Statistical Methods for Cost‐effectiveness Analyses That Use Observational Data: A Critical Appraisal Tool and Review of Current Practice

Abstract

Many cost-effectiveness analyses (CEAs) use data from observational studies. Statistical methods can only address selection bias if they make plausible assumptions. No quality assessment tool is available for appraising CEAs that use observational studies. We developed a new checklist to assess statistical methods for addressing selection bias in CEAs that use observational data. The checklist criteria were informed by a conceptual review and applied in a systematic review of economic evaluations. Criteria included whether the study assessed the 'no unobserved confounding' assumption, overlap of baseline covariates between the treatment groups and the specification of the regression models. The checklist also considered structural uncertainty from the choice of statistical approach. We found 81 studies that met the inclusion criteria: studies tended to use regression (51%), matching on individual covariates (25%) or matching on the propensity score (22%). Most studies (77%) did not assess the 'no observed confounding' assumption, and few studies (16%) fully considered structural uncertainty from the choice of statistical approach. We conclude that published CEAs do not assess the main assumptions behind statistical methods for addressing selection bias. This checklist can raise awareness about the assumptions behind statistical methods for addressing selection bias and can complement existing method guidelines for CEAs.

Psychometric Properties of Maternal Self-Efficacy Questionnaire in a Population of Iranian Mothers

“…The results indicate that KMO index of MSQ questionnaire represents a good factor analysis. In Bartlett's test, the null hypothesis is total independence (sphericity) of criteria (Kreif et al 2013). For MSQ questionnaire, the null hypothesis of data sphericity was rejected (p \ 0.05), and the KMO was confirmed.…”

“…Note that KMO index shouldn't be less than 0.5 (otherwise, it is unacceptable and negligible) (Kreif et al 2013). The results indicate that KMO index of MSQ questionnaire represents a good factor analysis.…”

Cost‐effectiveness of neoadjuvant concurrent chemoradiotherapy versus esophagectomy for locally advanced esophageal squamous cell carcinoma: A population‐based matched case‐control study

“…Patients were matched on the logit of PS using a caliper of 0.2 standard deviations of the logit of the PS via greedy matching . We used suggestions regarding database approach to CEA to guide our analysis and reporting . This study was approved by the Institutional Review Board (CMUH103‐REC‐005).…”

Addressing unmeasured confounding in comparative observational research

“…Despite of its importance, the impact of unmeasured confounding is rarely quantitatively assessed. Often, the potential existence of unmeasured confounding is simply mentioned as a limitation without attempting any quantitative analyses …”

“…Often, the potential existence of unmeasured confounding is simply mentioned as a limitation without attempting any quantitative analyses. 15 Schneeweiss 16 outlined strategies to control for unmeasured confounders in a 2006 research article. While a valuable resource as it guided researchers to potential methods based on available information, multiple new approaches for handling unmeasured confounding have been proposed and applied in the last decade.…”