COVID Mortality Variation in VA Due to Patient, Not Facility, Characteristics
January 29, 2021
Takeaway: Dr. Dawn Bravata collaborated with local VAMC leaders to develop an operational dataset to monitor COVID response and outcomes. As a result of this work and some of the questions posed by stakeholders, the project turned into a larger national study examining the substantial variability across facilities in terms of patient volume and outcomes. Of great interest to national VA leaders, (no one outside VA will know who these people are, so would drop, given that this is going on a public web site), differences in mortality reflect variations in patient characteristics (e.g., age), as opposed to facility-level characteristics.
Dr. Bravata, part of HSR&D’s Center for Health Information and Communication (CHIC) in Indianapolis, IN, and colleagues conducted a study (funded through QUERI) to answer two questions: 1) Were differences in facility COVID-19 mortality due to variation in COVID-19 patient characteristics, or do facility factors contribute to COVID-19 mortality differences? 2) Were strains on critical care capacity associated with increased COVID-19 mortality?
To address the first question, investigators identified all VA patients with a positive COVID-19 test date from March 1, 2020 to May 31, 2020 and followed them for 45 days after the positive test. Among 13,510 COVID-19 positive Veterans, cared for at 160 VA facilities, the facility 45-day all-cause mortality rate ranged from 0% to 30% (facility median: 9%). The facility mortality for COVID-19 inpatients ranged from 0% to 100% (facility median: 18%). The facility mortality for COVID-19 ICU patients ranged from 0% to 100% (facility median: 29%). The facility mortality for COVID-19 mechanical ventilator patients ranged from 0% to 100% (facility median: 53%). Marked heterogeneity in COVID-19 patient volume, patient characteristics, and mortality were observed across VA facilities; however, the majority of variation in facility mortality was attributable to differences in patient characteristics (e.g., age).
To address the second question, investigators examined 8,516 patients with COVID-19 (test date: March through August 2020) who were admitted to 88 VA facilities. COVID-19 ICU demand was defined as the mean number of patients with COVID-19 in the ICU during a patient’s stay divided by the maximum number of patients with COVID-19 in the ICU. The adjusted hazard ratio for all-cause mortality through 30-days post-discharge was 1.9 when the COVID-19 ICU demand was greater than 75%. (?? What does this mean??)
Facility-level data reports were provided to VA facilities upon request; these reports described an individual facility within the context of all other VAMCs. In some cases, these reports provided reassurance to VAMC leadership (about?? This sentence read a bit weird so wasn’t sure about the right word here) their diligent work to provide excellent care to patients with COVID-19, which was evident in lower than expected mortality.
Bravata D, Perkins A, Myers L, et al. Association of intensive care unit patient load and demand with mortality rates in US Department of Veterans Affairs Hospitals during the COVID-19 Pandemic. JAMA Network Open. January 19, 2021;4(1):e2034266.