Date of Award

January 2021

Document Type

Thesis

Degree Name

Medical Doctor (MD)

Department

Medicine

First Advisor

Shaili Gupta

Abstract

Objective:The aim of this study is to investigate sociodemographic characteristics and comorbidities associated with hospitalization and clinical outcomes among veterans diagnosed with COVID-19 infection. Methods: A chart review of veterans diagnosed with COVID-19 in six states of New England region (Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, and Vermont) was performed using the veterans Affairs healthcare system electronic medical record. Relevant sociodemographic information and comorbidities were extracted. Data on the clinical course including treatment and outcomes among hospitalized patients were collected. Univariate and multivariate analyses were conducted to determine risk factors associated with hospitalization, ICU admission and length of stay. In the outpatient setting, analyses were performed on outcomes including recovery from illness and time to recovery. Addresses were geocoded using the Census Bureau Lookup Tool. For Veterans residing in Connecticut, geographic characteristics such as neighborhood socioeconomic status, and their association with relevant outcomes were analyzed. Statistical analyses were performed using STATA. Results: A total of 214 veterans with confirmed COVID-19 infection were identified: 62 were hospitalized and 152 remained in the outpatient setting. In univariate analyses, age, BMI over 30, chronic heart disease, type II diabetes mellitus and COPD were predictors of hospitalization. After adjusting for covariates, increased age (OR:1.1 95% C.I (1.06 - 1.14) and COPD (OR: 2.8 95% C.I (0.97 – 8.1) were associated with an increased odd of hospitalization with COVID-19 infection. Patients with Type II diabetes mellitus were less likely to be hospitalized (OR:0,14 95% C.I: 0.04-0.46). ICU admission was associated with male gender, immunosuppression, and chronic liver disease. Chronic liver disease was the only significant predictor of length of stay. Overall, 6% of patients died during the study period. Among hospitalized patients, the mortality rate was 17%. In the outpatient setting, older age and chronic heart disease were significantly associated with prolonged time to recovery. Neighborhood socioeconomic scores were higher among hospitalized patients, but was not significantly associated with hospitalization after adjusting for age.

Discussion/Conclusion:This study identified several risk factors associated with poor clinical outcomes among veterans diagnosed with COVID-19. We found that old age, COPD and diabetic status were significant predictors of hospitalization from SARS-CoV-2 among veterans, in univariate and multivariate analyses. All factors considered, age was the leading predictor of clinical outcomes, the risk of hospitalization increasing by 10% with each one-yea increase in age and time to recovery increased by 0.1 days for every one-year increase in age in the outpatient setting. Adverse outcomes, such as ICU admission, were associated with male gender, immunosuppression and chronic liver disease. While hospitalizations were higher among patients from socioeconomically advantaged backgrounds in univariate analysis, this association was no longer present when adjusted for age. When NSES was classified in terciles, patients living in highest NSES neighborhoods were most likely to be hospitalized after adjusting for age.

Comments

This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.

Download

Share

COinS