Potentially inappropriate prescribing before and after nursing home admission among patients with and without dementia

This study was designed to describe changes in the prevalence of potentially inappropriate medication prescribing before and after nursing home admission, and to compare prevalence among residents with and without dementia. This paper extends the research on inappropriate medication prescribing among residents entering a nursing home, with the added feature of comparison by dementia status.

This retrospective cohort study was conducted using data from 59 randomly selected nursing homes in Maryland. Dually eligible (Medicare/Medicaid) residents aged > or = 65 years who were admitted to one of these nursing homes from 1992 to 1995 were eligible for inclusion in the cohort. An expert panel of physicians determined dementia status at admission. Potentially inappropriate prescribing, as defined by the 1997 Beers criteria, was compared using Medicaid prescription claims for up to 12 months before and after admission to characterize monthly prescribing patterns.

The study group included 546 dually eligible nursing home residents with > or = 1 paid prescription claim for the 12 months before or after their admission date. A total of 372 (68%) residents were white, 443 (81%) were unmarried, 408 (75%) were female, and 334 (61%) were diagnosed with dementia at admission. Before nursing home admission, the mean monthly prevalence of potentially inappropriate medications for residents with and without dementia was 20% and 23%, respectively. After admission, the mean monthly prevalence increased to 28% among residents without dementia and decreased to 19% among residents with dementia. After adjusting for the mean number of other prescriptions, sociodemographic factors, and number of comorbid conditions, residents with dementia were as likely as residents without dementia to receive a potentially inappropriate drug before admission (prevalence ratio, 0.97; 95% CI, 0.58-1.62). After admission, residents with dementia were 27% less likely than residents without dementia to receive a potentially inappropriate drug, although the difference did not reach statistical significance (prevalence ratio, 0.73; 95% CI, 0.53-1.01).

Inappropriate medication prescribing was similar before nursing home admission among patients with and without dementia. After admission, the prevalence was lower among residents with dementia, but it did not reach statistical significance.