Background:
Acute care transitions are particularly difficult for patients experiencing homelessness, yet there are limited data from the perspective of this high‐risk population to guide hospital‐based interventions to improve the quality of discharge care specifically.
Methods:
To better understand the experience of discharge care among homeless patients, we formed a strategic partnership with an area shelter, and together we recruited participants who reported an episode of acute care in the preceding 12 months. We performed structured interviews, with questions about sociodemo‐graphics and components of discharge care, including medication reconciliation and arrangement of follow‐up care. We performed multivariable logistic regression of survey data to determine whether housing assessment by any hospital staff predicted higher performance on any of these discharge components while adjusting for patient demographics (age, race, sex, and reported length of homeless‐ness) and inpatient care versus emergency department–only care. We hypothesized that patients would report a higher‐quality discharge plan if they reported their providers inquired about housing status.
Results:
Ninety‐eight home‐less individuals were enrolled in the study: 78 were male (80%), and reported race/ethnicity was 42% black, 41% white, and 16% Hispanic. Average age was 44 years, and average reported length of homelessness was 2.8 years. Fifty‐one participants (52%) reported their housing status was assessed by any hospital staff during their acute care episode. Of these, only 19 (37%) reported a physician assessed their housing status. The percentages of patients reporting components of high‐quality discharge included: medication reconciliation, 75%; advice on how to pay for medications, 54%; arrangement for follow‐up care with primary care, 53%; arrangement for follow‐up care with mental health provider, 36%; explanation of safe patient activity level, 50%; and transportation assistance for follow‐up visits, 20%. In the multivariable logistic regression we found significant associations between assessment of housing status and advice on how to pay for medications (OR, 4.4; 95% CI, 1.1–17.6) but not medication reconciliation; follow‐up arrangements with mental health provider (OR, 10.5; 95% CI, 1.2–96.1) but not primary care provider; and both explanation of postdischarge patient activity level (OR, 5.3; 95% CI, 1.6–18.4) and transportation assistance for follow‐up visits (OR, 10.1; 95% CI, 1.8–58.0).
Conclusions:
Homeless patients report variable levels of satisfaction with key discharge components during the transition from an acute care setting. Assessment of homelessness by hospital staff correlates with higher performance on some but not all of these measures. Greater attention to addressing the housing status of acute care patients may aid hospital‐level efforts to improve transitions in care for homeless patients.
Disclosures:
S. R. Greysen ‐ none; M. S. Rosenthal ‐ none; G. I. Lucas ‐ none; E. Wang ‐none