Applying the Bradford Hill Criteria to Assess the Independent Causal Roles of Aging and Medication in Frailty...

The scientific literature, including systematic reviews and meta-analyses, has frequently described associations between aging, medication use, and frailty, without evaluation of their independent causation. The Bradford Hill Criteria, a framework consisting of nine principles for assessing epidemiological causation, is ideally suited to unconfound and assess the independent causal effect of aging versus medication use, in frailty progression.

A systematic review following Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines, searched MEDLINE, EMBASE, and CENTRAL with no restrictions on date or study design. Studies were selected based on predefined inclusion criteria and assessed for quality using the Joanna Briggs Institute critical appraisal tool. Where appropriate, meta-analyses of collated data were performed in RStudio, including effect sizes accounting for minimum age and polypharmacy to reduce confounding bias. Causal relationships between aging, medication use, and frailty were then evaluated independently using the nine principles of the Bradford Hill Criteria.

Data from 105 moderate-to-high quality studies based on the Joanna Briggs Institute assessment were extracted, formatted, and compiled to allow evaluation via the Bradford Hill Criteria. Evidence supported a strong independent causal relationship between aging, medication use, and frailty progression across eight of the nine principles. Strength of association, consistency, and a clear biological gradient were observed, with frailty increasing alongside age and medication count, respectively. Temporality was addressed as aging and medication exposure often preceded frailty, while interventions reducing medication supported the experiment criterion. Biological plausibility, biological coherence, and analogy were reinforced by clear biological mechanism, scientific reasoning, and epidemiological patterns. However, specificity could not be fully met, as frailty is influenced by multiple factors beyond aging and medication use, making the relationship inherently non-specific.

An independent causal link between aging and frailty, as well as between medication use and frailty, is well supported by the framework of Bradford Hill Criteria. Given the limited availability of randomized controlled trials or interventional studies in older adults, these findings offer valuable insights where evidence has been lacking and serve as a strategic starting point for future investigations into factors driving frailty progression.

PROSPERO Registration Number CRD42024614144.