Delaying and reversing frailty and building resilience: developing and launching an optimum intervention for older adults in primary care

Background: Frailty is a vulnerability to external stressors that increases the risks of illness, dependency and mortality, while resilience is the ability to withstand stressors. Evidence for effective intervention in primary care is lacking, especially for those who have pre-frailty or mild frailty, and the rate of intervention is low. There is a need to provide person-centered primary care interventions to support older people in combatting frailty and building resilience. Hypothesis: Frailty can be reversed and resilience enhanced with optimised primary care intervention. Methods: Systematic review of scientific databases was conducted for controlled studies on primary care frailty interventions. Effectiveness was scored in terms of change of frailty status and ease of implementation in terms of human resource, marginal cost and time requirements. Meta-analysis was performed using the random effects model. An intervention was co-designed with public and patient involvement (PPI) in three stages. Eighteen over 65-year-olds helped co-design an exercise intervention in two group discussions. Ninety-four contributed feedback in one-on-one interviews during a nine-month feasibility assessment. Ten helped optimise the intervention in three online workshops. Multidisciplinary team input supported co-design. Feasibility of the exercise intervention was assessed using the Bowen model for acceptability; participation; demand; implementation; practicality; adaptation; integration; expansion; and limited-efficacy. The definitive intervention was tested in a multicentre, randomised-controlled, parallel-arm trial in six primary care practices in Ireland. 168 adults, aged 65 and over with Clinical Frailty Scale score =5, presenting to primary care from December 2020 to April 2021 were randomised to intervention or usual care. A home-based exercise regime, emphasising strength, and dietary protein guidance was provided by their GP with one month check-in call and three-month follow-up. Effectiveness was measured by comparing SHARE-Frailty Instrument between groups, on intention-to-treat basis. Secondary outcomes included muscle mass, bone mass and biological age measured by bioelectrical impedance analysis (BIA). Results: 925 studies satisfied review search criteria and 46 were assessed (15,690 participants). Interventions with both muscle strength training and protein supplementation consistently placed highest for effectiveness and ease of implementation. 31 studies with 4,794 participants were included in meta-analysis. Interventions using predominantly resistance-based exercise and nutrition supplementation improved frailty status versus control (RR = 0.62 (CI 0.48–0.79), I2 = 0%). Home-based resistance exercises and dietary protein guidance were co-designed with 112 older people. The intervention satisfied all Bowen feasibility criteria. 359 adults were assessed for RCT eligibility and 168 enrolled. 156 (92.9%) attended follow-up (mean age 77.1 years (SD 5.2); 67.3% women; 79 intervention, 77 control). At baseline, frailty proportions were 17.7% in the intervention and 16.9% in the control group. At follow-up, they were 6.3% and 18.2%, respectively. The odds ratio (OR) of being frail for the intervention group was 0.23 times that of the control group (95% confidence interval (CI): 0.07 to 0.72; P=0.011), adjusting for age, gender and site. The absolute risk reduction was 11.9% (95% CI: 0.8% to 22.9%; P=0.018). The number needed to treat was 8.4. All BIA parameters improved, bone mass significantly (P=0.040). 66.2% of participants found the intervention easy, 69.0% reported feeling better. Conclusions: A co-designed primary care intervention of exercises and dietary protein guidance significantly reduced frailty and improved self-reported health. Systematic review, meta-analysis, PPI and feasibility assessment all contributed to optimisation. This intervention appeared to reverse frailty in over one in nine participants.