Population-based multidimensional assessment of older people in UK general practice: a cluster-randomised factorial trial.

BACKGROUND: The benefit of multidimensional assessment and management of older people remains controversial. Most trials have been too small to produce adequate evidence to inform policy. We aimed to measure the effects of different approaches to assessment and management of older people. METHODS: We undertook a cluster-randomised factorial trial in 106 general practices (43219 eligible patients aged 75 years and older, 78% participation), comparing (1) universal versus targeted assessment and (2) subsequent management by hospital outpatient geriatric team versus the primary-care team. All participants received a brief multidimensional assessment followed by a nurse-led in-depth assessment in the universal group, whereas in the targeted group the in-depth assessment was offered only to those with problems established at the brief assessment. Referrals to the randomised team (geriatric management or primary care), other medical or social services, health-care workers, or agencies, and emergency referrals to the general practitioner were based on a standard protocol at the in-depth assessment. The primary endpoints were mortality, admissions to hospital and institution, and quality of life. Analysis was by intention to treat and per protocol. This trial has been assigned the International Standardised Randomised Controlled Trial Number ISRCTN23494848. FINDINGS: Mortality and hospital or institutional admissions did not differ between groups. During 3 years' follow-up, significant improvements in quality of life resulted from universal versus targeted assessment in terms of homecare, and from management by geriatric team versus primary-care team, in terms of mobility, social interaction, and morale. However, only the result for social interaction was consistent with a small but important effect. INTERPRETATION: The different forms of multidimensional assessment offered almost no differences in patient outcome.