Racial Disparities Shift in Observation Status in Hospital Admissions for Avoidable Conditions

Being admitted to a hospital for something that might have been avoided is sometimes used as a marker for the presence of quality primary care in a patient’s life. Compared with whites, racial and ethnic minorities have much higher rates of potentially avoidable hospitalizations. And blacks are less likely than whites to have a source of usual primary care.

Hospital readmissions are tracked through the Hospital Readmission Reduction Program (HRRP), a value-based purchasing program established in 2010 as part of the Affordable Care Act. The HRRP pays hospitals based on their readmissions. While early data indicated that hospital readmissions were falling, other studies raised concerns that the reduction stemmed from hospitals changing how they use observation status and diagnostic codes.

When classified under observation status, a Medicare beneficiary may be responsible for a larger and more variable share of their hospital bills. Instead of being billed under the hospital benefit (Part A) as an officially admitted patient, under observation status, a patient is considered an outpatient under Part B benefits. In addition, time spent under observation also does not count towards the 3-day inpatient hospital stay requirement in order for Medicare to pay for skilled-nursing facility coverage after a hospitalization.

A study released Monday used national Medicare data to try and understand disparities between black and white patients in avoidable hospitalizations for ambulatory care—sensitive conditions. Among the questions asked: If possibly avoidable hospitalizations are increasing, is it because hospitals are switching to observation status instead? And are there certain types of health care conditions that are influencing any improvement?

Writing in the June issue of Health Affairs, the researchers found that upon first look, it did appear that the gap between black and white patients had begun to narrow.

However, further analysis showed that black patients were being coded as admitted under observation status at a higher rate over time than white patients.

The researchers used Medicare data for 2011-2015, using a 20% national sample from the Inpatient File and the Outpatient File and a linear regression model to estimate yearly risk-adjusted rates for black and white beneficiaries. Time was treated as a categorical predictor of avoidable hospitalizations and observation stays. In addition, to quantify the changes in rates over time and to allow to comparisons by race, they estimated risk-adjusted slopes by performing another linear regression model using time as a continuous predictor, using race and its interaction with time.

At the outset, they found that black patients, as compared with white patients, were more likely to be younger and more likely to be dually eligible for Medicaid and Medicare, and they had a higher number of chronic conditions, including diabetes, dementia, chronic kidney disease, and history of stroke or transient ischemic attack.

While avoidable hospitalizations due chronic conditions that respond to outpatient care decreased, there was an increase in the gap of avoidable observation stays. The authors said that suggests that circumstances in ambulatory care settings did not change. Rather, it seems that an increase in switches to observation status appeared to be creating a reduction in racial disparities in avoidable hospitalizations.

The researchers noted that it is unknown why black Medicare beneficiaries were more likely to be switched to observation status, but the change has a negative financial impact on these patients. In addition, it raises policy implications as more risk-based alternative payment models take hold; the authors cautioned that pressure to reduce hospital admission should not come at the expense of patients.

Reference

Figueroa JF, Burke LG, Horneffer KE, Zheng J, Orav EJ, Jha AK, Avoidable hospitalizations and observation stays: shifts in racial disparities. Health Aff (Millwood). 2020; (39)6;1065-1071. doi: 10.1377/hlthaff.2019.01019