Department of Family and Preventive Medicine , College of Medicine , UAMS ; ^b Arkansas Center for Health Improvement , Little Rock , AR ; ^c Department of Health Policy and Management , Fay W . Boozman College of Public Health , UAMS , Little Rock , AR ; ^d Arkansas Department of Health , Little Rock , AR ; ^e Department of

Epidemiology , Fay W . Boozman College of Public Health , UAMS , Little Rock , AR ; ^f Department of Pediatrics , College of Medicine , UAMS , Little Rock , AR ;

Abstract

confirmed cases of COVID-19 in Arkansas . In this retrospective case cohort study , we assess the preexisting conditions associated with poor outcomes ( hospitalization , ICU admission , intubation , and death ) in Arkansans infected with COVID-19 using historical health care administrative data contained in the Arkansas All-Payer Claims Database . Our findings of elevated risk for severe outcomes among COVID-19 patients with a history of kidney failure , immunocompromised conditions , serious cardiovascular diseases , and diabetes mellitus are consistent with those documented by the CDC . In addition , we found an increased risk of death among COVID-19 patients with dementia .

Background

The ongoing coronavirus disease pandemic of 2019 ( COVID-19 ), caused by the 2019 novel coronavirus ( SARS-CoV-2 ), is one of the greatest public health challenges in the past century . As of August 24 , 2020 , there were 57,635 confirmed cases of COVID-19 in Arkansas , including an estimated 4,109 hospitalizations , 1,489 intensive care unit ( ICU ) admissions , 540 patient intubations , and 733 deaths .

On July 17 , 2020 , the Center for Disease Control and Prevention ( CDC ) published a list of underlying health conditions that may contribute to increased risk of COVID-19 case severity . Much of the research cited by the CDC was based on outcomes observed in Wuhan , China . In this confirmatory study , we identified health conditions that are prognostic for severe outcomes among Arkansas COVID-19 patients requiring hospitalization , ICU admission , intubation , or that result in death . Data from the Arkansas Department of Health ( ADH ) registry of Arkansans who have been tested for COVID-19 was linked with historical health care administrative data contained in the Arkansas All-Payer Claims Database ( APCD ).

The CDC has reported that there is strong evidence that serious heart conditions ( e . g ., heart failure , coronary artery disease , or cardiomyopathies ), cancer , chronic kidney disease , chronic obstructive pulmonary disease ( COPD ), obesity , sickle cell disease , solid organ transplantation , and type 2 diabetes mellitus increase the risk of more deleterious outcomes among COVID-19 patients . ¹ There is mixed evidence that asthma , cerebrovascular disease , hypertension , pregnancy , smoking , and use of corticosteroids or other immunosuppressive medications are associated with severe outcomes . The Arkansas population is burdened with a high prevalence of many of these underlying health conditions .

Analysis from the most recent 2018 Behavioral Risk Factor Surveillance System ( BRFSS ) identifies that Arkansas adults have among the highest rates of type 2 diabetes mellitus ( 13.9 %), kidney disease ( 3.7 %), hypertension ( 38.5 %), COPD ( 9.8 %), and self-reported fair or poor health ( 24.7 %) in the U . S . Also , U . S . Cancer Statistics reports that Arkansas has among the highest prevalence rates of cancer ( 471.4 per 100,000 ). ² Estimating the elevated risk of COVID-19 acuity due to these conditions can help shape state policies and communication to this vulnerable population .

A focal aim of this study was to assess and document conditions associated with poor outcomes in Arkansans infected with COVID-19 . In addition , we have demonstrated a real-time application of an APCD to conduct a retrospective cohort study on COVID-19 patients .

Method

We conducted a retrospective case cohort study on Arkansans who tested positive for COVID-19 by linking two databases — the ADH

COVID-19 registry and the APCD contained within the Arkansas Healthcare Transparency Initiative ( AHTI ).

The ADH COVID-19 registry contains demographic and geographic characteristics ( e . g ., race , gender , county of residence ) of COVID-19 patients in a secure database maintained at ADH . In 2015 , the Arkansas General Assembly established the AHTI to address the state ’ s data needs to improve health and support research . This initiative requires health care insurers — including Medicaid , Medicare , and commercial insurers throughout the state — to submit medical , pharmacy , and dental claims inclusive of diagnostic information from Arkansans to the APCD . This large-scale database is securely administered by the Arkansas Center for Health Improvement ( ACHI ) and governed by the Arkansas Insurance Department ( AID ). Conditions that individuals were diagnosed with prior to the coronavirus infection were identified using the APCD , which contains claims starting in 2013 for approximately 80 % of Arkansans .

In order to protect any personally identifiable information ( PII ) within the data , the APCD uses an anonymous and unique 44-character identifier — called a HASH ID — that incorporates a person ’ s last name and date of birth . This HASH ID is further combined with gender to improve data linkage accuracy . A HASH ID was constructed for individuals within the ADH COVID-19 registry and was linked to APCD enrollment , medical , and pharmaceutical claims records from 2013 through June 2019 . Medicare data were only available through December 2017 .

To ensure that the study sample was limited to individuals with sufficient clinical history in the APCD to observe diagnoses related to chronic risk conditions , we applied a set of inclusion criteria that included : ( 1 ) COVID-19 positive individuals in ADH COVID-19 database with valid age and gender information as of August

114 • The Journal of the Arkansas Medical Society www . ArkMed . org