Cargando…

Gerofit Decreases Medication Use Among Older Veterans

We examined whether Veterans enrolled in Gerofit for at least 12 months between 10/2012-3/2017 saw reductions in their medication utilization (5 sites, 226 Veterans). VA outpatient pharmacy data was used to identify medications used 12-months prior, and 12-months following Gerofit enrollment. Seven...

Descripción completa

Detalles Bibliográficos
Autores principales: Intrator, Orna, Li, Jiejin, Veazie, Peter, Wesgate, Sara
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Oxford University Press 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7742638/
http://dx.doi.org/10.1093/geroni/igaa057.2789
Descripción
Sumario:We examined whether Veterans enrolled in Gerofit for at least 12 months between 10/2012-3/2017 saw reductions in their medication utilization (5 sites, 226 Veterans). VA outpatient pharmacy data was used to identify medications used 12-months prior, and 12-months following Gerofit enrollment. Seven drug classes were identified (cardiovascular, diabetic, lipid, mental health, opioids, vitamins, other medications). Nearest-neighbor propensity-matched analyses was conducted with exact match on number of baseline medications and site. At baseline, Gerofit participants were taking, on average, 11.6 medications (1.7 cardiovascular, 0.6 diabetes, 0.7 lipid lowering, 0.7 mental health, 0.5 opioids, 0.6 vitamins, 7.0 other). At 12-month follow-up, Gerofit patients were taking fewer medications: any (-2.7 [-4.4, -1.0]); cardiovascular (-0.5 [-0.7, -0.3]); diabetes (-0.7 [-1.2, -0.2]); mental health (-0.9 [-1.5, -0.2]); and other (-2.4 [-3.5, -1.4]) compared to matched comparisons. No significant differences at 12-months were found for lipid lowering, opioids, or vitamins. Conclusion: Gerofit participation reduced medication use.