Abstract

Background: COVID-19 pandemic put an unprecedented burden on the hospital system in the USA, however, limited data is available from southern states. Objective: The current study was conducted to evaluate the health outcomes and inpatient costs of COVID-19 patients in a community hospital in central Alabama. Methods: The demographic and clinical data were obtained for patients admitted from March 2020 to March 2022, with a primary diagnosis of COVID-19. Mortality rate, Length of Stay (LOS) at the hospital, total hospital cost and discharge disposition were analyzed against demographic and clinical characteristics of patients. Results: A cohort of 2135 adult patients with a total of 2222 admissions were selected. Mean age of the cohort was 61.27 years, with 51.33% females. Average LOS was 10.06 days and mean hospital cost was $20226. African Americans had shorter LOS and lower total hospital cost. Patients with older age, obesity, more comorbidities, and End-Stage Renal Disease (ESRD) had significantly longer LOS and total higher hospital cost. White patients, older patients, and patients with comorbidities, as well as patients with obesity or ESRD, had higher mortality rate and greater likelihood to be discharged to rehab or skilled nursing facility, instead of home care. Conclusion: The study reveals how demographics and comorbidities are associated with longer LOS, higher hospital cost and higher mortality, among the patients admitted to a community hospital in central Alabama. The results would facilitate early assessment of disease prognosis upon admission and efficient resource allocation and discharge planning.

Keywords: COVID-19; Community hospital; Race; Length of stay; Inpatient cost

Introduction

COVID-19 was first identified in Wuhan, China, in November 2019. Since then, it has spread globally, claiming over 6,927,000 lives, with 1,127,928 deaths in the US alone [1-4]. According to Centers for Medicare and Medicaid Services (CMS), the COVID-19 pandemic has driven up healthcare spending by 9.7% in the year 2020, an unprecedented impact on the healthcare system [5]. As a result, the pandemic has created tremendous pressure on US hospitals, involving shortages in supply chains and staffing, and loss in revenue [6]. Identification of risk factors influencing the prognosis of COVID-19 infections and associated treatment costs could assist in efficient resource allocation and care strategy [7, 8]. Conditions such as hypertension, diabetes, Chronic Kidney Disease (CKD), and Chronic Obstructive Pulmonary Disease (COPD) have been identified as risk factors, and obesity and End-Stage Renal Disease (ESRD) have also been found to be associated with poorer outcomes from COVID-19 infections [915].

Length of stay (LOS) and hospital cost are often used as measures of patient outcome [16-18]. Several studies have been conducted to examine these parameters to analyze COVID-19 outcomes using nationwide as well as international data [19-22]. Notably, the healthcare systems of the southern US states have been historically set apart from the rest of the nation, ranked in the bottom 20% of the US in terms of healthcare quality, which has been connected to geographic, demographic, and socioeconomic differences [23,24]. Disparity between southern healthcare and the general US has also been observed with cancer mortality and HIV treatment [25-28]. Therefore, gaining perspective on COVID-19 patient outcomes in southern community hospitals could facilitate identifying unique issues in this demographic and promote efficient resource distribution. In this study, we compared the LOS and hospital costs of patients hospitalized in Jackson Hospital, a community hospital in Montgomery, AL, based on their demographic and clinical factors, while also comparing the mortality rate and discharge dispositions between these patients.

Methods

Data Source and Population

Adult patients (age 18 or above) who were admitted to Jackson Hospital (Montgomery, Alabama, USA) from March 1st, 2020 to March 1st, 2022 with the primary diagnosis of COVID-19 (ICD-10 code U071) were identified via Vizient Clinical Data Base. Personally identifiable information was removed from patient data before sharing with the investigators. The study was exempted from Jackson Hospital IRB committee. All data are fully compliant with the Health Insurance Portability and Accountability Act (HIPAA).

Outcomes and Measures

Primary outcomes of this study included LOS in the hospital measured in days, total inpatient hospital cost in US dollars rounded to the nearest whole dollar amounts, and mortality rate. In addition, discharge disposition, defined as chances of discharging a patient home, including home health, versus discharging to a rehab or Skilled Nursing Facility (SNF), was also evaluated. Outcomes were compared among patient population based on their gender, race, age, and other medical conditions. ICD10 diagnosis codes for hypertension, diabetes, COPD, CKD, obesity and ESRD were used to identify specific comorbidities and risk factors.

Statistical Analysis

Descriptive analyses reported means and standard deviation for continuous variables and frequency (%) for categorical variables. The association between the independent variables and dependent variables (LOS, cost) were evaluated using a generalized linear model with log link and variance as a gamma function. A multivariable analysis was used to measure the least square mean difference of LOS and costs between categories of an independent variable after adjusting for other covariates. A multivariable logistic regression analysis was conducted to find the association between various socio-demographic, personal, and clinical factors, and the likelihood of death. All statistical analyses were conducted using SAS version 9.4 and p-value less than 0.05 was considered as statistically significant. Graphs were generated using GraphPad Prism version 9.5.1 for Windows, GraphPad Software, San Diego, California USA.

Results

Study population

A cohort of 2135 adult patients with a collective 2222 admissions between March 2020 and March 2022, were selected for the current study. Overall, 51.33% of patients were female, 37.47% were non-Hispanic white and 56.3% were non-Hispanic black. The overall mean age was 61.27 ± 16.99 years, while that of non-Hispanic black patients was 59.01 years and non-Hispanic white patients was 65.87 years. Hypertension (74.9%), diabetes (44.1%), CKD (16.02%), and COPD (22.20%) were the most common comorbidities among the patient population. 26.7% of the patient cohort also had a diagnosis of obesity, while 8.34% of patients had a diagnosis of ESRD. The distribution of the cohort is summarized in Table 1.

Table 1: Patient demographics and concurrent medical conditions. There are 2135 adult patients in the cohort with a total of 2222 admissions. Mean age for the entire cohort was 61.27 ± 1 6.99 years (95% CI: 60.55-61.99 years).

Association between socio-demographic, clinical factors, and LOS

The average LOS for patients with a primary diagnosis of COVID-19 was 10.06 days. Mean LOS was significantly higher for males compared to females by 8.5%. Multivariable analysis found that the mean LOS for non-Hispanic Black is significantly less compared to non-Hispanic white by about 10%. After adjusting for other covariates, the mean LOS for non-Hispanic Black is less compared to non-Hispanic white by 1.16 days (Table 2). Mean LOS increased by 12.1% for every 10-year increase in age. Compared to those with no comorbidities, a gradual increase in mean LOS was observed with increasing number of comorbidities (Table 3) and the association was found to be statistically significant (Table 2). Similarly, significant association was observed between LOS and presence of obesity and ESRD, with an increase in mean LOS by 3.56 days (42.6% increase) and 4.89 days (50.2% increase), respectively (Tables 2 and 3).

Table 2. Association between LOS, hospital cost and demographics and clinical conditions. The unadjusted mean and median values for LOS and total hospital cost is shown for the entire cohort stratified based on the independent variables sex, demographics, and risk factors. The association between LOS, cost and the independent variables is represented by the coefficient calculated using a multivariate analysis as described in the methods section.

Table 3: Adjusted least square mean difference in LOS and inpatient cost between categories of independent variables.

Association between socio-demographic, clinical factors, and total hospital cost

The mean inpatient total cost for the entire cohort was $20226. The mean inpatient cost for males was significantly higher compared to females by 8.1% (Table 2). For every 10-year increase in age, the mean cost increased significantly by 9.6%. The mean inpatient total cost was significantly less by 13.8% for non-Hispanic black patients compared to non-Hispanic white patients. The multivariable analysis with a generalized linear model showed that after controlling for demographic, personal, and clinical characteristics , the adjusted mean inpatient total cost was statistically significantly higher by $2802.06 (95% CI: 683.12-4921.00, p =0.0096) for non-Hispanic white compared to non-Hispanic black people. Statistically significant association was also observed between total cost and increasing number of comorbidities (Table 2) and the mean inpatient total cost increased in ascending order as the number of comorbidities increased (Table 3). The multivariable analysis also showed that the mean inpatient total cost was significantly higher for those who were obese compared to those who were not by at least 46.4% (1.46, 95% CI: 1.35-1.59, p<0.0001) and the adjusted mean inpatient total cost was significantly higher by $7800 (95% CI: $5528.28$10072, p<0.0001). The presence of ESRD also increased the mean inpatient total cost significantly. For those with ESRD, the mean inpatient total cost was 78.3% higher (1.78, 95% CI: 1.572.02, p<0.0001) (Table 2 and 3).

Association between socio-demographic, clinical factors, and odds of death

The odds of death were not significantly different between males and females (Figure 1). Compared to non-Hispanic white, the odds of death were significantly lower for non-Hispanic blacks by 36% (Odds ratio = 0.64, 95% CI: 0.50-0.81, p = 0.0364). With every year increase in age, the odds of death also increased by 4.3% (Odds ratio = 1.04, 95% CI: 1.03-1.05, p <0.001). Compared to those with no comorbidities, the odds of death were significantly higher by at least 42.5% (Odds ratio = 2.35, 95% CI: 1.43-3.86, p = 0.0003) for those with at least three comorbidities. Obesity had a significant association with the odds of death. The odds of death were significantly higher for obese patients by 67.6% (Odds ratio = 1.68, 95% CI: 1.28-2.19, p = 0.0002) compared to non-obese patients. The presence of ESRD increased the odds of death by almost 104% (Odds ratio = 2.04, 95% CI: 1.42-2.92, p = 0.0001) (Figure 1).

Association between socio-demographic, clinical factors, and discharge disposition

For non-Hispanic black patients, the odds of the final disposition being SNF/Rehab instead of home was 34% less compared to non-Hispanic whites (Figure 1). As age increased by one year, the odds of the final disposition being SNF/rehab instead of home increased by 8%. Compared to those with no comorbidity, the odds of the final disposition being SNF/Rehab instead of home significantly increased by 2, 2.5, and 3.3 folds for those with one, two, or at least three comorbidities, respectively. The odds of the final disposition being SNF/Rehab instead of home was increased by 3.7 folds for those with ESRD compared to those without (Figure 1).

Figure 1: Odds of death and discharge disposition among patients grouped by demographic and clinical conditions. The Forest plots represent odds ratio (± 95% CI) of (A) death; and (B) discharge to skilled nursing facility or inpatient rehab facility compared to home care (routine or organized home health service).