Social Risk Factors are Associated with Risk for Hospitalization in Home Health Care: A Natural Language Processing Study.
Mollie Hobensack, Jiyoun Song, Sungho Oh and 8 others
PMID 37553081WHAT IT FOUND
Home health notes mentioning social environment, physical environment, education or literacy, food insecurity, or access to care were linked to hospitalization or emergency visits.
These factors may help clinicians spot risk, but they do not prove that treating them helps.
Key findings
01Documented social environment had 42% higher odds of hospitalization or emergency visit; physical environment had 31%, education or literacy had 24%, food insecurity had 13%, and access to care had 11%.
02At least one social risk factor was documented in 91,250 notes (3.9% of 2,341,018 notes), representing approximately 33% of episodes of care (28,628).
03The automated note reader had overall F-score 0.91, precision 0.93, and recall 0.89, but education and literacy had the lowest F-score, 0.77.
STILL TO COME
How it was doneWhat they foundWhat it means for PTsWhat it means for OTsWhat it means for RNs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The study is observational and retrospective, so it shows association, not that documenting or treating a social risk factor caused or prevented hospitalization or emergency visit. It used one home health agency in New York, so results may not apply to other agencies or rural settings. The unit of analysis was episodes of care, not unique patients. Social risk factors came from clinical notes and natural language processing, and documentation may vary by clinician, note type, and over time. Inter-rater agreement for manual annotation was moderate (kappa 0.49), and education and literacy had the lowest automated performance (F-score 0.77). The authors state that the contribution of each social factor to hospitalization or emergency visit risk cannot be determined.
Declared interests
The supplied text does not report a conflicts-of-interest declaration. The publication types list U.S. government Public Health Service and NIH extramural research support.
The easy way to misread this
Do not read these odds ratios as proof that documenting or treating a social risk factor prevents hospitalization. The study is retrospective, one agency, and cannot separate the contribution of each factor from other care or patient characteristics.