Frailty screening in older patients in primary care using routine care data

Abstract

Background: Primary care for frail older people is reported to be suboptimal. A transition toward proactive patient-centred care is needed. We investigated the effectiveness of U-PRIM, a frailty screening intervention based on routine care data, and of U-PRIM followed by U-CARE, a nurse-led personalised care intervention, on daily functioning of frail older people in primary care. Methods: A single-blind, three-armed, cluster-randomised controlled trial including 3092 older patients recruited from 39 general practices was conducted between October 2010 and March 2012, including one-year follow-up. The general practices were randomly assigned to the U-PRIM, U-PRIM + U-CARE, or control groups. The primary outcome of the study was daily functioning measured on the Katz-15 ADL/IADL scale. The secondary outcomes were quality of life (RAND-36), EuroQol (EQ5-D), primary care consultations, hospital admissions, emergency department visits, nursing home admissions and mortality. Analysis was by intention to treat. This trial is registered as NTR2288. Findings: Patients in both intervention groups demonstrated better preservation of daily functioning than those in the control group at 12 months (mean Katz-15 (95% confidence interval): U-PRIM 1·87 (1·77-1·97), U-PRIM+U-CARE 1·88 (1·80-1·96), and control group 2·03 (1·92-2·13); p = 0·03). In pre-specified subgroup analyses, higher educational level positively affected outcomes for patients in the U-PRIM+U-CARE group. No overall differences in quality of life were observed. The patients in the U-PRIM+U-CARE intervention group consulted their general practice more often by telephone compared to patients in the other groups. Interpretation: A frailty screening intervention (U-PRIM) and U-PRIM followed by a nurse-led personalised care intervention (U-CARE) led to better preservation of daily functioning compared to the control group. More highly educated older people had additional benefits from U-CARE, indicating that the effect is dependent on individual patient characteristics. Further refinement is necessary to optimise the U-CARE intervention to a heterogeneous group of frail older people.