SOURCE : Oliver Reich | PLOS ONE 2014
Methods: Using health care claims data of four health insurers for a sample of managed care patients 65 years of age and older to compare persons on PIM with persons not on PIM. Beers’ 2012 and PRISCUS criteria were used to determine the potential inappropriateness of prescribed medications. The sample included 169490 elderly patients on PIM and 339178 patients not on PIM in the time period of January 1, 2008 through December 31, 2012. Prevalence estimates are standardized to the population of Switzerland. Associations between PIM and hospitalizations were examined by multivariate Cox regression analyses controlling for possible confounding variables.
Results: The estimated prevalence of PIM use in our managed care sample was 22.5%. Logistic regression analysis showed that number of different medications used in the previous year, total costs in the previous year and hospitalization in the previous year all significantly increased the likelihood of receiving PIM. Multiple Cox regression analysis revealed that those on cumulative levels of PIM use acted significantly as a factor related to greater hospitalization rates: the adjusted HR was 1.13 (95% CI 1.07–1.19) for 1 PIM, 1.27 (95% CI 1.19–1.35) for 2 PIM, 1.35 (95% CI 1.22–1.50) for 3 PIM, and 1.63 (95% CI 1.40–1.90) for more than 3 PIM compared to no PIM use.
Conclusions: The prevalence of PIM in managed care health plans are widely found but seem to be much lower than rates of non-managed care plans. Furthermore, our study revealed a significant association with adverse outcomes in terms of hospitalizations. These findings stress the need for further development of interventions to decrease drug-related problems and manage patients with multiple chronic conditions.