Anticholinergic burden in older adults with falls: beyond functional assessment

Background
Falls are a common presentation within frailty services and frequently coexist with multimorbidity, polypharmacy and cognitive impairment. The Elderly Care Assessment Service (ECAS) at St David’s Hospital, Cardiff, assesses patients with frailty syndromes using comprehensive geriatric assessment (CGA) delivered by a multidisciplinary team. Physiotherapy-led functional measures inform falls risk and rehabilitation planning; however, medication-related vulnerability, particularly anticholinergic burden, may contribute to falls risk through mechanisms not fully captured by functional testing.

Aim
To describe the clinical, functional and medication-related characteristics of patients undergoing CGA within ECAS, and to explore relationships between anticholinergic burden, functional falls risk and reported falls.

Methods
We conducted a retrospective service evaluation of community-dwelling adults aged ≥60 years who completed a physiotherapy assessment with a recorded pre-intervention Tinetti score within ECAS in 2024. Extracted variables included age, sex, Clinical Frailty Scale (CFS), falls in the preceding 12 months, cognitive impairment (yes/no), number of regular medications and anticholinergic burden (ACB) score. Increased falls risk was defined as a Tinetti score ≤23. Analyses were descriptive with exploratory Spearman correlations.

Results
107 patients were included (mean age 79.5 years; median 81; 56[52.3%] female). Increased falls risk was present in 70/107(65.4%). Median falls in the preceding 12 months was 2. 15/107(14.0%) had cognitive impairment. High anticholinergic burden (ACB≥3) occurred in 20/107(18.7%), including 5/37(13.5%) classified as low falls risk (Tinetti≥24). Correlations between Tinetti score, ACB and falls were weak (Spearman ρ = −0.26 to +0.09).

Conclusion
Within a CGA-based elderly care assessment service, anticholinergic burden was distributed across functional performance levels and showed limited correlation with balance scores or falls history. These findings suggest that medication-related vulnerability may remain hidden when falls assessment relies on functional measures alone and support integrating structured medication review into multidisciplinary falls pathways.