Our Communications Manager, Katie M. Kanagawa, asked our Spring 2020 Epidemiology & Clinical Research PhD Program graduates three questions about their current work (post-graduation), their research in the Department of Epidemiology & Population Health, and if they have a message they would like to share with our academic community. Here are their responses.
What have you been up to since graduating from the Epidemiology & Clinical Research PhD Program?
- Katherine Holsteen: Since graduating, I have enjoyed spending the majority of my time taking care of my son, who just turned 1. We moved from Stanford to the Denver, CO area in June, and bought a house in August, so my husband and I have also spent time working on many home improvement projects. In my spare time, I am working from home as a part-time epidemiology consultant. I helped Prof. Lori Nelson and Lesley Park to design a new Big Data Methods course, reconnected with my undergraduate research advisor to contribute to his model robustness work, and, most recently, I am excited about new opportunities to help the Stanford WELL research team launch a wearable study and to begin a long-term role for an epi consulting firm.
- Eileen Leary: I’ve been improving our understanding of sleep medicine through patient centric, real world evidence research. As a Clinical Scientist at Jazz Pharmaceuticals, I leverage my sleep knowledge to identify and prioritize knowledge gaps related to sleep disorders and their treatments. Then I design and conduct phase IV studies to fill those gaps. As a consultant, I provide scientific support to teams bringing innovation to the sleep field.
- Stelios Serghiou: As of August, I have been working as an AI (Artificial Intelligence) Resident at Google Health. This is a fixed 1.5 year position during which I am expected to join any groups within Google Health of interest to me and produce research at the intersection of medicine and machine learning. At the moment, I work with a team interested in using data uniquely available to Google to study and utilize social determinants of health to empower all patients, clinicians and public health officials to achieve and promote high quality and equitable health and wellbeing. I also remain involved with Stanford, where I have been helping in activities related to promoting more open, rigorous and reproducible research, which are initiatives close to my heart!
- Yuan Jin Tan: After graduating in June 2020, I took a couple of months off and dabbled in several pandemic hobbies, including breadmaking and gardening. In August 2020, I joined Keystone Strategy as a Consultant. Keystone is a boutique consulting firm that focuses on tech and digital transformation. We work directly with clients seeking to innovate using tech-centric strategy and business models, both within the tech industry and in adjacent industries such as healthcare. Notable clients include Microsoft, Intel, Verizon, Pfizer, Roche, and Pepsico. So far, I have worked on cases in gaming, communications, and advertising, and am involved in building out the firm's digital healthcare practice. I'm also glad to be putting my grad school knowledge to good use at work - in my day to day, I frequently have to run analyses in R, interpret regressions, draft reports, and manage projects.
- Christophe Tchakouté: I joined the 23andMe research team two months before defending my dissertation. I help develop methods that enable the use of non-genetic (“E”) data to improve predictions and inference at 23andMe and I also support the therapeutic team in their clinical trials' designs.
Can you please give a brief summary of your PhD work.
- Katherine: My dissertation title was "Strengthening the study of migraine headache triggers through mobile health tools and n-of-1 methods." My project consisted of designing, executing, and analyzing a 90-day study of 290 patients with migraine. We collected daily survey data through an iPhone app, steps/heart rate data through an Apple watch, and daily weather data. The primary goals were to (1) try to predict migraine onset based on potential trigger factors and (2) assess correlations between steps/heart rate metrics and migraine occurrence. The results were mostly null, but the personal results, measured by my own learning and growth, were very significant (p = 0.000002) ;-)
- Eileen: For my dissertation, I conducted an in-depth investigation on the relationship between REM sleep and mortality and explored whether another sleep stage could be a better predictor of mortality. I used traditional statistics and machine learning for the main analyses, then ran multiple sensitivity analyses to rule out alternative explanations. Next, I evaluated the generalizability by replicating the final models in two independent cohorts. Finally, I performed a meta-analysis to provide a global quantification of the hazard ratios.
- I found a strong, consistent association between lower levels of REM sleep and mortality rate in middle-aged and older adults. The finding persisted for different causes of death (e.g. cardiovascular, cancer) with similar results in the independent cohorts. The meta-analysis yielded an overall hazard ratio of 1.13 (95% CI 1.10–1.17) for all-cause mortality and 1.10 (95% CI 1.03–1.16) for cardiovascular mortality for every 5% reduction in REM sleep. While it would be premature to make any clinical recommendations based on this study, the results support the critical role sleep plays in health outcomes.
- Stelios: As a PhD student, I worked with Prof. John Ioannidis on developing computational tools to improve our ability to study the biomedical literature at scale. Amongst others, our work extracted indicators of transparency, such as sharing data and code or reporting conflicts of interest, across 2.75 million articles creating maps of transparency across the entire open access literature on PubMed.
- Yuan Jin: During grad school, I worked on a variety of metaresearch projects with John Ioannidis, Manuel Blum, a former Epi MS student, and Ryan Crowley, a BS/MS student in Biomedical Computation. These projects involved evaluating research practices and risk of bias in several fields, including veterinary comparative clinical trials, sensitivity analyses in observational studies, and machine learning diagnostic tools. The work has been published in Scientific Reports, International Journal of Epi, and JAMIA.
- Christophe: The first chapter of my dissertation used a target trial approach to assess the impact of in-utero HIV exposure on child mortality and morbidity in sub-Saharan Africa. The second chapter leveraged pharmacy records to predict immunological and virological outcomes among people living with HIV at Kaiser Permanente Northern California. Finally, my third chapter was a method paper in which Prof. Mike Baiocchi and I showed that when combining the transportability framework with Empirical Bayes, we improve the estimation of the average treatment effect in a population with heterogeneous treatment effects.
Is there anything you would like to say to the E&PH community? Any messages or advice you would like me to share?
- Katherine: I am thankful for the opportunity to complete a PhD in Epi at Stanford: to be part of a small and supportive department, to have knowledgeable and caring professors, and many opportunities for growth as a researcher and as a person. Thank you, E&PH! I'd like to encourage current students to focus on learning from the process rather than on the results, to proactively seek help when you need it, and to not give up!
- Eileen: Start building your network and brand now. Be known for who you are and what you do, not where you work. Make sure your web presence is consistent.
- Stelios: I thoroughly enjoyed my time at Stanford and if I were to share any pieces of advice, those would be to do an internship while a PhD student whether you are certain you want to be an academic or not, and to also venture out of the coursework available within our department and experience the world through the lens of other data-driven fields, like econometrics, statistics, and computer science. I still feel part of the Stanford family and look forward to getting to know all students and being of help wherever I can!
- Yuan Jin: Thank you for the hard work that you do, especially in the midst of a global pandemic, during which the impact of your work is especially apparent and tangible. These days, I think a lot about the responsibility we all have to not only communicate effectively about facts and science, but also to reach beyond academic circles and speak directly with the general public. To the current students in the program, I'm excited for what lies ahead for you. Though school can be difficult at times, I'm sure you are all bound for great successes. I also want to encourage you to treasure your time at Stanford, to step outside of the classroom once in a while, to broaden your horizons and to leverage all the opportunities that are available to you. School should be a time for meeting people, for exploration, for making mistakes, and for learning from them - and I'd urge you to do so boldly and with insatiable hunger.
- Christophe: I am so grateful for my time at Stanford E&PH. I had the opportunity to take classes in Engineering, Economics, and Computer Science. I believe those experiences helped me ask better scientific questions and work better with people from other disciplines. I also want to say a big thank you to all my professors in the E&PH department, especially Prof. Sainani and Prof. Baiocchi. My advice to any trainee in the department is that there are multiple ways to make an impact, and life is never a linear process. When it feels scary to jump, that is exactly when you jump.
Click on the images below to read more
Katherine Holsteen, PhD, Independent Research Consultant, Epidemiology.
Eileen Leary, PhD, RPSGT, Clinical Scientist at Jazz Pharmaceuticals.
Stylianos (Stelios) Serghiou, PhD, MS, AI Resident (Healthcare) at Google.
Yuan Jin Tan, PhD, Consultant at Keystone Strategy.
Christophe Tchakouté, PhD, Scientist/Biostatistician at 23andme.