Onchee Yu, MS, is a biostatistician who has contributed her extensive experience in statistical applications to electronic health records (EHR) data to studies related to women’s health, pharmacoepidemiology, and vaccine safety and effectiveness. Ms. Yu has been a key member of Kaiser Permanente Washington Health Research Institute’s (KPWHRI) immunization research program for 20 years. Her work focuses on applying statistical methods to evaluate vaccine effectiveness, side effects, and safety. In collaboration with KPWHRI biostatisticians Jennifer Nelson, PhD, and Andrea Cook, PhD, Ms. Yu developed and improved statistical methods for monitoring the safety of postmarketing vaccines in the Vaccine Safety Datalink project.
Much of Ms. Yu's recent research is in pharmacoepidemiology, which is studying how drugs are used in a population and their impact on public health. She is an expert in statistical analysis in a complex, clinically important area—determining if taking medicine for one condition (for example, cardiovascular medications) affects risk of other illnesses (for example, cancer outcomes).
Ms. Yu also contributes to women’s health. Using extensive EHR data and in collaboration with University of Washington clinician and KPWHRI affiliate researcher Susan D. Reed, MD, MPH, Ms. Yu has estimated incidences and prevalences, validated diagnosis codes, and developed automated case-finding algorithms for women’s health conditions including uterine fibroids, adenomyosis and endometriosis.
Ms. Yu obtained her MS in biostatistics from the University of Washington in 1999. She is a member of the American Statistical Association and the International Society for Pharmacoepidemiology. Her statistical methodological expertise includes classification and regression tree analysis, and survival analysis.
Survival analysis; classification and regression tree analysis
Biostatistics; medication use and cancer outcomes
Biostatistics; vaccine safety and efficacy; postmarketing vaccine safety study design and analysis
Biostatistics; incidence and prevalence estimations; validation of diagnosis codes; automated case-finding algorithms
Biostatistics; medication use and cancer outcomes; postmarketing drug and vaccine safety study design and analysis; safety signal detection methods
Chen L, Chubak J, Yu O, Pocobelli G, Ziebell RA, Aiello Bowles EJ, Fujii MM, Sterrett AT, Boggs JM, Burnett-Hartman AN, Ritzwoller DP, Hubbard RA, Boudreau DM. Changes in use of opioid therapy after colon cancer diagnosis: a population-based study. Cancer Causes Control. 2019 Dec;30(12):1341-1350. doi: 10.1007/s10552-019-01236-5. Epub 2019 Oct 30. PubMed
Calip GS, Yu O, Boudreau DM, Shao H, Oratz R, Richardson SB, Gold HT. Diabetes and differences in detection of incident invasive breast cancer. Cancer Causes Control. 2019;30(5):435-441. doi: 10.1007/s10552-019-01166-2. Epub 2019 Apr 4. PubMed
Boudreau DM, Chen L, Yu O, Bowles EJA, Chubak J. Risk of second breast cancer events with chronic opioid use in breast cancer survivors. Pharmacoepidemiol Drug Saf. 2019 May;28(5):740-753. doi: 10.1002/pds.4779. Epub 2019 Apr 3. PubMed
Yu O, Reed SD, Schulze-Rath R, Grafton J, Hansen K, Scholes D. Identification of incident uterine fibroids using electronic medical record data. EGEMS (Wash DC). 2019;7(1):5. doi: 10.5334/egems.264. PubMed
Bowles EJA, Yu O, Ziebell R, Chen L, Boudreau DM, Ritzwoller DP, Hubbard RA, Boggs JM, Burnett-Hartman AN, Sterrett A, Fujii M, Chubak J. Cardiovascular medication use and risks of colon cancer recurrences and additional cancer events: a cohort study. BMC Cancer. 2019;19(1):270. doi: 10.1186/s12885-019-5493-8. PubMed
Researchers find a relationship between prescribed central nervous system-active medications and increased risk of falling among older people with dementia.
Dr. David Arterburn discusses reassuring news from his PCORnet study of the most widely used anti-obesity drug in the United States.
Dr. Paula Lozano explains how a Learning Health System project finds Kaiser Permanente Washington members who could benefit most from preventive services.