With background in clinical pharmacy and pharmacoepidemiology, Dr. Desai's research has naturally gravitated towards assessment of the use and outcomes of prescription medications in routine care patient populations. Broadly, his research falls within the following three areas:


(1) Chronic disease pharmacoepidemiology: Dr. Desai is actively involved in understanding the use of medications and the resulting outcomes in patients with chronic diseases such as rheumatoid arthritis (RA), inflammatory bowel disease (IBD), chronic kidney disease (CKD), and chronic heart failure (CHF). He has extensively studied the comparative safety of treatment with disease modifying therapies in RA and IBD using non-randomized studies as well as meta-analytic methods and have uncovered important differences in the risk of clinical endpoints, including infections, coronary artery disease, and venous thromboebolism, across various treatments. In other investigations, he has also compared the effectiveness of medications with respect to cardiovascular events including antiplatelet medications in patients with acute coronary syndrome, angiotensin receptor blockers for CHF, and phosphate binders for CKD. Publications from this line of work have appeared in top journals of the field, including the British Medical Journal, Arthritis & RheumatologyPLOS Medicine, and JAMA Internal Medicine.


(2) Methodological research to improve inference from observational studies: Dr. Desai has conducted several methodological investigations to improve confounding adjustment in non-randomized studies of medication effects. In a simulation study, he proposed and tested a novel weighting approach for confounding control that uses fine-stratification based on propensity scores for circumstances when exposure is infrequent and events rare - a typical situation when evaluating outcomes of newly marketed medications. In another methods study, he evaluated the performance of disease risk scores as confounding control tools in nested case-control studies. He has served as the Principal Investigator of a U01 grant from the FDA that aimed to improve confounding adjustment methods in post-marketing studies of generic drugs. He has also led multiple FDA Sentinel Task Order workgroups that have helped develop risk adjustment methods for active post-marketing surveillance of newly marketed medications.


(3) Research focusing on populations underrepresented in clinical trials: Dr. Desai has a special interest in geenrating a robust evidence base regarding medication outcomes in patient populations typically underrepresented in trials including pregnant women, the elderly, and patients with multiple chronic conditions. In a widely cited publication, he found increasing use of opioid analgesics during pregnancy in women covered by a Medicaid in the US. In a follow-on study, he evaluated the risk of neonatal withdrawal symptoms as a result of this practice and identified relevant maternal characteristics that increased this risk. He has also actively participated in safety evaluations of other clinically relevant medication classes in pregnancy, including anti-rheumatic drugs, anti-depressants, antipsychotics, and statins. He has led multiple studies related to medication errors and their prevention in nursing home patients, where ensuring safe medication use is critical because of the highly prevalent poly-pharmacy. In these studies, he focused on identifying factors related to errors in using many important drug classes, including anti-coagulants and opioid analgesics, in nursing homes with the goal of helping to minimize these errors.