Fellow Spotlight: Meet Jenna Wong, PhD, MSc
Meet Jenna Wong, PhD, MSc, a post-doctoral research fellow and Pyle Fellowship Awardee. Dr. Wong received her MSc in Epidemiology from the University of Ottawa and her PhD in Epidemiology from McGill University. Her research focuses on predictive modelling applications in epidemiology using electronic and linked administrative health data.
Under the mentorship of Professor Darren Toh, she is exploring the use of machine learning to extract information from unstructured electronic health data to enhance pharmacoepidemiologic research. But did you know she almost didn't pursue a career in academia? She told us about the path she almost chose, about her work and the impact she hopes to make, and the important role that focus has played in her academic career.
Q: Tell us about your path to becoming a post-doctoral fellow at DPM. What was your journey like?
A: “Fortuitous” is the word that comes to mind when I think back on my path to becoming a post-doctoral fellow at DPM. At the time, I was in the midst of finishing up my doctoral work at McGill University when I received an email from one of my professors (thanks, Dr. [Robert] Platt!) circulating a job posting for a post-doctoral fellowship with Professor Darren Toh at DPM. The job posting was seeking a candidate interested in analyzing complex electronic healthcare data and leading comparative effectiveness and safety studies or patient-centered outcomes projects. I had been looking for an opportunity to continue my doctoral work on off-label antidepressant use, and this fellowship sounded like an ideal position for me. I quickly put everything on hold to apply, which was followed by a brief phone interview soon after and an extensive in-person interview several months later. During my in-person interview, I met several faculty and post-docs and was impressed by the level of professionalism and camaraderie among people in the department, which left a lasting impression on me. When I was offered the fellowship shortly after, I was excited and grateful for the opportunity. Accepting this position was a big decision for a number of reasons, but looking back, I know without a doubt that it was the right decision for me.
Q: What kind of research do you do? What sort of impact do you hope that it will make?
A: A major theme throughout all my research has been the use of predictive modeling techniques to leverage information in electronic health data to help clinicians and institutions make more informed decisions for patients. Prior to my post-doctoral fellowship, my research focused on building prediction models using structured electronic health data. I am now broadening the scope of my research to build prediction models that automatically extract information from unstructured clinical text, which represents a rich but more complex data type that is underutilized in epidemiologic research.
Whether the impacts of my research on helping inform healthcare decisions are direct (e.g., creating a model that continually updates a patient’s daily risk of death during a hospitalization so clinicians and institutions can better forecast and mobilize resources to meet their patients’ changing needs) or indirect (e.g., creating models that predict important missing study covariates to improve the validity of epidemiologic analyses informing patient care), I hope my research enhances the capacity of clinicians and epidemiologists through models that provide them with valuable information from data that would otherwise not be available without the use of such predictive modeling methods.
I hope my research enhances the capacity of clinicians and epidemiologists through models that provide them with valuable information from data that would otherwise not be available without the use of such predictive modeling methods.
Q: Tell us about your fellowship project(s). What inspired this/these project(s)? How will these projects help you meet your career goals?
A: In a single sentence, my post-doctoral work is focused on enabling the use of electronic health record data to compare the effectiveness and safety of approved (“on-label”) and unapproved (“off-label”) drugs for insomnia through building models that extract key covariates and outcomes from unstructured clinical text using state-of-the-art machine learning techniques from the field of natural language processing. My interest in off-label drug use for insomnia arose through my doctoral work, which focused on antidepressant use for off-label indications and creating models to predict antidepressant treatment indications (a key variable not directly captured in most conventional data sources used in large-scale pharmacoepidemiologic studies). I was struck by one of my main findings–that trazodone use for insomnia was the most common off-label use for antidepressants despite little evidence to support its effectiveness and safety for insomnia compared to other approved medications, like zolpidem. As many of the key covariates and outcomes needed to evaluate insomnia drugs are more likely captured in unstructured clinical text rather than conventional structured electronic health data, I realized that building prediction models to automatically extract this information from unstructured clinical text could make invaluable contributions to enabling large-scale pharmacoepidemiologic studies on off-label drug use for insomnia. I am grateful to be working on a project that I feel so passionate about, both methodologically and substantively, and the skills and lessons learned from this project will help me move toward my career goals of using advanced predictive modeling techniques with structured and unstructured real-world data to enhance the capacity to study “data-challenging” research questions, including those related to off-label drug use.
I am grateful to be working on a project that I feel so passionate about, both methodologically and substantively, and the skills and lessons learned from this project will help me move toward my career goals.
Q: Which Institute faculty member(s) is/are you working with? How did you come to meet them?
A: I am working with Professor Darren Toh. Although he may not know it, my first encounter with him was actually at a conference during my early PhD days when I happened to attend one of his talks on the FDA Mini-Sentinel program. Although we never met personally at the time, I immediately remembered his presentation once I read the job posting for my post-doctoral fellowship.
I formally met Darren for the first time during my in-person interview at DPM, where I pitched my research ideas on off-label drug use and insomnia. At the time, I was unsure of how he would receive these ideas, but to my great relief, he was incredibly supportive. Since that day, he has continually supported and encouraged me in pursuing this research, providing me with excellent guidance and going above and beyond to help me obtain the resources I need to do this work. Beyond our projects together, he has also demonstrated on many occasions that he also respects and cares for me as an individual and a post-doc trying to find my way in the academic world. I feel incredibly fortunate to have Darren as my mentor. Words cannot express how grateful I am!
Q: What advice do you have for those interested in pursuing a doctorate, or who are in the early stages of a doctoral program? What are some resources that might be helpful?
A: Because everyone’s academic journey is different, I realize the lessons from my own experiences may not be as helpful for another person. That being said, I hope at least someone will find the perspectives from my own journey helpful for theirs.
The first piece of advice I would offer is to always choose your research intentionally, keeping in mind how it broadly relates to your bigger story. However, because it is natural, even expected, that your plans may change and evolve, it is equally important to be flexible and open to broadening your research portfolio when the opportunities arise, but still always keeping in mind how the work you do will add to your story.
The second piece of advice I would offer is that focusing your efforts on producing your best work will pay dividends in the end. I know this advice sounds so simple, almost silly to point out. But with the number of academic opportunities, activities, and conferences that you will encounter, one can quickly feel pressured to participate in all of them, with the hopes of staying competitive for a job application or academic scholarship. To clarify, I am not recommending anyone say ‘no’ to all these activities. Rather, I am suggesting that keeping your core research projects (i.e., your thesis work!) at the forefront of your priorities will always be one of the wisest investments of your time and efforts.
Finally, in terms of helpful resources, I have found that oftentimes, other people have been the most useful resources to me. Whether it be emailing an author of a paper for more details on their research or programming code, capitalizing on the advice and wisdom of your mentor, or connecting with a colleague of a colleague to collaborate, I have always found these interpersonal interactions to be incredibly valuable for my research. So, don’t be afraid to reach out to others for help or advice, including me!
Choose your research intentionally, keeping in mind how it broadly relates to your bigger story.
Q: What is something interesting about you that your colleagues might not know?
A: I almost became a professional musician instead of an academic. I played piano and violin competitively for nearly 15 years growing up, so when it came time to apply for university, I seriously considered going into music full time. In the end, I decided to keep music as a pastime instead of a career so I could enjoy the best of both worlds. I have never regretted that decision!