical data sources towards predicting future decisions. We trained an association rule engine for clinical orders (e.g., labs, imaging, medications) using structured inpatient data from a tertiary academic hospital. Comparing top order associations per admission diagnosis from training data in 2009 vs. 2012, we find practice variability from unstable diagnoses with rank biased overlap (RBO)<0.35 (e.g., pneumonia) to stable admissions for planned procedures (e.g., chemotherapy, surgery) with comparatively high RBO>0.6. Predicting admission orders for future (2013) patients with associations trained on recent (2012) vs. older (2009) data improved accuracy evaluated by area under the receiver operating characteristic curve (ROC-AUC) 0.89 to 0.92, precision at ten (positive predictive value of the top ten predictions against actual orders) 30% to 37%, and weighted recall (sensitivity) at ten 2.4% to 13%, (P<10(-10)). Training with more longitudinal data (2009-2012) was no better than only using recent (2012) data. Secular trends in practice patterns likely explain why smaller but more recent training data is more accurate at predicting future practices.

  View details for PubMedID 26776186

• Prospective Analysis of Health and Mortality Risk in Veteran and Non-Veteran Participants in the Women's Health Initiative WOMENS HEALTH ISSUES Weitlauf, J. C., LaCroix, A. Z., Bird, C. E., Woods, N. F., Washington, D. L., Katon, J. G., LaMonte, M. J., Goldstein, M. K., Bassuk, S. S., Sarto, G. E., Stefanick, M. L. 2015; 25 (6): 649-657

  Abstract

  The health of postmenopausal women veterans is a neglected area of study. A stronger empirical evidence base is needed, and would inform the provision of health care for the nearly 1 million U.S. women veterans currently 50 years of age or older. To this end, the present work compares salient health outcomes and risk of all-cause mortality among veteran and non-veteran participants of the Women's Health Initiative (WHI).This study features prospective analysis of long-term health outcomes and mortality risk (average follow-up, 8 years) among the 3,706 women veterans and 141,009 non-veterans who participated in the WHI Observational Study or Clinical Trials. Outcome measurements included confirmed incident cases of cardiovascular disease (CVD), cancer, diabetes, hip fractures, and all-cause mortality.We identified 17,968 cases of CVD, 19,152 cases of cancer, 18,718 cases of diabetes, 2,817 cases of hip fracture, and 13,747 deaths. In Cox regression models adjusted for age, sociodemographic variables, and health risk factors, veteran status was associated with significantly increased risk of all-cause mortality (hazard ratio [HR], 1.13; 95% CI, 1.03-1.23), but not with risk of CVD (HR, 1.00; 95% CI, 0.90-1.11), cancer (HR, 1.04; 95% CI, 0.95-1.14), hip fracture (HR, 1.16; 95% CI, 0.94-1.43), or diabetes (HR, 1.00; 95% CI, 0.89-1.1).Women veterans' postmenopausal health, particularly risk for all-cause mortality, warrants further consideration. In particular, efforts to identify and address modifiable risk factors associated with all-cause mortality are needed.

  View details for DOI 10.1016/j.whi.2015.08.006

  View details for Web of Science ID 000368247700010

  View details for PubMedCentralID PMC4641800

• Defibrillator implantations for primary prevention in the United States: Inappropriate care or inadequate documentation: Insights from the National Cardiovascular Data ICD Registry. Heart rhythm Kaiser, D. W., Tsai, V., Heidenreich, P. A., Goldstein, M. K., Wang, Y., Curtis, J., Turakhia, M. P. 2015; 12 (10): 2086-2093

  Abstract

  Prior studies have reported that more than 20% of implantable cardioverter-defibrillator (ICD) implantations in the United States do not adhere to trial-based criteria.We sought to investigate the patient characteristics associated with not meeting the inclusion criteria of the clinical trials that have demonstrated the efficacy of primary prevention ICDs.Using data from the National Cardiovascular Data Registry's ICD Registry, we identified patients who received ICDs for primary prevention from January 2006 to December 2008. We determined whether patients met the inclusion criteria of at least 1 of the 4 ICD primary prevention trials: Multicenter Automatic Defibrillator Implantation Trial (MADIT), MADIT-II, Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT), and the Multicenter Unsustained Tachycardia Trial (MUSTT).Among 150,264 patients, 86% met criteria for an ICD implantation based on trial data. The proportion of patients who did not meet trial-based criteria increased as age decreased. In multivariate analysis, the significant predictors for not meeting trial criteria included prior cardiac transplantation (odds ratio [OR] 2.1), pediatric electrophysiology operator (OR 2.0), and high-grade atrioventricular conduction disease (OR 1.4).Among National Cardiovascular Data Registry registrants receiving first-time ICDs for primary prevention, the majority met trial-based criteria. Multivariate analyses suggested that many patients who did not meet the trial-based criteria may have had clinical circumstances that warranted ICD implantation. These findings caution against the use of trial-based indications to determine site quality metrics that could penalize sites that care for younger patients. The planned incorporation of appropriate use criteria into the ICD registry may better characterize patient- and site-level quality and performance.

  View details for DOI 10.1016/j.hrthm.2015.05.010

  View details for PubMedID 25982720

• A validated risk model for 1-year mortality after primary prevention implantable cardioverter defibrillator placement. American heart journal Heidenreich, P. A., Tsai, V., Curtis, J., Wang, Y., Turakhia, M. P., Masoudi, F. A., Varosy, P. D., Goldstein, M. K. 2015; 170 (2): 281-289 e2

  Abstract

  We sought to determine survival for patients with heart failure after an implantation of an implantable cardioverter defibrillator (ICD) for primary prevention in the United States and to develop a simple model that would predict mortality risk.Clinical trials have found that patients with heart failure with a 1-year mortality risk near 20% may not benefit from an ICD.We identified patients from the ICD Registry of the National Cardiovascular Disease Registries who underwent ICD implantation for primary prevention from 2007 to 2009. Two risk scores for mortality were developed in 2 cohorts: one limited to those with a B-type natriuretic peptide (BNP) value and a second for all patients. The scores were obtained from derivation datasets and tested in a validation sets using logistic regression models and classification and regression trees.In a primary prevention population with BNP available (18,725) the 6 variables most predictive of 1-year mortality were age ≥75, BNP ≥700 pg/mL, chronic lung disease, dialysis, blood urea nitrogen ≥30 mg/dL, and systolic blood pressure <120 mmHg. Patients with zero risk factors had a 3.3% one-year mortality compared to a 66.7% one-year mortality for those with all 6 risk factors. Those with ≥3 risk factors (24.0% of the population) had a 25.8% one-year mortality. A second score using a larger cohort that did not consider BNP identified similar risk factors.A simple validated risk score can identify patients at high and low risk for death within a year after ICD placement. A large fraction of those currently implanted with an ICD in the United States have a high 1-year mortality and may not benefit from ICD therapy.

  View details for DOI 10.1016/j.ahj.2014.12.013

  View details for PubMedID 26299225

• Does Age Influence Cardiac Resynchronization Therapy Use and Outcome? JACC. Heart failure Heidenreich, P. A., Tsai, V., Bao, H., Curtis, J., Goldstein, M., Curtis, L., Hernandez, A., Peterson, P., Turakhia, M. P., Masoudi, F. A. 2015; 3 (6): 497-504

  Abstract

  This study sought to describe the use of CRT-D and its association with survival for older patients.Many patients who receive cardiac resynchronization therapy with defibrillator (CRT-D) in practice are older than those included in clinical trials.We identified patients undergoing ICD implantation in the National Cardiovascular Disease Registry (NCDR) ICD registry from 2006 to 2009, who also met clinical trial criteria for CRT, including left ventricular ejection fraction (LVEF) ≤35%, QRS ≥120 ms, and New York Heart Association (NYHA) functional class III or IV. NCDR registry data were linked to the social security death index to determine the primary outcome of time to death from any cause. We identified 70,854 patients from 1,187 facilities who met prior trial criteria for CRT-D. The mean age of the 58,147 patients receiving CRT-D was 69.4 years with 6.4% of patients age 85 or older. CRT use was 80% or higher among candidates in all age groups. Follow-up was available for 42,285 patients age ≥65 years at 12 months.Receipt of CRT-D was associated with better survival at 1 year (82.1% vs. 77.1%, respectively) and 4 years (54.0% vs. 46.2% , respectively) than in those receiving only an ICD (p < 0.001). The CRT association with improved survival was not different for different age groups (p = 0.86 for interaction).More than 80% of older patients undergoing ICD implantation who were candidates for a CRT-D received the combined device. Mortality in older patients undergoing ICD implantation was high but was lower for those receiving CRT-D.

  View details for DOI 10.1016/j.jchf.2015.01.012

  View details for PubMedID 25982109

• Support for Management of Multiple Comorbidities through a Performance Measure Clinical Dashboard for Primary Care Goldstein, M. K., Tu, S. W., Martins, S. B., Oshiro, C., Yuen, K., Hwang, T., Wang, D., Furman, A., Ashcraft, M., Heidenreich, P. WILEY-BLACKWELL. 2015: S208

  View details for Web of Science ID 000352578900584

• Automating Identification of Multiple Chronic Conditions in Clinical Practice Guidelines. AMIA Joint Summits on Translational Science proceedings AMIA Summit on Translational Science Leung, T. I., Jalal, H., Zulman, D. M., Dumontier, M., Owens, D. K., Musen, M. A., Goldstein, M. K. 2015; 2015: 456-460

  Abstract

  Many clinical practice guidelines (CPGs) are intended to provide evidence-based guidance to clinicians on a single disease, and are frequently considered inadequate when caring for patients with multiple chronic conditions (MCC), or two or more chronic conditions. It is unclear to what degree disease-specific CPGs provide guidance about MCC. In this study, we develop a method for extracting knowledge from single-disease chronic condition CPGs to determine how frequently they mention commonly co-occurring chronic diseases. We focus on 15 highly prevalent chronic conditions. We use publicly available resources, including a repository of guideline summaries from the National Guideline Clearinghouse to build a text corpus, a data dictionary of ICD-9 codes from the Medicare Chronic Conditions Data Warehouse (CCW) to construct an initial list of disease terms, and disease synonyms from the National Center for Biomedical Ontology to enhance the list of disease terms. First, for each disease guideline, we determined the frequency of comorbid condition mentions (a disease-comorbidity pair) by exactly matching disease synonyms in the text corpus. Then, we developed an annotated reference standard using a sample subset of guidelines. We used this reference standard to evaluate our approach. Then, we compared the co-prevalence of common pairs of chronic conditions from Medicare CCW data to the frequency of disease-comorbidity pairs in CPGs. Our results show that some disease-comorbidity pairs occur more frequently in CPGs than others. Sixty-one (29.0%) of 210 possible disease-comorbidity pairs occurred zero times; for example, no guideline on chronic kidney disease mentioned depression, while heart failure guidelines mentioned ischemic heart disease the most frequently. Our method adequately identifies comorbid chronic conditions in CPG recommendations with precision 0.82, recall 0.75, and F-measure 0.78. Our work identifies knowledge currently embedded in the free text of clinical practice guideline recommendations and provides an initial view of the extent to which CPGs mention common comorbid conditions. Knowledge extracted from CPG text in this way may be useful to inform gaps in guideline recommendations regarding MCC and therefore identify potential opportunities for guideline improvement.

  View details for PubMedID 26306285

• Classification of Contextual Use of Left Ventricular Ejection Fraction Assessments. Studies in health technology and informatics Kim, Y. n., Garvin, J. n., Goldstein, M. K., Meystre, S. M. 2015; 216: 599–603

  Abstract

  Knowledge of the left ventricular ejection fraction is critical for the optimal care of patients with heart failure. When a document contains multiple ejection fraction assessments, accurate classification of their contextual use is necessary to filter out historical findings or recommendations and prioritize the assessments for selection of document level ejection fraction information. We present a natural language processing system that classifies the contextual use of both quantitative and qualitative left ventricular ejection fraction assessments in clinical narrative documents. We created support vector machine classifiers with a variety of features extracted from the target assessment, associated concepts, and document section information. The experimental results showed that our classifiers achieved good performance, reaching 95.6% F1-measure for quantitative assessments and 94.2% F1-measure for qualitative assessments in a five-fold cross-validation evaluation.

  View details for PubMedID 26262121

• Prospective Analysis of Health and Mortality Risk in Veteran and Non-Veteran Participants in the Women's Health Initiative. Women's health issues : official publication of the Jacobs Institute of Women's Health Weitlauf, J. C., LaCroix, A. Z., Bird, C. E., Woods, N. F., Washington, D. L., Katon, J. G., LaMonte, M. J., Goldstein, M. K., Bassuk, S. S., Sarto, G. E., Stefanick, M. L. 2015; 25 (6): 649–57

  Abstract

  The health of postmenopausal women veterans is a neglected area of study. A stronger empirical evidence base is needed, and would inform the provision of health care for the nearly 1 million U.S. women veterans currently 50 years of age or older. To this end, the present work compares salient health outcomes and risk of all-cause mortality among veteran and non-veteran participants of the Women's Health Initiative (WHI).This study features prospective analysis of long-term health outcomes and mortality risk (average follow-up, 8 years) among the 3,706 women veterans and 141,009 non-veterans who participated in the WHI Observational Study or Clinical Trials. Outcome measurements included confirmed incident cases of cardiovascular disease (CVD), cancer, diabetes, hip fractures, and all-cause mortality.We identified 17,968 cases of CVD, 19,152 cases of cancer, 18,718 cases of diabetes, 2,817 cases of hip fracture, and 13,747 deaths. In Cox regression models adjusted for age, sociodemographic variables, and health risk factors, veteran status was associated with significantly increased risk of all-cause mortality (hazard ratio [HR], 1.13; 95% CI, 1.03-1.23), but not with risk of CVD (HR, 1.00; 95% CI, 0.90-1.11), cancer (HR, 1.04; 95% CI, 0.95-1.14), hip fracture (HR, 1.16; 95% CI, 0.94-1.43), or diabetes (HR, 1.00; 95% CI, 0.89-1.1).Women veterans' postmenopausal health, particularly risk for all-cause mortality, warrants further consideration. In particular, efforts to identify and address modifiable risk factors associated with all-cause mortality are needed.

  View details for PubMedID 26432346

• Using a Clinical Knowledge Base to Assess Comorbidity Interrelatedness Among Patients with Multiple Chronic Conditions. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Zulman, D. M., Martins, S. B., Liu, Y., Tu, S. W., Hoffman, B. B., Asch, S. M., Goldstein, M. K. 2015; 2015: 1381-1389

  Abstract

  Decision support tools increasingly integrate clinical knowledge such as medication indications and contraindications with electronic health record (EHR) data to support clinical care and patient safety. The availability of this encoded information and patient data provides an opportunity to develop measures of clinical decision complexity that may be of value for quality improvement and research efforts. We investigated the feasibility of using encoded clinical knowledge and EHR data to develop a measure of comorbidity interrelatedness (the degree to which patients' co-occurring conditions interact to generate clinical complexity). Using a common clinical scenario-decisions about blood pressure medications in patients with hypertension-we quantified comorbidity interrelatedness by calculating the number of indications and contraindications to blood pressure medications that are generated by patients' comorbidities (e.g., diabetes, gout, depression). We examined properties of comorbidity interrelatedness using data from a decision support system for hypertension in the Veterans Affairs Health Care System.

  View details for PubMedID 26958279

• Health Outcomes Associated with Polypharmacy in Community-Dwelling Older Adults: A Systematic Review JOURNAL OF THE AMERICAN GERIATRICS SOCIETY Fried, T. R., O'Leary, J., Towle, V., Goldstein, M. K., Trentalange, M., Martin, D. K. 2014; 62 (12): 2261-2272

  Abstract

  To summarize evidence regarding the health outcomes associated with polypharmacy, defined as number of prescribed medications, in older community-dwelling persons.Systematic review of MEDLINE (OvidSP 1946 to May, Week 3, 2014).Community.Observational studies examining health outcomes according to number of prescription medications taken.Association between number of medications and health outcomes. Because of the importance of comorbidity as a potential confounder of the relationship between polypharmacy and health outcomes, articles were assessed regarding the quality of their adjustment for confounding.Of the 50 studies identified, the majority that were rated good in terms of their adjustment for comorbidity demonstrated relationships between polypharmacy and a range of outcomes, including falls, fall outcomes, fall risk factors, adverse drug events, hospitalization, mortality, and measures of function and cognition. However, a number of these studies failed to demonstrate associations, as did a substantial proportion of studies rated fair or poor.Data are mixed regarding the relationship between polypharmacy, considered in terms of number of medications, and adverse outcomes in community-dwelling older persons. Because of the challenge of confounding, randomized controlled trials of medication discontinuation may provide more-definitive evidence regarding this relationship than observational studies can provide.

  View details for DOI 10.1111/jgs.13153

  View details for Web of Science ID 000346497600004

  View details for PubMedID 25516023

  View details for PubMedCentralID PMC4270076

• The Effects of Comorbidity on the Benefits and Harms of Treatment for Chronic Disease: A Systematic Review PLOS ONE Fried, T. R., O'Leary, J., Towle, V., Goldstein, M. K., Trentelange, M., Martin, D. K. 2014; 9 (11)

  Abstract

  There are concerns about the potential for unintentional harms when clinical practice guidelines are applied to patients with multimorbidity. The objective was to summarize the evidence regarding the effect(s) of comorbidity on the outcomes of medication for an index chronic condition.A systematic review was conducted of studies published in MEDLINE and Cochrane Trials before May 2012. The search strategy was constructed to identify articles indexed with "comorbidity" or a related term or by a given condition and one or more additional specified comorbid conditions. The search yielded 3252 articles, of which 37 passed the title/abstract screening process, and 22 were included after full-text review. An additional 23 articles were identified by screening the reference lists for included articles. Information was extracted on study design; population; therapy; comparison groups; outcome(s); main findings.Indexing of articles was inconsistent, with no term for "multimorbidity," and rare use of "comorbidity". Only one article examined the effects of comorbidity per se, finding no benefit of tight control of DM among persons with high comorbidity, defined using a comorbidity index. The remainder examined pairs of conditions, the majority of which were post-hoc analyses of randomized controlled trials and which found no difference in outcomes according to whether a comorbid condition was present. Several demonstrated no difference or an increased risk of adverse outcome among persons with DM and tight control of HTN as compared to usual control. Several demonstrated lack of benefit of statins among persons with end-stage renal disease.There is limited evidence regarding the effects of multiple comorbidities on treatment outcomes. The majority of studies demonstrated no effect of a single comorbid condition on outcomes. Additional studies examining a broad range of comorbidity are required, along with clear and consistent indexing to allow for improved synthesis of the evidence.

  View details for DOI 10.1371/journal.pone.0112593

  View details for Web of Science ID 000345158700056

  View details for PubMedID 25402463

  View details for PubMedCentralID PMC4234418

• Mental illness and intensification of diabetes medications: an observational cohort study BMC HEALTH SERVICES RESEARCH Frayne, S. M., Holmes, T. H., Berg, E., Goldstein, M. K., Berlowitz, D. R., Miller, D. R., Pogach, L. M., Laungani, K. J., Lee, T. T., Moos, R. 2014; 14

  Abstract

  Mental health condition (MHC) comorbidity is associated with lower intensity care in multiple clinical scenarios. However, little is known about the effect of MHC upon clinicians' decisions about intensifying antiglycemic medications in diabetic patients with poor glycemic control. We examined whether delay in intensification of antiglycemic medications in response to an elevated Hemoglobin A1c (HbA1c) value is longer for patients with MHC than for those without MHC, and whether any such effect varies by specific MHC type.In this observational study of diabetic Veterans Health Administration (VA) patients on oral antiglycemics with poor glycemic control (HbA1c ≥8) (N =52,526) identified from national VA databases, we applied Cox regression analysis to examine time to intensification of antiglycemics after an elevated HbA1c value in 2003-2004, by MHC status.Those with MHC were no less likely to receive intensification: adjusted Hazard Ratio [95% CI] 0.99 [0.96-1.03], 1.13 [1.04-1.23], and 1.12 [1.07-1.18] at 0-14, 15-30 and 31-180 days, respectively. However, patients with substance use disorders were less likely than those without substance use disorders to receive intensification in the first two weeks following a high HbA1c, adjusted Hazard Ratio 0.89 [0.81-0.97], controlling for sex, age, medical comorbidity, other specific MHCs, and index HbA1c value.For most MHCs, diabetic patients with MHC in the VA health care system do not appear to receive less aggressive antiglycemic management. However, the subgroup with substance use disorders does appear to have excess likelihood of non-intensification; interventions targeting this high risk subgroup merit attention.

  View details for DOI 10.1186/1472-6963-14-458

  View details for Web of Science ID 000343734700001

  View details for PubMedCentralID PMC4282515

• Quality of care for patients with multiple chronic conditions: the role of comorbidity interrelatedness. Journal of general internal medicine Zulman, D. M., Asch, S. M., Martins, S. B., Kerr, E. A., Hoffman, B. B., Goldstein, M. K. 2014; 29 (3): 529-537

  Abstract

  Multimorbidity-the presence of multiple chronic conditions in a patient-has a profound impact on health, health care utilization, and associated costs. Definitions of multimorbidity in clinical care and research have evolved over time, initially focusing on a patient's number of comorbidities and the associated magnitude of required care processes, and later recognizing the potential influence of comorbidity characteristics on patient care and outcomes. In this article, we review the relationship between multimorbidity and quality of care, and discuss how this relationship may be mediated by the degree to which conditions interact with one another to generate clinical complexity (comorbidity interrelatedness). Drawing on established theoretical frameworks from cognitive engineering and biomedical informatics, we describe how interactions among conditions result in clinical complexity and may affect quality of care. We discuss how this comorbidity interrelatedness influences the value of existing quality guidelines and performance metrics, and describe opportunities to quantify this construct using data widely available through electronic health records. Incorporating comorbidity interrelatedness into conceptualizations of multimorbidity has the potential to enhance clinical and research efforts that aim to improve care for patients with multiple chronic conditions.

  View details for DOI 10.1007/s11606-013-2616-9

  View details for PubMedID 24081443

• Choosing a physician depends on how you want to feel: The role of ideal affect in health-related decision making. Emotion Sims, T., Tsai, J. L., Koopmann-Holm, B., Thomas, E. A., Goldstein, M. K. 2014; 14 (1): 187-192

  Abstract

  When given a choice, how do people decide which physician to select? Although significant research has demonstrated that how people actually feel (their "actual affect") influences their health care preferences, how people ideally want to feel (their "ideal affect") may play an even greater role. Specifically, we predicted that people trust physicians whose affective characteristics match their ideal affect, which leads people to prefer those physicians more. Consistent with this prediction, the more participants wanted to feel high arousal positive states on average (ideal HAP; e.g., excited), the more likely they were to select a HAP-focused physician. Similarly, the more people wanted to feel low arousal positive states on average (ideal LAP; e.g., calm), the more likely they were to select a LAP-focused physician. Also as predicted, these links were mediated by perceived physician trustworthiness. Notably, while participants' ideal affect predicted physician preference, actual affect (how much people actually felt HAP and LAP on average) did not. These findings suggest that people base serious decisions on how they want to feel, and highlight the importance of considering ideal affect in models of decision making preferences.

  View details for DOI 10.1037/a0034372

  View details for PubMedID 24188062

• Organizational factors affecting implementation of the ATHENA-Hypertension clinical decision support system during the VA’s nation-wide information technology restructuring: a case study Health System Shluzas, L. M., Cronkite, R. C., Hoffman, B. B., Breeling, J., Musen, M. A., Owens, D. K., Goldstein, M. K. 2014

  View details for DOI 10.1057/hs.2014.5

• ANXIETY IS ASSOCIATED WITH MULTIMORBIDITY IN OLDER ADULTS IN THE HEALTH AND RETIREMENT STUDY Gould, C. E., Goldstein, M. K., O'Hara, R., Beaudreau, S. A. OXFORD UNIV PRESS INC. 2013: 288

  View details for Web of Science ID 000327442103405

• Using Natural Language Processing for Extracting Information from Portable Chest X-Ray Reports Wang, D. Y., Hwang, T. S., Rubin, D., Chambers, J., South, B. R., Goldstein, M. K. WILEY-BLACKWELL. 2013: S103

  View details for Web of Science ID 000317187600293

• AGE DIFFERENCES IN CHOOSING EMOTIONALLY-FRAMED HEALTH CARE OPTIONS ACROSS CULTURES Sims, T., Tsai, J. L., Goldstein, M. K. OXFORD UNIV PRESS INC. 2012: 271–271

  View details for Web of Science ID 000312888202405

• Automated extraction of ejection fraction for quality measurement using regular expressions in Unstructured Information Management Architecture (UIMA) for heart failure JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION Garvin, J. H., DuVall, S. L., South, B. R., Bray, B. E., Bolton, D., Heavirland, J., Pickard, S., Heidenreich, P., Shen, S., Weir, C., Samore, M., Goldstein, M. K. 2012; 19 (5): 859-866

  Abstract

  Left ventricular ejection fraction (EF) is a key component of heart failure quality measures used within the Department of Veteran Affairs (VA). Our goals were to build a natural language processing system to extract the EF from free-text echocardiogram reports to automate measurement reporting and to validate the accuracy of the system using a comparison reference standard developed through human review. This project was a Translational Use Case Project within the VA Consortium for Healthcare Informatics.We created a set of regular expressions and rules to capture the EF using a random sample of 765 echocardiograms from seven VA medical centers. The documents were randomly assigned to two sets: a set of 275 used for training and a second set of 490 used for testing and validation. To establish the reference standard, two independent reviewers annotated all documents in both sets; a third reviewer adjudicated disagreements.System test results for document-level classification of EF of <40% had a sensitivity (recall) of 98.41%, a specificity of 100%, a positive predictive value (precision) of 100%, and an F measure of 99.2%. System test results at the concept level had a sensitivity of 88.9% (95% CI 87.7% to 90.0%), a positive predictive value of 95% (95% CI 94.2% to 95.9%), and an F measure of 91.9% (95% CI 91.2% to 92.7%).An EF value of <40% can be accurately identified in VA echocardiogram reports.An automated information extraction system can be used to accurately extract EF for quality measurement.

  View details for DOI 10.1136/amiajnl-2011-000535

  View details for Web of Science ID 000307934600025

  View details for PubMedID 22437073

  View details for PubMedCentralID PMC3422820

• Willingness to Pay in Dermatology: Assessment of the Burden of Skin Diseases JOURNAL OF INVESTIGATIVE DERMATOLOGY Seidler, A. M., Bayoumi, A. M., Goldstein, M. K., Cruz, P. D., Chen, S. C. 2012; 132 (7): 1785-1790

  Abstract

  Willingness to pay (WTP) is a monetary, preference-based, burden-of-disease measure with a potential role in dermatology, where many conditions are temporary and/or mild, and many treatments are inexpensive and one might be able to imagine paying out of pocket. We assessed construct validity by interviewing 254 consecutive dermatology patients at Stanford Medical Center, Grady Hospital, and Parkland Hospital. Instruments asked about an individual's own health status and elicited WTP, time-trade-off (TTO) utilities, and health status quality of life (QOL). We measured WTP cure (short treatment course to eliminate disease) and WTP control (lifelong medication). Our data indicate greater construct validity in non-Medicaid (n=163) than Medicaid (n=91) patients. Non-Medicaid subjects had greater WTP as percent of income for cure (median: 2%) than control (median: 1.6%), P<0.01; Medicaid WTP amounts for control and cure did not differ. Non-Medicaid subjects with verrucae had little QOL impact, no measurable burden by TTO, and a correspondingly low WTP. Medicaid subjects with basal cell carcinoma had a strong, negative QOL impact and high burden by TTO, but had relatively moderate WTP. WTP appears promising in certain income categories. More studies are needed for conclusions about specific diagnoses.

  View details for DOI 10.1038/jid.2012.50

  View details for Web of Science ID 000305492900009

  View details for PubMedID 22418874

• A NOVEL CONCEPTUAL MODEL FOR CLINICAL DECISION COMPLEXITY Zulman, D. M., Martins, S. B., Asch, S., Goldstein, M. K. SPRINGER. 2012: S106–S107

  View details for Web of Science ID 000209142900020

• A Clinically Guided Approach for Improving Performance Measurement for Hypertension MEDICAL CARE Steinman, M. A., Lee, S. J., Peterson, C. A., Fung, K. Z., Goldstein, M. K. 2012; 50 (5): 399-405

  Abstract

  Performance measures often fail to account for legitimate reasons why patients do not achieve recommended treatment targets.We tested a novel performance measurement system for blood pressure (BP) control that was designed to mimic clinical reasoning. This clinically guided approach focuses on (1) exempting patients for whom tight BP control may not be appropriate or feasible and (2) assessing BP over time. Trained abstractors conducted structured chart reviews of 201 adults with hypertension in 2 VA health care systems. Results were compared with traditional methods of performance measurement.Among 201 veterans, 183 (91%) were male, and the mean age was 71±11 years. Using the clinically guided approach, 61 patients (30%) were exempted from performance measurement. The most common reasons for exemption were inadequate opportunity to manage BP (35 patients, 17%) and the use of 4 or more antihypertensive medications (19 patients, 9%). Among patients eligible for performance measurement, there was little agreement on the presence of controlled versus uncontrolled BP when comparing the most recent BP (the traditional approach) with an integrated assessment of BP control (κ 0.14). After accounting for clinically guided exemptions and methods of BP assessment, only 15 of 72 patients (21%) whose last BP was ≥140/90 mm Hg were classified as problematic by the clinically guided approach.Many patients have legitimate reasons for not achieving tight BP control, and the methods used for BP assessment have marked effects on whether a patient is classified as having adequate or inadequate BP control.

  View details for DOI 10.1097/MLR.0b013e318245a147

  View details for Web of Science ID 000314456400007

  View details for PubMedID 22627680

  View details for PubMedCentralID PMC3368331

• Knowing Loved Ones' End-of-Life Health Care Wishes: Attachment Security Predicts Caregivers' Accuracy HEALTH PSYCHOLOGY Turan, B., Goldstein, M. K., Garber, A. M., Carstensen, L. L. 2011; 30 (6): 814-818

  Abstract

  At times, caregivers make life-and-death decisions for loved ones. Yet very little is known about the factors that make caregivers more or less accurate as surrogate decision makers for their loved ones. Previous research suggests that in low stress situations, individuals with high attachment-related anxiety are attentive to their relationship partners' wishes and concerns, but get overwhelmed by stressful situations. Individuals with high attachment-related avoidance are likely to avoid intimacy and stressful situations altogether. We hypothesized that both of these insecure attachment patterns limit surrogates' ability to process distressing information and should therefore be associated with lower accuracy in the stressful task of predicting their loved ones' end-of-life health care wishes.Older patients visiting a medical clinic stated their preferences toward end-of-life health care in different health contexts, and surrogate decision makers independently predicted those preferences. For comparison purposes, surrogates also predicted patients' perceptions of everyday living conditions so that surrogates' accuracy of their loved ones' perceptions in nonstressful situations could be assessed.Surrogates high on either type of insecure attachment dimension were less accurate in predicting their loved ones' end-of-life health care wishes. It is interesting to note that even though surrogates' attachment-related anxiety was associated with lower accuracy of end-of-life health care wishes of their loved ones, it was associated with higher accuracy in the nonstressful task of predicting their loved ones' everyday living conditions.Attachment orientation plays an important role in accuracy about loved ones' end-of-life health care wishes. Interventions may target emotion regulation strategies associated with insecure attachment orientations.

  View details for DOI 10.1037/a0025664

  View details for Web of Science ID 000297029200020

  View details for PubMedID 22081941

  View details for PubMedCentralID PMC3228368

• PREDICTORS OF TREATMENT PREFERENCE AGREEMENT AMONG OLDER PATIENTS, SURROGATES, AND PHYSICIANS Woodhead, E., English, T., Goldstein, M., Garber, A., Carstensen, L. OXFORD UNIV PRESS INC. 2011: 450

  View details for Web of Science ID 000303602003039

• Influence of Age on Perioperative Complications Among Patients Undergoing Implantable Cardioverter-Defibrillators for Primary Prevention in the United States CIRCULATION-CARDIOVASCULAR QUALITY AND OUTCOMES Tsai, V., Goldstein, M. K., Hsia, H. H., Wang, Y., Curtis, J., Heidenreich, P. A. 2011; 4 (5): 549-U111

  Abstract

  BACKGROUND- The majority of current implantable cardioverter-defibrillator (ICD) recipients are significantly older than those in the ICD trials. Data on periprocedural complications among the elderly are insufficient. We evaluated the influence of age on perioperative complications among primary prevention ICD recipients in the United States. METHODS AND RESULTS- Using the National Cardiovascular Data's ICD Registry, we identified 150 264 primary prevention patients who received ICDs from January 2006 to December 2008. The primary end point was any adverse event or in-hospital mortality. Secondary end points included major adverse events, minor adverse events, and length of stay. Of 150 264 patients, 61% (n=91 863) were 65 years and older. A higher proportion of patients ≥65 years had diabetes, congestive heart failure, atrial fibrillation, renal disease, and coronary artery disease. Approximately 3.4% of the entire cohort had any complication, including death, after ICD implant. Any adverse event or death occurred in 2.8% of patients under 65 years old; 3.1% of 65- to 69-year-olds; 3.5% of 70- to 74-year-olds; 3.9% of 75- to 79-year-olds, 4.5% of 80- to 84-year-olds; and 4.5% of patients 85 years and older. After adjustment for clinical covariates, multivariate analysis found an increased odds of any adverse event or death among 75- to 79-year-olds (1.14 [95% confidence interval, 1.03 to 1.25], 80-to 84-year-olds (1.22 [95% confidence interval, 1.10 to 1.36], and patients 85 years and older (1.15 [95% confidence interval, 1.01 to 1.32], compared with patients under 65 years old. CONCLUSIONS- Older patients had a modestly increased-but acceptably safe-risk of periprocedural complications and in-hospital mortality, driven mostly by increased comorbidity.

  View details for DOI 10.1161/CIRCOUTCOMES.110.959205

  View details for Web of Science ID 000295028800012

  View details for PubMedID 21878667

• Age Differences in Primary Prevention Implantable Cardioverter-Defibrillator Use in US Individuals JOURNAL OF THE AMERICAN GERIATRICS SOCIETY Tsai, V., Goldstein, M. K., Hsia, H. H., Wang, Y., Curtis, J., Heidenreich, P. A. 2011; 59 (9): 1589-1595

  Abstract

  To estimate the potentially inappropriate use of implantable cardioverter-defibrillator ICDs in older U.S. adults.Retrospective study.The National Cardiovascular Data ICD Registry.Forty-four thousand eight hundred five individuals in the National Cardiovascular Data's ICD Registry(™) who had received ICDs for primary prevention from January 2006 to December 2008. Individuals with a prior myocardial infarction and ejection fraction less than 30% were included.Mortality risk was categorized using the Multicenter Automatic Defibrillator Implantation (MADIT) II risk-stratification system. Low-risk and very-high-risk individuals were considered potentially inappropriate recipients.Of 44,805 individuals, 67% (n = 29,893) were aged 65 and older, of whom 51% were aged 75 and older. A significant proportion of ICD recipients had a low risk of death (16%, n = 6,969) or very high risk of nonarrhythmic death (8%, n = 3,693). Potentially inappropriate ICD use was 10% in those aged 75 and older, much less than in younger groups (40%, <65; 21%, 65-74, P < .001). Although age was associated with a high risk of nonarrhythmic death, its influence was markedly attenuated after adjusting for comorbidities and timing of ICD implantation (odds ratio = 1.02, 95% confidence interval = 1.02-1.03, P < .001).Potentially inappropriate ICD use appears significantly less--and at modest rates--in older Americans than in younger age groups. Overall, almost one-quarter of individuals may have received ICDs inappropriately based on their risk of death. Physicians appear to be conservatively referring older adults and wisely deferring those with high comorbid burden.

  View details for DOI 10.1111/j.1532-5415.2011.03542.x

  View details for Web of Science ID 000295615000003

  View details for PubMedID 21883101

• Home Blood Pressure Management and Improved Blood Pressure Control Results From a Randomized Controlled Trial ARCHIVES OF INTERNAL MEDICINE Bosworth, H. B., Powers, B. J., Olsen, M. K., McCant, F., Grubber, J., Smith, V., Gentry, P. W., Rose, C., van Houtven, C., Wang, V., Goldstein, M. K., Oddone, E. Z. 2011; 171 (13): 1173-1180

  Abstract

  To determine which of 3 interventions was most effective in improving blood pressure (BP) control, we performed a 4-arm randomized trial with 18-month follow-up at the primary care clinics at a Veterans Affairs Medical Center.Eligible patients were randomized to either usual care or 1 of 3 telephone-based intervention groups: (1) nurse-administered behavioral management, (2) nurse- and physician-administered medication management, or (3) a combination of both. Of the 1551 eligible patients, 593 individuals were randomized; 48% were African American. The intervention telephone calls were triggered based on home BP values transmitted via telemonitoring devices. Behavioral management involved promotion of health behaviors. Medication management involved adjustment of medications by a study physician and nurse based on hypertension treatment guidelines.The primary outcome was change in BP control measured at 6-month intervals over 18 months. Both the behavioral management and medication management alone showed significant improvements at 12 months-12.8% (95% confidence interval [CI], 1.6%-24.1%) and 12.5% (95% CI, 1.3%-23.6%), respectively-but not at 18 months. In subgroup analyses, among those with poor baseline BP control, systolic BP decreased in the combined intervention group by 14.8 mm Hg (95% CI, -21.8 to -7.8 mm Hg) at 12 months and 8.0 mm Hg (95% CI, -15.5 to -0.5 mm Hg) at 18 months, relative to usual care.Overall intervention effects were moderate, but among individuals with poor BP control at baseline, the effects were larger. This study indicates the importance of identifying individuals most likely to benefit from potentially resource intensive programs.clinicaltrials.gov Identifier: NCT00237692.

  View details for Web of Science ID 000292658100006

  View details for PubMedID 21747013

• Developing Heart Failure (HF) Clinical Decision Support for Patients with Comorbidities. Goldstein, M. K., Martins, S. B., Tu, S. W., Heidenreich, P., Massie, B., Furmaga, E. WILEY-BLACKWELL. 2011: S146

  View details for Web of Science ID 000289524600424

• Behavioral medicine perspectives on the design of health information technology to improve decision-making, guideline adherence, and care coordination in chronic pain management. Translational behavioral medicine Midboe, A. M., Lewis, E. T., Cronkite, R. C., Chambers, D., Goldstein, M. K., Kerns, R. D., Trafton, J. A. 2011; 1 (1): 35-44

  Abstract

  Development of clinical decision support systems (CDSs) has tended to focus on facilitating medication management. An understanding of behavioral medicine perspectives on the usefulness of a CDS for patient care can expand CDSs to improve management of chronic disease. The purpose of this study is to explore feedback from behavioral medicine providers regarding the potential for CDSs to improve decision-making, care coordination, and guideline adherence in pain management. Qualitative methods were used to analyze semi-structured interview responses from behavioral medicine stakeholders following demonstration of an existing CDS for opioid prescribing, ATHENA-OT. Participants suggested that a CDS could assist with decision-making by educating providers, providing recommendations about behavioral therapy, facilitating risk assessment, and improving referral decisions. They suggested that a CDS could improve care coordination by facilitating division of workload, improving patient education, and increasing consideration and knowledge of options in other disciplines. Clinical decision support systems are promising tools for improving behavioral medicine care for chronic pain.

  View details for DOI 10.1007/s13142-011-0022-6

  View details for PubMedID 24073031

  View details for PubMedCentralID PMC3717707

• Designing an automated clinical decision support system to match clinical practice guidelines for opioid therapy for chronic pain IMPLEMENTATION SCIENCE Trafton, J. A., Martins, S. B., Michel, M. C., Wang, D., Tu, S. W., Clark, D. J., Elliott, J., Vucic, B., Balt, S., Clark, M. E., Sintek, C. D., Rosenberg, J., Daniels, D., Goldstein, M. K. 2010; 5

  Abstract

  Opioid prescribing for chronic pain is common and controversial, but recommended clinical practices are followed inconsistently in many clinical settings. Strategies for increasing adherence to clinical practice guideline recommendations are needed to increase effectiveness and reduce negative consequences of opioid prescribing in chronic pain patients.Here we describe the process and outcomes of a project to operationalize the 2003 VA/DOD Clinical Practice Guideline for Opioid Therapy for Chronic Non-Cancer Pain into a computerized decision support system (DSS) to encourage good opioid prescribing practices during primary care visits. We based the DSS on the existing ATHENA-DSS. We used an iterative process of design, testing, and revision of the DSS by a diverse team including guideline authors, medical informatics experts, clinical content experts, and end-users to convert the written clinical practice guideline into a computable algorithm to generate patient-specific recommendations for care based upon existing information in the electronic medical record (EMR), and a set of clinical tools.The iterative revision process identified numerous and varied problems with the initially designed system despite diverse expert participation in the design process. The process of operationalizing the guideline identified areas in which the guideline was vague, left decisions to clinical judgment, or required clarification of detail to insure safe clinical implementation. The revisions led to workable solutions to problems, defined the limits of the DSS and its utility in clinical practice, improved integration into clinical workflow, and improved the clarity and accuracy of system recommendations and tools.Use of this iterative process led to development of a multifunctional DSS that met the approval of the clinical practice guideline authors, content experts, and clinicians involved in testing. The process and experiences described provide a model for development of other DSSs that translate written guidelines into actionable, real-time clinical recommendations.

  View details for DOI 10.1186/1748-5908-5-26

  View details for Web of Science ID 000277618600001

  View details for PubMedID 20385018

  View details for PubMedCentralID PMC2868045

• When tight blood pressure control is not for everyone: a new model for performance measurement in hypertension. Joint Commission journal on quality and patient safety / Joint Commission Resources Steinman, M. A., Goldstein, M. K. 2010; 36 (4): 164-172

  Abstract

  Many patients with hypertension have legitimate reasons to forego standard blood pressure targets yet are nonetheless included in performance measurement systems. An approach to performance measurement incorporating clinical reasoning was developed to determine which patients to include in a performance measure.A 10-member multispecialty advisory panel refined a taxonomy of situations in which the balance of benefits and harms of anti-hypertensive treatment does not clearly favor tight blood pressure control (< 140/90 mm Hg).The panel identified several broad categories of reasons for exempting a patient from performance measurement for blood pressure control. These included (1) patients who have suffered adverse effects from multiple classes of antihypertensive medications; (2) patients already taking four or more antihypertensive medications; (3) patients with terminal disease, moderate to severe dementia, or other conditions that overwhelmingly dominate the patient's clinical status; and (4) other patient factors, including comfort care orientation and poor medication adherence despite attempts to remedy adherence difficulties. Several general principles also emerged. Performance measurement should focus on patients for whom the benefits of treatment clearly outweigh the harms and should incorporate a longitudinal approach. In addition, the criteria for exempting a patient from performance measurement should be more strict in patients at higher risk of adverse health outcomes from hypertension and more lenient for patients at lower risk.Incorporating "real world" clinical principles and judgment into performance measurement systems may improve targeting of care and, by accounting for patient case mix, allow for better comparison of performance between institutions.

  View details for PubMedID 20402373

  View details for PubMedCentralID PMC2858350

• Following Your Heart or Your Head: Focusing on Emotions Versus Information Differentially Influences the Decisions of Younger and Older Adults JOURNAL OF EXPERIMENTAL PSYCHOLOGY-APPLIED Mikels, J. A., Loeckenhoff, C. E., Maglio, S. J., Carstensen, L. L., Goldstein, M. K., Garber, A. 2010; 16 (1): 87-95

  Abstract

  Research on aging has indicated that whereas deliberative cognitive processes decline with age, emotional processes are relatively spared. To examine the implications of these divergent trajectories in the context of health care choices, we investigated whether instructional manipulations emphasizing a focus on feelings or details would have differential effects on decision quality among younger and older adults. We presented 60 younger and 60 older adults with health care choices that required them to hold in mind and consider multiple pieces of information. Instructional manipulations in the emotion-focus condition asked participants to focus on their emotional reactions to the options, report their feelings about the options, and then make a choice. In the information-focus condition, participants were instructed to focus on the specific attributes, report the details about the options, and then make a choice. In a control condition, no directives were given. Manipulation checks indicated that the instructions were successful in eliciting different modes of processing. Decision quality data indicate that younger adults performed better in the information-focus than in the control condition whereas older adults performed better in the emotion-focus and control conditions than in the information-focus condition. Findings support and extend extant theorizing on aging and decision making as well as suggest that interventions to improve decision-making quality should take the age of the decision maker into account.

  View details for DOI 10.1037/a0018500

  View details for Web of Science ID 000276369400007

  View details for PubMedID 20350046

• Evaluation of the Acceptability and Usability of a Decision Support System to Encourage Safe and Effective Use of Opioid Therapy for Chronic, Noncancer Pain by Primary Care Providers PAIN MEDICINE Trafton, J., Martins, S., Michel, M., Lewis, E., Wang, D., Combs, A., Scates, N., Tu, S., Goldstein, M. K. 2010; 11 (4): 575-585

  Abstract

  To develop and evaluate a clinical decision support system (CDSS) named Assessment and Treatment in Healthcare: Evidenced-Based Automation (ATHENA)-Opioid Therapy, which encourages safe and effective use of opioid therapy for chronic, noncancer pain.CDSS development and iterative evaluation using the analysis, design, development, implementation, and evaluation process including simulation-based and in-clinic assessments of usability for providers followed by targeted system revisions.Volunteers provided detailed feedback to guide improvements in the graphical user interface, and content and design changes to increase clinical usefulness, understandability, clinical workflow fit, and ease of completing guideline recommended practices. Revisions based on feedback increased CDSS usability ratings over time. Practice concerns outside the scope of the CDSS were also identified.Usability testing optimized the CDSS to better address barriers such as lack of provider education, confusion in dosing calculations and titration schedules, access to relevant patient information, provider discontinuity, documentation, and access to validated assessment tools. It also highlighted barriers to good clinical practice that are difficult to address with CDSS technology in its current conceptualization. For example, clinicians indicated that constraints on time and competing priorities in primary care, discomfort in patient-provider communications, and lack of evidence to guide opioid prescribing decisions impeded their ability to provide effective, guideline-adherent pain management. Iterative testing was essential for designing a highly usable and acceptable CDSS; however, identified barriers may limit the impact of the ATHENA-Opioid Therapy system and other CDSS on clinical practices and outcomes unless CDSS are paired with parallel initiatives to address these issues.

  View details for Web of Science ID 000276223500015

  View details for PubMedID 20202142

• Natural Language Processing for Lines and Devices in Portable Chest X-Rays Proc AMIA Annual Symposium 2010 Rubin D, Wang D, Chambers DA, Chambers JG, South BR, Goldstein MK 2010: 692-696
• The morningside initiative: collaborative development of a knowledge repository to accelerate adoption of clinical decision support. The open medical informatics journal Greenes, R., Bloomrosen, M., Brown-Connolly, N. E., Curtis, C., Detmer, D. E., Enberg, R., Fridsma, D., Fry, E., Goldstein, M. K., Haug, P., Hulse, N., Hongsermeier, T., Maviglia, S., Robbins, C. W., Shah, H. 2010; 4: 278-290

  Abstract

  The Morningside Initiative is a public-private activity that has evolved from an August, 2007, meeting at the Morningside Inn, in Frederick, MD, sponsored by the Telemedicine and Advanced Technology Research Center (TATRC) of the US Army Medical Research Materiel Command. Participants were subject matter experts in clinical decision support (CDS) and included representatives from the Department of Defense, Veterans Health Administration, Kaiser Permanente, Partners Healthcare System, Henry Ford Health System, Arizona State University, and the American Medical Informatics Association (AMIA). The Morningside Initiative was convened in response to the AMIA Roadmap for National Action on Clinical Decision Support and on the basis of other considerations and experiences of the participants. Its formation was the unanimous recommendation of participants at the 2007 meeting which called for creating a shared repository of executable knowledge for diverse health care organizations and practices, as well as health care system vendors. The rationale is based on the recognition that sharing of clinical knowledge needed for CDS across organizations is currently virtually non-existent, and that, given the considerable investment needed for creating, maintaining and updating authoritative knowledge, which only larger organizations have been able to undertake, this is an impediment to widespread adoption and use of CDS. The Morningside Initiative intends to develop and refine (1) an organizational framework, (2) a technical approach, and (3) CDS content acquisition and management processes for sharing CDS knowledge content, tools, and experience that will scale with growing numbers of participants and can be expanded in scope of content and capabilities. Intermountain Healthcare joined the initial set of participants shortly after its formation. The efforts of the Morningside Initiative are intended to serve as the basis for a series of next steps in a national agenda for CDS. It is based on the belief that sharing of knowledge can be highly effective as is the case in other competitive domains such as genomics. Participants in the Morningside Initiative believe that a coordinated effort between the private and public sectors is needed to accomplish this goal and that a small number of highly visible and respected health care organizations in the public and private sector can lead by example. Ultimately, a future collaborative knowledge sharing organization must have a sustainable long-term business model for financial support.

  View details for DOI 10.2174/1874431101004010278

  View details for PubMedID 21603282

• Preferences for dialysis withdrawal and engagement in advance care planning within a diverse sample of dialysis patients NEPHROLOGY DIALYSIS TRANSPLANTATION Tamura, M. K., Goldstein, M. K., Perez-Stable, E. J. 2010; 25 (1): 237-242

  Abstract

  Rates of dialysis withdrawal are higher among the elderly and lower among Blacks, yet it is unknown whether preferences for withdrawal and engagement in advance care planning also vary by age and race or ethnicity. DESIGN, SETTING, PARTICIPANTS AND METHODS: We recruited 61 participants from two dialysis clinics to complete questionnaires regarding dialysis withdrawal preferences in five different health states. Engagement in advance care planning (end-of-life discussions), completion of advance directives and 'do not resuscitate' or 'do not intubate' (DNR/DNI) orders were ascertained by a questionnaire and from dialysis unit records.The mean age was 62 +/- 15 years; 38% were Black, 11% were Latino, 34% were White and 16% of participants were Asian. Blacks were less likely to prefer dialysis withdrawal as compared with Whites (odds ratio 0.16, 95% confidence interval 0.03-0.88) and other race/ethnicity groups, and this difference was not explained by age, education, comorbidity and other confounders. In contrast, older age was not associated with preferences for withdrawal. Rates of engagement in end-of-life discussions were higher than for documentation of advance care planning for all age and most race/ethnicity groups. Although younger participants and minorities were generally less likely to document treatment preferences as compared with older patients and Whites, they were not less likely to engage in end-of-life discussions.Preferences for withdrawal vary by race/ ethnicity, whereas the pattern of engagement in advance care planning varies by age and race/ethnicity. Knowledge of these differences may be useful for improving communication about end-of-life preferences and in implementing effective advance care planning strategies among diverse haemodialysis patients.

  View details for DOI 10.1093/ndt/gfp430

  View details for Web of Science ID 000273113100039

  View details for PubMedID 19734137

  View details for PubMedCentralID PMC2910327

• Natural language processing for lines and devices in portable chest x-rays. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Rubin, D., Wang, D., Chambers, D. A., Chambers, J. G., South, B. R., Goldstein, M. K. 2010; 2010: 692-696

  Abstract

  Radiology reports are unstructured free text documents that describe abnormalities in patients that are visible via imaging modalities such as X-ray. The number of imaging examinations performed in clinical care is enormous, and mining large repositories of radiology reports connected with clinical data such as patient outcomes could enable epidemiological studies, such as correlating the frequency of infections to the presence or length of time medical devices are present in patients. We developed a natural language processing (NLP) system to recognize device mentions in radiology reports and information about their state (insertion or removal) to enable epidemiological research. We tested our system using a reference standard of reports that were annotated to indicate this information. Our system performed with high accuracy (recall and precision of 97% and 99% for device mentions and 91-96% for device insertion status). Our methods are generalizable to other types of radiology reports as well as to other information extraction tasks and could provide the foundation for tools that enable epidemiological research exploration based on mining radiology reports.

  View details for PubMedID 21347067

• Ability of expert physicians to structure clinical guidelines: reality versus perception JOURNAL OF EVALUATION IN CLINICAL PRACTICE Shalom, E., Shahar, Y., Taieb-Maimon, M., Martins, S. B., Vaszar, L. T., Goldstein, M. K., Gutnik, L., Lunenfeld, E. 2009; 15 (6): 1043-1053

  Abstract

  Structuring Textual Clinical Guidelines (GLs) into a formal representation is a necessary prerequisite for supporting their automated application. We had developed a collaborative guideline-structuring methodology that involves expert physicians, clinical editors and knowledge engineers, to produce a machine-comprehensible representation for automated support of evidence-based, guideline-based care. Our goals in the current study were: (1) to investigate the perceptions of the expert physicians and clinical editors as to the relative importance, for the structuring process, of different aspects of the methodology; (2) to assess, for the clinical editors, the inter-correlations among (i) the reported level of understanding of the guideline structuring ontology's (knowledge scheme's) features, (ii) the reported ease of structuring each feature and (iii) the actual objective quality of structuring.A clinical consensus regarding the contents of three guidelines was prepared by an expert in the domain of each guideline. For each guideline, two clinical editors independently structured the guideline into a semi-formal representation, using the Asbru guideline ontology's features. The quality of the resulting structuring was assessed quantitatively. Each expert physician was asked which aspects were most useful for formation of the consensus. Each clinical editor filled questionnaires relating to: (1) the level of understanding of the ontology's features (before the structuring process); (2) the usefulness of various aspects in the structuring process (after the structuring process); (3) the ease of structuring each ontological feature (after the structuring process). Subjective reports were compared with objective quantitative measures of structuring correctness.Expert physicians considered having medical expertise and understanding the ontological features as the aspects most useful for creation of a consensus. Clinical editors considered understanding the ontological features and the use of the structuring tools as the aspects most useful for structuring guidelines. There was a positive correlation (R = 0.87, P < 0.001) between the reported ease of understanding ontological features and the reported ease of structuring those features. However, there was no significant correlation between the reported level of understanding the features - or the reported ease of structuring by using those features - and the objective quality of the structuring of these features in actual guidelines.Aspects considered important for formation of a clinical consensus differ from those for structuring of guidelines. Understanding the features of a structuring ontology is positively correlated with the reported ease of using these features, but neither of these subjective reports correlated with the actual objective quality of the structuring using these features.

  View details for DOI 10.1111/j.1365-2753.2009.01241.x

  View details for Web of Science ID 000273067800026

  View details for PubMedID 20367704

• Patient education and provider decision support to control blood pressure in primary care: A cluster randomized trial 78th Annual Scientific Session of the American-Heart-Association Bosworth, H. B., Olsen, M. K., Dudley, T., Orr, M., Goldstein, M. K., Datta, S. K., McCant, F., Gentry, P., Simel, D. L., Oddone, E. Z. MOSBY-ELSEVIER. 2009: 450–56

  Abstract

  Less than one third of the 65 million Americans with hypertension have adequate blood pressure (BP) control. This study examined the effectiveness of 2 interventions for improving patient BP control.This was a 2-level (primary care provider and patient) cluster randomized trial with 2-year follow-up occurring among patients with hypertension enrolled from a Veterans Affairs Medical Center primary care clinic. Primary care providers (n = 17) in the intervention received computer-generated decision support designed to improve guideline concordant medical therapy at each visit; control providers (n = 15) received a reminder at each visit. Patients received usual care or a bimonthly tailored nurse-delivered behavioral telephone intervention to improve hypertension treatment. The primary outcome was proportion of patients who achieved a BP <140/90 mm Hg (<130/85 for diabetic patients) over the 24-month intervention.Of the 816 eligible patients contacted, 190 refused and 38 were excluded. The 588 enrolled patients had a mean age of 63 years, 43% had adequate baseline BP control, and 482 (82%) completed the 24-month follow-up. There were no significant differences in amount of change in BP control in the 3 intervention groups as compared to the hypertension reminder control group. In secondary analyses, rates of BP control for all patients receiving the patient behavioral intervention (n = 294) improved from 40.1% to 54.4% at 24 months (P = .03); patients in the nonbehavioral intervention group improved from 38.2% to 43.9% (P = .38), but there was no between-group differences at the end of the study.The brief behavioral intervention showed improved outcomes over time, but there were not significant between group differences.

  View details for DOI 10.1016/j.ahj.2008.11.003

  View details for Web of Science ID 000264214700008

  View details for PubMedID 19249414

• Use of a Technological Instrument to Facilitate Educational Program Development in Medicine. Bao, Y., Huh, J., Goldstein, M. K. WILEY-BLACKWELL PUBLISHING, INC. 2009: S38

  View details for Web of Science ID 000264611900106

• Can Physicians Structure Clinical Guidelines? Experiments with a Mark-Up-Process Methodology Workshop on Knowledge Management for Health Care Procedures (K4HelP 2008) held at the 18th European Conference on Artificial Intelligence Shalom, E., Shahar, Y., Taieb-Maimon, M., Bar, G., Martins, S. B., Young, O., Vaszar, L., Liel, Y., Yarkoni, A., Goldstein, M. K., Leibowitz, A., Marom, T., Lunenfeld, E. SPRINGER-VERLAG BERLIN. 2009: 67–80

  View details for Web of Science ID 000270024300006

• Simple counts of ADL dependencies do not adequately reflect older adults' preferences toward states of functional impairment JOURNAL OF CLINICAL EPIDEMIOLOGY Sims, T., Holmes, T. H., Bravata, D. M., Garber, A. M., Nelson, L. M., Goldstein, M. K. 2008; 61 (12): 1261-1270

  Abstract

  To use unweighted counts of dependencies in activities of daily living (ADLs) to assess the impact of functional impairment requires an assumption of equal preferences for each ADL dependency. To test this assumption, we analyzed standard gamble (SG) utilities of single and combination ADL dependencies among older adults.Four hundred older adults used multimedia software (FLAIR1) to report SG utilities for their current health and hypothetical health states of dependency in each of 7 ADLs and 8 of 30 combinations of ADL dependencies.Utilities for health states of multiple ADL dependencies were often greater than for states of single ADL dependencies. Dependence in eating, which is the ADL dependency with the lowest utility rating of the single ADL dependencies, ranked lower than 7 combination states. Similarly, some combination states with fewer ADL dependencies had lower utilities than those with more ADL dependencies. These findings were consistent across groups by gender, age, and education.Our results suggest that the count of ADL dependencies does not adequately represent the utility for a health state. Cost-effectiveness analyses and other evaluations of programs that prevent or treat functional dependency should apply utility weights rather than relying on simple ADL counts.

  View details for DOI 10.1016/j.jclinepi.2008.05.001

  View details for Web of Science ID 000261219700010

  View details for PubMedID 18722749

  View details for PubMedCentralID PMC2596888

• A quantitative assessment of a methodology for collaborative specification and evaluation of clinical guidelines JOURNAL OF BIOMEDICAL INFORMATICS Shalom, E., Shahar, Y., Taieb-Maimon, M., Bar, G., Yarkoni, A., Young, O., Martins, S. B., Vaszar, L., Goldstein, M. K., Liel, Y., Leibowitz, A., Marom, T., Lunenfeld, E. 2008; 41 (6): 889-903

  Abstract

  We introduce a three-phase, nine-step methodology for specification of clinical guidelines (GLs) by expert physicians, clinical editors, and knowledge engineers and for quantitative evaluation of the specification's quality. We applied this methodology to a particular framework for incremental GL structuring (mark-up) and to GLs in three clinical domains. A gold-standard mark-up was created, including 196 plans and subplans, and 326 instances of ontological knowledge roles (KRs). A completeness measure of the acquired knowledge revealed that 97% of the plans and 91% of the KR instances of the GLs were recreated by the clinical editors. A correctness measure often revealed high variability within clinical editor pairs structuring each GL, but for all GLs and clinical editors the specification quality was significantly higher than random (p<0.01). Procedural KRs were more difficult to mark-up than declarative KRs. We conclude that given an ontology-specific consensus, clinical editors with mark-up training can structure GL knowledge with high completeness, whereas the main demand for correct structuring is training in the ontology's semantics.

  View details for DOI 10.1016/j.jbi.2008.04.009

  View details for Web of Science ID 000261220500004

  View details for PubMedID 18550447

• Using health information technology to improve hypertension management CURRENT HYPERTENSION REPORTS Goldstein, M. K. 2008; 10 (3): 201-207

  Abstract

  High-quality medical care requires implementing evidence-based best practices, with continued monitoring to improve performance. Implementation science is beginning to identify approaches to developing, implementing, and evaluating quality improvement strategies across health care systems that lead to good outcomes for patients. Health information technology has much to contribute to quality improvement for hypertension, particularly as part of multidimensional strategies for improved care. Clinical reminders closely aligned with organizational commitment to quality improvement may be one component of a successful strategy for improving blood pressure control. The ATHENA-Hypertension (Assessment and Treatment of Hypertension: Evidence-based Automation) system is an example of more complex clinical decision support. It is feasible to implement and deploy innovative health information technologies for clinical decision support with features such as clinical data visualizations and evidence to support specific recommendations. Further study is needed to determine the optimal contexts for such systems and their impact on patient outcomes.

  View details for Web of Science ID 000256532700009

  View details for PubMedID 18765090

• The impact of chronic hepatitis C and co-morbid illnesses on health-related quality of life QUALITY OF LIFE RESEARCH Kwan, J. W., Cronkite, R. C., Yiu, A., Goldstein, M. K., Kazis, L., Cheung, R. C. 2008; 17 (5): 715-724

  Abstract

  Determine the relative impact of chronic hepatitis C (CHC) and co-morbid illnesses on health-related quality of life (HRQoL) in 3023 randomly selected veterans with known hepatitis C virus antibody (anti-HCV) status who previously completed a veteran-specific HRQoL questionnaire (SF-36V).Multiple regression analyses were performed to measure the relative contribution of anti-HCV status, four demographic variables, and ten common medical and six psychiatric co-morbidities to HRQoL between 303 anti-HCV(+) and 2720 anti-HCV(-) patients.Anti-HCV(+) veterans were younger, reported a lower HRQoL on seven of eight 36-Item Short Form Health Survey for Veterans (SF-36V) subscales (P < or = 0.001) and the mental component summary (MCS) scale (P < 0.001). The ten medical and six psychiatric co-morbidities had variable impact on predicting lower HRQoL in both groups. After adjusting for demographic variables and co-morbid illnesses, we found that anti-HCV(+) patients reported a significantly lower MCS score (P < 0.001) and a trend toward a lower physical component summary (PCS) score (P < 0.07) compared to anti-HCV(-) veterans. Among the anti-HCV(+) veterans, co-morbid medical illnesses contributed to impaired PCS but not to MCS.Veterans with CHC were younger than HCV(-) veterans and hence less likely to have other co-morbid medical illnesses. Medical co-morbidities seen in those veterans with CHC contribute to impaired PCS but not MCS. Anti-HCV(+) status negatively affects HRQoL, particularly MCS, independently of medical and psychiatric co-morbidities.

  View details for DOI 10.1007/s11136-008-9344-3

  View details for Web of Science ID 000256522600007

  View details for PubMedID 18427949

• Evaluation of an architecture for intelligent query and exploration of time-oriented clinical data ARTIFICIAL INTELLIGENCE IN MEDICINE Martins, S. B., Shahar, Y., Goren-Bar, D., Galperin, M., Kaizer, H., Basso, L. V., McNaughton, D., Goldstein, M. K. 2008; 43 (1): 17-34

  Abstract

  Evaluate KNAVE-II, a knowledge-based framework for visualization, interpretation, and exploration of longitudinal clinical data, clinical concepts and patterns. KNAVE-II mediates queries to a distributed temporal-abstraction architecture (IDAN), which uses a knowledge-based problem-solving method specializing in on-the-fly computation of clinical queries.A two-phase, balanced cross-over study to compare efficiency and satisfaction of a group of clinicians when answering queries of variable complexity about time-oriented clinical data, typical for oncology protocols, using KNAVE-II, versus standard methods: both paper charts and a popular electronic spreadsheet (ESS) in Phase I; an ESS in Phase II. The measurements included the time required to answer and the correctness of answer for each query and each complexity category, and for all queries, assessed versus a predetermined gold standard set by a domain expert. User satisfaction was assessed by the Standard Usability Score (SUS) tool-specific questionnaire and by a "Usability of Tool Comparison" comparative questionnaire developed for this study.In both evaluations, subjects answered higher-complexity queries significantly faster using KNAVE-II than when using paper charts or an ESS up to a mean of 255 s difference per query versus the ESS for hard queries (p=0.0003) in the second evaluation. Average correctness scores when using KNAVE-II versus paper charts, in the first phase, and the ESS, in the second phase, were significantly higher over all queries. In the second evaluation, 91.6% (110/120) of all of the questions asked within queries of all levels produced correct answers using KNAVE-II, opposed to only 57.5% (69/120) using the ESS (p<0.0001). User satisfaction with KNAVE-II was significantly superior compared to using either a paper chart or the ESS (p=0.006). Clinicians ranked KNAVE-II superior to both paper and the ESS.An evaluation of the functionality and usability of KNAVE-II and its supporting knowledge-based temporal-mediation architecture has produced highly encouraging results regarding saving of physician time, enhancement of accuracy of clinical assessment, and user satisfaction.

  View details for DOI 10.1016/j.artmed.2008.03.006

  View details for Web of Science ID 000256327300002

  View details for PubMedID 18442899

  View details for PubMedCentralID PMC2853917

• Implementing Effective Hypertension Quality Improvement Strategies: Barriers and Potential Solutions JOURNAL OF CLINICAL HYPERTENSION Walsh, J. M., Sundaram, V., McDonald, K., Owens, D. K., Goldstein, M. K. 2008; 10 (4): 311-316

  Abstract

  Many quality improvement strategies have focused on improving blood pressure control, and these strategies can target the patient, the provider, and/or the system. Strategies that seem to have the biggest effect on blood pressure outcomes are team change, patient education, facilitated relay of clinical information, and promotion of self-management. Barriers to effective blood pressure control can affect the patient, the physician, the system, and/or "cues to action."We review the barriers to achieving blood pressure control and describe current and potential creative strategies for optimizing blood pressure control. These include home-based disease management, combined patient and provider education, and automatic decision support systems. Future research must address which components of quality improvement interventions are most successful in achieving blood pressure control.

  View details for Web of Science ID 000261099600008

  View details for PubMedID 18401229

• The effect of continuous positive airway pressure therapy on falls in elderly patients with obstructive sleep apnea: Methodology of a phone survey study Javaheri, A., Goldstein, M., Sarinas, P. BLACKWELL PUBLISHING. 2008: S117

  View details for Web of Science ID 000254840300342

• New methods for performance measurement in hypertension: Results from an expert panel Steinman, M., Goldstein, M. K. SPRINGER. 2008: 351

  View details for Web of Science ID 000254237100713

• Information technology to support improved care for chronic illness State of the Art Conference on Managing Complexity in Chronic Care Young, A. S., Chaney, E., Shoai, R., Bonner, L., Cohen, A. N., Doebbeling, B., Dorr, D., Goldstein, M. K., Kerr, E., Nichol, P., Perrin, R. SPRINGER. 2007: 425–430

  Abstract

  In populations with chronic illness, outcomes improve with the use of care models that integrate clinical information, evidence-based treatments, and proactive management of care. Health information technology is believed to be critical for efficient implementation of these chronic care models. Health care organizations have implemented information technologies, such as electronic medical records, to varying degrees. However, considerable uncertainty remains regarding the relative impact of specific informatics technologies on chronic illness care.To summarize knowledge and increase expert consensus regarding informatics components that support improvement in chronic illness care.A systematic review of the literature was performed. "Use case" models were then developed, based on the literature review, and guidance from clinicians and national quality improvement projects. A national expert panel process was conducted to increase consensus regarding information system components that can be used to improve chronic illness care.The expert panel agreed that informatics should be patient-centered, focused on improving outcomes, and provide support for illness self-management. They concurred that outcomes should be routinely assessed, provided to clinicians during the clinical encounter, and used for population-based care management. It was recommended that interactive, sequential, disorder-specific treatment pathways be implemented to quickly provide clinicians with patient clinical status, treatment history, and decision support.Specific informatics strategies have the potential to improve care for chronic illness. Software to implement these strategies should be developed, and rigorously evaluated within the context of organizational efforts to improve care.

  View details for DOI 10.1007/s116O6-007-0303-4

  View details for Web of Science ID 000251543500010

  View details for PubMedID 18026812

  View details for PubMedCentralID PMC2150622

• Runtime application of Hybrid-Asbru clinical guidelines JOURNAL OF BIOMEDICAL INFORMATICS Young, O., Shahar, Y., Liel, Y., Lunenfeld, E., Bar, G., Shalom, E., Martins, S. B., Vaszar, L. T., Marom, T., Goldstein, M. K. 2007; 40 (5): 507-526

  Abstract

  Clinical guidelines are a major tool in improving the quality of medical care. However, to support the automation of guideline-based care, several requirements must be filled, such as specification of the guidelines in a machine-interpretable format and a connection to an Electronic Patient Record (EPR). For several different reasons, it is beneficial to convert free-text guidelines gradually, through several intermediate representations, to a machine-interpretable format. It is also realistic to consider the case when an EPR is unavailable. We propose an innovative approach to the runtime application of intermediate-represented Hybrid-Asbru guidelines, with or without an available EPR. The new approach capitalizes on our extensive work on developing the Digital electronic Guideline Library (DeGeL) framework. The new approach was implemented as the Spock system. For evaluation, three guidelines were specified in an intermediate format and were applied to a set of simulated patient records designed to cover prototypical cases. In all cases, the Spock system produced the expected output, and did not produce an unexpected one. Thus, we have demonstrated the capability of the Spock system to apply guidelines encoded in the Hybrid-Asbru intermediate representation, when an EPR is not available.

  View details for DOI 10.1016/j.jbi.2006.12.004

  View details for Web of Science ID 000249876200006

  View details for PubMedID 17276145

• Association of antihypertensive Therapy and diastolic Hypotension in chronic kidney disease HYPERTENSION Peralta, C. A., Shlipak, M. G., Wassel-Fyr, C., Bosworth, H., Hoffman, B., Martins, S., Oddone, E., Goldstein, M. K. 2007; 50 (3): 474-480

  Abstract

  The extent to which chronic kidney disease (CKD) affects achievement of blood pressure targets is not comprehensively understood. We evaluated the effects of CKD (estimated glomerular filtration rate: <60 mL/min per 1.73 m(2)) on achievement of blood pressure control (nondiabetic: <140/90 mm Hg; diabetic: <130/85 mm Hg) using data from the Guidelines for Drug Therapy of Hypertension Trial. This 15-month study obtained outpatient blood pressures from 3 Veteran's Affairs institutions. Among 9985 subjects with hypertension, we evaluated the association of CKD with achieved control and antihypertensive medication use. We also explored the association between the number of antihypertensives and systolic, diastolic, and pulse pressure. After 15 months, 41% of participants met blood pressure targets. CKD was not associated with control (adjusted odds ratio: 1.04; 95% CI: 0.93 to 1.15). However, CKD was associated with higher odds of use of >or=3 medications among nondiabetic subjects (odds ratio: 1.46; 95% CI: 1.25 to 1.71) and diabetic subjects (odds ratio: 1.40; 95% CI: 1.17 to 1.66). A significant interaction was observed between CKD and the number of antihypertensives as determinants of diastolic and pulse pressures. Among non-CKD participants, a greater number of antihypertensives (0 compared with 4) was associated with wider pulse pressure (Delta5.2 mm Hg; P<0.001), mainly because of higher systolic pressures (Delta3.6 mm Hg; P=0.001). Among participants with CKD, although greater numbers of antihypertensives were associated with even wider pulse pressures (Delta8.3 mm Hg; P<0.001), this was primarily because of lower diastolic pressures (Delta4.8 mm Hg; P<0.01). Among participants with CKD, greater use of antihypertensives was associated with lower diastolic pressures. Given recent evidence suggesting adverse effects of diastolic hypotension, these results suggest potential risks in patients with CKD from aggressive attempts to control systolic blood pressure.

  View details for DOI 10.1161/HYPERTENSIONAHA.107.088088

  View details for Web of Science ID 000248945100010

  View details for PubMedID 17664397

• Clinical reminders attached to echocardiography reports of patients with reduced left ventricular ejection fraction increase use of beta-blockers - A randomized trial CIRCULATION Heidenreich, P. A., Gholami, P., Sahay, A., Massie, B., Goldstein, M. K. 2007; 115 (22): 2829-2834

  Abstract

  Although beta-blockers are known to prolong survival for patients with reduced left ventricular ejection fraction, they are often underused. We hypothesized that a reminder attached to the echocardiography report would increase the use of beta-blockers for patients with reduced left ventricular ejection fraction.We randomized 1546 consecutive patients with a left ventricular ejection fraction <45% found on echocardiography at 1 of 3 laboratories to a reminder for use of beta-blockers or no reminder. Patients were excluded from analysis if they died within 30 days of randomization (n=89), did not receive medications through the Veterans Affairs system after 30 days (n=180), or underwent echocardiography at >1 laboratory (n=6). The primary outcome was a prescription for an oral beta-blocker between 1 and 9 months after randomization. The mean age of the 1271 included patients was 69 years; 60% had a history of heart failure, and 51% were receiving treatment with beta-blockers at the time of echocardiography. More patients randomized to the reminder had a subsequent beta-blocker prescription (74%, 458 of 621) compared with those randomized to no reminder (66%, 428 of 650; P=0.002). The effect of the reminder was not significantly different for subgroups based on patient location (inpatient versus outpatient) or prior use of beta-blockers.A reminder attached to the echocardiography report increased the use of beta-blockers in patients with depressed left ventricular systolic function.

  View details for DOI 10.1161/CIRCULATIONAHA.106.684753

  View details for Web of Science ID 000246972600008

  View details for PubMedID 17515459

• Hypertension Intervention Nurse Telemedicine Study (HINTS): Testing a multifactorial tailored behavioral/educational and a medication management intervention for blood pressure control AMERICAN HEART JOURNAL Bosworth, H. B., Olsen, M. K., McCant, F., Harrelson, M., Gentry, P., Rose, C., Goldstein, M. K., Hoffman, B. B., Powers, B., Oddone, E. Z. 2007; 153 (6): 918-924

  Abstract

  Only 31% of Americans with hypertension have their blood pressure (BP) under effective control. We describe a study that tests 3 different interventions in a randomized controlled trial using home BP telemedicine monitoring.A sample of hypertensive patients with poor BP control at baseline (N = 600) are randomized to 1 of 4 arms: (1) control group--a group of hypertensive patients who receive usual care; (2) nurse-administered tailored behavioral intervention; (3) nurse-administered medication management according to a hypertension decision support system; (4) combination of the 2 interventions. The interventions are triggered based on home BP values transmitted via telemonitoring devices over standard telephone lines. The tailored behavioral intervention involves promoting adherence with medication and health behaviors. Patients randomized to the medication management or the combined arm have their hypertension regimen changed by the study team using a validated hypertension decision support system based on evidence-based hypertension treatment guidelines and individualized to patients' comorbid illnesses. The primary outcome is BP control: < or = 140/90 mm Hg (nondiabetic) and < or = 130/80 mm Hg (diabetics) measured at 6-month intervals over 18 months (4 total measurements).Given the increasing prevalence of hypertension and our inability to achieve adequate BP control using traditional models of care, testing novel interventions in patients' homes may improve access, quality, and outcomes.

  View details for DOI 10.1016/j.ahj.2007.03.004

  View details for Web of Science ID 000247540600005

  View details for PubMedID 17540191

• Does cancer risk affect health-related quality of life in patients with Barrett's esophagus? Digestive Disease Week Meeting/106th Annual Meeting of the American-Gastroenterological-Association Gerson, L. B., Ullah, N., Hastie, T., Goldstein, M. K. MOSBY-ELSEVIER. 2007: 16–25

  Abstract

  Health-related quality of life is decreased in patients with GERD and Barrett's esophagus (BE).To determine whether time-tradeoff (TTO) values would differ in patients with BE when patients were asked to trade away the potential risk of esophageal adenocarcinoma rather than chronic heartburn symptoms.A prospective clinical trial.Subjects with biopsy-proven BE.Custom-designed computer program to elicit health-state utility values, quality of life in reflux and dyspepsia (QOLRAD), and Medical Outcomes Survey short form-36 surveys.TTO utility values for the annual cancer-risk-associated current health state and for hypothetical scenarios of dysplasia and esophageal cancer.We studied 60 patients in the cancer-risk cohort (57 men, 92% veteran; mean age [standard deviation; SD], 65 years [11 years], mean GERD duration 17 years [12 years]). The heartburn cohort included 40 patients with GERD and BE with TTO values derived for GERD symptoms. The mean (SD) utility for nondysplastic BE was 0.91 (0.13) compared with 0.90 (0.12) for the heartburn cohort (P = .7). The mean utility values were significantly lower for scenarios of low-grade dysplasia (0.85 [0.12], P = .02) and high-grade dysplasia (0.77 [0.14], P < .005). The mean TTO was 0.67 (0.19) for the scenario of esophageal cancer. There was no correlation between the utility scores and the disease-specific survey scores.TTO values were hypothetical for states of dysplasia and cancer.TTO utility values based on heartburn symptoms or annual risk of cancer in patients with nondysplastic BE are roughly equivalent. However, TTO utility values are significantly lower for health states with increasing cancer risks.

  View details for DOI 10.1016/j.gie.2006.05.018

  View details for Web of Science ID 000243361000005

  View details for PubMedID 17185075

• Quality improvement strategies for hypertension management - A systematic review MEDICAL CARE Walsh, J. M., McDonald, K. M., Shojania, K. G., Sundaram, V., Nayak, S., Lewis, R., Owens, D. K., Goldstein, M. K. 2006; 44 (7): 646-657

  Abstract

  Care remains suboptimal for many patients with hypertension.The purpose of this study was to assess the effectiveness of quality improvement (QI) strategies in lowering blood pressure.MEDLINE, Cochrane databases, and article bibliographies were searched for this study.Trials, controlled before-after studies, and interrupted time series evaluating QI interventions targeting hypertension control and reporting blood pressure outcomes were studied.Two reviewers abstracted data and classified QI strategies into categories: provider education, provider reminders, facilitated relay of clinical information, patient education, self-management, patient reminders, audit and feedback, team change, or financial incentives were extracted.Forty-four articles reporting 57 comparisons underwent quantitative analysis. Patients in the intervention groups experienced median reductions in systolic blood pressure (SBP) and diastolic blood pressure (DBP) that were 4.5 mm Hg (interquartile range [IQR]: 1.5 to 11.0) and 2.1 mm Hg (IQR: -0.2 to 5.0) greater than observed for control patients. Median increases in the percentage of individuals achieving target goals for SBP and DBP were 16.2% (IQR: 10.3 to 32.2) and 6.0% (IQR: 1.5 to 17.5). Interventions that included team change as a QI strategy were associated with the largest reductions in blood pressure outcomes. All team change studies included assignment of some responsibilities to a health professional other than the patient's physician.Not all QI strategies have been assessed equally, which limits the power to compare differences in effects between strategies.QI strategies are associated with improved hypertension control. A focus on hypertension by someone in addition to the patient's physician was associated with substantial improvement. Future research should examine the contributions of individual QI strategies and their relative costs.

  View details for Web of Science ID 000238806300006

  View details for PubMedID 16799359

• The impact of chronic hepatitis C and comorbid psychiatric illnesses on health-related quality of life JOURNAL OF CLINICAL GASTROENTEROLOGY Lim, J. K., Cronkite, R., Goldstein, M. K., Cheung, R. C. 2006; 40 (6): 528-534

  Abstract

  To determine the relative impact of chronic hepatitis C (CHC) and comorbid psychiatric illness on the health-related quality of life (HRQoL).Psychiatric conditions are more common among patients with CHC but their relative influence on HRQoL is not well understood.We identified 864 veterans who had previously completed a veteran-specific HRQoL questionnaire (SF-36V) as part of the 1999 VA Large Health Survey with known HCV antibody (anti-HCV) status before the survey. For 201 anti-HCV(+) and 663 anti-HCV(-) patients, we compared the HRQoL status and the prevalence of 6 major psychiatric diagnoses. We conducted multiple regression analyses to measure the effect of anti-HCV status and psychiatric comorbidity.Compared with the anti-HCV(-) group, anti-HCV(+) veterans were more likely to have alcohol dependence (P<0.001), depression (P=0.01), or posttraumatic stress disorder (PTSD) (P<0.004). The anti-HCV(+) group also reported lower HRQoL on 4 of the 8 SF-36V subscales (P<0.01) and the mental component summary scale (P<0.001). Even after adjusting for demographic variables and comorbid psychiatric illness, anti-HCV(+) patients reported a significantly lower mental component summary score (P<0.01) than did anti-HCV(-) patients. Multiple regression analysis found that depression and PTSD predicted lower HRQoL scores for all 8 HRQoL subscales (P<0.01) and both the physical (P<0.001) and mental component (P<0.03) summary scales independent of anti-HCV status.The HRQoL is significantly impaired in veterans with CHC, particularly the mental health components of HRQoL. In contrast, comorbid depression and PTSD are associated with both lower physical and mental components of HRQol, independent of CHC.

  View details for Web of Science ID 000239136700012

  View details for PubMedID 16825936

• Task analysis of writing hospital admission orders: evidence of a problem-based approach. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Johnson, C. D., Zeiger, R. F., Das, A. K., Goldstein, M. K. 2006: 389-393

  Abstract

  As part of a broader project to improve the usability of computerized physician order entry (CPOE)systems, we set out to study the cognitive tasks physicians undertake to write "admission orders" when admitting a patient to the hospital. In particular, we evaluate the hypothesis that physicians' mental model of diagnostic and therapeutic planning is problem based, whereas both paper-based ordering and CPOE are typically organized around functional categories of orders such as those reflected in the mnemonic ADCVAANDIML. A task analysis was performed which included think-aloud observations of physicians writing orders in clinical care settings and for fictional case-scenarios, as well as a semistructured questionnaire. Our work finds core tasks of admitting a patient to hospital and conflicts between physicians' mental model and traditional ordering systems. Based on our study, we suggest improvements to traditional CPOE systems.

  View details for PubMedID 17238369

• Offline testing of the ATHENA Hypertension decision support system knowledge base to improve the accuracy of recommendations. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Martins, S. B., Lai, S., Tu, S., Shankar, R., Hastings, S. N., Hoffman, B. B., Dipilla, N., Goldstein, M. K. 2006: 539-543

  Abstract

  ATHENA-HTN is a clinical decision support system (CDSS) that delivers guideline-based patient-specific recommendations about hypertension management at the time of clinical decision-making. The ATHENA-HTN knowledge is stored in a knowledge-base (KB). Changes in best-practice recommendations require updates to the KB. We describe a method of offline testing to evaluate the accuracy of recommendations generated from the KB. A physician reviewed 100 test cases and made drug recommendations based on guidelines and the "Rules" (descriptions of encoded knowledge). These drug recommendations were compared to those generated by ATHENA-HTN. Nineteen drug-recommendation discrepancies were identified: ATHENA-HTN was more complete in generating recommendations (15); ambiguities in the Rules misled the physician (3); and content in the Rules was not encoded (1). Three new boundaries were identified. Three updates were made to the KB based on the results. The offline testing method was successful in identifying areas for KB improvement and led to improved accuracy of guideline-based recommendations.

  View details for PubMedID 17238399

• Knowledge-based method for building patient decision-analytic tools. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Das, A. K., Ahmed, B. A., Garten, Y., Robin, J. I., Goldstein, M. K. 2006: 175-179

  Abstract

  Numerous health decision aids (HDAs) have been developed to increase the participation of patients in shared decision-making, but many have limited accessibility and narrow applicability in clinical care. In the Health e-Decision project, we address these limitations in our work on building general HDAs targeted for older adults. Our approach uses a decision-support software architecture that enables principled methods for HDAs. We have formalized a novel knowledge-based decision model (KBDM), using Protégé OWL, that developers and clinicians can instantiate to tailor the components of the architecture for a particular health problem. In this paper, we present the methods used in the architecture and the knowledgebase design; the latter encompasses influence-diagram concepts, specific health problems, health outcome states, and probabilistic relationships. We discuss how this approach improves upon prior HDA methods. We also show that our use of computer-interpretable knowledge provides a structured, customizable means of enabling patient-centered decision support.

  View details for PubMedID 17238326

• Racial differences in blood pressure control: potential explanatory factors. American journal of medicine Bosworth, H. B., Dudley, T., Olsen, M. K., Voils, C. I., Powers, B., Goldstein, M. K., Oddone, E. Z. 2006; 119 (1): 70 e9-15

  Abstract

  Poor blood pressure control remains a common problem that contributes to significant cardiovascular morbidity and mortality, particularly among African Americans. We explored antihypertensive medication adherence and other factors that may explain racial differences in blood pressure control.Baseline data were obtained from the Veteran's Study to Improve The Control of Hypertension, a randomized controlled trial designed to improve blood pressure control. Clinical, demographic, and psychosocial factors relating to blood pressure control were examined.A total of 569 patients who were African American (41%) or white (59%) were enrolled in the study. African Americans were more likely to have inadequate baseline blood pressure control than whites (63% vs 50%; odds ratio = 1.70; 95% confidence interval [CI] 1.20-2.41). Among 20 factors related to blood pressure control, African Americans also had a higher odds ratio of being nonadherent to their medication, being more functionally illiterate, and having a family member with hypertension compared with whites. Compared with whites, African Americans also were more likely to perceive high blood pressure as serious and to experience the side effect of increased urination compared with whites. Adjusting for these differences reduced the odds ratio of African Americans having adequate blood pressure control to 1.59 (95% confidence interval 1.09-2.29).In this sample of hypertensive patients who have good access to health care and medication benefits, African Americans continued to have lower levels of blood pressure control despite considering more than 20 factors related to blood pressure control. Interventions designed to improve medication adherence need to take race into account. Patients' self-reports of failure to take medications provide an opportunity for clinicians to explore reasons for medication nonadherence, thereby improving adherence and potentially blood pressure control.

  View details for PubMedID 16431192

• Deciphering the code AHRQ WebM&M [serial online]. Goldstein MK 2006
• Use of declarative statements in creating and maintaining computer-interpretable knowledge bases for guideline-based care. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Tu, S. W., Hrabak, K. M., Campbell, J. R., Glasgow, J., Nyman, M. A., McClure, R., McClay, J., Abarbanel, R., Mansfield, J. G., Martins, S. M., Goldstein, M. K., Musen, M. A. 2006: 784-788

  Abstract

  Developing computer-interpretable clinical practice guidelines (CPGs) to provide decision support for guideline-based care is an extremely labor-intensive task. In the EON/ATHENA and SAGE projects, we formulated substantial portions of CPGs as computable statements that express declarative relationships between patient conditions and possible interventions. We developed query and expression languages that allow a decision-support system (DSS) to evaluate these statements in specific patient situations. A DSS can use these guideline statements in multiple ways, including: (1) as inputs for determining preferred alternatives in decision-making, and (2) as a way to provide targeted commentaries in the clinical information system. The use of these declarative statements significantly reduces the modeling expertise and effort required to create and maintain computer-interpretable knowledge bases for decision-support purpose. We discuss possible implications for sharing of such knowledge bases.

  View details for PubMedID 17238448

• The effect of geriatrics evaluation and management on nursing home use and health care costs - Results from a randomized trial MEDICAL CARE Phibbs, C. S., Holty, J. E., Goldstein, M. K., Garber, A. M., Wang, Y. J., Feussner, J. R., Cohen, H. J. 2006; 44 (1): 91-95

  Abstract

  The Geriatric Evaluation and Management study was developed to assess the impact of a comprehensive geriatric assessment service on the care of the elderly.We sought to evaluate the cost and clinical impact of inpatient units and outpatient clinics for geriatric evaluation and management.We undertook a prospective, randomized, controlled trial using a 2x2 factorial design, with 1-year follow-up.A total of 1388 participants hospitalized on either a medical or surgical ward at 11 participating Veterans Affairs medical centers were randomized to receive either inpatient geriatric unit (GEMU) or usual inpatient care (UCIP), followed by either outpatient care from a geriatric clinic (GEMC) versus usual outpatient care (UCOP).We measured health care utilization and costs.Patients assigned to the GEMU had a significantly decreased rate of nursing home placement (odds ratio=0.65; P=0.001). Neither the GEMU nor GEMC had any statistically significant improvement effects on survival and only modest effects on health status. There were statistically insignificant mean cost savings of $1027 (P=0.29) per patient for the GEMU and $1665 (P=0.69) per patient for the GEMC.Inpatient or outpatient geriatric evaluation and management units didn't increase the costs of care. Although there was no effect on survival and only modest effects on SF-36 scores at 1-year follow-up, there was a statistically significant reduction in nursing home admissions for patients treated in the GEMU.

  View details for Web of Science ID 000234342600013

  View details for PubMedID 16365618

• Identifying barriers to hypertension guideline adherence using clinician feedback at the point of care. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Lin, N. D., Martins, S. B., Chan, A. S., Coleman, R. W., Bosworth, H. B., Oddone, E. Z., Shankar, R. D., Musen, M. A., Hoffman, B. B., Goldstein, M. K. 2006: 494-498

  Abstract

  Factors contributing to low adherence to clinical guidelines by clinicians are not well understood. The user interface of ATHENA-HTN, a guideline-based decision support system (DSS) for hypertension, presents a novel opportunity to collect clinician feedback on recommendations displayed at the point of care. We analyzed feedback from 46 clinicians who received ATHENA advisories as part of a 15-month randomized trial to identify potential reasons clinicians may not intensify hypertension therapy when it is recommended. Among the 368 visits for which feedback was provided, clinicians commonly reported they did not follow recommendations because: recorded blood pressure was not representative of the patient's typical blood pressure; hypertension was not a clinical priority for the visit; or patients were nonadherent to medications. For many visits, current quality-assurance algorithms may incorrectly identify clinically appropriate decisions as guideline nonadherent due to incomplete capture of relevant information. We present recommendations for how automated DSSs may help identify "apparent" barriers and better target decision support.

  View details for PubMedID 17238390

• Improving adherence to guidelines for hypertension drug prescribing: Cluster-randomized controlled trial of general versus patient-specific recommendations AMERICAN JOURNAL OF MANAGED CARE Goldstein, M. K., Lavori, P., Coleman, R., Advani, A., Hoffman, B. B. 2005; 11 (11): 677-685

  Abstract

  To determine whether an intervention focusing clinician attention on drug choice for hypertension treatment improves concordance between drug regimens and guidelines.Cluster-randomized controlled trial comparing an individualized intervention with a general guideline implementation in geographically diverse primary care clinics of a university-affiliated Department of Veterans Affairs healthcare system.Participants were 36 attending physicians and nurse practitioners (16 in the general group and 20 in the individualized group), with findings based on 4500 hypertensive patients. A general guideline implementation for all clinicians, including education about guideline-based drug recommendations and goals for adequacy of blood pressure control, was compared with addition of a printed individualized advisory sent to clinicians at each patient visit, indicating whether or not the patient's antihypertensive drug regimen was guideline concordant. We measured change from baseline to end point in the proportion of clinicians' patients whose drug therapy was guideline concordant.The individualized intervention resulted in an improvement in guideline concordance more than twice that observed for the general intervention (10.9% vs 3.8%, t = 2.796, P = .008). Bootstrap analysis showed that being in the individualized group increased the odds of concordance 1.5-fold (P = .025). The proportion of patients with adequate blood pressure control increased within each study group; however, the difference between groups was not significant.An individualized advisory regarding drug therapy for hypertension given to the clinician at each patient visit was more effective in changing clinician prescribing behavior than implementation of a general guideline.

  View details for Web of Science ID 000233132300002

  View details for PubMedID 16268751

• ACE inhibitor reminders attached to echocardiography reports of patients with reduced left ventricular ejection fraction AMERICAN JOURNAL OF MEDICINE Heidenreich, P. A., Chacko, M., Goidstein, M. K., Atwood, J. E. 2005; 118 (9): 1034-1037

  View details for DOI 10.1016/j.amjmed.2004.12.028

  View details for Web of Science ID 000231807300018

  View details for PubMedID 16164891

• The veterans' study to improve the control of hypertension (V-STITCH): Design and methodology CONTEMPORARY CLINICAL TRIALS Bosworth, H. B., Olsen, M. K., Goldstein, M. K., Orr, M., Dudley, T., McCant, F., Gentry, P., Oddone, E. Z. 2005; 26 (2): 155-168

  Abstract

  Among the 60 million Americans with hypertension, only approximately 31% have their blood pressure (BP) under control (<140/90 mm Hg). Despite the damaging impact of hypertension and the availability of evidence-based target values for BP, interventions to improve BP control have had limited success.A randomized controlled health services intervention trial with a split-plot design is being conducted to improve BP control. This 4-year trial evaluates both a patient and a provider intervention in a primary care setting among diagnosed hypertensive veterans.In a cluster-randomization, 30 primary care providers in the Durham VAMC Primary Care Clinic were randomly assigned to receive the provider intervention or control. The provider intervention is a patient-specific electronically generated hypertension decision support system (DSS) delivering guideline-based recommendations to the provider at each patient's visit, designed to improve guideline-concordant therapy. For these providers, a sample of their hypertensive patients (n=588) was randomly assigned to receive a telephone-administered patient intervention or usual care. The patient intervention incorporates patients' need assessments and involves tailored behavioral and education modules to promote medication adherence and improve specific health behaviors. All modules are delivered over the telephone bi-monthly for 24 months. In this trial, the primary outcome is the proportion of patients who achieve a BP < or =140/90 mm Hg at each outpatient clinic visit over 24 months.Despite the known risk of poor BP control, a majority of adults still do not have their BP controlled. This study is an important step in testing the effectiveness of a patient and provider intervention to improve BP control among veterans in the primary care setting.

  View details for DOI 10.1016/j.cct.2004.12.006

  View details for Web of Science ID 000229144700005

  View details for PubMedID 15837438

• Invariance and inconsistency in utility ratings 22nd Annual Meeting of the Society-for-Medical-Decision-Making Bravata, D. M., Nelson, L. M., Garber, A. M., Goldstein, M. K. SAGE PUBLICATIONS INC. 2005: 158–67

  Abstract

  To assess utilities of composite health states for dependence in activities of daily living (ADLs) for invariance (i.e., when subjects provide a utility of 1 for all health states) and order inconsistency (i.e., when subjects order their utilities such that their utility for a combination of ADL dependencies is greater than their utility for any subset of the combination).Each of the 400 subjects, age 65 y and older, enrolled in one of several regional medical centers of the Kaiser Permanente Medical Care Program of Northern California and provided standard-gamble utilities for single ADL dependencies (e.g., bathing, dressing, continence) and for dependence in 8 other combinations of ADL dependencies. For order-inconsistent responses, the authors calculated the maximum magnitude of inconsistency as the maximum difference between the utility for the combined ADL dependence health state and that of its inconsistent subset.A total of 76 subjects (19%) gave a utility of 1.0 for all health states presented to them; 19 (5%) gave the same utility other than 1.0 for all health states; 130 (33%) gave at least 1 utility < 1.0 and had no order inconsistencies; and 175 (44%) had at least 1 order inconsistency. Invariance was associated with a Mini-Mental Status Examination score < 28.6 (P = 0.01), with education < 12 y (P = 0.004), with race/ethnicity other than non-Hispanic White/Caucasian (P = 0.001), and with shorter time spent on the utility elicitation task (P < 0.0001). Among the inconsistent subjects, 69% had a maximal magnitude of inconsistency that was within 1 standard deviation of the mean utilities. The maximal magnitude of inconsistency was associated with longer time spent on the elicitation task (P < 0.0001) and race/ethnicity other than non-Hispanic White/Caucasian (P = 0.005). The mean (s) utility for dependence in continence among consistent subjects who were not invariant (0.88 [0.24]) was higher than among inconsistent subjects (0.80 [0.27]; P = 0.01).Invariance and order inconsistencies in utility ratings for complex health states occur frequently. Utilities of consistent subjects may differ from those of inconsistent subjects. Utility assessments should attempt to measure and report these patterns.

  View details for DOI 10.1177/0272989X05275399

  View details for Web of Science ID 000228014000004

  View details for PubMedID 15800300

• Patient-derived health state utilities for gastroesophageal reflux disease AMERICAN JOURNAL OF GASTROENTEROLOGY Gerson, L. B., Ullah, N., Hastie, T., Triadafilopoulos, G., Goldstein, M. 2005; 100 (3): 524-533

  Abstract

  Gastroesophageal reflux disease is a chronic disease that adversely affects health-related quality of life. The purpose of this study was to derive health state utilities for patients with chronic heartburn symptoms.We used a custom-designed computer program in order to elicit utilities with the time-tradeoff and standard-gamble techniques. Patients with chronic (more than 6 months) symptoms of gastroesophageal reflux disease entered the study. Two interviews were performed in random sequence either initially on medications for heartburn that adequately controlled symptoms, or off of medications for 1 wk while the patient was symptomatic. We also collected data using visual-analog scales, quality of life in reflux and dyspepsia (QOLRAD), and Gastrointestinal Symptom Rating Scale (GSRS) scores.We invited 222 patients to participate; 158 (71%) patients (129 men, 29 women) completed the study. Barrett's esophagus was present in 40 (25%), erosive disease in 17 (11%), and 118 (74%) had comorbid conditions. The mean (+/-SD) utility ratings were 0.94 +/- 0.09 on medical therapy and 0.90 +/- 0.12 off medications for patients with reflux alone using time tradeoff (p= 0.004), and 0.94 +/- 8.0 both on and off of antireflux medications with standard-gamble assessment (p= 0.96). Mean time-tradeoff scores were also significantly lower off of medications for patients with other comorbid conditions (p= 0.002). There was no significant difference between mean utility scores for patients with or without Barrett's esophagus or erosive disease.Gastroesophageal reflux disease adversely affects health-related quality of life. Time-tradeoff utility for patients with reflux disease is substantially higher when patients are on medication than off medications.

  View details for DOI 10.1111/j.1572-0241.40588.x

  View details for Web of Science ID 000227697900005

  View details for PubMedID 15743346

• Leveraging point-of-care clinician feedback to study barriers to guideline adherence. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Chan, A. S., Shankar, R. D., Coleman, R. W., Matins, S. B., Hoffman, B. B., Goldstein, M. K. 2005: 915-?

  Abstract

  Studies of barriers to guideline adherence have generally surveyed clinicians temporally remote from the clinical scenario in which recommendations were delivered, potentially adversely biasing clinician observations. The user interface of ATHENA DSS, a guideline-based decision support system for hypertension, includes a point-of-care feedback window that accepts clinician-user comments during the display of recommendations. Analysis of this feedback has revealed a number of intriguing patient, provider, and technical barriers to adherence collected during real-time system use.

  View details for PubMedID 16779202

• Post-Fielding Surveillance of a Guideline-Based Decision Support System Advanced in Patient Safety: From Research to Implementation Chan A.S., Martins SB, Coleman R.W., Bosworth H.B., Oddone E.Z., Shlipak M.G., Tu S.W., Musen M.A., Hoffman B.B., Goldstein M.K. 2005; 1: 331-339
• Multimedia quality of life assessment: advances with FLAIR. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Sims, T. L., Garber, A. M., Miller, D. E., Mahlow, P. T., Bravata, D. M., Goldstein, M. K. 2005: 694-698

  Abstract

  Assessing impact of functional dependency on quality of life (QOL) among older adults can provide an in-depth understanding of health preferences. Utilities as a measure of preferences are necessary in conducting cost-effectiveness evaluations of healthcare interventions designed to improve overall QOL. We describe further development of a multimedia utility elicitation instrument that is highly portable and easily accessible. An earlier version, FLAIR1, introduced features designed for older adult, computer inexperienced users. FLAIR2 includes modifications such as migration to a web-based platform, consistency checks, audio/visual updates, and more response methods. As compared with FLAIR1, more FLAIR2 respondents (n=318) preferred using the computer and found the computer program to be enjoyable, easy to use, and easily understood. There were also fewer inconsistencies among FLAIR2 respondents. FLAIR2 enhancements have increased portability, minimized invariance and inconsistency, and produced a more user friendly design.

  View details for PubMedID 16779129

• Clinical Decision-Support for Managing Chronic Disease in Primary Care: ATHENA DSS FORUM Goldstein M.K., Oddone E, Hoffman BB 2005: 4
• Behavioral interventions and cost-effectiveness analysis PREVENTIVE MEDICINE Wagner, T. H., Goldstein, M. K. 2004; 39 (6): 1208-1214

  Abstract

  Behavioral health interventions are often gauged with a dichotomous outcome, "success" or "failure." Hidden by this dichotomy is a series of behavior changes that can be followed with the Transtheoretical Model (stages of change). There has been little consideration, however, about whether this information can and should be used in cost-effectiveness analysis. We review the stages of change model and its applications to behavioral health interventions. We then discuss analytical methods for including stages of change, or similar behavior change models, in cost-effectiveness analysis (CEA). This is typically not done but it may be critical for study design and for interpreting CEA results.

  View details for DOI 10.1016/j.ypmed.2004.04.052

  View details for Web of Science ID 000225496200021

  View details for PubMedID 15539057

• Clinician awareness of adherence to hypertension guidelines AMERICAN JOURNAL OF MEDICINE Steinman, M. A., Fischer, M. A., Shlipak, M. G., Bosworth, H. B., Oddone, E. Z., Hoffman, B. B., Goldstein, M. K. 2004; 117 (10): 747-754

  Abstract

  Little is known about how well clinicians are aware of their own adherence to clinical guidelines, an important indicator of quality. We compared clinicians' beliefs about their adherence to hypertension guidelines with data on their actual performance.We surveyed 139 primary care clinicians at three Veterans Affairs medical centers, asking them to assess their own adherence to hypertension guidelines. We then extracted data from the centers' clinical databases on guideline-concordant medication use and blood pressure control for patients cared for by these providers during a 6-month period. Data were collected for patients with hypertension and diabetes, hypertension and coronary disease, or hypertension with neither of these comorbid conditions.Eighty-six clinicians (62%) completed the survey. Each clinician saw a median of 94 patients with hypertension (mean age, 65 years). Patients were treated with an average of 1.6 antihypertensive medications. Overall, clinicians overestimated the proportion of their patients who were prescribed guideline-concordant medications (75% perceived vs. 67% actual, P <0.001) and who had blood pressure levels <140/90 mm Hg on their last visit (68% perceived vs. 43% actual, P <0.001). Among individual clinicians, there were no significant correlations between perceived and actual guideline adherence (r = 0.18 for medications, r = 0.14 for blood pressure control; P > or =0.10 for both). Clinicians with relatively low actual guideline performance were most likely to overestimate their adherence to medication recommendations and blood pressure targets.Clinicians appear to overestimate their adherence to hypertension guidelines, particularly with regards to the proportion of their patients with controlled blood pressure. This limited awareness may represent a barrier to successful implementation of guidelines, and could be addressed through the use of provider profiles and point-of-service feedback to clinicians.

  View details for DOI 10.1016/j.amjmed.2004.03.035

  View details for Web of Science ID 000225232900005

  View details for PubMedID 15541324

• Health-related quality of life in patients with pulmonary arterial hypertension CHEST Shafazand, S., Goldstein, M. K., Doyle, R. L., Hlatky, M. A., Gould, M. K. 2004; 126 (5): 1452-1459

  Abstract

  Patients with pulmonary arterial hypertension (PAH) often present with dyspnea and severe functional limitations, but their health-related quality of life (HRQOL) has not been studied extensively. This study describes HRQOL in a cohort of patients with PAH.Cross-sectional study.A tertiary care, university hospital-based, pulmonary hypertension (PH) clinic.We studied HRQOL in 53 patients with PAH (mean age, 47 years; median duration of disease, 559 days). Eighty-three percent were women, 53% received epoprostenol, and 72% reported moderate-to-severe functional limitations with a New York Heart Association class 3 or 4 at enrollment.We examined HRQOL by administering the Nottingham Health Profile, Congestive Heart Failure Questionnaire, and Hospital Anxiety and Depression Scale. We used the Visual Analog Scale and standard gamble (SG) techniques to measure preferences for current health (utilities). Compared with population norms, participants reported moderate-to-severe impairment in multiple domains of HRQOL, including physical mobility, emotional reaction, pain, energy, sleep, and social isolation. Mean SG utilities were 0.71, suggesting that, on average, participants were willing to accept a 29% risk of death in order to be cured of PH.PAH is a devastating condition that affects predominately young women in the prime of their life. Understanding HRQOL and preferences are important in the care and management of these patients. Compared with population norms, patients with PAH have substantial functional and emotional limitations that adversely affect their HRQOL.

  View details for Web of Science ID 000225214700016

  View details for PubMedID 15539712

• Translating research into practice: Organizational issues in implementing automated decision support for hypertension in three medical centers JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION Goldstein, M. K., Coleman, R. W., Tu, S. W., Shankar, R. D., O'Connor, M. J., Musen, M. A., Martins, S. B., Lavori, P. W., Shlipak, M. G., Oddone, E., Advani, A. A., Gholami, P., Hoffman, B. B. 2004; 11 (5): 368-376

  Abstract

  Information technology can support the implementation of clinical research findings in practice settings. Technology can address the quality gap in health care by providing automated decision support to clinicians that integrates guideline knowledge with electronic patient data to present real-time, patient-specific recommendations. However, technical success in implementing decision support systems may not translate directly into system use by clinicians. Successful technology integration into clinical work settings requires explicit attention to the organizational context. We describe the application of a "sociotechnical" approach to integration of ATHENA DSS, a decision support system for the treatment of hypertension, into geographically dispersed primary care clinics. We applied an iterative technical design in response to organizational input and obtained ongoing endorsements of the project by the organization's administrative and clinical leadership. Conscious attention to organizational context at the time of development, deployment, and maintenance of the system was associated with extensive clinician use of the system.

  View details for Web of Science ID 000223898000005

  View details for PubMedID 15187064

  View details for PubMedCentralID PMC516243

• A catalog of dermatology utilities: A measure of the burden of skin diseases JOURNAL OF INVESTIGATIVE DERMATOLOGY SYMPOSIUM PROCEEDINGS Chen, S. C., Bayoumi, A. M., Soon, S. L., Aftergut, K., Cruz, P., Sexton, S. A., McCall, C. O., Goldstein, M. K. 2004; 9 (2): 160-168

  Abstract

  Utilities are measures of quality of life that reflect the strength of individuals' preferences or values for a particular health outcome. As such, utilities represent a measure of disease burden. The aim of this article is to introduce the concept of utilities to the dermatology community and to present a catalog of dermatology utilities obtained from direct patient interviews. Our data are based on 236 total subjects from Grady Hospital (Atlanta, GA), Stanford Medical Center (Palo Alto, CA), and Parkland Hospital (Dallas, TX). The mean time trade-off utilities ranged from 0.640 for blistering disorders to 1.000 for alopecia, cosmetic, and urticaria. The mean utility across all diagnoses was 0.943. We present utilities for 17 diagnostic categories and discuss the underlying reasons for the significant disease burden that these utilities represent. We also present these dermatology categories relative to noncutaneous diseases to place the cutaneous utilities in perspective and to compare the burden of disease. We have demonstrated that skin diseases have considerable burden of disease and provided a preliminary repository of utility data for future researchers and policy makers.

  View details for Web of Science ID 000220652200014

  View details for PubMedID 15083784

• Treadmill scores in elderly men JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY Lai, S., Kaykha, A., Yamazaki, T., Goldstein, M., Spin, J. M., Myers, J., Froelicher, V. F. 2004; 43 (4): 606-615

  Abstract

  This study seeks to further characterize the role of exercise testing in the elderly for prognosis and diagnosis of coronary artery disease.Recent exercise testing guidelines have recognized that statements regarding the elderly do not have an adequate evidence-based quality because the studies they are based on have limitations in sample size and design. The Duke Treadmill Score has been recommended for risk stratification, but recent evidence has suggested that it does not function in the elderly.The study population consisted of male veterans (1872 patients >or=65 years; 3798 patients <65 years) who underwent routine clinical exercise testing with a mean follow-up of six years. A subset who underwent coronary angiography as clinically indicated (elderly, n = 405; younger, n= 809) were included. The primary outcome for all subjects was cardiovascular mortality with coronary angiographic findings as the outcome in those selected for angiography.In survival analysis, exercise-induced ST depression was prognostic in both age groups only when cardiovascular death was considered as the end point. Peak metabolic equivalents were the most significant predictor for both age groups only when all-cause death was considered as the end point. New age-specific prognostic scores were developed and found to be predictive for cardiovascular mortality in the elderly. Moreover, in the angiographic subset of the elderly, a specific diagnostic score provided significantly better discrimination than exercise ST measurements alone. For any new score, there is a need for validation in another elderly population.The mortality end point affected the choice of prognostic variables. This study demonstrates that exercise test scores can be helpful for the diagnosis and prognosis of coronary disease in the elderly.

  View details for DOI 10.1016/j.jacc.2003.07.051

  View details for Web of Science ID 000188944600017

  View details for PubMedID 14975471

• Modeling guidelines for integration into clinical workflow 11th World Congress on Medical Informatics Tu, S. W., Musen, M. A., Shankar, R., Campbell, J., Hrabak, K., McClay, J., Huff, S. M., Mcclure, R., Parker, C., Rocha, R., ABARBANEL, R., Beard, N., Glasgow, J., Mansfield, G., Ram, P., Ye, Q., Mays, E., Weida, T., Chute, C. G., McDonald, K., Mohr, D., Nyman, M. A., Scheitel, S., Solbrig, H., Zill, D. A., Goldstein, M. K. I O S PRESS. 2004: 174–178

  Abstract

  The success of clinical decision-support systems requires that they are seamlessly integrated into clinical workflow. In the SAGE project, which aims to create the technological infra-structure for implementing computable clinical practice guide-lines in enterprise settings, we created a deployment-driven methodology for developing guideline knowledge bases. It involves (1) identification of usage scenarios of guideline-based care in clinical workflow, (2) distillation and disambiguation of guideline knowledge relevant to these usage scenarios, (3) formalization of data elements and vocabulary used in the guideline, and (4) encoding of usage scenarios and guideline knowledge using an executable guideline model. This methodology makes explicit the points in the care process where guideline-based decision aids are appropriate and the roles of clinicians for whom the guideline-based assistance is intended. We have evaluated the methodology by simulating the deployment of an immunization guideline in a real clinical information system and by reconstructing the workflow context of a deployed decision-support system for guideline-based care. We discuss the implication of deployment-driven guideline encoding for sharability of executable guidelines.

  View details for PubMedID 15360798

• An intelligent case-adjustment algorithm for the automated design of population-based quality auditing protocols 11th World Congress on Medical Informatics Advani, A., Jones, N., Shahar, Y., Goldstein, M., Musen, M. A. I O S PRESS. 2004: 1003–1007

  Abstract

  We develop a method and algorithm for deciding the optimal approach to creating quality-auditing protocols for guideline-based clinical performance measures. An important element of the audit protocol design problem is deciding which guide-line elements to audit. Specifically, the problem is how and when to aggregate individual patient case-specific guideline elements into population-based quality measures. The key statistical issue involved is the trade-off between increased reliability with more general population-based quality measures versus increased validity from individually case-adjusted but more restricted measures done at a greater audit cost. Our intelligent algorithm for auditing protocol design is based on hierarchically modeling incrementally case-adjusted quality constraints. We select quality constraints to measure using an optimization criterion based on statistical generalizability coefficients. We present results of the approach from a deployed decision support system for a hypertension guideline.

  View details for Web of Science ID 000226723300202

  View details for PubMedID 15360963

• Evaluating provider adherence in a trial of a guideline-based decision support system for hypertension 11th World Congress on Medical Informatics Chan, A. S., Coleman, R. W., Martins, S. B., Advani, A., Musen, M. A., Bosworth, H. B., Oddone, E. Z., Shlipak, M. G., Hoffman, B. B., Goldstein, M. K. I O S PRESS. 2004: 125–129

  Abstract

  Measurement of provider adherence to a guideline-based decision support system (DSS) presents a number of important challenges. Establishing a causal relationship between the DSS and change in concordance requires consideration of both the primary intention of the guideline and different ways providers attempt to satisfy the guideline. During our work with a guideline-based decision support system for hypertension, ATHENA DSS, we document a number of subtle deviations from the strict hypertension guideline recommendations that ultimately demonstrate provider adherence. We believe that understanding these complexities is crucial to any valid evaluation of provider adherence. We also describe the development of an advisory evaluation engine that automates the interpretation of clinician adherence with the DSS on multiple levels, facilitating the high volume of complex data analysis that is created in a clinical trial of a guideline-based DSS.

  View details for Web of Science ID 000226723300026

  View details for PubMedID 15360788

• Evaluation of KNAVE-II: a tool for intelligent query and exploration of patient data 11th World Congress on Medical Informatics Martins, S. B., Shahar, Y., Galperin, M., Kaizer, H., Goren-Bar, D., McNaughton, D., Basso, L. V., Goldstein, M. K. I O S PRESS. 2004: 648–652

  Abstract

  We present the results of a preliminary evaluation of KNAVE-II, a distributed knowledge-based computational framework for visualization, interpretation, and exploration of longitudinal clinical data and of multiple levels of concepts derivable from these data. KNAVE-II uses a distributed architecture to access at run-time clinical time-oriented data, a domain-specific knowledge base containing properties of the clinical data, and a knowledge-based problem-solving method for computing on-the-fly interpretations of these data. The purpose of the evaluation was to compare efficiency and user satisfaction when answering clinical queries of variable complexity about clinical time-oriented data using KNAVE-II, versus using methods available in standard clinical settings: paper chart or electronic spreadsheet (ESS). Subjects answered high-complexity queries significantly faster using KNAVE-II than when using paper or ESS. User satisfaction with KNAVE-II was significantly superior compared to satisfaction using paper or ESS, based on a standard usability scale. Users also explicitly ranked KNAVE-II as superior to paper and the ESS.

  View details for Web of Science ID 000226723300131

  View details for PubMedID 15360893

• Academic detailing to improve anti hypertensive prescribing patterns AMERICAN JOURNAL OF HYPERTENSION Siegel, D., Lopez, J., Meier, J., Goldstein, M. K., Lee, S., Brazill, B. J., Matalka, M. S. 2003; 16 (6): 508-511

  Abstract

  Several studies indicate that treatment of hypertension in the United States does not follow recommendations from expert bodies. We thus implemented a program using academic detailers to increase practitioner compliance with antihypertensive treatment guidelines. Five Veterans Affairs medical facilities including academic medical centers and community based outpatient clinics were chosen for the intervention. Pharmacists were trained as academic detailers, and the intervention included lectures, educational materials, provider profiling, and meetings with 25 to 50 providers each. After intervention, the proportion of hypertensives receiving calcium antagonists decreased from 43% to 38% (P <.001), whereas the proportion receiving a beta blocker or thiazide diuretic increased from 58% to 64% (P <.001). For hypertensive subjects with diabetes mellitus or congestive heart failure, the proportion receiving an angiotensin converting enzyme inhibitor or angiotensin receptor blocker increased from 72% to 76% for the former and from 74% to 78% for the latter (P <.001 for both). Among hypertensive subjects with coronary artery disease an increase in beta blocker use was noted after intervention (P <.001 for change from baseline). Prescribing patterns after academic detailing more closely followed national recommendations.

  View details for DOI 10.1016/S0895-7061(03)00060-8

  View details for Web of Science ID 000183356400015

  View details for PubMedID 12799103

• A distributed, collaborative, structuring model for a clinical-guideline digital-library. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Shahar, Y., Shalom, E., Mayaffit, A., Young, O., Galperin, M., Martins, S., Goldstein, M. 2003: 589-593

  Abstract

  The Digital Electronic Guideline Library (DeGeL) is a Web-based framework and a set of distributed tools that facilitate gradual conversion of clinical guidelines from free text, through semi-structured text, to a fully structured, executable representation. Thus, guidelines exist in a hybrid, multiple-format representation The three formats support increasingly sophisticated computational tasks. The tools perform semantic markup, classification, search, and browsing, and support computational modules that we are developing, for run-time application and retrospective quality assessment. We describe the DeGeL architecture and its collaborative-authoring authorization model, which is based on (1) multiple medical-specialty authoring groups, each including a group manager who controls group authorizations, and (2) a hierarchical authorization model based on the different functions involved in the hybrid guideline-specification process. We have implemented the core modules of the DeGeL architecture and demonstrated distributed markup and retrieval using the knowledge roles of two guidelines ontologies (Asbru and GEM). We are currently evaluating several of the DeGeL tools.

  View details for PubMedID 14728241

• Graphical Displays to Improve Guideline-Based Therapy of Hypertension Hypertension Primer (eds J.L. Izzo and H.R.Black, Wms & Wilkins Goldstein, M.K., Hoffman, B.B. 2003
• A Web-Based system for interactive visualization and exploration of time-oriented clinical data and their abstractions. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Shahar, Y., Boaz, D., Tahan, G., Galperin, M., Goren-Bar, D., Kaizer, H., Basso, L. V., Martins, S. B., Goldstein, M. K. 2003: 1073-?

  Abstract

  In this theater-style demonstration, the speakers will demonstrate KNAVE-II, a Web-based distributed system for interactive visualization and exploration of large amounts of time-oriented clinical data from multiple sources, and of clinically meaningful concepts (abstractions) derivable from these data. The KNAVE-II system and its complete underlying architecture provide a solution to the data overload problem.

  View details for PubMedID 14728576

• Interactive visualization and exploration of time-oriented clinical data using a distributed temporal-abstraction architecture. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Shahar, Y., Boaz, D., Tahan, G., Galperin, M., Goren-Bar, D., Kaizer, H., Basso, L. V., Martins, S. B., Goldstein, M. K. 2003: 1004-?

  Abstract

  KNAVE-II is a system for visualization and exploration of large amounts of time-oriented clinical data and of multiple levels of clinically meaningful abstractions derivable from these data. KNAVE-II uses a distributed temporal-abstraction architecture that integrates a set of knowledge services, each interacting with a domain-specific knowledge source, a set of data-access services, each interacting with a clinical data source, and a computational service for deriving knowledge-based abstractions of the data.

  View details for PubMedID 14728507

• The Effects of CPOE on ICU workflow: an observational study. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Cheng, C. H., Goldstein, M. K., Geller, E., Levitt, R. E. 2003: 150-154

  Abstract

  Computerized physician order entry (CPOE) has had demonstrated benefits in error reduction and guideline adherence, but its implementation has often been complicated by disruptions in established workflow processes. We conducted an observational study of the healthcare team in an intensive care unit after the implementation of mandatory CPOE. We found that policies designed to increase flexibility and safety led to an increased coordination load on the healthcare team, and created opportunities for new sources of error. We attribute this in part to implicit assumptions in the CPOE system design that execution of physician orders is a linear work process. Observational workflow studies are an important tool to understand how to redesign CPOE systems so as to avoid harm and achieve the full potential of benefit for improved patient safety.

  View details for PubMedID 14728152

  View details for PubMedCentralID PMC1480350

• Hypertension Recordkeeping and Electronic Management Systems (eds) Izzo J.L. and Black H.R. Hypertension Primer: The Essentials of High Blood Pressure Goldstein M.K., Hoffman B.B. 2003: 393-396
• Desirability of Outcome States Symptom Research: Methods and Opportunities. eds M.B. Max and J. Lynn. NIH online interactive textbook Goldstein, M.K., Tsevat, J. 2003
• Developing quality indicators and auditing protocols from formal guideline models: knowledge representation and transformations. AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium Advani, A., Goldstein, M., Shahar, Y., Musen, M. A. 2003: 11-15

  Abstract

  Automated quality assessment of clinician actions and patient outcomes is a central problem in guideline- or standards-based medical care. In this paper we describe a model representation and algorithm for deriving structured quality indicators and auditing protocols from formalized specifications of guidelines used in decision support systems. We apply the model and algorithm to the assessment of physician concordance with a guideline knowledge model for hypertension used in a decision-support system. The properties of our solution include the ability to derive automatically context-specific and case-mix-adjusted quality indicators that can model global or local levels of detail about the guideline parameterized by defining the reliability of each indicator or element of the guideline.

  View details for PubMedID 14728124

• Patient safety in guideline-based decision support for hypertension management: ATHENA DSS Annual Meeting of the American-Medical-Information-Association Goldstein, M. K., Hoffman, B. B., Coleman, R. W., Tu, S. W., Shankar, R. D., O'Connor, M., Martins, S., Advani, A., Musen, M. A. B M J PUBLISHING GROUP. 2002: S11–S16

  View details for DOI 10.1197/jamia.M1218

  View details for Web of Science ID 000179005600004

• Utility elicitation software for older adults FLAIR2 for quality of life ratings of dependency in ADLS Goldstein, M., Garber, A., Miller, D. GERONTOLOGICAL SOCIETY AMER. 2002: 250

  View details for Web of Science ID 000179541400885

• Quality of life assessment software for computer-inexperienced older adults: Multimedia utility elicitation for activities of daily living Annual Symposium of the American-Medical-Informatics-Association Goldstein, M. K., Miller, D. E., Davies, S., Garber, A. M. HANLEY & BELFUS INC MED PUBLISHERS. 2002: 295–299

  Abstract

  Functional status as measured by dependencies in the Activities of Daily Living (ADLs) is an important indicator of overall health for older adults. Methodologies for outcomes-based medical-decision-making for public policy, such as decision modeling and cost-effectiveness analysis, require utilities for outcome health states. Utilities have been reported for many disease states, but have not been indexed by functional status, which is a strong predictor of outcome in geriatrics. We describe here a utility elicitation program developed specifically for use with computer-inexperienced older adults: Functional Limitation And Independence Rating (FLAIR1). FLAIR1 design features address common physical problems of the aged and computer attitudes of inexperienced users that could impede computer acceptance. We interviewed 400 adults ages 65 years and older with FLAIR1. In exit interviews with 154 respondents, 118 (76%) found FLAIR1 easy to use. Design features in FLAIR1 can be applied to other software for older adults

  View details for PubMedID 12463834

• Accuracy of computerized outpatient diagnoses in a veterans affairs general medicine clinic Annual Meeting of the Department-of-Veterans-Affairs-Health-Services-Research-and-Development Szeto, H. C., Coleman, R. K., Gholami, P., Hoffman, B. B., Goldstein, M. K. MANAGED CARE & HEALTHCARE COMMUNICATIONS LLC. 2002: 37–43

  Abstract

  Electronically available data, both administrative, such as outpatient encounter diagnostic data, and clinical, such as problem lists, are being used increasingly for outcome and quality assessment, risk adjustment, and clinical reminder systems.To determine the accuracy of outpatient primary care diagnostic information recorded in administrative and clinical files in a Veterans Affairs VISTA (Veterans Health Information Systems and Technology Architecture) database compared with medical chart notes.Cross-sectional medical chart review of 148 patients attending a general medicine clinic at a university-affiliated Veterans Affairs hospital for 9 diagnoses relevant to the choice of drug therapy for hypertension.An administrative file of encounter diagnoses, for a 2-year period, and a clinical file of the problem list maintained by the clinician were the sources of electronic diagnoses. We compared these sources with diagnoses abstracted by medical chart review. We estimated the sensitivity and specificity of each electronic data source for detecting medical chart note diagnoses.The sensitivity for 8 of the 9 study diagnoses was greater than 80% in the administrative file and 49% in the clinical problem list. The specificity was good for the administrative file (91% to 100%) and even better for the clinical file (98% to 100%).Outpatient encounter diagnoses relevant to hypertension recorded as electronic data had high specificity, and some codes had high sensitivity when collected over multiple visits. The administrative file was more sensitive but less specific than the clinical file. Administrative vs clinical files can be selected to minimize either the false-negative or the false-positive designations, respectively, as dictated by the needs of the quality assessment review.

  View details for Web of Science ID 000173442700002

  View details for PubMedID 11814171

• A framework for evidence-adaptive quality assessment that unifies guideline-based and performance-indicator approaches Annual Symposium of the American-Medical-Informatics-Association Advani, A., Goldstein, M., Musen, M. A. HANLEY & BELFUS INC MED PUBLISHERS. 2002: 2–6

  Abstract

  Automated quality assessment of clinician actions and patient outcomes is a central problem in guideline- or standards-based medical care. In this paper we describe a unified model representation and algorithm for evidence-adaptive quality assessment scoring that can: (1) use both complex case-specific guidelines and single-step population-wide performance-indicators as quality measures; (2) score adherence consistently with quantitative population-based medical utilities of the quality measures where available; and (3) give worst-case and best-case scores for variations based on (a) uncertain knowledge of the best practice, (b) guideline customization to an individual patient or particular population, (c) physician practice style variation, or (d) imperfect reliability of the quality measure. Our solution uses fuzzy measure-theoretic scoring to handle the uncertain knowledge about best-practices and the ambiguity from practice variation. We show results of applying our method to retrospective data from a guideline project to improve the quality of hypertension care.

  View details for Web of Science ID 000189418100001

  View details for PubMedID 12463775

  View details for PubMedCentralID PMC2244239

• Ethics Primary Care Geriatrics: A Case-Based Approach. eds R.J.Ham, P.D. Sloane, G.A. Warshaw. Mosby, Inc Goldstein, M.K. 2002
• What should be reported in a methods section on utility assessment? MEDICAL DECISION MAKING Stalmeier, P. F., Goldstein, M. K., Holmes, A. M., Lenert, L., Miyamoto, J., Stiggelbout, A. M., Torrance, G. W., Tsevat, J. 2001; 21 (3): 200-207

  Abstract

  The measurement of utilities, or preferences, for health states may be affected by the technique used. Unfortunately, in papers reporting utilities, it is often difficult to infer how the utility measurement was carried out.To present a list of components that, when described, provide sufficient detail of the utility assessment.An initial list was prepared by one of the authors. A panel of 8 experts was formed to add additional components. The components were drawn from 6 clusters that focus on the design of the study, the administration procedure, the health state descriptions, the description of the utility assessment method, the description of the indifference procedure, and the use of visual aids or software programs. The list was updated and redistributed among a total of 14 experts, and the components were judged for their importance of being mentioned in a Methods section.More than 40 components were generated. Ten components were identified as necessary to include even in an article not focusing on utility measurement: how utility questions were administered, how health states were described, which utility assessment method(s) was used, the response and completion rates, specification of the duration of the health states, which software program (if any) was used, the description of the worst health state (lower anchor of the scale), whether a matching or choice indifference search procedure was used, when the assessment was conducted relative to treatment, and which (if any) visual aids were used. The interjudge reliability was satisfactory (Cronbach's alpha = 0.85).The list of components important for utility papers may be used in various ways, for instance, as a checklist while writing, reviewing, or reading a Methods section or while designing experiments. Guidelines are provided for a few components.

  View details for Web of Science ID 000170801800005

  View details for PubMedID 11386627

• Patient safety in guideline-based decision support for hypertension management: ATHENA DSS Annual Symposium of the American-Medical-Informatics-Association (AMIA 2001) Goldstein, M. K., Hoffman, B. B., Coleman, R. W., Tu, S. W., Shankar, R. D., O'Connor, M., Martins, S., Advani, A., Musen, M. A. BMJ PUBLISHING GROUP. 2001: 214–218

  Abstract

  The Institute of Medicine recently issued a landmark report on medical error.1 In the penumbra of this report, every aspect of health care is subject to new scrutiny regarding patient safety. Informatics technology can support patient safety by correcting problems inherent in older technology; however, new information technology can also contribute to new sources of error. We report here a categorization of possible errors that may arise in deploying a system designed to give guideline-based advice on prescribing drugs, an approach to anticipating these errors in an automated guideline system, and design features to minimize errors and thereby maximize patient safety. Our guideline implementation system, based on the EON architecture, provides a framework for a knowledge base that is sufficiently comprehensive to incorporate safety information, and that is easily reviewed and updated by clinician-experts.

  View details for Web of Science ID 000172263400045

  View details for PubMedID 11825183

  View details for PubMedCentralID PMC2243380

• Integration of textual guideline documents with formal guideline knowledge bases Annual Symposium of the American-Medical-Informatics-Association (AMIA 2001) Shankar, R. D., Tu, S. W., Martins, S. B., Fagan, L. M., Goldstein, M. K., Musen, M. A. BMJ PUBLISHING GROUP. 2001: 617–621

  Abstract

  Numerous approaches have been proposed to integrate the text of guideline documents with guideline-based care systems. Current approaches range from serving marked up guideline text documents to generating advisories using complex guideline knowledge bases. These approaches have integration problems mainly because they tend to rigidly link the knowledge base with text. We are developing a bridge approach that uses an information retrieval technology. The new approach facilitates a versatile decision-support system by using flexible links between the formal structures of the knowledge base and the natural language style of the guideline text.

  View details for Web of Science ID 000172263400126

  View details for PubMedID 11825260

• Building an explanation function for a hypertension decision-support system 10th World Congress on Medical Informatics (MEDINFO 2001) Shankar, R. D., Martins, S. B., Tu, S. W., Goldstein, M. K., Musen, M. A. I O S PRESS. 2001: 538–542

  Abstract

  ATHENA DSS is a decision-support system that provides recommendations for managing hypertension in primary care. ATHENA DSS is built on a component-based architecture called EON. User acceptance of a system like this one depends partly on how well the system explains its reasoning and justifies its conclusions. We addressed this issue by adapting WOZ, a declarative explanation framework, to build an explanation function for ATHENA DSS. ATHENA DSS is built based on a component-based architecture called EON. The explanation function obtains its information by tapping into EON's components, as well as into other relevant sources such as the guideline document and medical literature. It uses an argument model to identify the pieces of information that constitute an explanation, and employs a set of visual clients to display that explanation. By incorporating varied information sources, by mirroring naturally occurring medical arguments and by utilizing graphic visualizations, ATHENA DSS's explanation function generates rich, evidence-based explanations.

  View details for Web of Science ID 000172901700127

  View details for PubMedID 11604798

• Ethical decision making and patient autonomy - A comparison of physicians and patients in Japan and the United States CHEST Ruhnke, G. W., Wilson, S. R., Akamatsu, T., Kinoue, T., Takashima, Y., Goldstein, M. K., Koenig, B. A., Hornberger, J. C., Raffin, T. A. 2000; 118 (4): 1172-1182

  Abstract

  Patient-centered decision making, which in the United States is typically considered to be appropriate, may not be universally endorsed, thereby harboring the potential to complicate the care of patients from other cultural backgrounds in potentially unrecognized ways. This study compares the attitudes toward ethical decision making and autonomy issues among academic and community physicians and patients of medical center outpatient clinics in Japan and the United States.A questionnaire requesting judgments about seven clinical vignettes was distributed (in English or Japanese) to sample groups of Japanese physicians (n = 400) and patients (n = 65) as well as US physicians (n = 120) and patients (n = 60) that were selected randomly from academic institutions and community settings in Japan (Tokyo and the surrounding area) and the United States (the Stanford/Palo Alto, CA, area). Responses were obtained from 273 Japanese physicians (68%), 58 Japanese patients (89%), 98 US physicians (82%), and 55 US patients (92%). Physician and patient sample groups were compared on individual items, and composite scores were derived from subsets of items relevant to patient autonomy, family authority, and physician authority.A majority of both US physicians and patients, but only a minority of Japanese physicians and patients, agreed that a patient should be informed of an incurable cancer diagnosis before their family is informed and that a terminally ill patient wishing to die immediately should not be ventilated, even if both the doctor and the patient's family want the patient ventilated (Japanese physicians and patients vs US physicians and patients, p < 0.001). A majority of respondents in both Japanese sample groups, but only a minority in both US sample groups, agreed that a patient's family should be informed of an incurable cancer diagnosis before the patient is informed and that the family of an HIV-positive patient should be informed of this disease status despite the patient's opposition to such disclosure (Japanese physicians and patients vs US physicians and patients, p < 0.001). Physicians in both Japan and the United States were less likely than patients in their respective countries to agree with physician assistance in the suicide of a terminally ill patient (Japanese physicians and patients vs US physicians and patients, p < 0.05). Across various clinical scenarios, all four respondent groups accorded greatest authority to the patient, less to the family, and still less to the physician when the views of these persons conflicted. Japanese physicians and patients, however, relied more on family and physician authority and placed less emphasis on patient autonomy than the US physicians and patients sampled. Younger respondents placed less emphasis on family and physician authority.Family and physician opinions are accorded a larger role in clinical decision making by the Japanese physicians and patients sampled than by those in the United States, although both cultures place a greater emphasis on patient preferences than on the preferences of the family or physician. Our results are consistent with the view that cultural context shapes the relationship of the patient, the physician, and the patient's family in medical decision making. The results emphasize the need for clinicians to be aware of these issues that may affect patient and family responses in different clinical situations, potentially affecting patient satisfaction and compliance with therapy.

  View details for Web of Science ID 000090008300046

  View details for PubMedID 11035693

• Consistency validation of standard gamble utilities for dependence in activities of daily living Bravata, D., Olkin, Cottle, R., Eaves, B. C., Garber, A., Goldstein, M. LIPPINCOTT WILLIAMS & WILKINS. 2000: S117

  View details for Web of Science ID 000088887000492

• Clinical decision-support systems: Evaluating the evaluation MEDICAL DECISION MAKING Hornberger, J., Goldstein, M. K. 2000; 20 (1): 130-131

  View details for Web of Science ID 000084667100016

  View details for PubMedID 10638545

• Applying Utility Assessment at the Bedside Decision Making in Health Care: Theory, Psychology, and Applications. eds G.Chapman and F. Sonnenberg. Cambridge University Press Goldstein, M.K., Tsevat, J. 2000
• Implementing clinical practice guidelines while taking account of changing evidence: ATHENA DSS, an easily modifiable decision-support system for managing hypertension in primary care Annual Symposium of the American-Medical-Informatics-Association Goldstein, M. K., Hoffman, B. B., Coleman, R. W., Musen, M. A., Tu, S. W., Advani, A., Shankar, R., O'Connor, M. HANLEY & BELFUS INC. 2000: 300–304

  Abstract

  This paper describes the ATHENA Decision Support System (DSS), which operationalizes guidelines for hypertension using the EON architecture. ATHENA DSS encourages blood pressure control and recommends guideline-concordant choice of drug therapy in relation to comorbid diseases. ATHENA DSS has an easily modifiable knowledge base that specifies eligibility criteria, risk stratification, blood pressure targets, relevant comorbid diseases, guideline-recommended drug classes for patients with comorbid disease, preferred drugs within each drug class, and clinical messages. Because evidence for best management of hypertension evolves continually, ATHENA DSS is designed to allow clinical experts to customize the knowledge base to incorporate new evidence or to reflect local interpretations of guideline ambiguities. Together with its database mediator Athenaeum, ATHENA DSS has physical and logical data independence from the legacy Computerized Patient Record System (CPRS) supplying the patient data, so it can be integrated into a variety of electronic medical record systems.

  View details for Web of Science ID 000170207500062

  View details for PubMedID 11079893

• Cost-comparison analysis versus cost-effectiveness analysis: An important difference JOURNAL OF THE AMERICAN ACADEMY OF DERMATOLOGY Chen, S., Bayoumi, A. M., Goldstein, M. K. 1999; 41 (6): 1050-1050

  View details for Web of Science ID 000083972700036

  View details for PubMedID 10570405

• Postoperative complications in Parkinson's disease JOURNAL OF THE AMERICAN GERIATRICS SOCIETY Pepper, P. V., Goldstein, M. K. 1999; 47 (8): 967-972

  Abstract

  Although Parkinson's disease is relatively common in America, with an average annual incidence of 20 cases per 100,000 population, little information exists about postoperative morbidity and mortality in those Parkinson's patients who undergo elective surgery.We performed a retrospective cohort study using the Veterans Affairs (VA) Austin database system (a cumulative index of admissions and discharges from all US VA Medical Centers) to identify 41,213 patients who underwent elective bowel resection, cholecystectomy, or radical prostatectomy between January 1, 1990, and December 31, 1995. We examined the study population using univariate analysis, acute length of stay with multivariate analysis, and postoperative complications with logistic regression.The selected surgeries were performed on 234 patients with a diagnosis of Parkinson's disease and 40,979 with no such diagnosis. In univariate analysis, patients with Parkinson's disease had significantly longer acute hospital stays than non-Parkinson's patients (11.4 +/- 15.9 days vs 8.8 +/- 9.0 days, P < .001). In addition, Parkinson's patients had a higher in-hospital mortality than non-Parkinson's patients (7.3% vs 3.8%, P = .006). After we adjusted for coexisting morbidity, age, admitting location, and gender, patients with Parkinson's disease had an average acute hospital stay 2.34 days longer than that of non-Parkinson's patients (P < .001). However, the mortality difference did not reach statistical significance in multivariate analysis (P = .098). Finally, Parkinson's patients had significantly increased incidences of urinary-tract infection (odds ratio 2.045, P < .001), aspiration pneumonia (odds ratio 3.825, P < .001), and bacterial infections (odds ratio 1.682, P < .001).Patients with Parkinson's disease are at greater risk for specific complications and longer hospital stay after elective bowel resection, cholecystectomy, or radical prostatectomy. Awareness of these complications may help caregivers to reduce postoperative mortality and morbidity and to decrease the length of hospitalization.

  View details for Web of Science ID 000081880800008

  View details for PubMedID 10443858

• The influence of physician race, age, and gender on physician attitudes toward advance care directives and preferences for end-of-life decision-making JOURNAL OF THE AMERICAN GERIATRICS SOCIETY Mebane, E. W., Oman, R. F., Kroonen, L. T., Goldstein, M. K. 1999; 47 (5): 579-591

  Abstract

  To determine whether physicians' preferences for end-of-life decision-making differ between blacks and whites in the same pattern as patient preferences, with blacks being more likely than whites to prefer life-prolonging treatments.A mailed survey.American Medical Association (AMA) and National Medical Association (NMA) databases. To enrich the sample of black physicians, we targeted physicians in the AMA database practicing in high minority area zip codes and graduates of the traditionally black medical schools.Self-reported physician attitudes toward end-of-life decision-making and preference of treatment for themselves in persistent vegetative state or organic brain disease compared by race, controlling for age and gender.The 502 physicians (28%) who returned the questionnaire included 280 white and 157 black physicians. With regard to attitudes toward patient care, 58% of white physicians agreed that tube-feeding in terminally ill patients is "heroic," but only 28 % of black physicians agreed with the statement (P < .001). White physicians were more likely than black physicians to find physician-assisted suicide an acceptable treatment alternative (36.6% vs 26.5% of black physicians) (P < .05). With regard to the physicians preferences for future treatment of themselves for the persistent vegetative state scenario, black physicians were more than six times more likely than white physicians to request aggressive treatments (cardiopulmonary resuscitation, mechanical ventilation, or artificial feeding) for themselves (15.4% vs 2.5%) (P < .001). White physicians were almost three times as likely to want physician-assisted suicide (29.3% vs 11.8%) (P < .001) in this scenario. For a state of brain damage with no terminal illness, the majority of all physicians did not want aggressive treatment, but black physicians were nearly five times more likely than white physicians (23.0% vs 5.0%) (P < .001) to request these treatments. White physicians, on the other hand, were more than twice as likely to request physician-assisted suicide (22.5% vs 9.9%), P < .001 in this scenario.Physicians preferences for end-of-life treatment follow the same pattern by race as patient preferences, making it unlikely that low socioeconomic status or lack of familiarity with treatments account for the difference. Self-denoted race may be a surrogate marker for other, as yet undefined, factors. The full spectrum of treatment preferences should be considered in development of guidelines for end-of-life treatment in our diverse society.

  View details for Web of Science ID 000080201100013

  View details for PubMedID 10323652

• Decisions at the end of life: Cultural considerations beyond medical ethics GENERATIONS-JOURNAL OF THE AMERICAN SOCIETY ON AGING Hallenbeck, J., Goldstein, M. K. 1999; 23 (1): 24-29

  View details for Web of Science ID 000087030100005

  View details for PubMedID 17144035

• Integrating a modern knowledge-based system architecture with a legacy VA database: The ATHENA and EON projects at Stanford Annual Symposium of the American-Medical-Informatics-Association Advani, A., Tu, S., O'Connor, M., Coleman, R., Goldstein, M. K., Musen, M. BMJ PUBLISHING GROUP. 1999: 653–657

  Abstract

  We present a methodology and database mediator tool for integrating modern knowledge-based systems, such as the Stanford EON architecture for automated guideline-based decision-support, with legacy databases, such as the Veterans Health Information Systems & Technology Architecture (VISTA) systems, which are used nation-wide. Specifically, we discuss designs for database integration in ATHENA, a system for hypertension care based on EON, at the VA Palo Alto Health Care System. We describe a new database mediator that affords the EON system both physical and logical data independence from the legacy VA database. We found that to achieve our design goals, the mediator requires two separate mapping levels and must itself involve a knowledge-based component.

  View details for Web of Science ID 000170207300134

  View details for PubMedID 10566440

• The application of medical decision analysis to genetic testing: An introduction 49th Annual Meeting of the American-Society-of-Human-Genetics McConnell, L. M., Goldstein, M. K. MARY ANN LIEBERT INC. 1999: 65–70

  Abstract

  The availability of genetic tests to diagnose or predict Alzheimer disease (AD) causes a shift in the way people think about the condition and how they assess the options available to them. Decision analysis in a quantitative approach for dealing with the uncertainties inherent in many medical decisions, including decisions about genetic testing. Decision analysis does not guarantee a good outcome, but aims to yield better overall average results by providing a framework for people to evaluate their options and minimize cognitive biases. We provide an overview of the decision analysis process, including the terms and tools commonly associated with it. We also use a recent example to demonstrate one way decision analysis has been applied to genetics in the medical literature. This paper is an introduction to subsequent papers that explore the specific question of whether decision analysis is a helpful tool for understanding the uncertainty inherent in probabilistic information about genetic risk for AD.

  View details for Web of Science ID 000087218200009

  View details for PubMedID 10464579

• Decision analysis and Alzheimer disease: Three case studies 49th Annual Meeting of the American-Society-of-Human-Genetics Holtzman, S., Ozanne, E., Carone, B., Goldstein, M. K., Steinke, G., Timbs, J. MARY ANN LIEBERT INC. 1999: 71–83

  Abstract

  Decision analysis may be useful to people facing Alzheimer disease (AD) decisions. The use of decision analysis in three such cases is reported. The first case involved a middle-aged person worried about early-onset AD and deciding whether to seek genetic testing. The analysis let the participant reject testing and consider innovative care options. The second case involved a middle-aged person concerned about later-onset AD. The analysis for her was more complex, and led to the assignment of some limited value on genetic testing for her. The third case revolved around a caregiver's treatment decisions for a patient with severe AD. It led her to recognize the importance of factors she had not previously considered. In each of the three cases, the intensive process of decision analysis appears to have improved the subject's decision.

  View details for Web of Science ID 000087218200010

  View details for PubMedID 10464580

• The effect of search procedures on utility elicitations 18th Annual Meeting of the Society-for-Medical-Decision-Making Lenert, L. A., Cher, D. J., Goldstein, M. K., Bergen, M. R., Garber, A. SAGE PUBLICATIONS INC. 1998: 76–83

  Abstract

  Elicited preferences for health states vary among scaling methods, manners of describing health states, and other features of the elicitation process. The authors examined the effects of changing the search procedure for a subject's utility on mean utility values.A randomized controlled trial of two search procedures (titration and "ping-pong") using two otherwise identical computer programs that describe health states related to Gaucher's disease, then measuring subjects' preferences.Paid, healthy volunteers recruited from the community through advertisements.The mean time tradeoff (TTO) and standard gamble (SG) utility values for life with severe anemia and splenomegaly and life with chronic bone pain from Gaucher's disease were between 0.10 and 0.15 higher with the titration search procedure than with the ping-pong procedure. Effects of the search procedure were additive with variability due to scaling methods, resulting in mean differences in utility ratings for the same health state of as much as 0.28 among procedures and scaling methods. Effects of search procedures on utility values persisted on repeated testing at week 2 and week 3; there was no evidence of convergence to a single "true" utility value over time.The procedure used to search for subjects' utility values strongly influences the results of preference-assessment experiments. Effects of search procedures persist on repeated testing. The results suggest that utility values are heavily influenced by, if not created during, the process of elicitation. Thus, utility values elicited using different search procedures may not be directly comparable.

  View details for Web of Science ID 000071405300015

  View details for PubMedID 9456212

• Measurement of the validity of utility elicitations performed by computerized interview MEDICAL CARE Lenert, L. A., Morss, S., Goldstein, M. K., Bergen, M. R., Faustman, W. O., Garber, A. M. 1997; 35 (9): 915-920

  Abstract

  The authors evaluate a measure of the validity of utility elicitations and study the potential effects of invalid elicitations on population utility values.The authors used a computerized survey to describe and measure preferences for three common side-effects of anti-psychotic drugs (tardive dyskinesia [TD], akathesia [AKA], pseudo-parkinsonism). The authors compared the validity of elicitations in 41 healthy volunteers to 22 schizophrenic patients. Preferences were measured using visual analog scale (VAS), pair-wise comparison (PWC), and the Standard Gamble (SG) methods. To assess the validity of each groups' responses, the authors compared the consistency of subjects' rank-order of the desirability of states across methods of preferences assessment (CAMPA).All healthy volunteers and 82% of patients completed the computer survey; of these subjects, 97% of healthy volunteers and 70% of patients indicated they thought they understood the task required of them. However, only 78% of healthy subjects and 44% of patients had a consistent rank ordering of preferences among VAS and PWC ratings; only 80% and 61%, respectively, had a consistent rank ordering preferences among SG and PWC ratings. For two of the three health states, inconsistent subjects had statistically higher SG utilities (for TD, 0.94 versus 0.87, and for AKA 0.92 versus 0.86; P < 0.05).The CAMPA test can identify potentially invalid preference ratings. Potentially invalid preference ratings may bias the "population" utilities for health states.

  View details for PubMedID 9298080

• The effect of assessment method and respondent population on utilities elicited for Gaucher Disease Annual Meeting of the Society-for-Medical-Decision-Making Clarke, A. E., Goldstein, M. K., Michelson, D., Garber, A. M., Lenert, L. A. RAPID SCIENCE PUBLISHERS. 1997: 169–84

  Abstract

  Measured preferences have been reported to vary with the method of elicitation and respondent population surveyed. We elicited utilities for Gaucher disease using a multimedia implementation of the time trade-off, standard gamble, and a conceptually different, largely untested approach, the risk-risk trade-off, from those who are healthy, those with a chronic illness and those with Gaucher disease. The risk-risk trade-off produced significantly lower utilities than the other two preference assessment methods and had the poorest test-retest reliability. The respondent's self-rated current health state utility was the most important determinant of utility values elicited by the time trade-off and standard gamble for the hypothetical health states. Our results do not support the use of our implementation of the risk-risk trade-off method. In eliciting preferences for hypothetical health states from the general population, the subjective rating of a respondent's own health state should be considered in determining representative population groups.

  View details for Web of Science ID A1997WW41600009

  View details for PubMedID 9161117

• Cultural considerations of death and dying in the United States CLINICS IN GERIATRIC MEDICINE Hallenbeck, J., Goldstein, M. K., Mebane, E. W. 1996; 12 (2): 393-?

  Abstract

  Culture is not something apart from us. It is always here, and we, like fish in an ocean, may be blind to the water in which we swim. We recommend that, in addition to learning as much as possible about people from cultures other than our own, we consider carefully the cultural roots of our own thoughts and actions. A combination of respectful behavior and a conscientious effort to understand individuals from another culture will go far in enabling us to work well with dying patients from many backgrounds.

  View details for Web of Science ID A1996UL98800013

  View details for PubMedID 8799356

• Comprehensive geriatric assessment: is it too comprehensive for compliance and cost-effectiveness? journal of the American Board of Family Practice Goldstein, M. K. 1995; 8 (4): 337-340

  View details for PubMedID 7572302

• DEVELOPING AND TESTING A MULTIMEDIA PRESENTATION OF A HEALTH-STATE DESCRIPTION MEDICAL DECISION MAKING Goldstein, M. K., Clarke, A. E., Michelson, D., Garber, A. M., Bergen, M. R., Lenert, L. A. 1994; 14 (4): 336-344

  Abstract

  Quality-adjustment weights for health states are an essential component of cost-utility analysis (CUA). Quality-adjustment weights are obtained by presenting large numbers of subjects with multiattribute descriptions of health states for rating. Comprehending multiattribute health states is a difficult task for most respondents. The authors hypothesized that multimedia (MM) presentation using computers might facilitate this task better than would a paper-based text (Text). To test this hypothesis, they developed closely matched MM and Text descriptions of health states in the first-person narrative style, and developed a method of testing the presentation of a health state. Subjects were randomized to exposure to either MM or Text and subject recall of the health state and recognition of features of the health state were tested. How well defined the preferences of the subjects were after each presentation method was assessed by having the subjects mark on a double-anchored visual-analog scale the "best" and "worst" they believed the quality of life in the health state might be. MM subjects had better recall (11.85 vs 9.44 of a total of 24 meaning units, p = 0.098) and better recognition (4.71 vs 4.22, p = 0.08). The average interval between the "best" and "worst" ratings was shorter for the MM subjects (2.19 cm vs 3.26 cm, p = 0.12).(ABSTRACT TRUNCATED AT 250 WORDS)

  View details for Web of Science ID A1994PL43300004

  View details for PubMedID 7528868

• INTENSITY OF TREATMENT IN MALNUTRITION - THE ETHICAL CONSIDERATIONS PRIMARY CARE Goldstein, M. K., Fuller, J. D. 1994; 21 (1): 191-206

  Abstract

  Decisions about initiation and withdrawal of life-sustaining artificial nutrition and hydration are complex and sometimes are agonizing to make. Artificial feeding is considered by most medical ethicists to be a medical intervention about which decisions should be made based on the benefits, risks, and burdens of the treatment. State law varies, particularly with regard to the inclusion of artificial feeding in advance directives and in laws about its discontinuance. Conditions in which it is common for patients or families to consider refusal of artificial feeding include terminal cancer, advanced dementia, and persistent vegetative state. Alternative approaches to nutrition, such as offering favorite foods or treating underlying depression, may obviate the need for tube feeding. Competent patients may refuse any medical treatment, including life-sustaining treatment. An appropriate proxy decision maker may also refuse treatment on behalf of a mentally incapacitated patient.

  View details for Web of Science ID A1994ND10400013

  View details for PubMedID 8197255

• ETHICAL DILEMMAS IN THE CRITICALLY ILL ELDERLY CLINICS IN GERIATRIC MEDICINE Clarke, D. E., Goldstein, M. K., Raffin, T. A. 1994; 10 (1): 91-101

  Abstract

  Elderly people in the United States often receive treatment through an enormous array of medical technology when they become critically ill. Some, or all, such interventions may be unwanted, and patients have the right to be informed about what prospects lie ahead. CPR, with survival rates of 2% to 20%, rarely has the effect for which it was intended, as studies over the last two decades have repeatedly demonstrated. Although it is not clear that elderly people are at higher risk for poor outcomes of CPR, it is clear most of them do not possess sufficient knowledge about either CPR or its usual outcome to make an informed choice. Ideally, an individual should choose his or her own resuscitation status, but evidence exists indicating surrogates more often than not make this decision. This is unfortunate, because both surrogates and physicians are poor judges of patients' resuscitation preferences. Advance directives, especially when coupled with effective physician-patient communication, will aid elderly persons in making decisions about life support. We encourage all physicians who care for the elderly to avert many of tomorrow's ethical dilemmas by communicating with their healthy patients today.

  View details for Web of Science ID A1994PK09100007

  View details for PubMedID 8168029

• WITHHOLDING AND WITHDRAWING NUTRITION AND HYDRATION - SURROGATES CAN MAKE THIS DECISION FOR INCOMPETENT PATIENTS CHEST Clarke, D. E., Goldstein, M. K., Raffin, T. A. 1993; 104 (6): 1646-1647

  View details for Web of Science ID A1993ML30800002

  View details for PubMedID 8252934

• A multimedia preference-assessment tool for functional outcomes. Proceedings / the ... Annual Symposium on Computer Application [sic] in Medical Care. Symposium on Computer Applications in Medical Care Goldstein, M. K., Michelson, D., Clarke, A. E., Lenert, L. A. 1993: 844-848

  Abstract

  Functional outcomes of clinical trials are often reported as number of dependencies in activities of daily living (ADLs). Quality-weighting for the ADLs has not been reported. We designed and pilot-tested ADLIB (ADL Index Builder), a multimedia computer program, that presents ADL health states to subjects and elicits from subjects a rating for the quality of life of each health state. Subjects, who were patients over age 50 without previous computer experience, found the program easy to use. Health care professionals specializing in geriatrics confirmed that the ADL presentations used in the program are in accord with typical practice in scoring ADLs. We plan to use the program to obtain population-based preference ratings that can be used to assess efficacy of clinical trials and to provide quality-weights for cost-effectiveness analysis.

  View details for PubMedID 8130598

• DURABLE POWER OF ATTORNEY FOR HEALTH-CARE - A SURVEY OF SENIOR CENTER PARTICIPANTS ARCHIVES OF INTERNAL MEDICINE Roe, J. M., Goldstein, M. K., Massey, K., Pascoe, D. 1992; 152 (2): 292-296

  Abstract

  Fifty-nine healthy senior center participants were interviewed to determine how, when, and why the durable power of attorney for health care is being used. The 21 users of the durable power of attorney for health care executed a form for the expected reasons; however, a majority had not given a copy of the form to their physician and few had discussed details of their preferences with the proxy. Of the 38 nonusers, the most frequent reasons for not executing a durable power of attorney for health care were: lack of awareness of the form, procrastination, and difficulty choosing a proxy. The new Patient Self-Determination Act requiring hospitals to inform patients of advance directives will help to overcome some of the obstacles in use of the durable power of attorney for health care; however, community education must still be encouraged.

  View details for Web of Science ID A1992HE06900010

  View details for PubMedID 1739356

• DURABLE POWER OF ATTORNEY FOR HEALTH-CARE - ARE WE READY FOR IT WESTERN JOURNAL OF MEDICINE Goldstein, M. K., VALLONE, R. P., PASCOE, D. C., Winograd, C. H. 1991; 155 (3): 263-268

  Abstract

  Health care professionals need to be well informed about advance directives for medical care in the event a patient becomes incapacitated. The Patient Self-Determination Act requires that all patients be advised of their options at the time of hospital admission. Hospitals and health care professionals will need to work together to plan for implementing this law. We surveyed 215 physicians, nurses, and social workers at a Veterans Affairs Medical Center about the California advance directive, the Durable Power of Attorney for Health Care. Attitudes were generally positive. All of the social workers had heard of the durable power of attorney directive, but 36% of physicians and nurses had never heard of it and an additional 20% had no experience with one. For respondents who had heard of the directive, the mean knowledge score was 6.35 of a possible 10 (5 predicted by chance). Respondents brought up the issue of durable power of attorney with patients before a crisis only 19% of the time and determined whether one had been signed for only 16% of older patients in hospital. The most commonly cited reasons for failure to discuss this with patients were lack of proper forms, pamphlets, or a place to refer a patient. Of those who had ever seen such a document in use, 42% were aware of a problem with it at some time. Whereas attitudes toward advance directives are positive, many physicians and nurses had little knowledge of the Durable Power of Attorney for Health Care and were poorly equipped to discuss it with patients. We encourage educating hospital staff to prepare for the enactment of the Patient Self-Determination Act. We also recommend that the concerns raised by professionals about the use of a durable power of attorney be addressed.

  View details for Web of Science ID A1991GF77400005

  View details for PubMedID 1949773

• SCREENING FOR FRAILTY - CRITERIA AND PREDICTORS OF OUTCOMES JOURNAL OF THE AMERICAN GERIATRICS SOCIETY Winograd, C. H., Gerety, M. B., Chung, M., Goldstein, M. K., Dominguez, F., Vallone, R. 1991; 39 (8): 778-784

  Abstract

  To determine the reliability of rapid screening by clinically derived geriatric criteria in predicting outcomes of elderly hospitalized patients.Prospective cohort study of 985 patients screened at the time of hospital admission and followed for 1 year with respect to the outcomes of mortality, hospital readmission, and nursing home utilization.Palo Alto Veterans Affairs Medical Center, a tertiary care teaching hospital.Male patients 65 years of age and older admitted to the Medical and Surgical services during the period from October 1, 1985 through September 30, 1986.Patients were grouped by specific screening criteria into three groups of increasing frailty: Independent, Frail, and Severely Impaired. Each criterion focused on a geriatric condition and was designed to serve as a marker for frailty. Increasing frailty was significantly correlated with increasing length of hospital stay (P less than 0.0001), nursing home utilization (P less than 0.0001), and mortality (P less than 0.0001). Multivariate analyses revealed that the clinical groups were more predictive of mortality and nursing home utilization than were age or Diagnosis-Related Groups (DRGs). Rehospitalization was unrelated to age, clinical group, or DRG, suggesting that utilization may not be driven by the clinical factors measured in this study.Rapid clinical screening using specific geriatric criteria is effective in identifying frail older subjects at risk for mortality and nursing home utilization. Our findings suggest that geriatric syndromes are more predictive of adverse outcomes than diagnosis per se. This well operationalized screening process is inexpensive as well as effective and could easily be introduced into other hospital settings.

  View details for Web of Science ID A1991GA27300006

  View details for PubMedID 1906492

• LONG-TERM ENTERAL FEEDING - THE BRITISH VIEW JOURNAL OF THE AMERICAN GERIATRICS SOCIETY Goldstein, M. K. 1991; 39 (7): 732-733

  View details for Web of Science ID A1991FU96900020

  View details for PubMedID 1905736

• MISUSE OF DURABLE POWER OF ATTORNEY FOR HEALTH-CARE JOURNAL OF THE AMERICAN GERIATRICS SOCIETY Goldstein, M. K., Shadlen, M. F. 1991; 39 (7): 730-730

  View details for Web of Science ID A1991FU96900017

  View details for PubMedID 2061542

• Ethical considerations in pharmacotherapy of the aged. Drugs & aging Goldstein, M. K. 1991; 1 (2): 91-97

  View details for PubMedID 1794011

• Gynecologic factors in sexual dysfunction of the older woman. Clinics in geriatric medicine Goldstein, M. K., Teng, N. N. 1991; 7 (1): 41-61

  Abstract

  Older women may experience sexual dysfunction due to many different causes. Some problems related to menopausal hormonal change may be easily treated with estrogen supplements. Other problems involve intricate interpersonal relations between the woman and her sexual partner and may require a combination of medical therapy and sexual counseling. Gynecologic cancer and cancer treatments are often accompanied by problems in sexual functioning. These problems may then impair relations and self-image, leading to a vicious circle of deteriorating social function. Some recommendations for the clinician follow. The clinician should maintain an attitude of openness to the possibility of sexual concerns in older women. Such concerns should be taken seriously and should not be dismissed as part of aging. Routine periodic health examinations can include a question such as "Do you have any concerns about your sexual life that you would like to discuss?" In follow-up visits for procedures with a high likelihood of causing sexual dysfunction, questions that would open the door to a discussion of sexuality should be asked. Sexual dysfunction should be recognized as a couple-oriented phenomenon. A woman's anxiety about her appearance, postoperative depression, or dyspareunia may be perceived by her partner as a sexual rejection and may initiate a cycle of decreasing contact or may even lead to erectile dysfunction. Sexual counseling should include both partners. When a surgical procedure that will probably have an impact on sexual function is contemplated, provide the patient and her partner with advance counseling. Descriptions of surgery should not be simply a statement of body parts to be removed but should specifically address the anticipated sexual effects. Counseling should include a description of basic anatomy and function of the genital organs. Illustrations and appropriate demonstration during the physical examination should be used to ensure the patient's understanding. Descriptions should be accurate without being either frightening or falsely reassuring. The patient should be counseled about the benefits of including her partner in discussions. Then, when possible, the sexual partner of the patient should be invited to sessions of advance counseling on contemplated procedures. Clinicians should remain open to the possibility that the sexual partner will be a nontraditional one, e.g., an unmarried male partner or another woman. The clinician should be alert to remediable causes of dysfunction. For example, decreased vaginal lubrication may be managed with use of water-soluble lubricants.(ABSTRACT TRUNCATED AT 400 WORDS)

  View details for PubMedID 2004290

• FECAL INCONTINENCE IN AN ELDERLY MAN - STANFORD-UNIVERSITY GERIATRICS CASE CONFERENCE JOURNAL OF THE AMERICAN GERIATRICS SOCIETY Goldstein, M. K., Brown, E. M., Holt, P., Gallagher, D., Winograd, C. H. 1989; 37 (10): 991-1002

  View details for Web of Science ID A1989AV89600012

  View details for PubMedID 2794323

• ETHICAL CARE OF THE ELDERLY - PITFALLS AND PRINCIPLES GERIATRICS Goldstein, M. K. 1989; 44 (3): 101-?

  Abstract

  Ethical problems have become intrinsic to geriatric medicine. Physician's familiarity with ethical principles of medical decision-making, such as beneficence, autonomy, and justice, can be helpful. These principles can lead, however, to conflicting conclusions, and the decisions must be personal. An informed approach can help physicians resolve these increasingly commonplace dilemmas.

  View details for Web of Science ID A1989T612300007

  View details for PubMedID 2917715