A multifactorial intervention for frail older people is more than twice as effective among those who are compliant: complier average causal effect analysis of a randomised trial.

What is the effect of a multifactorial intervention on frailty and mobility in frail older people who comply with their allocated treatment? Secondary analysis of a randomised, controlled trial to derive an estimate of complier average causal effect (CACE) of treatment. A total of 241 frail community-dwelling people aged ≥ 70 years. Intervention participants received a 12-month multidisciplinary intervention targeting frailty, with home exercise as an important component. Control participants received usual care. Primary outcomes were frailty, assessed using the Cardiovascular Health Study criteria (range 0 to 5 criteria), and mobility measured using the 12-point Short Physical Performance Battery. Outcomes were assessed 12 months after randomisation. The treating physiotherapist evaluated the amount of treatment received on a 5-point scale. 216 participants (90%) completed the study. The median amount of treatment received was 25 to 50% (range 0 to 100). The CACE (ie, the effect of treatment in participants compliant with allocation) was to reduce frailty by 1.0 frailty criterion (95% CI 0.4 to 1.5) and increase mobility by 3.2 points (95% CI 1.8 to 4.6) at 12 months. The mean CACE was substantially larger than the intention-to-treat effect, which was to reduce frailty by 0.4 frailty criteria (95% CI 0.1 to 0.7) and increase mobility by 1.4 points (95% CI 0.8 to 2.1) at 12 months. Overall, compliance was low in this group of frail people. The effect of the treatment on participants who comply with allocated treatment was substantially greater than the effect of allocation on all trial participants. Australian and New Zealand Trial Registry ANZCTRN12608000250336. [Fairhall N, Sherrington C, Cameron ID, Kurrle SE, Lord SR, Lockwood K, Herbert RD (2016) A multifactorial intervention for frail older people is more than twice as effective among those who are compliant: complier average causal effect analysis of a randomised trial.Journal of Physiotherapy63: 40-44].