Pharmacist-guided deprescribing for frail older adults in nursing homes using STOPPFrail

Abstract

Abstract Introduction Frail older adults in nursing homes are often prescribed inappropriate medications. Whilst STOPPFrail can aid deprescribing in such persons with limited life expectancy, there has been insufficient research showing pharmacists’ application of this tool.[1,2] Aim To evaluate the impact of pharmacist-led application of STOPPFrail to frail older nursing home residents. Methods Nursing home residents aged ≥65 years with advanced frailty and a poor one-year survival prognosis were eligible to partake.[1] A convenience sample of nursing homes was recruited via nursing home GPs within the research team’s professional network. After recruitment by their general practitioner (GP), a research pharmacist (EH) reviewed participants using STOPPFrail to identify potentially inappropriate medications (PIMs). Deprescribing recommendations for each participant were provided to GPs via email and discussed via videoconference, whereby recommendations were implemented if GPs perceived they were clinically appropriate. Outcome measures were recorded at baseline, post intervention, and at six months. Descriptive and inferential statistics were performed, where p<0.05 represents a statistically significant difference from baseline. Results Ninety-nine patients from six nursing homes were recruited (76% female). The mean age was 86 years (standard deviation [SD]: 6.9), and the mean number of comorbidities was 15 (SD: 5.3). Nearly all patients (94%) were prescribed ≥1 PIM at baseline. From 349 clinically relevant STOPPFrail recommendations to deprescribe 329 PIMs, 58% were accepted (n=203) and 55% were implemented (n=193), resulting in 176/329 PIMs (53%) being deprescribed – with 167/176 (95%) persisting at six months. Recommendations most commonly concerned medications without a clear indication (29.5%), vitamin D (16.9%), antipsychotics (8.9%), memantine (7.2%), and proton pump inhibitors (7.2%). Relating to baseline, post intervention, and at six months: i)the mean ± SD number of prescribed medications was 16.0±6.1, 14.6±5.7 (p<0.01), and 15.4±5.5 (p=0.03); ii)the monthly mean ± SD cost of patients’ medications was €186.8±123.7, €172.7±119.0 (p<0.01), and €186.4±121.2 (p=0.949); iii)the mean ± SD Modified Medication Appropriateness Index (MMAI) was 2.19±0.73, 2.11±0.74 (p<0.01), and 2.08±0.71 (p<0.01); iv)the median [interquartile range] Drug Burden Index (DBI) was 1.03 [0.50-2.00], 0.93 [0.50-1.80] (p<0.01), and 0.93 [0.50-1.80] (p<0.01); v)the mean ± SD Anticholinergic Cognitive Burden (ACB) was 4.27±2.45, 3.86±2.54 (p<0.01), and 3.94±2.70 (p=0.03). The mean ± SD EQ-5D-5L QoL summary score was 0.183±0.286 at baseline and 0.159±0.312 (p=0.18) at six months. The mean ± SD EQ-5D-5L visual analogue score was 60.3±22.8 at baseline and 61.8±19.9 (p=0.45) at six months. Over two-thirds of residents (70/99) were alive at six months. There were 36 falls, one emergency department visit, and eight non-elective hospitalisations six months prior to baseline compared to 31, four, and seven respectively in the six months post intervention. Conclusion To the authors’ knowledge, this is the first study to apply STOPPFrail to residents of long-term care facilities and also the first to have a pharmacist do so. Although this study lacked a control group, the intervention significantly reduced PIMs, medication costs (initially, but not maintained at follow-up), and both the anticholinergic and sedative burden. The intervention maintained QoL and reduced fall frequency. Given these results, policymakers, nursing home facilities, and clinicians should consider the wider implementation of pharmacist-led STOPPFrail-guided reviews to optimise the medications of frail older adults.