Abstract
APMs have potential for adverse effects and mortality risk for NH residents with dementia. CMS initiated its “National Partnership” (March 2012) to reduce APM use, implemented public reporting an APM quality measure (July 2012), and revised NH surveyor guidelines and training to identify unnecessary APMs and compliance with new dementia care standards (May 2013). We determine NH response to these initiatives, by examining change in APM and physical restraint utilization among long-stay NH residents and whether these vary with resident cognitive functioning. Using data from 2011–2013 Minimum Data Set, we examined long-stay residents in free-standing facilities that did not have CMS-approved indications for APM (N≈8million). Linear probability models controlled for resident and facility characteristics and determined how restraint and APM rates change in response to each CMS initiative. Results were stratified into groups: no dementia/mental illness, dementia without symptoms (i.e., behavioral symptoms, delusions/hallucinations), dementia with symptoms, and severe mental illness consistent with APM utilization (e.g., bipolar, psychosis, and severe depression). Pre-initiative physical restraint use averaged 2.5%, with rates higher among those with dementia and severe mental illness. Among all groups, restraint use declined with each initiative. Pre-initiative APM use averaged 23.2%, though rates varied significantly with cognitive functioning (7.0%-66.2%). All groups saw declines in APM use with each CMS initiative; largest decline was among dementia residents without symptoms (27.8%), smallest among residents with mental illness (9.7%). Regulatory efforts to reduce APM prescribing had a significant impact, without an increase in physical restraint use; effectiveness varies with resident cognitive functioning.