The study addresses the significant gap in understanding the long-term effects of COVID-19, particularly in non-hospitalized older adults. While "long COVID" or Post-acute Sequelae of COVID-19 (PASC) has been studied in hospitalized patients, the experience of community-dwelling individuals remains less clear. Most previous studies have lacked non-COVID-19 control groups, making it difficult to isolate the effects of the virus from other factors. This research is crucial due to the large number of individuals who experienced COVID-19 in the community, without hospitalization. The study leverages the rich dataset from the CLSA, a large, well-established longitudinal cohort, to address these limitations. The aim is to determine the cumulative incidence of COVID-19 symptoms, analyze symptom persistence over time (1 and 3 months post-infection), and identify pre-pandemic factors associated with symptom severity and duration. This knowledge can guide the development of support services and interventions to improve recovery outcomes for individuals experiencing long COVID.

Existing research on long COVID predominantly focuses on hospitalized patients, with studies like Huang et al. reporting high rates of persistent symptoms (76% at 6 months), mainly fatigue, sleep problems, and anxiety/depression. However, the vast majority of COVID-19 cases were managed in the community, necessitating research on this less-studied population. Previous community-based studies often employed convenience samples, limiting generalizability, lacked control groups for comparison, and focused on specific subgroups or used limited data on sociodemographic and contextual factors. The lack of comprehensive studies with control groups hinders the ability to definitively attribute observed symptoms to SARS-CoV-2 infection and to differentiate from pre-existing conditions or other post-viral fatigue syndromes. This study aims to fill these gaps by analyzing data from a large, population-based cohort with pre-pandemic data and a non-COVID-19 control group.

This cross-sectional study utilized data from the Canadian Longitudinal Study on Aging (CLSA), a large population-based cohort of adults aged 50 and older. A COVID-19 questionnaire (September-December 2020) was administered to CLSA participants, collecting data on 23 COVID-related symptoms experienced since March 1, 2020. The questionnaire assessed symptom severity (mild, moderate, severe) and duration. Participants were categorized into COVID-19 and non-COVID-19 groups based on self-reported positive test or physician diagnosis. The study compared the cumulative incidence of symptoms between groups. For participants with COVID-19, symptom persistence was analyzed at 1 and 3 months post-infection. Multivariable regression models were used to identify factors associated with persistent symptoms, controlling for age, sex, and other relevant variables such as multimorbidity, depression/anxiety, and subjective social status. Pre-pandemic data from CLSA baseline and follow-up surveys (2011-2018) provided information on participants’ health status, functional abilities, and socioeconomic factors. The study used a large sample size (n=23,757) which included 121 participants who reported a COVID-19 diagnosis. Standardized differences and p-values were calculated to compare characteristics between COVID-19 and non-COVID-19 groups. The study included sensitivity analyses to assess the robustness of the findings.

The most prevalent symptoms reported by both groups were fatigue, dry cough, muscle/joint pain, sore throat, headache, and runny nose (over 25% in both groups). However, participants with COVID-19 reported significantly higher cumulative incidence of moderate/severe symptoms across various symptoms, with the absolute difference ranging from 16.8% (runny nose) to 37.8% (fatigue). Approximately 60% of males and 73% of females with COVID-19 reported at least one symptom persisting for more than one month. Symptom persistence beyond one month was significantly associated with female sex and multimorbidity in multivariable analyses. Persistence beyond three months was associated with a decrease of 15% with each unit increase in subjective social status after adjusting for age, sex, and multimorbidity. Interestingly, there was no consistent pattern of symptom persistence related to age. The study illustrated that many community-dwelling older adults experienced persistent symptoms for 1-3 months after COVID-19 infection, highlighting the need for additional support and rehabilitative care.

The findings demonstrate a substantial burden of persistent COVID-19 symptoms in the community-dwelling older adult population, even among those not requiring hospitalization. The significantly higher incidence of moderate/severe symptoms in the COVID-19 group underscores the long-term impact of the infection. The association of female sex and multimorbidity with longer symptom duration aligns with existing literature, highlighting vulnerability in these groups. The novel finding of an association between subjective social status and reduced symptom persistence suggests that social determinants of health may influence long COVID outcomes. These results have significant clinical and public health implications, advocating for increased access to rehabilitative services and social support to aid recovery. The large sample size, inclusion of a control group, and access to pre-pandemic health data strengthen this study's generalizability and robustness.

This study highlights the significant prevalence and persistence of COVID-19 symptoms in community-living older adults. Findings underscore the need for increased access to rehabilitative and supportive care to improve recovery outcomes. Further research should explore the role of social determinants of health in long COVID and the long-term impacts beyond three months post-infection.

The study's reliance on self-reported COVID-19 status and symptoms may introduce recall bias. The relatively small number of participants with confirmed COVID-19 infections may limit the statistical power of certain analyses and lead to wide confidence intervals. The lack of precise symptom onset dates could underestimate the duration of symptom persistence. Lastly, the study does not assess symptom persistence beyond three months post-infection.