LAUSR

OBJECTIVES An increasing number of post-acute care patients are admitted to skilled nursing facilities (SNFs) with behavioral symptoms such as wandering and rejecting care, which can interfere with care and place a patient at risk of social isolation or injury. This study examines whether increasing the qualifications of social service staff through using a greater proportion of qualified social workers (QSWs) instead of paraprofessionals improves patients' behavioral symptoms and reduces the use of antipsychotic medications. DESIGN Secondary data analysis of national data drawn from the Minimum Data Set, Medicare Beneficiary Summary File, and Certification and Survey Provider Enhanced Reporting surveys. SETTING AND PARTICIPANTS Newly admitted, Medicare fee-for-service patients aged ≥65 years who received post-acute care after a hospitalization from 2011-2015 (1,201,096 patient admissions in 5383 unique SNFs). METHODS A regression discontinuity approach is used to identify how changes in the qualifications of social service staffing affect changes in patient outcomes, through exploiting a federal regulation that requires 1 full-time QSW for SNFs with at least 121 beds. RESULTS SNFs with a greater proportion of QSWs improved behavioral symptoms that affected residents or others and reduced the use of antipsychotic medications. Marginally significant improvements were also found in patients' rejecting care, wandering, and having any behavioral symptoms. There was no statistically significant change in delusions. CONCLUSIONS AND IMPLICATIONS Although the focus of most research and effects on improving patient outcomes in SNFs has focused on nursing staff, our results underscore the importance of staffing in social services. Specifically, social service staff with higher qualifications are integral to improving care through reducing patients' behavioral symptoms and avoiding the use of antipsychotic medications. It may be time for SNFs to revisit efforts to improve staffing in social services.