Post COVID-19 condition diagnosis: A population-based cohort study of occurrence, associated factors, and healthcare use by severity of acute infection

Abstract

Background: The occurrence and healthcare use trajectory of post COVID-19 condition (PCC) is poorly understood. Our aim was to investigate these aspects in SARS-CoV-2-positive individuals with and without a PCC diagnosis.

Methods: We conducted a population-based cohort study of adults in Stockholm, Sweden, with a verified infection from 1 March 2020 to 31 July 2021, stratified by the severity of the acute infection. The outcome was a PCC diagnosis registered any time 90-360 days after a positive test. We performed Cox regression models to assess baseline characteristics associated with the PCC diagnosis. Individuals diagnosed with PCC were then propensity-score matched to individuals without a diagnosis to assess healthcare use beyond the acute infection.

Results: Among 204,805 SARS-CoV-2-positive individuals, the proportion receiving a PCC diagnosis was 1% among individuals not hospitalized for their COVID-19 infection, 6% among hospitalized, and 32% among intensive care unit (ICU)-treated individuals. The most common new-onset symptom diagnosis codes among individuals with a PCC diagnosis were fatigue (29%) among nonhospitalized and dyspnea among both hospitalized (25%) and ICU-treated (41%) individuals. Female sex was associated with a PCC diagnosis among nonhospitalized and hospitalized individuals, with interactions between age and sex. Previous mental health disorders and asthma were associated with a PCC diagnosis among nonhospitalized and hospitalized individuals. Among individuals with a PCC diagnosis, the monthly proportion with outpatient care was substantially elevated up to 1 year after acute infection compared to before, with substantial proportions of this care attributed to PCC-related care.

Conclusion: The differential association of age, sex, comorbidities, and healthcare use with the severity of the acute infection indicates different trajectories and phenotypes of PCC, with incomplete resolution 1 year after infection.

Keywords: COVID-19; SARS-CoV-2; long-term outcomes; post COVID-19 condition; sequelae.

Fig. 1

Monthly number of COVID‐19 cases (bars) and proportion subsequently diagnosed with post COVID‐19 condition (line) by the severity of acute infection. Note: The overall proportion diagnosed with a post COVID‐19 condition is presented as text in the upper part of each graph. Abbreviations: COVID‐19, coronavirus disease 2019; ICU, intensive care unit.

Fig. 2

Associations between age, sex, and post COVID‐19 condition by the severity of acute infection. Note: The model included age, sex, and the interaction between age and sex and these were stratified on the calendar month of the first positive test. Age was included as a continuous variable, using restricted cubic splines with four knots. Abbreviation: ICU, intensive care unit.

Fig. 3

Associations between comorbidities, healthcare use, days with sickness benefit, and sociodemographic variables with post COVID‐19 condition diagnosis. Note: Cox proportional hazards regression models stratified on the month of first positive SARS‐CoV‐2 test, adjusted for age (restricted cubic splines with four knots), sex, and the interaction between age and sex were used for each factor separately. Absence of each comorbidity, zero comorbidities, zero days with sickness benefit, zero primary care visits, zero outpatient specialist care visits, zero hospitalizations, area type 4, and being born in Sweden were used as references. Residential area type 1 represents the most deprived areas and type 5 represents the least deprived areas; see the supplement for details. Abbreviation: ICU, intensive care unit.

Fig. 4

Primary care, outpatient specialist care, and inpatient care use from 9 months before to 12 months after SARS‐CoV‐2 infection in individuals with post COVID‐19 condition diagnosis and matched controls. Note: The upper panels show the proportion with a primary care medical doctor visit per month. The middle panels show the proportion with an outpatient specialist care medical doctor visit per month. The lower panels show the proportion with an inpatient visit per month. For each month after the infection, only individuals with full follow‐up for each month were included per month. The visits where individuals with post COVID‐19 condition diagnosis were first diagnosed with post COVID‐19 condition were excluded from the analyses. Abbreviations: COVID‐19, coronavirus disease 2019; ICU, intensive care unit; SARS‐CoV‐2, severe acute respiratory syndrome coronavirus 2.