Liliya Sinyavskaya

Tell us a bit about your background and how you ended up studying in this field.

In my country of origin Kazakhstan, I graduated from medical school in 1994 and then practiced as a physician. Part of my work in the early 2000s was conducting research on genitourinary tuberculosis, and participating in the implementation of the World Health Organization (WHO) tuberculosis control and prevention program in Kazakhstan. For several years our medical team visited a network of TB clinics across the country and did monitoring and training. I received a scholarship from the US National Institute of Health (NIH) to attend the School of Public Health at New York State University in Albany where I obtained a Master of Public Health degree. I was on a student visa in the US, so after graduating, I returned to my country to work for several years.

In 2010, I decided to immigrate to Canada. When I arrived in Montreal my goal was to find a job in medical research which I found at the Centre for Clinical Epidemiology of Lady Davis Research Institute at the Jewish General Hospital. I worked at Lady Davis as a research assistant at first. The Centre for Clinical Epidemiology has been an inspiring environment that helped me clearly identify my future career path. As a research assistant, every day at work I communicated with experienced senior researchers who are also CNODES researchers. There were weekly presentations, journal clubs where we discussed questions related to pharmacoepi, and I interacted with professors, including Dr. Samy Suissa, Dr. Kristian Filion and others. My decision to undertake doctoral studies came through this inspiring work experience.

Please give us a short summary of the research and why it is important.

Treatment with oral anticoagulants is very important for prevention of stroke in people with atrial fibrillation (A-Fib). However, these drugs may increase the risk of serious bleedings. New direct oral anticoagulants were approved for use in this population based on the results from randomized controlled clinical trials. However, clinical trials are only conducted among highly selected patients who are also closely monitored through the limited trial period. In a broader population, there are concerns about the safety and effectiveness of these drugs.

To address these concerns, CNODES is running an A-Fib research project called “Safety and effectiveness of direct oral anticoagulants and warfarin for stroke prevention in non-valvular atrial fibrillation: a multi-database cohort study with meta-analysis.”  My PhD supervisor, Dr. Madeleine Durand, is the principal investigator, and my co-supervisor, Dr. Christel Renoux, is a site investigator of this project. The project is based on the analyses of large population-based healthcare databases that are commonly used to compare safety and effectiveness of oral anticoagulants in real world settings. However, studying oral anticoagulants with population-based data poses some methodological challenges, including misclassification of exposure, and influence of time-dependent factors that may bias the safety and effectiveness estimates.

My research explores methodological issues that arose while planning and conducting the CNODES A-Fib project. Results from my PhD work have been useful to Dr. Durand and her team in designing the vast CNODES A-Fib project; our work together serves as a lab to inform protocol decisions. I have also benefited from close contact and teaching from Dr. Mireille Schnitzer, the statistical lead for the CNODES A-Fib project and a causal inference expert.

Currently, we are investigating the risk of error when measuring exposure to oral anticoagulants using pharmaceutical dispensation records. In our previous research we characterized pharmacy dispensing patterns and found that gaps between dispensations were larger and had a greater variability in patients who were dispensed warfarin compared to patients dispensed new direct oral anticoagulants.

In clinical practice, physicians frequently adjust the dosage of warfarin over the course of treatment. Therefore, the true duration of warfarin exposure may be longer or shorter than that recorded in pharmaceutical dispensation databases. This is different for direct oral anticoagulants which have fixed doses. This research is important because errors in the measurement of exposure may influence the conclusions of a study comparing warfarin and direct oral anticoagulants.

Tell us about your experiences related with CNODES.

Thanks to CNODES, I have every chance of succeeding in my professional endeavors. I benefit from the great mentoring of CNODES professionals who have vast experience in pharmacoepidemiology and medicine. The CNODES semi-annual meetings are another opportunity to learn from leading experts, and to meet with professionals from other parts of Canada, and with other students. CNODES supports me financially to do my PhD research, and to attend the McGill pharmacoepidemiology classes and the causal inference course at Harvard T.H. Chan School of Public Health. CNODES also supported my participation in the 33th and 34th ICPE International Conference on Pharmacoepidemiology and Therapeutic Risk Management (ICPE) where I presented my PhD work in two poster presentations.

What excites you the most about the research you are doing or hope to do in the future?

The significance of my research for medical practice, and the contribution of our results to future research. My research explores the impact of methodological issues in the context of an important clinical question. Results of the previous observational studies can be reappraised based on our findings, and some heterogeneity can be explained. The findings can also be extrapolated to other clinical questions.

The manuscript of our research results is still under revision by our team. We haven’t submitted yet, so I am not sure when it will be published, but hopefully within a few months. I will be presenting our research here at the University of Montreal and I’ll be attending ICPE in Philadelphia in August to present some of the results as a poster presentation.

Are there aspects of the work that you find particularly challenging?

I conduct my PhD research in the Department of Medicine at the University of Montreal. My research topic and the research environment require me to fully understand epidemiology, statistics, as well as the characteristics and clinical management of my study population. Also, I must be very focused on coding with my statistical software. It is challenging work, and also very inspiring.

Any career goals you would like to share?

My goal is to become an independent researcher in the field of pharmacoepidemiology. For now, it’s interesting to be working on methods. Maybe in the future, I will pursue other kinds of research. I hope to find a job in the pharmaceutical industry. I want to work in industry because, for medicine, it’s important; it’s related to new treatment and it’s a big responsibility to make sure new products are safe. This work will support health and prolong life. I know in the future I will have to answer these big questions about my career, but for now, I am focused on the SAS coding for the project I’m working on today.

Outside of work and studies, what are you are passionate about?

As it is summertime, I like to care for my balcony flower garden. I love to plant fragrant flowers, like small roses or dianthus. It’s great to sit out on my balcony enjoying the flowers.

Thank you for your interest in my research. I wish you an enjoyable summer with your families and friends.

References

Publications:

Sinyavskaya L, Matteau A, Johnson S, Durand M. Methodological challenges in assessment of current use of warfarin among patients with atrial fibrillation using dispensation data from administrative health care databases. Pharmacoepidemiology and Drug Safety. 2018;27(9):979-86.

https://www.ncbi.nlm.nih.gov/pubmed/29869382

Published abstracts:

Sinyavskaya L , Matteau A, Johnson S, Le Lorier J, Durand M. Issues in assessment of current use of warfarin using dispensation data from administrative healthcare databases: the case of anticoagulation in atrial fibrillation. Pharmacoepidemiol Drug Saf. 2017;26(S2):295.

https://onlinelibrary.wiley.com/toc/10991557/2017/26/S2

Sinyavskaya L, Durand M. Assumptions about grace period in definition may impact comparative estimates on risk of gastrointestinal bleeding between oral anticoagulants. Pharmacoepidemiol Drug Saf. 2018;27(S2):223.

https://onlinelibrary.wiley.com/toc/10991557/2018/27/S2