Abstracts
|
|---|
| Title, Authors, Affiliations |
Abstract |
A novel Emergency Department split flow model increases throughput in response to overcrowding
Gregory Wallingford, MD, Nikita Joshi, MD, Patrice Callagy, RN, Jamie Stone, RN, Ian Brown, MD, Sam Shen, MD
1- Stanford University School of Medicine, Department of Emergency Medicine
|
Purpose
Emergency Department (ED) overcrowding is an issue that is affecting every emergency department and every hospital. The inability to maintain patient flow into and out of the ED paralyzes the ability to provide effective and timely patient care. In this article, the authors seek to describe and provide metrics for a patient flow methodology that targets ESI-3 patients in a vertical flow model.
Methods
In the Stanford ED, a vertical flow patient care area was created from existing ED space by structural and operational redesign. The model was launched and sustained through frequent interdisciplinary huddles, a detailed inclusion and exclusion criteria, scripted verbiage on how to promote the flow model to patients, and close analytics of metrics. This was a 6-month September 2014 to February 2015 retrospective pre and post intervention study that examined length of stay as a marker for effective launch and implementation of a vertical patient workflow model. Secondary outcome measures included anecdotal and opinion data from staff and patient satisfaction data of those triaged to the vertical flow area.
Results/Anticipated Results
The patients triaged to the Delta Zone in the study period tended to be younger than in the control time period (43 years vs. 52 years, p=0.00). There was a significant decrease in our primary endpoint, the total LOS for ESI 3 patients triaged to the Delta Zone vertical area (270 min vs 384 min, p=0.00).
Conclusion
fImplementation of a vertical patient flow strategy can decrease length of stay for the vertical ESI patients based upon the inclusion and exclusion criteria. Furthermore, this is accomplished with minimal financial investment within the physical constraints of an existing ED.
|
| A Quality Improvement Curriculum for Neurology Residents
Rebecca Miller-Kuhlmann, MD [1], Lironn Kraler, MD [1], Alex Frolov, MD [1], Nina Bozinov, MD [1], Carl Gold, MD [2], Kathryn Kvam, MD [2]
1- Stanford, Department of Neurology, Resident
2- Stanford, Deparment of Neurology, Hospitalist
|
Purpose
(1) build foundational knowledge in quality improvement
(2) develop skills in the practical application of QI tools to relevant clinical issues
Methods
Following a needs- assessment with a pretest administered to all neurology residents as well as additional resident focus groups, we developed a 2 part quality improvement curriculum comprised of a lecture series designed to dovetail with a monthly M&M case conference.
Results/Anticipated Results
Pretest:
18 residents out of 26 responded to the pretest. 61% had never participated in a quality improvement project. 83% felt that instruction in QI and patient safety is important, but the majority rated themselves low regarding their comfort with various QI skills such as writing a clear problem statement, analyzing and understanding medical errors, and identifying resources and measures for a QI project.
Curriculum:
1. 6 hour curriculum comprised of three two-hour sessions given during weekly neurology lecture time. Sessions were designed to be active and included lecture/workshops on patient safety and high value care followed by a guided, mentored working session applying A3 thinking to resident identified active QI issues from recent M&M conferences.
2. Monthly QI-focused Morbidity and Mortality conference where 2-3 residents create a problem statement and perform a root cause analysis of self-identified area of potential improvement raised by a recent case. Examples thus far include "safety of MRI in patients with pacemakers and implantable defibrillators" or "increased frequency of goals of care discussions with chronically ill patients."
Conclusion
QI conferences and M&M have been well-received per qualitative post-session feedback. Resident engagement and comfort with quality improvement has increased, with multiple formal and informal QI projects in process at present. When surveyed immediately after the QI A3 workshop, for example, all 5 groups were planning on carrying their projects forward. More formal post-test data at the 18 month mark (May 2017) will be collected shortly.
|
| A randomized controlled trial of a patient-centered intervention to reduce unnecessary dual-energy x-ray absorptiometry (DEXA) screening in younger and low-risk patients
Shreya Shah, MD, Andre Kumar, MD, Kirsti Weng, MD, MPH, Sang-ick Chang, MD, MPH, Jason Hom, MD
1 - Department of Medicine, Stanford University, Palo Alto, CA |
Purpose
The Choosing Wisely initiative recommends against DEXA screening for osteoporosis in women age 65 or men age70 with no risk factors. The U.S. Preventative Services Task Force (USPSTF) recommends using a 9.3% 10-year fracture risk threshold to screen, based on the fracture risk assessment(FRAX) score. We conducted a randomized controlled trial at six Stanford Health Care primary care clinics to determine if a patient-centered intervention resulted in lower rates of unnecessary DEXA scans.
Methods
The intervention consisted of flyers in the waiting area and screensavers on computers, with materials emphasizing when DEXA scans are indicated. Three clinics were in the intervention arm, and three were in the control arm. All women age 65 and men age 70 with DEXA scans ordered were included in the analysis. Patients were excluded if they had a prior diagnosis of osteoporosis or osteopenia.
Results/Anticipated Results
There was no significant difference in the proportion of appropriate scans when comparing the pre-intervention and post-intervention periods for the intervention group. However, there was a significant improvement in the primary outcome based on USPSTF guidelines (12.2%vs25.6%,p=0.02) and secondary outcome based on the Choosing Wisely recommendations (55.1%vs68.9%,p=0.05) when comparing the pre-intervention period and post-intervention period among the overall study cohort.
Conclusion
The patient-centered intervention itself did not lead to different ordering patterns. However, the overall proportion of appropriate DEXA scans increased significantly in the post-intervention period. Despite this overall improvement, there remains a large proportion of inappropriate osteoporosis screening post-intervention - 74.4% based on the USPSTF guidelines and 31.1% based on the Choosing Wisely recommendations.
|
| Adverse Events in Hospitalized Children with Medical Complexity
Brett Palama MD MS [1], Krisa Hoyle Elgin FNP MPH [2], Paul Sharek MD MPH [1,2]
1 - Division of Pediatric Hospital Medicine, Stanford School of Medicine
2 - Center for Quality and Clinical Effectiveness, Lucile Packard Children's Hospital Stanford |
Purpose
To determine the effect of medical complexity on rates, severity, and preventability of adverse events (AE) in patients admitted to a children's hospital using a previously published pediatric global trigger tool.
Methods
234 pediatric patients admitted to LPCH between July 2006- June 2012 were randomly selected for inclusion. Medical complexity was defined by the presence of a chronic complex condition (CCC). Suspected AEs were identified by a trained nurse reviewer using the Global Assessment of Pediatric Patient Safety (GAPPS) trigger tool; each suspected AE was reviewed independently by two trained physician reviewers to confirm the presence of an AE and to assess severity and preventability for each event. Disagreements were settled by consensus.
Results/Anticipated Results
119/234 patients (50.9%) had at least one CCC. Patients with a CCC experienced 32 AEs per 1,000 hospital days, vs. 23 AEs per 1,000 hospital days for children without a CCC. There was no difference between groups in percentage of AEs considered to be definitely or probably preventable. All 12 AEs classified as permanent or requiring intervention to sustain life were in patients with a CCC.
Conclusion
Children with a CCC have a higher rate of AEs per 1,000 hospital days and are more likely to experience AEs of higher severity. Further study is warranted to determine if these findings are replicated in other academic children's hospitals, as well as other settings in which children with medical complexity are hospitalized.
|
| An EMR Handoff Tool Improves Medicine Residents' Handoff Quality
Rebecca Tisdale, MD, MPA [1], Lisa Shieh, MD, PhD [1]
1 - Stanford School of Medicine, Department of Internal Medicine |
Purpose
Communication breakdown plays a part in the majority of adverse events in healthcare. Physician to physician handoffs are particularly prone to communication errors, yet have been shown to be more complete when systematized according to a standardized bundle. However, interventions that improve thoroughness of handoffs have not been widely studied. Therefore, the objective of this quality improvement project was to evaluate the completeness of written handoffs among internal medicine residents at baseline and after implementation of an EMR-based tool.
Methods
The authors evaluated all written handoffs from general and specialty (hematology, oncology, cardiology) internal medicine inpatient units staffed by internal medicine interns and residents from a randomly chosen representative sample of days in April and May 2015 at Stanford University Medical Center, focusing on content elements. The intervention was then implemented in June 2015 with post-intervention data collected in August-September 2016. The EMR-based tool included radio buttons that prompted users to select whether the illness severity categorized the patient as stable, a “watcherâ€, or unstable. It also automatically pulled in EMR data on the patient's 24-hour vitals, common lab tests, and code status. Finally, it provided text boxes labeled Active Issues, Action List (To-Dos), and If/Then for residents to fill in.
Results/Anticipated Results
Implementation of this EMR-based tool resulted in significant increases in handoff completeness both for handoff elements automatically pulled in and for those requiring active input by residents.
Conclusion
EMR-based interventions may be highly effective at improving handoffs, which in turn leads to fewer medical errors and better patient care.
|
| Building A Better Discharge: Use Of A Pre-Discharge Planner To Improve Patient Engagement In The Discharge Process
Justin Slade, MD [1], Lisa Zhu, MD [1], Camille Huwyler [2] , Jacqueline Aredo [3], Allison Kwong, MD [1], Michelle Chen, MD [4], Leon Castaneda, MD [1], Julia Chang, MD [1], Ginger Yang, MD [1], Alexandra Ruan, MD [1], Rebecca Tisdale, MD [1], Marta Almli, MD [1], Teresa Liu, MD [1], David Crichton [5], Margaret Smith [6] , Joseph Hopkins, MD [1], and Lisa Shieh, MD [1]
1 - Stanford University, School of Medicine, Department of Medicine
2 - Case Western Reserve University, School of Medicine
3 - Stanford University, School of Medicine
4 - Stanford University, School of Medicine, Department of Otolaryngology
5 - Stanford Health Care, Healthcare Quality Improvement
6 - Stanford Health Care, Performance Excellence |
Purpose
The complexity of modern healthcare has contributed towards poorly standardized hospital discharge processes that allow for failures in communication between health providers and their patients which may diminish patient and provider satisfaction and result in suboptimal hospital discharges. Our group created a patient-centered discharge planning questionnaire for patients admitted to internal medicine wards teams with the goal of proactively identifying and addressing common barriers to a safe and timely discharge.
Methods
The discharge planner asks patients to identify their mode of transportation from the hospital, identify their preferred pharmacy, list any questions or concerns regarding their medical care, and identify any members of their care team with whom they would like to speak prior to discharge. Of the five medicine wards teams, three were randomized to the intervention group and two to the control group. The primary outcomes for this pilot study include patient- and physician-perceived usefulness of the questionnaire and survey completion rate.
Results/Anticipated Results
Over the course of eleven weeks 72 pre-discharge patient questionnaires have been completed by patients admitted to the three general medicine teams participating in this intervention, accounting for approximately 2/3 of patient's discharged home from these groups. All patients completing this form have been surveyed, with 36 (56%) grading this intervention very helpful in preparing for discharge and an additional 26 (40%) grading it as somewhat helpful. The responses from interns surveyed that were involved in the care of these patients have been positive as well.
Conclusion
Our pre-discharge patient questionnaire has, in this initial investigation, proven itself to be a promising method for achieving increased patient engagement in the discharge planning process in order to improve upon patient preparedness for and satisfaction with their hospital discharge. Investigation of its impact on timeliness of hospital discharge is pending completion of this pilot study.
|
| Dysphagia Delaying Discharge in Acute Stroke Patients
Jennifer Shum, MD [1], Katherine Werbaneth, MD [1], Justin Ruey Tse, MD [2], Sandra Deane, MS, CCC-SLP, BCS-S [3], Maria Larrenaga, MA, CCC-SLP, MBS-IMP [3] , Nirali Vora, MD [1]
1 - Stanford University, School of Medicine, Department of Neurology & Neurologic Sciences
2 - Stanford University, School of Medicine, Department of Radiology
3 - Stanford Health Care, Speech Language Pathology |
Purpose
At Stanford Hospital, 62% of our acute stroke patients have dysphagia. Of those who go on to require G-tube placement, their median length of stay is prolonged at 17 days. We propose a novel protocol to standardize speech therapy evaluation among acute stroke patients with dysphagia with the goal of reducing delays to final speech therapy recommendation (PO diet, short term NG, or g-tube) and thereby reducing overall length of stay.
Methods
We looked retrospectively into prior patient data of all acute stroke patients admitted to neurology service in 2016 and measured baseline metrics of days from initial speech therapy evaluation to final speech therapy recommendation as well as overall length of stay. We then implemented our speech therapy evaluation protocol and re-measured our metrics.
Results/Anticipated Results
We anticipate that our speech therapy evaluation protocol will reduce time to final speech therapy recommendation and thereby reduce overall length of stay.
Conclusion
Standardizing the speech therapy evaluation reduces length of stay.
|
| Evidence of Pre Discharge Interventions Increasing Press Ganey Response Rates in Patients Discharged from the Emergency Department
Courtney Azevedo [1], I-Ching Lee [1], Megan Fee [1], Alexei Wagner, MD [1], Sam Shen, MD [1]
1- Stanford, School of Medicine, Department of Emergency Medicine |
Purpose
We investigated various approaches to increase Press Ganey survey response rates. The purpose of thisstudy is to evaluate whether various pre-discharge interventions would increase survey response rates.
Methods
A pre-discharge thank you card was created, informing patients that they will receive a patient experience survey via mail within a 2-3 day time period after being discharged from FT. A minimum of 75% of total patients seen per day in FT were given thank you cards and a verbal explanation before being discharged, performed by the scheduled scribe. Responses were analyzed using statistical modeling techniques.
Results/Anticipated Results
Over a 6-month duration (September 2016-February 2017) of practicing pre-discharge interventions, LTR scores remained the same (83.3%). Survey response rates increased 3%.
Conclusion
Pre-discharge thank you card distribution and a verbal explanation of receiving a patient satisfaction survey via mail significantly increased patient response rate. Based on these interpreted results, further study needs to be done to investigate the correlation between pre-discharge interventions and LTR scores
|
| Hepatitis B Vaccination in Inflammatory Bowel Disease(IBD) Patients Receiving Biologic Infusions
Warapan Nakayuenyongsuk, MD [1], Megan Christofferson, BA [1], Hayden Schwenk, MD [2], Dorsey Bass, MD [1]
1 - Lucile Packard Children's Hospital at Stanford, Department of Pediatrics, Division of Pediatric Gastroenterology, Hepatology and Nutrition
2 - Lucile Packard Children's Hospital at Stanford, Department of Pediatrics, Division of Pediatric Infectious Diseases |
Purpose
IBD patients receiving immune-suppressing medication can be at higher risk for infection. Hepatitis B virus (HBV) infection can be prevented with a safe, effective vaccine, although some patients may not havean adequate response to the primary vaccine series. We aimed to increase documented Hepatitis B immunity in IBD patients receiving biologic infusions at LPCH from 6% to 50% over a sixmonth period.
Methods
An interdisciplinaryteam from Gastroenterology, Infectious Diseases, and Pharmacy developed a laboratory and vaccination protocol. Patients receiving biologics (infliximab or vedolizumab) at our outpatient infusion unit were reviewed, and Hepatitis B titers were added to standard infusion labs if not previously documented. Adouble dose of Engerix-Bwas administered to HBV seronegative (anti-HBs <10mIU/mL) patientsat the next infusion visit,thenanti-HBs titerswere re-run;if stillnegative, two more double dose vaccines were administeredat the usual Hepatitis B vaccinationintervals.
Results/Anticipated Results
Of 80 patients eligible, five (6%) had documented immunity to HBV, five were not immune, and 70 had no documentation.Titers were run on these 70 patients, and 23 wereimmune. Of theremaining 52 patients withnegative titers,45received the first vaccination dose. Titers were rechecked on 29patients, and 22(76%) converted to positive. Five of the seven patients with persistent negative titers have received the second vaccination dose.
Conclusion
Documented immunity to Hepatitis B increased from 5 to 50 patients(63%) over five months. A single double-dose vaccination converted 76% of patients to HBV immune. This initiative is ongoing.
|
| How Do Residents Reach Nurses?
Kelly Fedoruk, MD [1], Adam X Sang, MD [2], Hilary H Wang, MD [3], Jessica X Yu, MD [4], Rebecca K Miller-Kuhlmann, MD [3], Haley Manella, MD [5], Michael Genualdi, MD [6], Lisa Shieh, MD PhD FHM [7]
1 - Department of Obstetrics/Anesthesia
2 - Department of Surgery
3 - Department of Neurology
4 - Division of Gastroenterology
5 - Department of Emergency Medicine
6 - Department of Medicine
7 - Medical Director for Quality, Department of Medicine
|
Purpose
Closed loop communication between nurses and physicians is key to good patient care. While the standard method of reaching MDs is paging, there is no standard method to reach nurses. Current options include calls to unit-based phones, Spectralinks, in-person meetings, Epic's message board feature, and nursing communication orders. There is no published data regarding which method is preferred or more effective. Our project investigates how our residents achieve RN communication and what barriers they face.
Methods
A survey including quantitative and qualitative items was distributed to residents. Focus groups were conducted both with house staff and nurses from an inpatient ward. Finally, Epic data was collected to measure how frequently nurses update their contact information.
Results/Anticipated Results
The survey had 11 respondents. Phone calls and nursing communication orders were the most frequently used methods to reach nurses. All respondents have faced some barriers in reaching a nurse. Barriers identified include not knowing 1) who the appropriate RN is, 2) how to reach the RN, and 3) whether the order has been acknowledged or completed. The focus groups revealed several ways in which MD to RN communication can be improved. Finally, the Epic data from five (5) nursing cohorts showed that only about 20% of nurses update both their name and contact information.
Conclusion
There are many barriers that residents face when trying to reach a nurse. Phone calls still appear to be the preferred method of reaching a nurse. However, only 20% of nurses provide their contact number in Epic.
|
| ICU care in the ED: The Emergency Critical Care Team Pilot Project
Julian Villar MD MPH [1], Sara Crager MD [2], Jason Nesbitt RN [3], Chris Cinkoski RN [3], Jim Mobley RN [3], Tsuyoshi Mitarai [4]
1 - Critical Care Medicine, Department of Medicine, Stanford
2 - Critical Care Medicine, Department of Anesthesia, Stanford
3 - FLEX RN Program, Emergency Department
4 - Department of Emergency Medicine, Stanford |
Purpose
Many patients admitted to the ICU from the ED spend many hours boarding in the ED while awaiting an in-patient bed. The purpose of this project was to provide ICU-level of care to critically-ill patients in the ED, provide real-time critical care consults, and contribute to ED flow by providing quicker dispositions for patients. We also hoped to reduce the number of patient needing ICU care by providing aggressive intensive care in the first few hours of their hospitalization.
Methods
Prospective pre-post interventional study. October 24, 2016 - December 18, 2016. Either Dr Crager or Dr Villar were in the ED every day from 2pm-2am during the study period. We evaluated and cared for all patients requiring admission to the MICU or consultation from the ICU Fellow. Our goal was to provide optimal, high-quality care and correct physiologic derangement as quickly as possible.
Results/Anticipated Results
We cared for 134 patients during the study period, 95 were admitted, 39 were consults. Of the 95 that were admitted, only 9 patients (9.5%) had ICU beds available at the time of admission, and 31 (33%) were downgraded to a floor service and never required to the ICU. 43 patients met criteria for Severe sepsis or septic shock, and of these 41 patients (95%) received 30cc/kg within 3 hours, and 35 (81%) received a reassessment within 6 hours.
Conclusion
The presence of the ECCT Fellows in the ED led to higher quality of care for critically ill patients, improved ED flow or critically ill patients, led to more prompt downgrades of improving patients, and to improved sepsis compliance. Comparison with pre- and post-intervention control groups is ongoing.
|
| Impact of medical scribes on throughput in mid- and low-acuity treatment zones of an academic emergency department
Keith Thomas [1], Joshua Marcum [1], Michael Kohn, MD [2], Alexei Wagner, MD [2]
1-Stanford Hospital & Clinics Emergency Department
2-Stanford School of Medicine Department of Emergency Medicine |
Purpose
The traditional role of medical scribes has been transcription of the patient encounter. Medical scribes may also be tasked with facilitating patient flow however these responsibilities have largely been ancillary with variable success. A pilot program was launched with an onus on executing patient flow initiatives as part of a multidisciplinary treatment team in addition to serving as documentation specialists.
Methods
A cross-sectional study comparing census and length of stay (LOS) for patients in pre- and post-scribe cohorts of two treatment zones designed to treat primarily ESI 3-5 patients at a tertiary academic emergency department was performed. Patients were stratified on treatment zone, acuity level, and disposition.
Results/Anticipated Results
The pre-scribe cohort contained 8769 patients and the post-scribe cohort contained 9827 patients. LOS for patients discharged from the mid-acuity Vertical zone decreased from 235 minutes to 223 minutes (p<0.0001) despite a 10% increase in census. With a 13% increase in Vertical admissions, LOS for admitted Vertical patients increased by 3 minutes (p=0.5324). A 14% increase in patients discharged from the low-acuity Fast Track zone was correlated with a LOS increase from 89 minutes to 95 minutes (p<0.0001). LOS for Fast Track admissions increased by 27 minutes (p=0.3351).
Conclusion
The use of medical scribes as a facilitator of patient flow is beneficial to patient throughput in a mid-acuity setting. In a low-acuity zone with faster patient turnover, utilizing scribes to execute measures to facilitate throughput in addition to documentation may be less impactful.
|
| Implementation of a Stroke Navigator: Using the EHR to improve patient care and clinical documentation
Kassi Kronfeld, MD [1], Addason McCaslin, MD [1], Brian Schenone, RN [2], Kyle Hobbs, MD [1]
1 - Stanford School of Medicine, Department of Neurology
2 - Digital Solutions Group, Stanford Health Care
|
Purpose
Comprehensive Stroke Centers (CSCs) provide the highest level of stroke care, and thus are held to a strict standard by the Joint Commission to maintain certification. Rigorous clinical metrics and documentation requirements ensure the best stroke patient care. The Stanford Stroke Navigator was designed to improve compliance with these measures, as well as reduce the need for addendum requests to resolve missing data.
Methods
The Stanford Stroke Navigator was developed by a team of physicians, nurses, and EPIC engineers. We designed a standardized electronic data form within EPIC to collect required clinical information based on the Joint Commission Stroke (C-STK) Measures at time of admission and discharge of stroke patients. We compared the number of addendum requests indicating missing documentation in the 3 months before and 3 months after the Stroke Navigator was implemented.
Results/Anticipated Results
Implementation of the Stroke Navigator reduced the number of addendum requests by 20%, indicating improved compliance with CSC requirements. In the pre-intervention group, the most common addendum request was reason the patient did not require inpatient rehabilitation. This remained the most common addendum request after the intervention.
Conclusion
The Stanford Stroke Navigator provides a streamlined work flow and improves stroke care through increased compliance with CSC metrics. Potential additional benefits include more efficient documentation by physicians, and less time manually extracting clinical data. There remains a challenge in documenting the reason a patient did not require inpatient rehabilitation after stroke.
|
| Implementing an Automated Electronic Calculator for Management of DKA/HHS
Madeline Grade, MPH [1], Sara Choi [1], Julie Chen, MD [1]
1- Stanford School of Medicine, Department of Endocrinology |
Purpose
Diabetes ketoacidosis (DKA) and hyperosmolar hyperglycemic state (HHS) require urgent action to correct fluid and electrolyte abnormalities. The current Stanford Healthcare (SHC) standard is a paper protocol outlining a multi-level algorithm for rate of insulin infusion based on prior blood glucose levels (BGLs), anion gap, and potassium levels, in addition to frequency of subsequent checks and/or management of complications. In emergency settings, this paper protocol is bulky, complex, and susceptible to human error. The objective of this study was to (1) assess the need and desire among SHC staff for a computerized DKA/HHS calculator in Epic, and (2) translate the paper protocol into a working computer model.
Methods
A survey was distributed to SHC nursing staff in March 2017 with Likert scale questions about perceived errors with the current protocol and potential utility of a calculator. The protocol was subsequently modeled in Excel and systematically tested with a variety of use cases, designed for ease of implementation as an Epic flowsheet.
Results/Anticipated Results
92% of surveyed nurses (56/61) agreed they would find an automated Epic calculator helpful for DKA management and 95% (58/61) agreed that the current paper protocol is susceptible to error with 100% believing this has safety ramifications. Further data regarding rates of error via retrospective chart review of DKA/HHS patients is forthcoming, and the calculator is under review by the SHC Epic team.
Conclusion
We identified a need and successfully developed an Epic tool for automated DKA/HHS management, which has implications for improved clinical outcomes and significant cost-savings for the hospital.
|
| Improving Advance Care Planning with a Letter Project
Ilana Yurkiewicz, MD, VJ Periyakoil, MD, Lori Klein, BCC, JD, MA, Lisa Shieh, MD, PhD
1 - Stanford Health Care, Department of Medicine
2 - Stanford Health Care, Department of Medicine and Palliative Care
3 - Stanford Health Care, Spiritual Care Service
3 - Stanford Health Care, Department of Medicine |
Purpose
Many patients lack advance care planning and spend their last days in ways not concordant with their values. This work assessed the use of a “What Matters Most†letter compared to a traditional advance directive in ascertaining patients’ end-of-life wishes.
Methods
From September to November 2016, all patients admitted to Stanford Hospital who expressed interest in advance care planning based on an initial screen were randomized to receive either a traditional advance directive or the “What Matters Most†letter. All patients were permitted to switch groups if they preferred an alternative form of advance care planning.
Results/Anticipated Results
Thirty-five “What Matters Most†letters and forty-seven traditional advance directives were completed. Patients expressed that family mattered most (71%), followed by religion (17%), hobbies (17%), and being home (17%). Life milestones mostly focused around family events (e.g. "son’s graduation," 74%), followed by travel aspirations (20%) and career ambitions (e.g. “open a barber shop,†14%). Most expressed a wish to be pain-free (81%), to be sedated if pain or distress were difficult to control (75%), and to die gently and naturally (66%). Half (49%) wanted hospice care. Most preferred the letter be used to guide their families (69%) compared to having their families override their wishes (23%).
Conclusion
Hospitalized patients were generally receptive to completing a “What Matters Most†letter in lieu of a traditional advance directive. The letter may offer a more holistic picture of patients’ values, family dynamics, and cultural backgrounds that can help providers honor end-of-life wishes.
|
| Improving Antibiotic Utilization for Late-Onset Sepsis and Necrotizing Enterocolitis in the NICU
Vidya V. Pai, MD [1], Hayden Schwenk [2], Yvonne Zorn, Pharm D [3], William Rhine, MD [1], Jenna F. Kruger, MPH [4], Betty P. Lee, Pharm D [3], Alexis S. Davis [1]
1- Stanford School of Medicine, Department of Pediatrics, Division of Neonatology
2- Stanford School of Medicine, Department of Pediatrics, Division of Infectious Diseases
3- Lucile Packard Children's Hospital, Department of Pharmacy
4- Lucile Packard Children's Hospital, Analytics and Clinical Effectiveness Department
|
Purpose
Overuse of antibiotics in the neonatal intensive care unit (NICU) has been associated with adverse clinical outcomes and the development of antibiotic resistance. For this reason, antibiotic stewardship is vital. In the Lucile Packard Children’s Hospital NICU, there is considerable variation in empiric antibiotic choices and duration of treatment for late-onset sepsis (LOS) and necrotizing enterocolitis (NEC). From January 2015 through September 2016, data obtained from the Pediatric Health Information System (PHIS) demonstrated that the NICU had nearly twice the national average of days of antibiotic therapy when compared to peer institutions. The goal of this project is to limit unnecessary antibiotic use in the NICU by establishing evidence-based antibiotic guidelines.
Methods
A team of neonatologists, pharmacists, and infectious disease specialists met to discuss an evidence-based approach to the antibiotic management of LOS and NEC in the NICU with the goal of establishing specific interventions to limit unnecessary broad-spectrum antibiotic exposure.
Results/Anticipated Results
These multidisciplinary discussions resulted in specific targets for intervention that will be implemented through an EPIC orderset, including: 1) establishing standardized antibiotic regimens for LOS and NEC, 2) instituting a 48-hour time-limit on antibiotics for initial sepsis-evaluation, and 3) limiting treatment duration to 5 days for culture-negative sepsis and 7 days for NEC.
Conclusion
With a multidisciplinary approach, it is possible to achieve evidence-based consensus around antibiotic stewardship. After implementation of these interventions, we will observe whether there is a reduction in antibiotic use. Ongoing resource analysis will demonstrate the impact of this quality improvement program on antibiotic use.
|
| Improving ICU Transfer Quality Using Standardized Documentation for Neurology Patients
Addason McCaslin MD [1], Aditya Joshi MD [1], Kassi Kronfeld MD [1], Nick Murray MD [1], Kyle Hobbs MD [1]
1 - Stanford, School of Medicine, Department of Neurology |
Purpose
When transferring complex patients out of the intensive care unit (ICU), key pieces of information may be incompletely described or overlooked, increasing the risk for mistakes in subsequent care. Standardizing the essential information communicated during transitions of care may smooth workflow, improve resident satisfaction, and augment patient safety.
Methods
We performed a retrospective assessment of the transfer summaries of all ICU patients transferred to the stroke neurology service over a 3-month period. Results were synthesized with a survey of neurology residents' experiences with ICU transfers to measure baseline quality of ICU transfer documentation. We will implement a standardized transfer document to be utilized for all ICU transfers to a primary neurology service. Another 3-month chart review and survey of residents' experiences post-implementation will be performed to determine the success of the intervention.
Results/Anticipated Results
Fifty patients with an intracranial vascular diagnosis were admitted to the Stanford ICU between 11/1/2016 and 2/1/2016. Twenty-eight patients were transferred from the medical ICU service onto the stroke neurology service. Only 3/28 patients had notes entitled "Transfer Summary", and only 1/28 used the note type "IP Transfer Summary". 16/17 residents stated that they had a near miss at least "rarely" due to insufficient transfer information. 11/17 residents at least "somewhat agree" that a shared transfer summary should be maintained.
Conclusion
Very few stroke neurology patients had formal transfer documentation. Most neurology residents had at least "rare" near-misses due to insufficient transfer information. A majority of residents favor maintaining a shared transfer summary.
|
| Improving Patient Education on Goals of Care
Natalie Htet, MD [1], Kevin Blackney, MD [2], Talha Mehmood, MD [1], Erin Hennessey, MD [2], Joshua Fronk, DO [1]
1 - Stanford, School of Medicine, Department of General Internal Medicine
2 - Stanford, School of Medicine, Department of Anesthesiology, Perioperative and Pain Medicine |
Purpose
Plans for end of life care is one of the most critical discussions between a physician and patient. Despite thorough explanations, often a knowledge gap remains for patients and families as to what cardiopulmonary resuscitation (CPR) and intubation entail. Thus, we are developing a short video demonstrating CPR and intubation. The goal is to improve the efficacy of goals of care discussions and ensure patients are making informed decisions.
Methods
This quality improvement project is early in development. We have partnered with Palliative Care and Oncology, and obtained input and acceptance from Ethics and the Patient Advisory Council. Our video will be less than 5 minutes and will include a physician explaining CPR and intubation as well as a demonstration of resuscitation and initiation of “life supportâ€Â. The video will conclude by encouraging patients to discuss their goals further with their providers. After Institutional Review Board (IRB) approval, we will utilize surveys to assess perceptions of the current goals of care process. We will then introduce the video to a select cohort of patients, and compare the intervention participants with controls on changes in code status, documented discussions about their preferences and knowledge scores.
Results/Anticipated Results
Currently, we do not have data to report.
Conclusion
We believe this video will facilitate patients in making a more informed decision on their code status and will allow providers to more easily advise patients on their end of life care.
|
| Improving Procedural Informed Consent Through Patient-Centric Videos
Greg Wallingford, MD [1], Haley Manella, MD [1], Nikita Joshi, MD [1], Samuel Shen, MD [1]
1 - Stanford Hospital, Department of Emergency Medicine |
Purpose
Informed procedural consent is considered both an ethical and legal prerequisite prior to procedural intervention in the Emergency Department (ED). The American College of Emergency Physicians (ACEP) offers guidelines for informed consent, but there is no standardized process of communicating to a patient the risks, benefits, alternatives, and indications for particular interventions. Given this lack of standardization, the general paucity of available useful visual aides, and the time constraints placed on ED providers, this is an area ripe for quality improvement.
Methods
We have developed several consent videos for a variety of procedures performed in the ED. We have begun to implement these videos as part of ED workflow and are actively seeking general feedback from physicians on this process. In the future we will gather data from physicians and patients on how these videos have improved the consent process in the ED.
Results/Anticipated Results
We hope this project will:
• Provide an accessible and efficient means for obtaining true informed consent
• Improve patient education regarding benefits and potential harms of procedures without increasing practitioner burden
• Decrease potential of litigation when there are bad outcomes
• Improve patient satisfaction and the patient experience
Conclusion
As this is an ongoing project we will continue to gather data on physician satisfaction with this style of consent. We will also gather patient data on satisfaction with how well they understood the procedure and how well it prepared them for what to expect. We will expand our procedural video library as well as translate videos into a variety of languages to better serve our varied patient population.
|
| Improving the quality of pediatric colonoscopy bowel preparations through pre-procedure patient education
Christine Yang, MD [1], Megan Christofferson, BA [1], Rachel Bensen, MD [1]
1 - Lucile Packard Children's Hospital, Division of Pediatric Gastroenterology, Hepatology, and Nutrition |
Purpose
To improve the rate of adequate bowel preps for patients undergoing outpatient colonoscopy at Lucile Packard Children’s Hospital by 10% over a 3 month period through patient education.
Methods
Adequacy of all outpatient colonoscopy preps was documented for all colonoscopies for 3 months to determine baseline state. 68% (38/47) of colonoscopy preps were adequate. During the next 3 months, a single pediatric gastroenterologist called all patients undergoing outpatient colonoscopies to provide telephone counseling the afternoon prior to the procedure. Counseling included confirming patients knew the correct amount of prep prescribed, and instructions to take additional polyethylene glycol (PEG) should stool not be clear by the evening; the same physician followed up the next day to document whether additional PEG was needed.
Results/Anticipated Results
3 weeks after beginning, an initial test of change found the adequacy rate declined to 33.3% (2/6), so content of education was modified to include education on what clear stool should look like, as well as consequences of inadequate preps. Over the remainder of the 3 months, adequate preps increased to 91% (32/35), with an overall adequacy of 83% (34/41). 51% (18/35) of patients reported they took extra PEG. Average phone call duration was 6 minutes 19 seconds (range: 2 minutes 34 seconds to 13 minutes 23 seconds).
Conclusion
Patient education in the form of brief phone counseling can greatly increase colonoscopy prep adequacy. Next steps include integration of pre-procedure telephone counseling into routine workflow, and monitoring sustainability of the intervention and rate of adequate preps for the next 3 months.
|
| Improving the Rate of Cervical Cancer Screening in the Stanford Internal Medicine Resident Clinic
Tony Cun, MD [1], Mary Christensen, RN [2], Pavianna Wilson, MA [2], Anuradha Phadke, MD [1, 2]
1 - Stanford Medicine, Department of Internal Medicine, Stanford, CA
2 - Stanford Medicine, Stanford Internal Medicine at Hoover Pavillion, Palo Alto, CA |
Purpose
To increase the rate of cervical cancer screening in the Stanford Internal Medicine Resident Clinic from a baseline rate of 46% to >65% by July 31, 2017.
Methods
We are presently using a LEAN A3 problem solving approach to analyze and address the problem of low rates of cervical cancer screening. To date, we have performed process observations, stakeholder interviews, and chart review to understand this problem. In collaboration with Stanford Internal Medicine West and Stanford Family Medicine, we launched a joint project to determine and address the root causes common to our three clinics, which are all primary care clinics at the Hoover Pavilion. Having completed this analysis, we are examining our clinics' workflow to identify areas for interventions.
Results/Anticipated Results
The five main causes of a low screening rate were no follow-up appointment was made for screening, outside medical records have been requested but not received, screening was not addressed by the provider, outside results are available but were not abstracted into our records, and screening was not appropriate. We are currently focusing on two interventions: standardizing the process of obtaining outside medical records using a tracking system and streamlining abstraction from outside records already received.
Conclusion
We have used a LEAN A3 approach to identify a number of root causes for the resident clinic’s low rate of cervical cancer screening. We are currently working on interventions in records' procurement and abstraction. Future efforts will include improving follow-up for patients who need to be scheduled for Pap smear and increasing resident provider and patient awareness of the importance of screening.
|
| Intraoperative Parathyroid Hormone Testing: The Cost of Surgical Success
Amelia Read, MD [1], Lisa Orloff, MD [1]
1 - Stanford, School of Medicine, Department of Otolaryngology- Head & Neck Surgery |
Purpose
Intraoperative parathyroid hormone testing (IOPTH) is reliably confirms successful surgery for hyperparathyroidism, utilizing the short half-life of PTH and sharp decline in PTH after resection of abnormal parathyroid glands. Point-of-care IOPTH assays have emerged, with claims of shortened operative times. This study aims to determine the efficiency of current IOPTH testing and its effect on patient care delivery at Stanford Ambulatory Surgery Center.
Methods
Retrospective review of institutional data
Results/Anticipated Results
Specimen transit, lab processing, and total result call-back times were collected for 164 IOPTH measurements over a four month period ( March - June, 2016). Average times were: transit, 20 minutes; processing, 24 minutes; total result call-back time, 45 minutes.
Conclusion
Current IOPTH testing at Stanford is performed via central laboratory system, wherein samples are transported via pneumatic tube system, processed in a multi-purpose laboratory and analyzer, and manually called into OR. Currently, average total turnaround time for IOPTH results is 45 minutes. A point-of-care IOPTH assay with reported turnaround time of 8 minutes is available, and is used at numerous academic institutions around the U.S.. Based on an estimated $20 per OR minute , adoption of this new technology would reduce the cost of parathyroid surgery by $740 per case of idle OR time. With progressively increasing parathyroid surgery volume at Stanford over the same time period and since, this assay, or adjustments to emulate its efficiency, would result in monthly savings of nearly $15,000.
|
| Novel-Web Based Algorithm to facilitate use of Emergency Contraception among Adolescents and Young Adults in the Emergency Room
Sujatha Seetharaman MD MPH, Dan Imler MD, Paula Hillard MD, Sophia Yen MD MPG
1- Division of Adolescent Medicine, Department of Pediatrics
2-Division of Emergency Medicine, Stanford University, CA
3-Department of Obstetrics and Gynecology, Stanford University, CA |
Purpose
Create a widely accessible resource for use of emergency contraception (EC) for emergency room(ER) providers, according to evidence based and expert guidelines.
Methods
This intervention was conducted in a single academic center with collaboration of Emergency Medicine, Obstetrics and Gynecology, Family planning and Adolescent medicine Departments. Evidence based guidelines were developed based on a systematic literature review of published studies and they were based on the 2016 CDC recommendations for EC. An Algorithm was created using Curbside (http://www.curbsideup.com), an open access website of clinical pathways. The process involved many steps to limit the risk of bias and incorporating expert opinions. Content was created for indications, effectiveness, advantages and disadvantages for use of the three methods of EC including the most effective Copper IUD, Ulipristal (Ella), and the second most effective EC method, followed by Levonorgestrel EC (Plan B) which is the third most effective EC. We are currently recording utilization of the EC protocol using Page views/day and use of EC among the ER providers
Results/Anticipated Results
Our web-based EC protocol facilitated a change in the hospital formulary to have access to Ulipristal in the ER and provided 24/7 access to the expert consultation of the family planning department for placement of Copper IUD in the ER. Content is freely available to anyone with internet access, as opposed to subscription based services, thereby, increasing its access. Format of the protocol allows for quick and easy update, as new evidence is available. From September 13th 2016 to April 24th 2017 the pathway was viewed by 344 unique individuals from 14 different countries (1.5/per day).
Conclusion
Integration within an Electronic Medical Record would allow study of whether use of this algorithm improved use of EC by using a multi department approach. We are currently recording utilization of the EC protocol using Page views/day and use of EC among the ER providers
|
| Optimizing Code Blue Activations
Yarl Balachandran, MD* [1], Bikram Iom, MD [2], Carter Neugarten, MD [2], Tiffany Sinclair, MD [1], Niharika Tipirneni, MD [2], Julian Vilar, MD [2], Melissa Vogelsong, MD [2], Libo Wang, MD [2]
1 - Stanford University Medical Center, Department of Surgery
2 - Stanford University Medical Center |
Purpose
Cardiopulmonary arrest (CPA) in hospitals is a true medical emergency that requires prompt deployment of hospital resources to optimize time to resuscitation. Based on prior data from a previous Resident Safety Council (RSC) Code Blue (CB) team and current SHC data, it is estimated only 25% to 40% of CB activations meet clinical criteria for a true CPA. Our goal is to increase the percentage of true CB activations by 20% come January 2018.
Methods
In order to investigate the discrepancy between CB activations and true CPAs, we attended several CB activations and asked the RN who called the code and the ICU Fellow present to complete a survey.
Results/Anticipated Results
Based on the RN survey, the majority of CBs were called in keeping with SHC policy. However, additional reasons were also identified, including needing more resources; lack of MD; rapid clinical deterioration; general “worryâ€; and vital sign abnormalities consistent with calling a Rapid Response Team (RRT). Only 63% of RNs were aware of the CB and RRT policies. The ICU Fellows judged that 85% of the CBs were true CBs.
Conclusion
Given these results, we are currently focused on 2 points of intervention: 1. Operator pages primary team for CBs and RRTs, which currently does not occur; and 2. RN educators disseminate SHC policy on CBs/RRTs and reinforce primary team notification. Our hope is that with these 2 interventions CBs will not be called when RRTs would suffice, and the overall percentage of CB activations will increase by 20% come January 2018.
|
| Provider perceptions and utilization of the What Matters Most Letter
Jessica Langston, MD MPH [1], Lori Klein [2], VJ Periyakoil, MD [3], Lisa Shieh, MD [1]
[1] Stanford Hospital, Department of Internal Medicine
[2] Stanford Hospital, Department of Spiritual Care
[3] VA Palo Alto, Department of Palliative Care |
Purpose
The What Matters Most Letter (WMML) is a Stanford-specific goals of care document meant to compliment existing resources such as POLST forms and Advanced Directives, and currently offered to all patients admitted to Stanford Hospital. Since April 2016, 96 WMMLs have been completed by hospitalized patients, though without follow-up evaluation of implementation. This project sought to characterize current awareness of the WMMLs by patient providers as well as identify barriers to their effective use in the inpatient and outpatient setting.
Methods
Surveys were completed with both inpatient providers whose patients completed a WMML during the month of January 2017 (n=19) and with primary care physicians whose patients have completed a WMML since project implementation (n=11). Questions focused on awareness of letter completion as well as utility of the letter in patient care.
Results/Anticipated Results
Our survey found significant provider interest in the WMML, with 83% of providers finding the letter moderately to extremely helpful to the care of their patient, and all outpatient providers reporting a favorable view of WMML as compared to a traditional Advanced Directive. However we also identified lack of notification and follow-through after letter completion, with only one provider having been aware that their patient had a WMML on file prior to being contacted for this survey.
Conclusion
Working with the Spiritual Care Team, we have since streamlined the notification process for both inpatient and outpatient physicians regarding completion of these documents as well as providing patient education focusing on alerting relevant stakeholders regarding their wishes.
|
| Quality Improvement Initiative to Develop a New Algorithm for Management of Fever in Children with Intestinal Failure and a Central Line
Jennifer Damman, MD* [1], Danielle Barnes, MD* [1], Colleen Nespor, RN, MSN* [1], John Kerner, MD* [1], Rachel Bensen, MD, MPH* [1]
1 - Stanford University, Department of Pediatrics, Division of Gastroenterology, Hematology, and Nutrition |
Purpose
Patients with intestinal failure (IF) are at increased risk of bacteremia and sepsis, particularly those with indwelling central venous catheters (CVC). There is a particularly increased risk of infection with gram negative (GN) organisms due to abnormal gut anatomy and small bowel bacterial overgrowth. Current standard of care for pediatric IF patients with CVC and fever at our large quaternary children’s hospital is a 48-hour hospitalization and broad spectrum gram positive and GN antibiotic coverage with vancomycin and ceftazidime. Exposure to broad spectrum antibiotics can cause adverse effects such as diarrhea, hypoglycemia with interruption of parenteral nutrition, and development of antibiotic resistance. Selection of antibiotics is crucial to promptly treat impending sepsis, reduce infectious complications, and shorten the length of hospitalization. The purpose of this quality improvement initiative is to develop an algorithm for rapid detection and treatment of sepsis with targeted antibiotic therapy in IF patients with central line and fever. The goal is to provide an approach that is standardized, evidence-based, and optimized to the specific organisms in our patient population in order to improve outcomes and reduce costs.
Methods
We retrospectively reviewed the data in IF patients with CVC and fever from January 2014 – March 2016. Data included antibiotics used and blood culture results, including organism and sensitivity. Inclusion criteria was all IF patients with indwelling CVC who were <18 years of age with fever >38.0°C in previous 24 hours. Exclusion criteria was any recent positive blood culture that had not been fully treated. Physicians from the pediatric divisions of GI, critical care, emergency department, and infectious disease created an algorithm based on this clinical data that is available from any web browser via the CurbSideUp.com program.
Results/Anticipated Results
A total of 180 cases were identified with hospitalization for fever in an IF patient with a CVC Ninety-two (51%) had a positive blood culture. Of these positive cultures, 52 grew gram positive organisms, 35 grew GN organisms, and 8 grew yeast. Of the 35 GN isolates (see figure 1), 19 were SPACE organisms (Serratia, Pseudomonas, Acinetobacter, Citrobacter, and Enterobacter). Of the 19 SPACE organisms, 7 were inadequately covered by ceftazidime, but would have been covered by the broader antibiotic cefepime. The remaining 10 SPACE organisms were sensitive to all cephalosporins. Of the 16 non-SPACE organisms, 14 were sensitive to all cephalosporins. We found that ceftazidime provided no benefit over the narrower antibiotic ceftriaxone in patients who grew non-SPACE organisms. Therefore, we concluded that for IF patients with no recent history of high-risk isolates, empiric coverage for GN organisms can be narrowed to ceftriaxone, with the ease of single daily dosing. Of those patients who grew SPACE organisms, many that were inadequately covered by ceftazidime would be more appropriately covered with cefepime. Therefore, for patients with history of high-risk isolates in previous 3 months, empiric coverage should be broadened to cefepime.
Conclusion
Our experience demonstrates continued high risk of bacteremia in children with IF and CVCs. A quality improvement approach can be used to standardize and optimize treatment for these high risk patients. Future plans include evaluation of outcomes after implementation of the algorithm and consideration of observation without empiric antibiotics in select IF patients with CVC and fever who are otherwise well appearing with normal vital signs, laboratory studies, and an alternate source of infection.
|
| Recognition, Diagnosis, and Management of Postpartum Urinary Retention
Rosa Liu, MD, Jussely Morfin, BS, Carolyn Cruz Kerr, MD, James McCarrick III, MD
Santa Clara Valley Medical Center, Department of Obstetrics and Gynecology |
Purpose
To create a best practice for postpartum bladder care pertaining to a county hospital population.
Methods
We created a guideline for residents, attendings, and postpartum nurses using available data on postpartum bladder function for management of postpartum urinary retention (PPUR). Elements included length of time after delivery after which a postpartum patient should void, interventions to aid in voiding, education to providers on PPUR symptoms, nursing competency training for bladder scanning, algorithm-guided management dependent on postvoid residual volume (PVR), standardized documentation, and creating order sets for coordination of care.
Results/Anticipated Results
We reviewed all 51 cases of PPUR from our institution between July 2015 to February 2017. Patient characteristics included vaginal delivery (86%), epidural (84%), any laceration (71%), labial swelling (47%), covert urinary retention (33%), and postpartum tubal ligation (6%). 43% of patients had PVR>1000ml, of which 12% failed voiding trial, requiring a temporary leg bag. Only 0.04% had positive urine cultures; therefore urine culture was removed from the algorithm. 97% of nurses deemed the protocol as a good resource for patient care on a survey distributed on the postpartum ward.
Conclusion
Our incidence of PPUR is among the lowest reported in literature at 0.9%. By implementing this PPUR guideline, we have heightened awareness of this issue department-wide and applied consistent management. After 20 months of utilization, we have identified risk factors pertaining to our population, fine-tuned our protocol to decrease unnecessary usage of resources, and have revised doctor-nurse communication for postpartum bladder care. .
|
| Rethinking and Redesigning the Anesthesiology Paper Record Using Quality Improvement Methodology
Felipe D. Perez [1], Dave Creighton [1], Andrew J. Giustini [1], Wendy Ma [1], Sean Paschall [1], Ashley Peterson [1], Noelle Wilson [1], Jocelyn Wong [1], Kristen Telischak [1], Ruth Fanning [1], Cliff Schmiesing [1]
1 - Stanford, School of Medicine, Department of Anesthesiology, Perioperative and Pain Medicine |
Purpose
The majority of academic medical practices utilize electronic medical charting and seldom rely upon paper records. The paper record is typically used only during electronic medical record (EMR) downtime and environmental emergency situations. Our current paper record needs to be updated to meet the current standards of practice with respect to patient care, medication reconciliation, billing and medical record documentation.
Methods
A group of eight anesthesiology residents with the guidance of the anesthesiology quality improvement faculty used the PDSA cycle to help create a new paper record. Using design-thinking principles, we have assessed stakeholders’ needs while soliciting feedback on the current draft. We will study the efficiency and ease of use, by videotaping residents and faculty using a standardized patient and comparing the current record with the new paper record. Completed paper records will be iteratively reviewed by stakeholders to create the optimal paper record.
Results/Anticipated Results
The quality improvement project is in progress at this time.
Conclusion
Like all modern practices, we are highly dependent on our EMR. In case of a natural disaster or during EMR downtime, existing paper records are time consuming to complete, distracting to patient care, subjective in their nature, and difficult for physicians caring for the patient at a later time to interpret. A new anesthesia paper record created with current standards in mind will address these significant patient safety concerns.
|
| Rounding Optimization, Satisfaction and Efficiency (R.O.S.E.) Initiative: An Implementation of Nursing Integrated Rounds in a Culturally Naive PICU
Michelle Olson, MD [1], Lindsey Troy, MD [1], Kirby Jacobs, RN, BSN, CCRN [2], Miranda Schmidt, MSN, RN, CCRN [2], Stefanie Zeihen RN, APNP [2], Saraswati Kache, MD [1], Kate Steffen, MD [1]
1- Stanford, School of Medicine, Department of Pediatrics, Division of Critical Care
2- Lucile Packard Children's Hospital Stanford, Pediatric Intensive Care Unit |
Purpose
Daily rounds in our pediatric ICU were inefficiency with frequent interruptions and numerous clarifications after rounds leading to poor staff satisfaction and less than optimal patient care. Our specific aims are in 6 months to reduce rounding times by 5% and improve satisfaction by 30%.
Methods
A self-identified, multidisciplinary team assembled to address the problems of poor communication and inefficiency. Multiple assessments occurred through direct observation, interviews and a survey to providers to understand current attitudes regarding efficiency, participation, team trust, content and comprehension of rounds. A root cause analysis was performed using a fishbone diagram. Taking into account the culture and organization of our unit, we designed a nursing integrated rounding system to develop a team shared mental model. The bedside nurse will provides current objective data followed by the frontline provider performing an assessment and plan by organ system. Initiation occurred after extensive education.
Results/Anticipated Results
Average rounding time was 127 minutes prior to nursing integrated rounds. Implementation was in January 2017 and we are performing PDSA cycles to improve flow and structure of individual presentations as well as unit flow. After three months, we are seeing a high rate of compliance on a daily basis.
Conclusion
We have received positive and constructive feedback from all levels of the team and plan on performing a 6 month follow up survey to evaluate provider satisfaction. We do not yet have evidence of improved efficiency though our findings are promising that the PICU is developing a culture that values a team-based approach.
|
| Standardization of Labor and Delivery Stroke Codes to Reduce Time to Initial CT Scan
Nina Bozinov, MD [1], Bryan Lanzman, MD [2] , Maurice Druzin, MD [3], Sarah Lee, MD [1]
1 - Stanford, Department of Neurology
2 - Stanford, Department of Neuroradiology
3 - Stanford, Department of Obstetrics & Gynecology
|
Purpose
Pregnant and postpartum patients are at risk for hemorrhagic stroke, and timely diagnosis can significantly affect management and outcomes.1 Stroke codes activated by labor and delivery are run by SHC neurology residents despite the patient being admitted under LPCH, which can lead to delays to neuroimaging due to the lack of intercommunication between LPCH and SHC EPIC systems. With input from obstetrics, neuroradiology and stroke neurology, we set out to create a standardized stroke code pathway to reduce time to CT scan.
Methods
We conducted a survey of neurology residents to determine their current level of familiarity with L&D stroke codes. The obstetrics department provided input on current practices in patients with new neurologic deficits.
Results/Anticipated Results
Pretest: Of the 17/26 neurology residents who completed the survey, 82% knew where L&D was located and had a login for LPCH EPIC. Only 18% had ever run an L&D stroke code. 53% were unsure where to obtain neuroimaging, and 71% did not feel confident about imaging options in this patient population.
Conclusion
After evaluating potential solutions, it was determined the best immediate pathway was for all antepartum and postpartum patients to be taken to SHC for urgent CT imaging. Orders can be placed either as an outpatient or a paper order. Ongoing collaborations with the radiology department and EPIC support team are underway to determine the best method of obtaining imaging in SHC while patients are admitted to LPCH.
|
| Streamlining a Blood Center and Hospital Transfusion Service Supply-Chain with an Informatics Vendor-Managed Inventory Solution
Hamilton Tsang, MD* [1], David Lancaster [2], Dianne Geary [2], Robert Scott [2], Anh Thu Nguyen [2], Adam Garcia [2], Raina Shankar [2], Leslie Buchanan [2], Tho Pham, MD [1]
1 - Stanford, School of Medicine, Department of Pathology, Transfusion Medicine
2 - Stanford Blood Center |
Purpose
The current ordering process at our institution involves twice-daily shipments from the blood center(BC) to the hospital transfusion service(HTS), with shipments predicated upon current stock levels. Manually census is time-consuming and error-prone. We aim to enhance inventory management by developing an informatics solution to streamline the ordering process and reallocate tech time.
Methods
The general inventory accounts for >50 product categories based on component, blood-type, irradiation, and CMV-serology status. Over a 5-month calibration period, we algorithmically optimized an electronic ordering process to reduce labor time. We created in-house programs on Visual Basic for Applications to determine inventory levels and implemented a 3-month pilot period.
Results/Anticipated Results
Our system showed non-inferiority while saving labor time. The average weekly transfused:stocked ratio for cryoprecipitate, plasma, and RBC, respectively, were 1.03, 1.21, and 1.48 before the pilot period compared with 0.88, 1.17, and 1.40 during (p=0.28). There were 27(before) and 31(during) average stat units ordered per week (p=0.86). The number of monthly wasted products due to expiration were 226(before) and 196(during) units, respectively (p = 0.28). An estimated 7 hours per week of technologist time was reallocated to other tasks. This translates to 0.175 FTE and $18,200 per year saved from labor costs per year if permanently adopted.
Conclusion
We created an in-house electronic ordering system to enhance information fidelity, re-allocate technologist time, and further standardize ordering. Our system showed non-inferiority to the labor-intensive manual system. This is achieved while freeing up over 360 hours of staff time per year.
|
| The effect of predicted toxin reporting with PCR on management of suspected C. difficile infection
Matthew Hitchcock, MD, MPH [1], Marisa Holubar, MD [1], Richard Wilson [2], Niaz Banaei, MD [1, 3]
1 - Stanford University, School of Medicine, Department of Medicine, Division of Infectious Diseases and Geographic Medicine
2 - Stanford Health Care
3 - Stanford University, School of Medicine, Department of Pathology |
Purpose
Detection of toxin genes cannot differentiate C. difficile colonization from infection (CDI), but literature suggests that toxin detection can. At Stanford, the Xpert tcdB PCR is used to detect C. difficile. In-house studies showed that toxin positivity can be predicted with high sensitivity at a predefined Xpert cycle threshold. In October 2016, toxin prediction was added to the PCR result. When negative, a comment recommends against CDI therapy. We evaluated the impact of this reporting change on management and outcomes in PCR-positive patients.
Methods
All patients tested from Oct. to Dec. 2016 were included. Chart review was performed and patients were followed until Jan. 2017. Clinical data was evaluated for differences at testing between those with predicted toxin positive (pTox+) and negative (pTox-) results. Variables were analyzed with the Chi square and t-tests using SPSS and α was 0.05.
Results/Anticipated Results
Eighty-nine of 574 (15.5%) tests were positive, with 54 (60.7%) pTox+ and 35 (39.3%) pTox-. All pTox+ patients received therapy, as did 26 (74.3%) pTox- patients. All untreated patients had resolution of diarrhea and there were no readmissions. There was no difference in mean WBC, creatinine, or stools/day at testing, though these observations were limited by sample size and missing data.
Conclusion
During this study period, therapy was avoided in 25.7% of pTox- patients. No poor outcomes were noted in follow-up. Most pTox- patients still received CDI therapy. Alteration in clinician behavior to avoid unnecessary therapy will likely require active stewardship intervention.
|
| Thrombophilia Testing in the Inpatient Setting: An Educational Intervention
Henry Kwang, MD[1], Eric Mou, MD [1], Ilana Richman, MD [1], Andre Kumar, MD [1], Caroline Berube, MD, FRCP(C) [2], Neera Ahuja, MD [1], Stephanie Harman [1], Tyler Johnson, MD [3], Neil Shah, MD [4], Mehran Teymourtash [5], Ronald Witteles, MD [6], Lisa Shieh*, MD [1], Jason Hom, MD [1]
1) Stanford, School of Medicine, Department of Medicine
2) Stanford, School of Medicine, Department of Medicine, Division of Hematology
3) Stanford, School of Medicine, Department of Medicine, Division of Oncology
4) Stanford, School of Medicine, Department of Pathology
5) Stanford Health Care, Department of Finance
6) Stanford, School of Medicine, Department of Medicine, Division of Cardiology
|
Purpose
Thrombophilia testing is frequently ordered in the inpatient setting. Testing is costly and can be misleading with acute thromboembolism or concurrent anticoagulation use.
Methods
We conducted an educational intervention with a randomized cross-over design for Internal Medicine interns at Stanford during academic year 2014. Interns were randomized into two groups that received interactive sessions during the first or second half of the year (early and late intervention). The lecture focused on evidence-based guidelines regarding thrombophilia workups.
We performed chart review of all inpatient thrombophilia tests from academic year 2013 to 2015. We analyzed 1) intervention group ordering, 2) intern ordering during the prior academic year (historical control) and 3) non-intervention services ordering services during the intervention year (contemporaneous control).
Results/Anticipated Results
Breaking down discrete periods of pre-early intervention, pre-late intervention, and post-both interventions, the early-intervention group had 40.0%, 36.4%, 16.1% inappropriate orders while the late-intervention group had 40.0%, 38.1%, 23.3% inappropriate orders. Analyzing early and later intervention groups collectively, inappropriate ordering decreased from 20/51 orders (39.2%) pre-intervention to 23/96 orders (24.0%) post-intervention.
Both historical and contemporaneous control groups did not demonstrate improvement with the former ordering 35.7%, 33.78%, 40.0% inappropriately and the latter ordering 42.2%, 44.1%, 42.8% inappropriately over their respective time periods.
Conclusion
During the period after the early intervention but before the late intervention, there was no difference in ordering patterns between intervention groups. However, both early intervention and late intervention groups had improved significantly over time during the final follow-up period, an improvement that was not seen in either control groups.
|
| Transfers of Care from the E.D. to the Floor: How Sick is This Patient?
Ilana Yurkiewicz, MD, Frank Chen, MD, Katie Doering, MD, Kathleen Jia, MD, Silvia Loica-Mersa, MD, Anisha Mazloom, MD, Jason Qian, MD, Josi Schwann, MD, Arpeet Shah, MD, William Shomali, MD, Chris Winstead-Derlega, MD, Michael Zheng, MD
Stanford Health Care, Resident Safety Council |
Purpose
Transfers of care are a frequent occurrence in Stanford Hospital and take many forms. One of the most prominent – and frequent – is from the emergency room to the medicine floors. While these transfers of care are common, they sometimes entail miscommunication and misconceptions of a patient’s clinical status. The purpose of this work was to look into current use of IPASS as a validated tool to perform patient handoffs.
Methods
We evaluated the current use of IPASS during handsoffs from physicians in the emergency room to internal medicine residents. Forty sign-outs were observed during a two week block of night shifts.
Results/Anticipated Results
Among 40 sign-outs, only the “P†(patient summary) portion of IPASS was communicated 100 percent of the time. The other aspects of IPASS performed significantly worse. Illness severity was only communicated in 5 cases (12.5%), action list in 12 cases (30%), situation awareness in 5 cases (12.5%), and synthesis by receiver in 10 cases (25%).
Conclusion
While all residents are trained in IPASS use as a way to do handoffs, this only occurs a fraction of the time. Better implementation of IPASS has the potential to improve transfers of care and reduce miscommunication among different medical teams.
|
| Use of PHQ-9 screening at the Stanford Adolescent and Eating Disorder Clinics as a screening tool for depression
Sujatha Seetharaman MD MPH, Neville Golden MD, Cynthia Kapphahn MD MPH, Bonnie Halpern-Felsher PhD, Jennifer Carlson MD, Rachel Goldstein MD, Jonathan Avila MD
1- Division of Adolescent medicine, Department of Pediatrics, Stanford University |
Purpose
The goal of this quality improvement project was to implement depression screening into the Stanford Teen and Young Adult Clinic and Eating Disorders Clinics using the PHQ-9 questionnaires for all new patient and health maintenance visits.
Methods
With the help of the Epic team, a tab on the Epic flow sheet section for patient health questionnaires was created to record the questions and answers of the PHQ-9 screening questionnaire completed by the patient. The total score of the PHQ-9 screen was automatically calculated by Epic and was pulled into the Synopsis section along with the vitals. Clinicians could then easily view this information. A tab was created to add the PHQ-9 total score and whether or not social work was consulted in to the patient’s progress note. Providers and staff in the clinic were trained on the PHQ-9 screening procedure and documentation.
Results/Anticipated Results
During the per-intervention period (June 1st 2016 - October 30th 2016), there were 34 PHQ-9 screens/100 new patient visits (N = 29 of 85 medical visits), and during the post-intervention period (November 1st 2016- March 30th 2017), there were 93 PHQ-9 screens/100 medical visits (N = 89 of 96 medical visits; X2 = 65.6, df = 1, p < 0.00001). Among the patients who were screened during the post-intervention period, the PHQ-9 screen was reviewed by the provider, documented in the Epic progress notes for 100% of the new patient visits, and used to inform treatment for 98.7% of the new patients.
Conclusion
Screening for depression in adolescents at new patient and health maintenance visits utilizing the PHQ-9 questionnaire is feasible and can facilitate referral of depressed adolescents to mental health services
|
| Utilizing Precise Locator Technology to Measure Physician-Nurse Rounding Habits
Adam X Sang MD[1], Becca L Tisdale MD[2], Julia E Noel MD[3], Silvia Loica-Mersa MD[2], Ian Chong MD[4], Travis Miller MD[5], Blake M Read MD[1], Lisa Shieh, MD PhD FHM[6]
1 - Department of Surgery, Stanford
2 - Department of Internal Medicine, Stanford
3 - Department of Otolaryngology, Head & Neck Surgery, Stanford
4 - Department of Emergency Medicine, Stanford
5 - Division of Plastic Surgery, Stanford
6 - Medical Director for Quality, Department of Internal Medicine, Stanford |
Purpose
The quality of communication between physicians and nurses plays an important role in patient safety and provider job satisfaction. The purpose of this project was to utilize existing hospital technologies to accurately and quantitatively assess MD and RN rounding habits. Based on our focus groups, time/workload constraints and coordination challenges are the main drivers of poor MD-RN overlap during morning rounds. By studying MD and RN rounding habits precisely, we will be able to design specific, targeted interventions to improve MD-RN face time.
Methods
Hospitalist physicians and nurses at our academic institution were given small, portable locator devices to wear. These devices turn on a specific colored light (white for nurses and green for physicians) when the wearer enters the room. In addition, our information technologists are able to precisely and automatically record when each locator device enters and leaves a specific patient room, thereby generating a temporal-spatial map of both MD and RN activity. We recorded and analyzed the rounding habits of MDs and RNs in three separate in-patient wards over a period of 90 consecutive days.
Results/Anticipated Results
Over the 90-day period, 739 MD rounding events were captured. The average MD rounding time was 7.3 minutes (±0.35 min, range 1 to 41 min). Importantly, of these rounding events, 42.1% occurred with an RN at bedside. The average length of MD-RN face time overlap was 4.1 minutes. Interestingly, the distance from the nursing station (located at the center of each linear ward) to the individual patient rooms did not have a noticeable effect on the frequency of RNs rounding together with the MDs (p = 0.37, 0.23, 0.48 for each ward). In addition, the length of rounds in which an RN was present (8.0±0.29 min) was longer than rounds with MDs only (6.2±0.22 min) (p < 0.0001), although this difference is likely associative rather than causal.
Conclusion
This approach has successfully generated precise, quantitative data on MD and RN rounding habits that have been previously unattainable. Altogether, these novel data directly contribute to our understanding of MD-RN rounding and overlap habits, and can be used to both guide future quality improvement projects as well as quantitatively measure their impact.
|
| Variability of Human Prion Disease Policies and Procedures in Leading US Neurological Institutions
Praveen Tummalapalli, BS [1], Katherine Werbaneth, MD [2], Carl A. Gold, MD, MS [2]
1 - Stanford University School of Medicine, Palo Alto, CA
2 - Department of Neurology & Neurological Sciences, Stanford University, Palo Alto, CA |
Purpose
Prion diseases are progressive, fatal encephalopathies spread by prions. These prions can be transmitted through contact with infected tissue or contaminated medical instruments. Little is known about human-to-human transmission via CSF exposure. Typical sterilization techniques fail to inactivate prions, meaning healthcare facilities must take special care to ensure that samples, specimens, and instruments from known or suspected prion disease patients are handled safely to prevent the transmission of prion diseases to healthcare workers and patients. Although the WHO and CDC have issued guidelines to prevent iatrogenic transmission, it is unclear whether these guidelines have been universally adopted and to what degree institutions differ regarding prion disease infection control policies.
Methods
We will request policies and procedures for patients with suspected prion disease from US News and World Report top 50 Neurology/Neurosurgery institutions in 2016. These protocols will be compared regarding precautions for surgical procedures and lumbar puncture. Policies will also be analyzed for variance in sterilization methods for instruments used in such procedures, the handling and disposal of tissue and CSF specimens of these patients, and precautions for the autopsy and embalming of the bodies of deceased prion disease patients.
Results/Anticipated Results
The study is in progress. In the pilot data, there is great variability in precaution guidelines for lumbar puncture. Further information will become available as we gather more data.
Conclusion
Institutions differ significantly in their protocols regarding the handling of prion disease patients, specimens, and instruments. This has important implications for potential iatrogenic transmission of prion disease to healthcare workers and patients.
|
| WikiConsult: a crowdsourced online encyclopedia of best practices for clinical consults
Ron C. Li, MD [1], Saurabh Gombar, MD PhD [2], Robert Rope, MD [1], Kai Swenson, MD [1], Laurence Katznelson, MD [1]
1 - Stanford School of Medicine, Department of Medicine
2 - Stanford School of Medicine, Department of Pathology |
Purpose
Clinical consultation is a critical component of patient care and housestaff training. Consultations provide housestaff education about specialties outside of their area of practice and foster skills in team-based care. We created WikiConsult, an online encyclopedia maintained by residents and fellows to share information about common consult questions. We aim to improve the efficiency and educational value of the consultation process by increasing transparency regarding how to effectively ask consult questions.
Methods
WikiConsult is built on the Stanford medwiki platform (medwiki.stanford.edu/display/wikiconsult). Content is submitted by chief residents and fellows and curated by the WikiConsult team. The website will be available for all housestaff to use at the start of the next academic year as a resource when calling consults. Pre-intervention data on housestaff attitudes towards the consult process were collected via anonymous, online surveys. Post-intervention data will consist of email-based surveys, user comments, site use metrics, and selected in-person interviews.
Results/Anticipated Results
367 Stanford housestaff responded to the pre-intervention survey. Responses were measured on a 5 point Likert scale (1 = never, 5 = always). 15% of respondents sometimes, rarely, or never felt adequately prepared to formulate and present a consult question to a consultant and 34% recalled that consultants sometimes, usually, or always declined or delayed seeing consults because they were told that the consult was inappropriate or incomplete. When answering consults, 84% of respondents recalled that crucial information was sometimes, usually, or always missing from a consult.
Conclusion
Inadequate understanding of how to effectively frame consult questions among housestaff may negatively impact the efficiency and educational value of clinical consultations. WikiConsult may improve this process by providing an online space to capture and share useful knowledge for calling effective consultations.
|