Factors impacting sleep center no-show rates post hospital discharge utilizing geospatial coding in Appalachia

Sunil Sharma; Robert Stansbury; Edward Rojas; Priyanka Srinivasan; Kassandra Olgers; Scott Knollinger; Wes Kimble; Brian Hendricks; Timothy Dotson; Brian A Witrick

doi:10.5664/jcsm.11494

Factors impacting sleep center no-show rates post hospital discharge utilizing geospatial coding in Appalachia

J Clin Sleep Med. 2024 Dec 12. doi: 10.5664/jcsm.11494. Online ahead of print.

Authors

Affiliations

¹ Division of Pulmonary, Critical Care, and Sleep Medicine, West Virginia University, Morgantown, WV.
² University of Pittsburgh Department of Medicine.
³ Department of Respiratory Care, Ruby Memorial hospital, Morgantown, WV.
⁴ West Virginia Clinical and Translational Sciences Institute, Morgantown WV.
⁵ Clemson University, Department of Public Health Sciences, Clemson, SC.

PMID: 39663925
DOI: 10.5664/jcsm.11494

Abstract

Study objectives: Screening for early detection of sleep-disordered breathing (SDB) in hospitalized patients has been shown to reduce readmission rates. However, post-discharge polysomnography for confirmation of diagnosis is required. We analyzed factors for "no-shows" using geospatial techniques.

Methods: Data were obtained between September 2019 and September 2023. The outcome for the study was patient's no-show rate (non-compliant) for polysomnography post hospital discharge. Predictors included patient age, gender, BMI, health literacy, distressed community index (DCI) score and distance to sleep center for the patient's zip code of residence. Logistic regression was applied to estimate odds of patient compliance at the patient-level utilizing geospatial mapping technique. Geographically weighted logistic regression was applied to estimate the odds of a zip-code having compliant patients.

Results: Of the 1318 hospitalized patients established as high risk for SDB and referred for an overnight sleep study who were able to be geocoded, 228 were compliant and 1130 were non-compliant. In non-spatial regression analyses, health literacy (aOR = 1.06, 95%CI = 1.03, 1.09), age (aOR=0.99, 95% CI=0.98, 0.99), and drive time (aOR=0.95, 95% CI = 0.92, 0.97) were identified as statistically significant predictors of patient compliance. Spatial regression analyses identified areas that had high and low predictive probability of patient compliance, as well as which community level factors were co-occurring in those areas.

Conclusions: The findings suggest that both patient-level factors and the community where patients live may impact no-show rates. Health literacy was identified as a key modifiable predictor at the patient level. At the community level, we found that predicted probability of patient compliance varied throughout the state. Efforts should focus on enhancing patient education at the individual level and understanding geographical factors to improve compliance.

Keywords: geospatial mapping; health literacy; hospital sleep medicine; no-show rate; sleep health disparities; sleep-disordered breathing.