Background Adjusting for large numbers of covariates ascertained from patients’ health care claims data may improve control of confounding, as these variables may collectively be proxies for unobserved factors. Here we develop and test an algorithm that empirically identifies candidate covariates, prioritizes covariates, and integrates them into a propensity-score-based confounder adjustment model. Methods We developed a multi-step algorithm to implement high-dimensional proxy adjustment in claims data. Steps include (1) identifying data dimensions, e.g. diagnoses, procedures, and medications, (2) empirically identifying candidate covariates, (3) assess recurrence of codes, (4) prioritizing covariates, (5) selecting covariates for adjustment, (6) estimating the exposure propensity score, and (7) estimating an outcome model. This algorithm was tested in Medicare claims data, including a study on the effect of Cox-2 inhibitors on reduced gastric toxicity compared to nonselective nonsteroidal anti-inflammatory drugs (NSAIDs). Results In a population of 49,653 new users of Cox-2 inhibitors or nonselective NSAIDs, a crude relative risk (RR) for upper GI toxicity (RR = 1.09 [95% confidence interval = 0.91–1.30]) was initially observed. Adjusting for 15 predefined covariates resulted in a possible gastroprotective effect (0.94[0.78–1.12]). A gastroprotective effect became stronger when adjusting for an additional 500 algorithm-derived covariates (0.88[0.73–1.06]). Results of a study on the effect of statin on reduced mortality were similar. Using the algorithm adjustment confirmed a null finding between influenza vaccination and hip fracture (1.02[0.85–1.21]). Conclusion In typical pharmacoepidemiologic studies, the proposed high-dimensional propensity score resulted in improved effect estimates compared with adjustment limited to predefined covariates, when benchmarked against results expected from randomized trials.

If confirmed, these results suggest that conventional antipsychotic medications are at least as likely as atypical agents to increase the risk of death among elderly persons and that conventional drugs should not be used to replace atypical agents discontinued in response to the FDA warning.

No abstract

Context.-Although clinical trials have demonstrated the benefits of lipidlowering therapy, little is known about how these drugs are prescribed or used in the general population. Objective.-To estimate predictors of persistence with therapy for lipid-lowering drug regimens in typical populations of patients in the United States and Canada. Design.-A cohort study defining all prescriptions filled for lipid-lowering drugs during 1 year, as well as patients' demographic and clinical characteristics. Setting.-New Jersey's Medicaid and Pharmacy Assistance for the Aged and Disabled programs and Quebec's provincial medical care program. Patients.-All continuously enrolled patients older than 65 years who filled 1 or more prescriptions for lipid-lowering drugs (N = 5611 in the US programs, and N = 1676 drawn from a 10% sample in Quebec). Main Outcome Measures.-Proportion of days during the study year for which patients had filled prescriptions for lipid-lowering drugs; predictors of good vs poor persistence with therapy. Results.-In both populations, patients failed to fill prescriptions for lipid-lowering drugs for about 40% of the study year. Persistence rates with 3-hydroxy-3methylglutaryl coenzyme A reductase inhibitors were significantly higher than those seen with cholestyramine (64.3% vs 36.6% of days with drug available, respectively). Patients with hypertension, diabetes, or coronary artery disease had significantly higher rates of persistence with lipid-lowering regimens. In New Jersey, multivariable analysis indicated that the poorest patients (those enrolled in Medicaid) had lower rates of drug use than less indigent patients (those enrolled in Pharmacy Assistance for the Aged and Disabled) after adjusting for possible confounders, despite virtually complete drug coverage in both programs. When rates of use were measured in the US population for the 5 years following the study year, only 52% of surviving patients who were initially prescribed lipid-lowering drugs were still filling prescriptions for this drug class. Conclusion.-In all populations studied, patients who were prescribed lipidlowering drug regimens remained without filled prescriptions for over a third of the study year on average. Rates of persistence varied substantially with choice of agent prescribed, comorbidity, and socioeconomic status, despite universal coverage of prescription drug costs. After 5 years, about half of the surviving original cohort in the United States had stopped using lipid-lowering therapy altogether.