A Systematic Review of Methods Used for Confounding Adjustment in Observational Economic Evaluations in Cardiology Conducted between 2013 and 2017.

Background. Observational economic evaluations (i.e., economic evaluations in which treatment allocation is not randomized) are prone to confounding bias. Prior reviews published in 2013 have shown that adjusting for confounding is poorly done, if done at all. Although these reviews raised awareness on the issues, it is unclear if their results improved the methodological quality of future work. We therefore aimed to investigate whether and how confounding was accounted for in recently published observational economic evaluations in the field of cardiology. Methods. We performed a systematic review of PubMed, Embase, Cochrane Library, Web of Science, and PsycInfo databases using a set of Medical Subject Headings and keywords covering topics in "observational economic evaluations in health within humans" and "cardiovascular diseases." Any study published in either English or French between January 1, 2013, and December 31, 2017, addressing our search criteria was eligible for inclusion in our review. Our protocol was registered with PROSPERO (CRD42018112391). Results. Forty-two (0.6%) out of 7523 unique citations met our inclusion criteria. Fewer than half of the selected studies adjusted for confounding (n = 19 [45.2%]). Of those that adjusted for confounding, propensity score matching (n = 8 [42.1%]) and other matching-based approaches were favored (n = 8 [42.1%]). Our results also highlighted that most authors who adjusted for confounding rarely justified their methodological choices. Conclusion. Our results indicate that adjustment for confounding is often ignored when conducting an observational economic evaluation. Continued knowledge translation efforts aimed at improving researchers' knowledge regarding confounding bias and methods aimed at addressing this issue are required and should be supported by journal editors.

Methodological challenges in assessment of current use of warfarin among patients with atrial fibrillation using dispensation data from administrative health care databases.

PURPOSE: Algorithms to define current exposure to warfarin using administrative data may be imprecise. Study objectives were to characterize dispensation patterns, to measure gaps between expected and observed refill dates for warfarin and direct oral anticoagulants (DOACs). METHODS: Retrospective cohort study using administrative health care databases of the Régie de l'assurance-maladie du Québec. We identified every dispensation of warfarin, dabigatran, rivaroxaban, or apixaban for patients with AF initiating oral anticoagulants between 2010 and 2015. For each dispensation, we extracted date and duration. Refill gaps were calculated as difference between expected and observed dates of successive dispensation. Refill gaps were summarized using descriptive statistics. To account for repeated observations nested within patients and to assess the components of variance of refill gaps, we used unconditional multilevel linear models. RESULTS: We identified 61 516 new users. Majority were prescribed warfarin (60.3%), followed by rivaroxaban (16.4%), dabigatran (14.5%), apixaban (8.8%). Most frequent recorded duration of dispensation was 7 days, suggesting use of pharmacist-prepared weekly pillboxes. The average refill gap from multilevel model was higher for warfarin (9.28 days, 95%CI:8.97-9.59) compared with DOACs (apixaban 3.08 days, 95%CI: 2.96-3.20, dabigatran 3.70, 95%CI: 3.56-3.84, rivaroxaban 3.15, 95%CI: 3.03-3.27). The variance of refill gaps was greater among warfarin users than among DOAC users. CONCLUSIONS: Greater refill gaps for warfarin may reflect inadequate capture of the period covered by the number of dispensed pills recorded in administrative data. A time-dependent definition of exposure using dispensation data would lead to greater misclassification of warfarin than DOACs use.

A Discrete Event Simulation Model to Assess the Economic Value of a Hypothetical Pharmacogenomics Test for Statin-Induced Myopathy in Patients Initiating a Statin in Secondary Cardiovascular Prevention.

BACKGROUND: Statin (HMG-CoA reductase inhibitor) therapy is the mainstay dyslipidemia treatment and reduces the risk of a cardiovascular (CV) event (CVE) by up to 35%. However, adherence to statin therapy is poor. One reason patients discontinue statin therapy is musculoskeletal pain and the associated risk of rhabdomyolysis. Research is ongoing to develop a pharmacogenomics (PGx) test for statin-induced myopathy as an alternative to the current diagnosis method, which relies on creatine kinase levels. The potential economic value of a PGx test for statin-induced myopathy is unknown. METHODS: We developed a lifetime discrete event simulation (DES) model for patients 65 years of age initiating a statin after a first CVE consisting of either an acute myocardial infarction (AMI) or a stroke. The model evaluates the potential economic value of a hypothetical PGx test for diagnosing statin-induced myopathy. We have assessed the model over the spectrum of test sensitivity and specificity parameters. RESULTS: Our model showed that a strategy with a perfect PGx test had an incremental cost-utility ratio of 4273 Canadian dollars ($Can) per quality-adjusted life year (QALY). The probabilistic sensitivity analysis shows that when the payer willingness-to-pay per QALY reaches $Can12,000, the PGx strategy is favored in 90% of the model simulations. CONCLUSION: We found that a strategy favoring patients staying on statin therapy is cost effective even if patients maintained on statin are at risk of rhabdomyolysis. Our results are explained by the fact that statins are highly effective in reducing the CV risk in patients at high CV risk, and this benefit largely outweighs the risk of rhabdomyolysis.

A model to assess the cost-effectiveness of pharmacogenomics tests in chronic heart failure: the case of ivabradine.

Pharmacogenomics (PGx) tests have the potential of improving the effectiveness of expensive new drugs by predicting the likelihood, for a particular patient, to respond to a treatment. The objective of this study was to develop a pharmacoeconomic model to determine the characteristics and the cost-effectiveness of a hypothetical PGx test, which would identify patients who are most likely to respond to an expensive treatment for chronic heart failure. For this purpose, we chose the example of ivabradine. Our results suggest that the use of a PGx test that could select a subgroup of patients to be treated with an expensive drug has the potential to provide more efficient drug utilization.