A5004416630- Health Systems, Economic Evaluations, Quality of Life
- Advanced Causal Inference Techniques
- Statistical Methods in Clinical Trials
- Diabetes Treatment and Management
- Medication Adherence and Compliance
- Pharmaceutical Economics and Policy
- Pharmaceutical Practices and Patient Outcomes
- Healthcare Policy and Management
- Pharmacovigilance and Adverse Drug Reactions
- Atrial Fibrillation Management and Outcomes
- Pharmaceutical studies and practices
- Blood Pressure and Hypertension Studies
- Meta-analysis and systematic reviews
- Rheumatoid Arthritis Research and Therapies
- Heart Failure Treatment and Management
- Antiplatelet Therapy and Cardiovascular Diseases
- Schizophrenia research and treatment
- Metabolism, Diabetes, and Cancer
- Primary Care and Health Outcomes
- Diabetes Management and Research
- Inflammatory mediators and NSAID effects
- Chronic Disease Management Strategies
- Venous Thromboembolism Diagnosis and Management
- Statistical Methods and Bayesian Inference
- Dermatology and Skin Diseases
Brigham and Women's Hospital
2016-2025
Harvard University
2016-2025
Mass General Brigham
2018-2025
AstraZeneca (United Kingdom)
2023-2024
Massachusetts General Hospital
2023-2024
Pharmac
2006-2023
RELX Group (United States)
2023
Boehringer Ingelheim (Germany)
2018-2022
Novem (Netherlands)
2018-2022
Aetion (United States)
2016-2022
Despite the growing popularity of propensity score (PS) methods in epidemiology, relatively little has been written epidemiologic literature about problem variable selection for PS models. The authors present results two simulation studies designed to help epidemiologists gain insight into a analysis. illustrate how choice variables that are included model can affect bias, variance, and mean squared error an estimated exposure effect. suggest unrelated but related outcome should always be...
In Brief Background: Adjusting for large numbers of covariates ascertained from patients' health care claims data may improve control confounding, as these variables collectively be proxies unobserved factors. Here, we develop and test an algorithm that empirically identifies candidate covariates, prioritizes integrates them into a propensity-score-based confounder adjustment model. Methods: We developed multistep to implement high-dimensional proxy in data. Steps include (1) identifying...
Recently, the Food and Drug Administration (FDA) issued an advisory stating that atypical antipsychotic medications increase mortality among elderly patients. However, did not apply to conventional medications; risk of death with these older agents is known.We conducted a retrospective cohort study involving 22,890 patients 65 years age or who had drug insurance benefits in Pennsylvania began receiving medication between 1994 2003. Analyses rates Cox proportional-hazards models were used...
Comorbidity is an important confounder in epidemiologic studies. The authors compared the predictive performance of comorbidity scores for use research with administrative databases. Study participants were British Columbia, Canada, residents aged >or=65 years who received angiotensin-converting enzyme inhibitors or calcium channel blockers at least once during observation period. Six computed all 141,161 baseline year (1995-1996). Endpoints death and health care utilization a 12-month...
Adherence to medications that are prescribed after myocardial infarction is poor. Eliminating out-of-pocket costs may increase adherence and improve outcomes.
Abstract Background Large health care utilization databases are frequently used to analyze unintended effects of prescription drugs and biologics. Confounders that require detailed information on clinical parameters, lifestyle, or over‐the‐counter medications often not measured in such datasets, causing residual confounding bias. Objective This paper provides a systematic approach sensitivity analyses investigate the impact pharmacoepidemiologic studies use databases. Methods Four basic...
Background— Although cyclooxygenase-2 inhibitors (coxibs) were developed to cause less gastrointestinal hemorrhage than nonselective nonsteroidal antiinflammatory drugs (NSAIDs), there has been concern about their cardiovascular safety. We studied the relative risk of acute myocardial infarction (AMI) among users celecoxib, rofecoxib, and NSAIDs in Medicare beneficiaries with a comprehensive drug benefit. Methods Results— conducted matched case-control study 54 475 patients 65 years age or...
Abstract Here we identify hundreds of new drug-disease associations for over 900 FDA-approved drugs by quantifying the network proximity disease genes and drug targets in human (protein–protein) interactome. We select four network-predicted to test their causal relationship using large healthcare databases with 220 million patients state-of-the-art pharmacoepidemiologic analyses. Using propensity score matching, two network-based predictions are validated patient-level data: carbamazepine is...
Frailty is a key determinant of health status and outcomes care interventions in older adults that not readily measured Medicare data. This study aimed to develop validate claims-based frailty index (CFI).We used data from Current Beneficiary Survey 2006 (development sample: n = 5,593) 2011 (validation 4,424). A CFI was developed using the claims approximate survey-based (SFI) calculated survey as reference standard. We compared combined comorbidity (CCI) ability predict death, disability,...
ABSTRACT Background Among the large number of cohort studies that employ propensity score matching, most match patients 1:1. Increasing matching ratio is thought to improve precision but may come with a trade‐off respect bias. Objective To evaluate several methods in through simulation and empirical analyses. Methods We simulated cohorts 20 000 exposure prevalence 10%–50%. five dichotomous continuous confounders. estimated scores matched using digit‐based greedy (“greedy”), pairwise nearest...
In pharmacoepidemiology, routinely collected data from electronic health records (including primary care databases, registries, and administrative healthcare claims) are a resource for research evaluating the real world effectiveness safety of medicines. Currently available guidelines reporting using non-randomised, data—specifically REporting studies Conducted Observational Routinely Data (RECORD) Strengthening Reporting OBservational in Epidemiology (STROBE) statements—do not capture...
Although it is known that rheumatoid arthritis associated with an increased risk of cardiovascular disease (CVD), the pattern this not clear. This study investigated relative myocardial infarction, stroke and CVD mortality in adults compared without across age groups, sex prior event status.We conducted a cohort among all residents aged >or=18 years residing British Columbia between 1999 2003. Residents who had visited doctor at least thrice for (International Classification Disease = 714)...
The authors followed 532 consecutive patients with episodic migraine (<15 days/month) for 1 year. Sixty-four (14%) developed chronic headache (>/=15 days/month). odds ratios developing CH were 20.1 (95% CI 5.7 to 71.5) comparing a "critical" (10 14 vs "low" (0 4 and 6.2 1.7 26.6) in an "intermediate" (6 9 frequency 19.4 8.7 43.2) without medication overuse.
<b>Background:</b> Public health advisories have warned that the use of atypical antipsychotic medications increases risk death among elderly patients. We assessed short-term mortality in a population-based cohort people British Columbia who were prescribed conventional and medications. <b>Methods:</b> used linked care utilization data all BC residents to identify aged 65 years older began taking between January 1996 December 2004 free cancer. compared 180-day all-cause those <b>Results:</b>...
Epidemiologic studies are increasingly used to investigate the safety and effectiveness of medical products interventions. Appropriate adjustment for confounding in such is challenging because exposure determined by a complex interaction patient, physician, healthcare system factors. The challenges control particularly acute using utilization databases where information on many potential factors lacking meaning variables often unclear. We discuss advantages disadvantages different approaches...
Aprotinin (Trasylol) is used to mitigate bleeding during coronary-artery bypass grafting (CABG). Accumulating evidence suggests that this practice increases mortality.Using electronic administrative records of the Premier Perspective Comparative Database, we studied hospitalized patients with operating-room charges for use aprotinin (33,517 patients) or aminocaproic acid (44,682 on day CABG was performed. We tabulated numbers a hospital-discharge status death and performed three types...
Postmarketing observational studies of the safety and effectiveness prescription medications are critically important but fraught with methodological problems. The data sources available for such research often lack information on indications other confounders drug exposure under study. Instrumental variable methods have been proposed as a potential approach to control confounding by indication in nonexperimental treatment effects; however, good instruments hard find.We propose an instrument...