Research
Our Research
Dr. Kazi has authored over 175 publications (>100 original research publications; >45,000 citations; h-index 51), including in JAMA, NEJM, Annals of Internal Medicine, BMJ, JACC, Circulation, JAMA Cardiology, Lancet Global Health, and others. Our findings have influenced clinical practice and are regularly cited in international clinical guidelines, Congressional documents, and the general press. In particular, our findings have meaningfully influenced drug pricing and access (e.g., producing approximately 800 million dollars in annual savings for Medicare due to the price reduction for PCSK9 inhibitors).
Key Themes
Guiding the uptake and price of novel cardiovascular therapies by using computational models to quantify population health impact, costs, and cost-effectiveness.
We have developed numerous simulation models to project long-term clinical and economic outcomes of strategies to prevent or treat various forms of cardiovascular disease. These models synthesize inputs from dozens of data sources, including clinical trials, registries, claims, epidemiologic studies, publications from the Food and Drug Administration and World Health Organization, peer-reviewed literature, vital statistics, and census data. Our findings have been extensively cited in clinical guidelines (lipid management, heart failure, chronic coronary disease) and reports to the US Congress. Our publications on the value of PCSK9 inhibitors motivated an unprecedented 60% price reduction of these therapies, resulting in improved access and lower patient out-of-pocket costs, as well as substantial savings for Medicare Part D (approximately 800 million dollars per year).
Characterizing the magnitude and causes of health disparities using diverse datasets and statistical approaches.
A major focus of our work has been to quantify the burden of disparities in cardiovascular outcomes in underserved populations, so as to make a public health and economic case to invest in strategies that advance health equity. The overarching goal is to do more than describe disparities – our work aims to shed light on the underlying causes of these disparities so as to identify potential solutions.
Examining the comparative effectiveness of novel therapies in real-world populations by applying novel causal inference methods to diverse, linked datasets.
Using statistical techniques to account for observed and unobserved confounders, we have analyzed large datasets to evaluate the clinical and economic burden of cardiovascular disease, and the effectiveness of novel clinical interventions such as drugs, devices, and policies. Additionally, we have evaluated heterogeneity of treatment effect in underserved populations, and used real-world data to predict risk of adverse events. These projects have relied on data from the Centers for Medicare and Medicaid Services, the Agency for Healthcare Research and Quality’s Healthcare Cost and Utilization Project, IBM Marketscan, electronic health records, and an integrated dataset of claims, laboratory results, and pharmacy information from Kaiser Permanente. Our work has shown that observational data can be reliably used to derive meaningful epidemiologic and clinical insights, and to develop models to accurately predict outcomes in patients with cardiovascular disease.
Active Research Funding:
Comparative Effectiveness and Value of Cardiac Rehabilitation in Diverse Populations (NHLBI R01HL175300)
A Multi-Dimensional Linked Registry to Identify Biological, Clinical, Health System, and Socioeconomic Risk Factors for COVID-19-Related Cardiovascular Events (NIH/NHLBI R01HL157530)
Implementation Strategies for Self-Measured Blood Pressure Monitor 
ing in Racially and Ethnically Diverse Populations (InS2PiRED) (AHRQ R18HS029817)
Effects of DASH Groceries on Blood Pressure in Black Residents of Urban Food Deserts Without Treated Hypertension: Groceries for Black Residents of Boston to Stop Hypertension (Go Fresh) (AHA 878488)