[Estimation of the risk of drug-related problems in nursing home residents : A retrospective analysis of routine data].

Background: Polypharmacy and drug-related problems are major challenges in the care and treatment of nursing home residents. Many interventional studies showed disappointing results, which lead to the question if this could also be due to the selection of the target parameters of these studies.

Material And Methods: A routine data set from six long-term care facilities was retrospectively analyzed. The question is if the recently validated medication risk score (MERIS) is suitable for carrying out a risk assessment in a population of nursing home residents. Associations between MERIS and the dependent variables hospital admissions and falls over 12 months and a weight loss of ≥ 5% over 3 months were examined.

Results: Out of 495 residents 38.6% (n = 191) have a high risk of drug-related problems according to MERIS. A univariate regression analysis showed a significantly increased risk of hospital admissions (OR 2.2; p < 0.001) and weight loss of ≥ 5% (OR 1.95; p = 0.041) with high MERIS, but no significant association with falls. In the multivariate regression the risk of hospitalization was increased by diabetes mellitus (OR 1.88; p = 0.004), falls in the same period (OR 1.91; p = 0.001), positive MERIS (OR 1.75; p = 0.006) and decreased with stable weight (OR 0.88; p = 0.004).

Conclusion: The results indicate the potential of the score for future research projects and individual risk assessment; however, due to the limitations of retrospective secondary analyses further studies are required.