Disparities in Access to Medical Care After Hospitalization for Severe COVID-19 Pneumonia

Abstract

RATIONALE: Communities of color are bearing a disproportionate burden of coronavirus disease 2019 (COVID-19) morbidity and mortality. Social determinants of health have resulted in higher prevalence and severity of COVID-19 among minority groups. Published work on COVID-19 disparities has focused on higher transmission, hospitalization, and mortality risk among people of color, but studies on disparities in the post-acute care setting are scarce. Our aim was to identify socioeconomic disparities in health resource utilization after hospital discharge. METHODS: This was a retrospective study. We identified adult patients who were hospitalized at CUIMC or the Allen Hospital from March 1st through April 30th 2020, had a positive RT-PCR for severe acute respiratory syndrome coronavirus 2 (SARS-COV-2), developed severe hypoxemic respiratory failure requiring invasive mechanical ventilation, and were successfully discharged from the hospital without need for ventilator support. Patients who received a tracheostomy and were weaned off the ventilator prior to discharge were included. Exclusion criteria included transfer from or to another institution, prior tracheostomy, in-hospital death, and discharge with a ventilator. RESULTS: We identified 195 patients meeting inclusion criteria. The median age was 59 (IQR 47-67), and 135 (66.5%) were men. There were 25 (12.8%) patients who were uninsured and 116 (59.5%) patients who had public insurance. There were 121 (62%) Hispanic, 34 (17%) Black, and 18 (9%) White patients. Uninsured patients within our cohort were more likely to be Hispanic and Spanish-speaking (p=0.027;p<0.001, respectively). Uninsured patients were also more likely to be discharged to home (p<0.001) than to a rehabilitation facility. 8.8% of patients were readmitted to CUIMC within 30 days and 41.5% saw a medical provider at CUIMC within 30 days of discharge. Insurance status did not predict 30-day re-hospitalization or completion of outpatient follow-up, although our study was underpowered to answer these questions. CONCLUSION: Our study demonstrated that race/ethnicity and primary language are associated with insurance status with Hispanic and Spanish-speaking patients being more likely to be uninsured. Uninsured patients were more likely to be discharged home after hospitalization, rather than to facility for further care and rehabilitation. We did not demonstrate any short-term differences in 30-day re-hospitalization rates or follow-up visits but we suspect socioeconomic disparities represent a significant barrier to adequate follow-up care in the long term. We plan to investigate this further with longitudinal follow-up and survey data.