Population‐based assessment of risks for severe COVID‐19 disease outcomes

Abstract Among approximately 4.6 million members of Kaiser Permanente Northern California, we examined associations of severe COVID‐19 with demographic factors and comorbidities. As of July 23, 2021, 16 182 had been hospitalized, 2416 admitted to an ICU, and 1525 died due to COVID‐19. Age was strongly associated with hospitalization, ICU admission, and death. Black persons and Hispanic ethnicity had higher risk of death compared with Whites. Among the comorbidities examined, Alzheimer's disease was associated with the highest risk for hospitalization (aHR 3.19, CI: 2.88–3.52) and death (aHR 4.04, CI: 3.32–4.91). Parkinson's disease had the second highest risk of death (aHR = 2.07, CI: 1.50–2.87).


| INTRODUCTION
Information on risk of severe coronavirus disease 2019 (COVID-19) is important to identify whom might benefit most from COVID-19 prevention including vaccination and assertive treatment. Previous studies reported that older age 1,2 and comorbidities including asthma, chronic obstructive pulmonary disease (COPD), diabetes, and heart and renal disease are associated with severe COVID-19. [3][4][5][6][7] However, association between some neurological condition such as Parkinson's disease, demyelinating disorders, and epilepsy have not been examined extensively. Furthermore, non-White racial and ethnic persons are reportedly disproportionally affected by COVID-19. 6,[8][9][10][11][12] Previous studies on risk factors were conducted when vaccines were not widely available. This study expands on knowledge from earlier studies by using population-based information from Kaiser Permanente in Northern California (KPNC) where comprehensive databases permit assessment of risks of testing positive for SARS-CoV-2, COVID-19 hospitalization, admission to an intensive care unit (ICU), and death in relation to demographic factors and comorbidities and assess whether risk factors changed in the presence of COVID-19 vaccines.

| Setting and population
This study was conducted at KPNC, an integrated healthcare system with a stable population of approximately 4.6 million members who receive nearly all their healthcare at KPNC facilities. Clinical databases include comprehensive information updated daily on diagnoses in the outpatient, emergency department, and inpatient settings and on prescriptions and laboratory tests.

| Study design
Using KPNC clinical databases, we conducted near-real-time surveillance of the pandemic, monitoring testing, hospitalization, ICU admission, and death. We report here on association between severe COVID-19 and risk factors included in our weekly surveillance from January 1, 2020, through July 23, 2021.

| COVID-19-related outcomes
We examined three indicators of severity (hospitalization, ICU admission, and mortality) stratified by the presence or absence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection confirmed by polymerase chain reaction (PCR).

| Statistical analysis
We described testing positive, COVID-19 hospitalizations, ICU admissions, and mortality by sex, age-group, race/ethnicity, and comorbidities. We used Cox regression to assess associations between testing positive, severe COVID-19 outcomes with demographic factors and comorbidities. To further assess association of severe COVID with demographic and comorbidities, we also conducted additional analyses restricted to people who tested positive and to those who were hospitalized only. We used Cox models conditioned on a calendar timeline and 14 geographic service areas.
To assess whether risk factors associated with severe COVID-19 evolved during the course of the pandemic, we conducted supplemental analyses by dividing the study into three time periods: The first period included data from January 1, 2020, through October   (Table S1).
Among people who tested positive for SARS-CoV-2, Blacks, Asian, multiracial, and Hispanic persons were at increased risk of hospitalization, admission to an ICU, and death compared with White persons after adjustment for age and comorbidities (Table S2). Among hospitalized people who tested positive for SARS-CoV-2, Hispanic persons were at increased risk of admission to an ICU (aHR = 1.47, CI 1.30-1.67) and death (aHR = 1.37, CI 1.18-1.58) compared with White persons (Table S3).

| DISCUSSION
In this large population-based study, increasing age was strongly associated with risk of hospitalization, ICU admission, and death. Non-White racial and ethnic groups were at markedly higher risk of severe COVID-19. The risk of severe COVID was strongly associated with some comorbidities but not with others. Even after wide availability of COVID-19 vaccines, the pattern of associations of severe COVID-19 with the risk factors was stable over time.
Although our results are consistent with previous studies reporting associations of age and race/ethnicity with severe COVID-19, [12][13][14] our results pertain to three levels of severity-hospitalization, ICU use, and death-and were adjusted for age, race/ethnicity, sex, comorbidities, and service area. Unlike a meta-analysis suggesting that the association of age with severe COVID-19 is much weaker when adjusted for agerelated comorbidities, 10

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1111/irv.12901.

DATA AVAILABILITY STATEMENT
Research data are not shared because of privacy concerns.