Objectives: To quantify the association between including specific medications deemed potentially inappropriate in the surveyors' interpretive guidelines for nursing homes and the prevalence of use.
Design: Quasi-experimental.
Setting: One thousand one hundred forty-one nursing homes in four U.S. states.
Participants: Residents living in one of the included nursing homes in operation during 1997 (before Beers; n=130,250) and 2000 (after Beers; n=164,889).
Intervention: Inclusion of specific medications deemed potentially inappropriate in the surveyors' interpretive guidelines for nursing homes.
Measurements: Logistic regression models adjusting for clustering effects of residents residing in homes provided estimates of the relationship between the survey process and use of any medications targeted as potentially inappropriate as part of the survey process, as well as those deemed inappropriate but not included.
Results: The use of any potentially inappropriate medication decreased from 42.5% in 1997 to 39.8% in 2000. After adjustment for resident characteristics, residents were less likely to receive any potentially inappropriate medication (odds ratio (OR)=0.85, 95% confidence interval (95% CI)=0.84-0.87), those considered high-severity drugs (those with a high likelihood of a clinically significant adverse event) (OR=0.67, 95% CI=0.65-0.69), or Beers' medications not included in the surveyors' guidelines (OR=0.76, 95% CI=0.74-0.79) in 2000 than in 1997 after the changes to the drug regulations and interpretive guidelines.
Conclusion: Targeting specific drugs in the surveyor's interpretive guidelines as a method to reduce potentially inappropriate medication use may not produce desired gains in medication-use quality improvement. Alternative strategies for nursing homes should be evaluated.