Are high-care nursing home residents at greater risk of unplanned hospital admission than other elderly patients when exposed to Beers potentially inappropriate medications?

Abstract

To compare the risk of unplanned hospitalization in high-care nursing home residents taking Beers potentially inappropriate medications (PIM) against that of other elderly. Using an enhanced case-time-control design and conditional logistic regression applied to the pharmaceutical claims and other linked data of 245436 Western Australians aged ≥ 65 years (1993-2005), the study derived odds ratios for unplanned hospitalization in each group, from which attributable fractions, numbers, proportions and rates of PIM-related admissions were derived. Overall, 383150 unplanned hospitalizations were identified. PIM exposure was associated with a similar proportional increase in unplanned hospitalizations in high-care nursing home residents as in other older people; adjusted OR 1.21 (95% CI 1.10-1.34; attributable fraction 17.5%) versus OR 1.19 (95% CI 1.16-1.21; attributable fraction 15.7%). However, high-care nursing home residents had much higher estimated rates of hospitalizations attributed to Beers medications than other elderly (3951 vs 1394 per 100000 person-years). The relative risk of unplanned hospitalization rose similarly in both groups with increasing numbers of different PIM taken (OR 5.1 for 10 vs 0 PIM), but was affected more markedly by 3-month PIM consumption in nursing home residents (OR 4.85, 95% CI 2.40-9.83 for 900 vs 0 PIM daily doses) than in other older adults (OR 2.10, 95% CI 1.73-2.55). High-care nursing home residents do not appear to have a greater relative risk of unplanned hospitalization when given PIM, but do incur a higher absolute burden than other older adults. Physicians should exert caution with Beers medications in all older patients, restricting the number of different PIM and PIM quantity prescribed whenever possible.