Analysis of Observational Studies in the Presence of Treatment Selection Bias

Abstract

N THE FACE OF THE FINANCIAL, practical, and ethical challenges inherent in undertaking randomized clinical trials (RCTs), investigators often use observational data to compare the outcomes of different therapies. These comparisons may be biased due to prognostically important baseline differences among patients, often as a result of unobserved treatment selection biases. Unmeasurable clinical and social interactions in the diagnostic-treatment pathway, and physicians' knowledge of unmeasured prognostic variables, may affect treatment decisions and outcomes. Physicians are frequently risk averse in case selection, performing interventions on lower-risk patients despite greater clinical benefit to higher-risk patients. [1][2][3] In some cases, especially when data are collected on detailed clinical risk factors, these differences can be controlled using standard statistical methods. In other cases, when unmeasured patients characteristics affect both the decision to treat and the outcome, these For editorial comment see p 314.

The assessment and appraisal of regenerative medicines and cell therapy products: an exploration of methods for review, economic evaluation and appraisal

“…A summary of the studies that have looked at methods of adjustment for confounding bias in NRSs and how reliable they are [62][63][64][65][66][67][68][69][70] is presented in Appendix 3. Overall, it is unclear which methods are most appropriate in certain circumstances and further research is needed.…”

“…63,67 Crosby et al 63 found that results from regression analysis and instrumental variable methods differed somewhat and suggested that instrumental variable methods are potentially superior. Stukel et al 67 compared all three methods of adjustment (regression analysis, instrumental variables and propensity scoring) and concluded that instrumental variables may lead to less biased estimates of treatment effects. Although the evidence on instrumental variables is limited it nevertheless suggests that instrumental variables may offer advantages over other methods and may produce the least biased estimates.…”

“…In addition to the seven empirical studies identified, [62][63][64][65][66][67]69 a discussion paper by Biondi-Zoccai et al 70 provides a useful overview of the alternative methods of adjustment. Biondi-Zoccai et al 70 concluded that there is no clearly superior method, noting that 'both standard multivariable methods and propensity scores have key limitations, and none is able to take into account unknown confounders' (p. 736).…”

Effect of β‐Blockers Beyond 3 Years After Acute Myocardial Infarction

“…Although we adjusted for significant risk factors, unmeasured covariates, such as instrumental variables, may have influenced the study outcomes. Although some statistical techniques can be used to eliminate unmeasured covariates,33 only well‐designed randomized controlled clinical trials are likely to be free of measured and unmeasured confounders. Furthermore, because of attrition of sample size in populations 2 and 3, we cannot exclude the possibility of type 1 error for the neutral association between β‐blockers and mortality.…”

From Medical Records to Clinical Science

“…But it does seem worthwhile to propose steps that should be employed unless there are reasons not to do so. For an example following most of these steps, see Newgard et al 5 Similar approaches can be found in Stukel at al, 6 Sun et al, 7 Smeeth et al, 8 Rutten et al, 9 Polsky et al, 10 and Pollack et al 11 Not all of the latter studies were equally careful to exclude Treatment Influences from propensity scores.…”