To What Extent Do Local Nursing Home Prescribing Patterns Relate to Psychotropic Prescribing in Assisted Living?

The assisted living sample came from seven states chosen for expected prescribing variability, so it may not represent other states. Many invited assisted living communities refused or remained pending, so the sample may be biased, although participating and non-participating sites did not differ by size. Assisted living case-mix data, such as dementia and Medicaid shares, were self-reported by administrators, so estimates may be imprecise. The assisted living antipsychotic measure did not exclude residents with Huntington's disease or a six-month prognosis as the nursing home measure did, although Huntington's disease prevalence is expected to be low. The study compared assisted living communities with nearby nursing homes but could not validly compare assisted living communities with each other by distance.

The article lists research support from NIH and the U.S. government (non-P.H.S.); the authors declared no conflicts of interest.

Do not read these prescribing patterns as proof that changing nursing home prescribing will change assisted living prescribing. This observational study only describes medication records and adjusted associations.