Investigating Racial Disparities in Length of Stay and Discharge Severity


Investigating Racial Disparities in Length of Stay and Discharge Severity
May 30, 2024 MHO

Investigating Racial Disparities in Length of Stay and Discharge Severity

Previous research in healthcare contexts has shown race-based disparities in treatment. Research specifically in psychiatric contexts also shows disparities, with fewer Black and Latino patients receiving adequate care than their White counterparts. During the COVID-19 pandemic, for example, White patients were less likely to say they had “unmet [behavioral health] needs” compared to Black, Hispanic, or Asian patients [1]. As a result, The Joint Commission has recently called attention to, and issued requirements to help fight, racial disparities in healthcare [2]. CMS similarly has begun focusing on disparities.

Previously, MHO found racial and ethnic disparities in seclusion and restraint interventions. We wondered, then, if disparities also exist in indicators of care adequacy, namely length of stay (LOS) and/or discharge severity on patient self-report outcomes assessments. To investigate this, four years of data from Adult Inpatient stays at 254 MHO-contracted facilities were surveyed (NWhite = 1,011,766; NBlack = 309,163).

Overall, no differences in LOS were observed by race, where each group’s median LOS was 7 days with little variation year to year. Similarly, LOS differences were not found by diagnostic category, with one exception. In patients diagnosed with “Behavioral syndromes with physical factors”, Black patients had a median LOS of 7 days versus 13 for White patients 1. This diagnostic category is mostly constituted by eating disorders, post-pregnancy disorders, and sleeping disorders. A closer inspection of this category found eating disorders made up over 80% of the “Behavioral syndromes with physical factors” category, and it was the disorder with the longest median LOS. Importantly, 85% of White patients in this category were diagnosed with eating disorders versus only 28% of Black patients. Therefore, once adjusted for diagnostic subcategory, LOS differences disappear.

To further assess potential disparities, discharge severity on patient self-report outcomes assessments was investigated (NWhite = 519,091; NBlack = 144,647). Again, no racial differences were found across a variety of assessments, suggesting both groups are discharged at a similar level of symptom severity.

Together, these analyses suggest no racial disparities in the duration of inpatient treatment or patient self-report of symptom severity at discharge. In other words, regardless of race, patients are leaving hospitals in a similar place of “wellness” and this “wellness” is achieved in a similar amount of time. Future analysis is needed to assess disparities in other ways, such as investigating whether Black versus White patients report differential satisfaction after discharge, and whether these patterns hold for other racial and ethnic groups.

1. This difference was statistically supported by a Brown-Mood’s median test (Z = -5.94, p < .001).


[1] Thomeer, M. B., Moody, M. D., & Yahirun, J. (2023). Racial and Ethnic Disparities in Mental Health and Mental Health Care During The COVID-19 Pandemic. Journal of racial and ethnic health disparities, 10(2), 961 – 976.
[2] The Joint Commission (2022). New requirements to reduce health care disparities. R3 Report: Requirement, rationale, reference, 36.