Evidence of potential bias in a comparison of β blockers and calcium channel blockers in patients with chronic obstructive pulmonary disease and acute coronary syndrome: results of a multinational study

Machine learning for improving high-dimensional proxy confounder adjustment in healthcare database studies: an overview of the current literature

Controlling for large numbers of variables that collectively serve as 'proxies' for unmeasured factors can often improve confounding control in pharmacoepidemiologic studies utilizing administrative healthcare databases. There is a growing body of evidence showing that data-driven machine learning algorithms for high-dimensional proxy confounder adjustment can supplement investigator-specified variables to improve confounding control compared to adjustment based on investigator-specified variables alone. Consequently, there has been a recent focus on the development of data-driven methods for high-dimensional proxy confounder adjustment. In this paper, we discuss the considerations underpinning three areas for data-driven high-dimensional proxy confounder adjustment: 1) feature generation-transforming raw data into covariates (or features) to be used for proxy adjustment; 2) covariate prioritization, selection and adjustment; and 3) diagnostic assessment. We survey current approaches and recent advancements within each area, including the most widely used approach to proxy confounder adjustment in healthcare database studies (the high-dimensional propensity score or hdPS). We also discuss limitations of the hdPS and outline recent advancements that incorporate the principles of proxy adjustment with machine learning extensions to improve performance. We further discuss challenges and avenues of future development within each area. This manuscript is endorsed by the International Society for Pharmacoepidemiology (ISPE). This article is protected by copyright. All rights reserved.

Beta-Blockers in COPD: A Methodological Review of the Observational Studies

The use of beta-blockers in patients with chronic obstructive pulmonary disease (COPD) has received much attention. Several observational studies reported important reductions in mortality and exacerbations with these drugs, but the extent of bias in these studies is unclear. Nevertheless, the large ongoing randomized trial (βLOCK-COPD) was initiated specifically to evaluate these effects. We searched the literature to identify all observational studies investigating the effectiveness of beta-blockers in COPD patients on major outcomes, including death and COPD exacerbation. We identified 18 observational studies, with 10 studies affected by confounding bias and six by immortal time bias, while two addressed these biases. Reductions in all-cause mortality with beta-blocker use were observed among the studies with confounding bias (pooled rate ratio 0.72; 95% CI 0.59-0.88) and those with immortal time bias (pooled rate ratio 0.64; 95% CI 0.53-0.77). A large five-database study that addressed these two biases reported hazard ratios of 0.90 (95% CI: 0.78-1.02) for death and 0.54 (95% CI: 0.47-0.61) for COPD hospitalization. However, this latter estimate was the same as for the first 30 days after treatment initiation, thus indicating that important residual confounding cannot be ruled out. Observational studies, important to provide evidence from real-world data on medication effects, are unsupportive for beta-blockers in COPD. Even if immortal time bias is properly avoided, confounding bias cannot be fully controlled due to their relative contraindication in COPD. In the case of beta-blockers, randomized trials such as βLOCK-COPD are necessary to eliminate the uncertainty from residual confounding bias.

Bias in pharmacoepidemiologic studies using secondary health care databases: a scoping review

BACKGROUND

The availability of clinical and therapeutic data drawn from medical records and administrative databases has entailed new opportunities for clinical and epidemiologic research. However, these databases present inherent limitations which may render them prone to new biases. We aimed to conduct a structured review of biases specific to observational clinical studies based on secondary databases, and to propose strategies for the mitigation of those biases.

METHODS

Scoping review of the scientific literature published during the period 2000-2018 through an automated search of MEDLINE, EMBASE and Web of Science, supplemented with manually cross-checking of reference lists. We included opinion essays, methodological reviews, analyses or simulation studies, as well as letters to the editor or retractions, the principal objective of which was to highlight the existence of some type of bias in pharmacoepidemiologic studies using secondary databases.

RESULTS

A total of 117 articles were included. An increasing trend in the number of publications concerning the potential limitations of secondary databases was observed over time and across medical research disciplines. Confounding was the most reported category of bias (63.2% of articles), followed by selection and measurement biases (47.0% and 46.2% respectively). Confounding by indication (32.5%), unmeasured/residual confounding (28.2%), outcome misclassification (28.2%) and "immortal time" bias (25.6%) were the subcategories most frequently mentioned.

CONCLUSIONS

Suboptimal use of secondary databases in pharmacoepidemiologic studies has introduced biases in the studies, which may have led to erroneous conclusions. Methods to mitigate biases are available and must be considered in the design, analysis and interpretation phases of studies using these data sources.

β-blockers after acute myocardial infarction in patients with chronic obstructive pulmonary disease: A nationwide population-based observational study

BACKGROUND

Patients with chronic obstructive pulmonary disease (COPD) less often receive β-blockers after acute myocardial infarction (AMI). This may influence their outcomes after AMI. This study evaluated the efficacy of β-blockers after AMI in patients with COPD, compared with non-dihydropyridine calcium channel blockers (NDCCBs) and absence of these two kinds of treatment.

METHODS AND RESULTS

We conducted a nationwide population-based cohort study using data retrieved from Taiwan National Health Insurance Research Database. We collected 28,097 patients with COPD who were hospitalized for AMI between January 2004 and December 2013. After hospital discharge, 24,056 patients returned to outpatient clinics within 14 days (the exposure window). Those who received both β-blockers and NDCCBs (n = 302) were excluded, leaving 23,754 patients for analysis. Patients were classified into the β-blocker group (n = 10,638, 44.8%), the NDCCB group, (n = 1,747, 7.4%) and the control group (n = 11,369, 47.9%) based on their outpatient prescription within the exposure window. The β-blockers group of patients had lower overall mortality risks (adjusted hazard ratio [95% confidence interval]: 0.91 [0.83-0.99] versus the NDCCB group; 0.88 [0.84-0.93] versus the control group), but the risk of major adverse cardiac events within 1 year was not statistically different. β-blockers decreased risks of re-hospitalization for COPD and other respiratory diseases by 12-32%.

CONCLUSIONS

The use of β-blockers after AMI was associated with a reduced mortality risk in patients with COPD. β-blockers did not increase the risk of COPD exacerbations.

Synergies From Integrating Randomized Controlled Trials and Real-World Data Analyses

Analyses using administrative claims databases or national registries provide estimates of benefits and harms of medications in real-world settings for large and diverse patient populations. Whereas claims-based nonrandomized studies and randomized-controlled trials (RCTs) have distinct limitations, their strengths are complementary. Integrating RCT and claims data offers substantial synergies. We propose obtaining routinely collected longitudinal claims data from RCT participants and discuss the added value of the novel evidence that can be derived from this "information overlap."