Abstract
Objective
Fatigue, headache, problems sleeping and numerous other symptoms have been reported to be associated with long COVID. However, many of these symptoms coincide with symptoms reported by the general population, possibly exacerbated by restrictions/precautions experienced during the COVID-19 pandemic. This study examines the symptoms reported by individuals who tested positive for COVID-19 compared with those who tested negative.
Design
Observational study.
Setting
The study was conducted on adult residents in Alberta, Canada, from October 2021 to February 2023.
Participants
We evaluated self-reported symptoms in 7623 adults with positive COVID-19 tests and 1520 adults who tested negative, using surveys adapted from the internationally standardised International Severe Acute Respiratory and emerging Infection Consortium (ISARIC)-developed COVID-19 long-term follow-up tools. These individuals had an index COVID-19 test date between 1 March 2020 and 31 December 2022 and were over 28 days post-COVID-19 testing.
Primary outcome measures
The primary outcomes were to identify the symptoms associated with COVID-19 positivity and risk factors for reporting symptoms.
Results
Fatigue was the top reported symptom (42%) among COVID-19-positive respondents, while headache was the top reported symptom (32%) in respondents who tested negative. Compared with those who tested negative, COVID-19-positive individuals reported 1.5 times more symptoms and had higher odds of experiencing 31 out of the 40 listed symptoms during the postinfectious period. These symptoms included olfactory dysfunction, menstruation changes, cardiopulmonary and neurological symptoms. Female sex, middle age (41–55 years), Indigeneity, unemployment, hospital/intensive care unit (ICU) admission at the time of testing and pre-existing health conditions independently predicted a greater number and variety of symptoms.
Conclusions
Our results provide evidence that COVID-19 survivors continue to experience a significant number and variety of symptoms. These findings can help inform targeted strategies for the unequally affected population. It is important to offer appropriate management for symptom relief to those who have survived the acute COVID-19 illness.
Keywords: COVID-19, EPIDEMIOLOGY, Post-Acute COVID-19 Syndrome, PUBLIC HEALTH
STRENGTHS AND LIMITATIONS OF THIS STUDY.
This comprehensive study is the largest survey completed in Alberta, Canada, assessing symptomology beyond 28 days post-COVID-19 testing on COVID-19-positive and COVID-19-negative individuals via internationally standardised survey tools.
Relatively few studies have been conducted on a large sample of individuals, covering those who received COVID-19 testing over 34 months during the pandemic.
Response bias remains a major confounder in survey studies. However, as a similar demographic distribution was observed among positive and negative respondents, the reported excess symptoms were likely unrelated to the responder bias.
The possibility of misclassification due to false-negative and false-positive COVID-19 results cannot be entirely ruled out.
Introduction
The COVID-19 pandemic has significantly disrupted people’s well-being, community stability and healthcare systems.1 2 The exponential spread of COVID-19 directly resulted in unprecedented morbidity and mortality globally, making COVID-19 one of the leading causes of death in the US between March 2020 and October 2021.3 Despite recovering from acute COVID-19, persistent symptoms and/or delayed long-term complications, commonly referred to as long COVID or post COVID-19 condition, have been reported by several studies.4–6 Symptoms related to the main body systems including the respiratory, circulatory and neurological systems have been reported, which may be a result of cellular damage from direct viral attacks and/or a prolonged immune response.7 Indirectly, due to lockdowns, social distancing, service disruption and economic hardship, it has been reported that the pandemic has considerably altered social, economic and behavioural responses,8 9 which in turn has significantly impacted mental health, lifestyle choices and health services use.10 Over and above the estimated 9.08 million years of life lost resulted from the COVID-19 pandemic through March 2021 in the USA, three times more pandemic-associated excess mortality was observed worldwide between January 2020 and December 2021, compared with the number of COVID-19 deaths.11 12 In Alberta, Canada, an 11% excess mortality rate was reported from March 2020 to December 2021, with 43.1% of total excess deaths attributed to non-COVID-19 causes.13 In addition, high rates of symptoms associated with anxiety, depression and stress have been reported globally.14
A number of studies that aimed to characterise long COVID outcomes in acute COVID-19 survivors have reported constellations of symptoms including fatigue, headache and insomnia that may be a result of living in the stressful pandemic environment.5 14 While we are gaining an understanding of patient characteristics from various jurisdictions, we also need to better understand the types and groups of symptoms that people are experiencing. It is also important to separate these from an increase in bodily distress symptoms because of societal changes,15 supporting the development of a universally accepted long COVID definition. Furthermore, understanding the magnitude of the health burden associated with the pandemic, planning for additional health-related care, is urgently required.
In this study, we surveyed residents across the province of Alberta, Canada who underwent COVID-19 testing and were beyond 28 days after the COVID-19 test to compare symptoms experienced between those who tested positive and those who only ever tested negative for COVID-19 before survey completion. The primary objective of this study was to determine the association of COVID-19-positive status with self-reported symptoms during the COVID-19 pandemic. The secondary objective was to characterise symptoms experienced by Alberta residents during the COVID-19 pandemic.
Methods
Data sources
Following ethical approval (Ethics ID: Pro00112053), the Alberta POST-COVID Follow-up Study surveyed Alberta residents who were aged 18 years and older, tested for COVID-19 via COVID-19 PCR or rapid antigen tests between 2020 and 2023, were at least 28 days post-COVID-19 testing and consented to participate using surveys adapted for online and telephone use from the standardised comprehensive COVID-19 long-term follow-up tools developed by ISARIC.16 Multiple advertising strategies were employed to recruit study participants, including targeted mail-out advertisements sent to randomly selected adults who underwent COVID-19 testing among approximately 3.9 million adult residents in Alberta. The study was also promoted online and through study posters and emails. Moreover, various survey modes, such as online surveys and phone surveys, were used to ensure the inclusion of a representative population.16 While completing the survey, the respondents provided their up-to-date COVID-19 test results before being classified into the COVID-19 positive or negative group. The first positive COVID-19 test date was used as the index COVID-19 test date for the respondents who tested positive for COVID-19, while the first negative test date was used as the index date for the respondents who did not go on to test positive. Respondents who had an index COVID-19 test date between 1 March 2020 and 31 December 2022 and completed the survey between 1 October 2021 and 28 February 2023 were included in the current study.
Measures
Designed to enable self-completion, the survey tool collected a wide range of data including demographics, hospitalisation status around the index COVID-19 test, pre-existing dyspnoea and pre-existing difficulties in functioning before the index testing through Washington Group Short Set on Functioning (online supplemental materials). The questionnaire included inquiries regarding 39 symptoms (headache, persistent cough, fatigue, chest pain, confusion/lack of concentration, constipation, problems passing urine, palpitations, dizziness, menstruation changes and others) that were present in the 7 days prior to survey completion. A free-text question was used to capture any symptoms outside of the list.
bmjopen-2023-078119supp002.pdf (9MB, pdf)
Statistical analysis
Categorical data were summarised as frequencies and percentages, while continuous data were reported as means with the corresponding SDs or medians with the corresponding IQRs. The standardised mean difference was used to measure the effect size between the respondents who tested positive or negative for COVID-19. We created models to examine the self-reported symptoms differences between groups adjusting for age, sex, education, occupation, ethnicity, self-reported hospitalisation, and breathlessness and functioning difficulty before the COVID-19 test. A negative binomial regression model was used to estimate symptom counts. Effect estimates are presented as incidence rate ratios (IRRs) and the corresponding 95% CI. Logistic regression, presented as ORs and the corresponding 95% CI, was used for binary outcomes. The adjusted ORs on symptoms are presented visually as heatmaps. Subgroup analyses with the Bonferroni test were performed to determine whether the effect of COVID-19 positive varied in subgroups. Hierarchical clustering analysis was performed to segment symptoms into clusters with hierarchical dendrograms for visualisation. Statistical analyses were performed by using SAS Enterprise Guide V.8.3. Statistical significance was taken at p≤0.05. Data visualisations were through Tableau Desktop V.2021.2.
Patient and public involvement
Patients and/or the public had no role in data analysis, or interpretation, or dissemination plans of the paper.
Results
Of the 9524 surveys collected through the Alberta POST-COVID Follow-up Study by 18 February 2023, 9143 (93.1%) surveys met the inclusion criteria and were included in the analysis, excluding 12 respondents not having a COVID-19 test dated within the study period and 369 respondents who partially completed the survey. Comparing with the adult population in Alberta in fiscal year 2022, survey respondents had a slightly higher likelihood to be female (online supplemental table 1). The respondents completed the survey during the period between 29 and 1036 days after the index COVID-19 test. 83.4% (7623) of the total 9143 respondents tested positive for COVID-19 and 1520 (16.6%) respondents reported never testing positive before survey completion.
bmjopen-2023-078119supp001.pdf (1.7MB, pdf)
The characteristics of the respondents are shown in table 1. Most respondents were females (62.8%), with a mean age of 48.7 (SD 15.5) years, and of predominantly white ethnicity (82.7%). 77.6% (7096/9143) of respondents reported having an education level beyond high school, and 57.8% (5287/9143) were working full time prior to the COVID-19 test. 4.6% of respondents were hospitalised at the time of testing. The majority of respondents did not report breathlessness (79.1%) or functioning difficulty (53.9%) prior to the index date.
Table 1.
Baseline characteristics of survey respondents comparing those who tested positive for COVID-19 with those who tested negative before survey completion
| Negative | Positive | Standardised difference | |
| Total N | 1520 | 7623 | |
| Age at survey, mean (SD) | 53.5 (15.8) | 47.7 (15.2) | −0.37 |
| Age group, n (%) | |||
| 18–40 | 362 (23.8) | 2679 (35.1) | 0.48 |
| 41–55 | 390 (25.7) | 2392 (31.4) | |
| 56–65 | 360 (23.7) | 1391 (18.2) | |
| 66–79 | 350 (23.0) | 951 (12.5) | |
| 80+ | 43 (2.8) | 109 (1.4) | |
| Unknown | 15 (1.0) | 101 (1.3) | |
| Sex, n (%) | |||
| Male | 576 (37.9) | 2791 (36.6) | 0.02 |
| Female | 943 (62.0) | 4802 (63.0) | |
| Unknown | <5* | 30 (0.4) | |
| Ethnic group, n (%) | |||
| White | 1305 (85.9) | 6258 (82.1) | 0.14 |
| Asian† | 91 (6.0) | 549 (7.2) | |
| Indigenous‡ | 47 (3.1) | 316 (4.1) | |
| Latin | 13 (0.9) | 116 (1.5) | |
| Mixed ethnicity | 19 (1.3) | 127 (1.7) | |
| Other/unknown§ | 45 (3.0) | 257 (3.4) | |
| Education, n (%) | |||
| Primary/secondary¶ | 37 (2.4) | 251 (3.3) | 0.15 |
| High school | 189 (12.4) | 1095 (14.4) | |
| Vocational** | 492 (32.4) | 2665 (35.0) | |
| Bachelor degree | 515 (33.9) | 2282 (29.9) | |
| Master/PhD degree | 223 (14.7) | 919 (12.1) | |
| Other/unknown†† | 64 (4.2) | 411 (5.4) | |
| Occupation Before the COVID-19 test, n (%) | |||
| Work full time | 734 (48.3) | 4553 (59.7) | 0.34 |
| Work part time | 194 (12.8) | 996 (13.1) | |
| Unemployed‡‡ | 69 (4.5) | 269 (3.5) | |
| Retired§§ | 430 (28.3) | 1148 (15.1) | |
| Other/unknown¶¶ | 93 (6.1) | 657 (8.6) | |
| Self-report hospitalisation at the COVID-19 test, n (%) | |||
| Non-hospitalised/missing | 1436 (94.5) | 7284 (95.6) | 0.10 |
| Hospitalised | 80 (5.3) | 251 (3.3) | |
| ICU admission | <5* | 88 (1.2) | |
| Breathlessness before the COVID-19 test, n (%) | |||
| Mild/missing | 1136 (74.7) | 6100 (80.0) | 0.12 |
| Moderate | 339 (22.3) | 1336 (17.5) | |
| Severe | 45 (3.0) | 187 (2.5) | |
| Functioning difficulty before the COVID-19 test, n (%) | |||
| None/missing | 771 (50.7) | 4155 (54.5) | 0.16 |
| Mild | 637 (41.9) | 3009 (39.5) | |
| Moderate | 97 (6.4) | 423 (5.5) | |
| Severe | 15 (1.0) | 36 (0.5) |
*A sample size less than five respondents in the cell.
†Asian refers to East Asian, South Asian, West Asian and Southeast Asian.
‡Indigenous refers to First Nations, Métis and Inuit.
§Ethnic group—other/unknown refers to black, Arab and other ethnicity with smaller sample size, prefer not to say and missing.
¶Primary/secondary refers to primary education and secondary education (3–10 years of school).
**Vocational refers to vocational/practical school and higher college/university.
††Education—other/unknown refers to not completed formal education/training, prefer not to say, others and missing.
‡‡Unemployed refers to unemployed and unable to work due to chronic illness.
§§Retired refers to retired and medically retired.
¶¶Occupation—other/unknown refers to full-time carer, student, prefer not to say and missing.
Compared with the respondents who tested negative, COVID-19-positive respondents were younger (47.7 years, SD 15.2 vs 53.5 years, SD 15.8) and had a similar sex distribution. They were also more likely to be working full time (59.7% vs 48.3% of the COVID-19-negative respondents). Moreover, COVID-19-positive respondents were less likely to report breathlessness (20.0% vs 25.3%) and functioning difficulties (45.5% vs 49.3%) prior to the testing but were more likely to report admission to an ICU (1.2%) at the testing compared with those who tested negative.
Figure 1 examines the report of symptoms by respondents. The top three symptoms among both COVID-19-positive and COVID-19-negative respondents were fatigue (41.7% and 28.4%), headache (40.3% and 32.4%) and problems sleeping (30.3% and 23.2%). After adjustment for the baseline characteristic variables, 10 symptoms resulted in ORs greater than 2.0. While eight symptoms ranged from an OR of 2.0 (95% CI 1.4 to 2.8) for pain on breathing to an OR of 2.8 (95% CI 2.0 to 3.8) for menstruation change following the COVID-19 test, the other two symptoms, altered taste (adjusted OR 6.0, 95% CI 4.0 to 9.0) and altered smell (adjusted OR 5.9, 95% CI 4.1 to 8.6), were associated with much higher ORs. In the COVID-19-positive respondents, those who completed the survey over 90 days after the test were slightly younger, more likely to self-identify as non-white, and more likely to be working full time (online supplemental table 2). Online supplemental figure 1 examines changes in reported symptoms among positive respondents over various time period following the initial positive test. It showed persistent cough, stomach pain, loss of appetite, feeling sick and weight loss with an OR less than 1.0 after 90 days post-COVID-19, while dizziness, shortness of breath, palpitations, diarrhoea and chest pains had an OR less than 1.0 only during the 90 days to 9 months period.
Figure 1.
Symptom prevalence and associations between individual symptoms and COVID-19 positivity. Bars represent the percentage of respondents in their category who reported experiencing each symptom. The forest plot shows a multivariate-adjusted association between the reports of symptoms related to testing positive for COVID-19. aAdjusted for age group, sex, ethnicity, education levels, occupation at the COVID-19 test date, hospitalisation status around the COVID-19 test date, breathlessness and functioning difficulty scales before the COVID-19 test is done, displaying the association of symptoms and COVID-19 positivity comparing with COVID-19 negativity. bPercentage of symptoms and ORs in the corresponding sex group.
Figure 2 (online supplemental figure 2) is a heatmap of adjusted ORs, depicting the relationship between symptoms and baseline characteristics. The hierarchical clustering identified 9 clusters of symptoms, ranging from 2 to 10 symptoms per cluster. The clusters grouped on symptoms involving the same body systems. For example, cluster 1 contains symptoms associated with olfactory dysfunction, while cluster 2 includes cardiopulmonary symptoms. However, neurological symptoms were scattered into clusters 3, 5, 6 and 9, essentially mixing with primarily gastrointestinal and musculoskeletal symptoms. Unlike cluster 1 (olfactory dysfunction) which was substantially different from other clusters based on the dendrogram branching, clusters 2–4 (cardiopulmonary, gastrointestinal and some neurological symptoms) were closer to each other in reporting pattern, and clusters 5–9 (musculoskeletal, integumentary, urological and some other neurological symptoms) had similarities with each other. Looking at the heatmap, in addition to the high likelihood of reporting 31 out of the 40 listed symptoms by the respondents who tested COVID-19-positive, female sex, Indigeneity, unemployment, ICU admission around the index COVID-19 test, pre-existing breathlessness and baseline functioning difficulty were associated with higher odds of reporting various symptoms in different clusters, while Asian ethnicity, higher education and retirement predicted lower odds.
Figure 2.
Association between individual symptoms and COVID-19 positivity and clustering analysis. Heatmap of adjusted ORs of associations between COVID-19 positivity and demographic characteristics and the report of each symptom with hierarchical clustering analysis on symptoms. aAdjusted for age group, sex, ethnicity, education levels, occupation at the COVID-19 test date, hospitalisation status around the COVID-19 test date, and breathlessness and functioning difficulty scales before the COVID-19 test done for COVID-19 positivity comparing with COVID-19 negativity. bAdjusted ORs in the corresponding sex group. ICU, Intensive Care Unit.
Online supplemental table 3 shows the count of symptoms reported by all participants, a higher percentage of respondents in the COVID-19-positive group reported experiencing more than four symptoms compared with the respondents who tested negative (44.7%, 3407/9143 vs 29.4%, 448/1520). The respondents who tested positive reported significantly more symptoms (median 3, IQR 0–7 vs median 1, IQR 0–4). Similar results were obtained in many baseline characteristic subgroups (online supplemental table 4).
After controlling for baseline demographics and conditions, our results showed that testing positive for COVID-19 was associated with reporting 1.6 times more symptoms than those testing negative (adjusted IRR 1.6, 95% CI 1.5 to 1.7; figure 3). In addition, age, sex, ethnicity, education, occupation, self-report hospitalisation, baseline breathlessness and baseline functioning difficulty were significant positive predictors for the number of symptoms. Respondents who were between 41 and 55 years of age, female, Indigenous, unemployed, admitted to hospitals, with pre-existing breathlessness, or with pre-existing functional difficulties reported a higher symptom count compared with their counterparts, with adjusted IRRs ranging from 1.2 (95% CI 1.0 to 1.4) in unemployed respondents to 2.3 (95% CI 1.8 to 2.9) in ICU-admitted patients. However, Asian respondents reported fewer symptoms compared with the white population (adjusted IRR 0.8, 95% CI 0.8 to 0.9); respondents who were retired also reported fewer symptoms than the full-time workers (adjusted IRR 0.8, 95% CI 0.7 to 0.9).
Figure 3.
Predictors for symptom count adjusting for baseline characteristics. The forest plot depicts the adjusted rate ratios of the risk of reporting more symptoms. aAdjusted for age group, sex, ethnicity, education levels, occupation at the COVID-19 test date, hospitalisation status around the COVID-19 test date, and breathlessness and functioning difficulty scales before the COVID-19 test was done. ICU, Intensive Care Unit.
Additional analyses found interactions between COVID-19 test results and sex, ethnic group and baseline functioning difficulty (figure 4), suggesting that the effect of COVID-19 positivity over the number of symptoms varies based on these three variables. Analyses with adjustment for baseline characteristics and the interactions yielded generally comparable findings on the predictors of the number of symptoms being reported (online supplemental figure 3), which showed 1.5 times more symptoms being reported by the respondents who tested positive for COVID-19 compared with those who tested negative (adjusted IRR 1.5, 95% CI 1.3 to 1.8). The interaction subgroup analyses indicated a higher risk of reporting more symptoms in respondents who tested positive for COVID-19 than those testing negative among females (adjusted IRR 2.0, 95% CI 1.5 to 2.7), Asian communities (adjusted IRR 2.4, 95% CI 1.6 to 3.4) and respondents without functioning difficulties (adjusted IRR 2.2, 95% CI 1.7 to 2.7; figure 4).
Figure 4.
Subgroup analysis of symptom count in COVID-19-positive respondents compared with the negative respondents. aAdjusted for age group, sex, ethnicity, education levels, occupation at the COVID-19 test date, hospitalisation status around the COVID-19 test date, and breathlessness and functioning difficulty scales before the COVID-19 test done for the association of symptom count and COVID-19 positivity comparing with COVID-19 negativity. The dotted vertical line indicates the overall HR in all the respondents.
Discussion
This cross-sectional analysis of data from a population-based provincial study found that testing positive for COVID-19, beyond 28 days from acute COVID-19 infection, was associated with the report of a larger number of symptoms, compared with respondents with negative COVID-19 test results, after accounting for baseline demographic and health characteristics. In addition to olfactory dysfunction (altered taste and altered smell), various symptoms across different systems, including menstruation changes, and cardiopulmonary, neurological, and musculoskeletal systems, were associated with previous COVID-19 positivity. The predictors for reporting higher numbers of symptoms included female sex, age 41–55 years, Indigenous population, unemployment, hospitalisation at the test, pre-existing breathlessness and baseline functioning difficulty.
Accounting for the overlap of physical symptoms reported by COVID-19 survivors and the general population,15 this study, which includes those who were COVID-19 negative, identified symptoms highly associated with having tested positive for COVID-19. Previous studies assessing long COVID have mostly been restricted to patients with a positive PCR COVID-19 test result, reporting symptom prevalence of 41%–83% in COVID-19 survivors, with symptoms including fatigue, headache and muscle pain.5 17 18 Similarly, we reported that the top-reported symptom, fatigue, affected 42% of COVID-19 survivors. However, many of these symptoms were also frequently reported in the population with negative COVID-19 test results, with the top symptom (headache) impacting 32% of negative respondents, followed by fatigue. These findings cannot be explained solely by a false COVID-19 negative result or the belief in having experienced COVID-19.19 Looking beyond the symptoms that have been reported after the acute COVID-19 infection,5 20 such as olfactory dysfunction, anosmia, fatigue, headache and cough, this study recognised a range of other symptoms, including menstruation changes, muscle weakness, problem speaking and balance problems, which were significantly associated with previous COVID-19 infection. In addition, our study on the pattern of symptoms reported by COVID-19-positive respondents agreed in part with previous reports showing the persistence of symptoms and the fluctuations of the symptoms over time, with loss of appetite and cough displaying a decreasing trend.21 While a decreased trend was noted in the report of persistent cough, stomach pain, loss of appetite, feeling sick and weight loss, the percentage of respondents reporting individual symptoms is still considerable. However, as this is a cross-sectional study, subsequent studies that follow changes in symptoms over time within an individual are required to better understand the changes in post-COVID-19 symptoms over time. Furthermore, we showed the report of more symptoms and groups of symptoms in those who previously tested positive for COVID-19, with 45% of respondents who tested positive reporting the presence of at least four symptoms.
In addition to prior COVID-19 positivity, the predictive effects of females sex, Indigeneity, unemployment and pre-existing health conditions may be explained by studies showing that females, people facing a financial crisis, disabled adults and ethnic minorities have been disproportionately impacted economically and socially by the pandemic; and tended to be more vulnerable to stressors, demonstrating a higher rate of developing psychological symptoms.14 22 On the other hand, Asian ethnicity, higher education levels and retirement were negatively associated with the experience of multiple symptoms. Interestingly, despite that a higher number of symptoms was reported by respondents who were hospitalised around the COVID-19 testing, no specific symptoms were specifically associated with hospitalised patients. This is consistent with previous research showing a great risk of new clinical sequelae after COVID-19 despite no pre-existing conditions or hospitalisation at COVID-19.23 This highlights the need to target services to vulnerable populations.
Although we did not attempt to investigate the aetiology or underlying mechanisms for symptoms, exploring concurrent symptoms (symptom clusters), in the current study may provide implications for future studies on the underlying mechanisms leading to symptom clusters and the management of symptom clusters instead of focusing on individual symptoms. Despite the growing list of long COVID symptoms, no distinct symptom has been identified, and no clear mechanistic explanation has been established for the full range of symptoms.5 24 However, at symptom cluster levels, several hypotheses have been generated around smell and taste dysfunction including the slow regeneration of the COVID-19-triggered loss or damaged olfactory receptor neurons and the alteration of gene expression.25 26 Moreover, post-COVID-19 cardiopulmonary outcomes, such as chronic respiratory failure, myocardial infarction and arrhythmias, have gained much attention and have been suggested as a result of a chronic inflammatory response and autoimmune response-induced tissue damage.20 23 27 In addition, besides being highly associated with previous COVID-19 positivity, the current study suggested that smell and taste dysfunction were more commonly present in females and the Indigenous population, while cardiopulmonary symptoms were more commonly associated with females, Indigenous people, ICU admission around the index COVID-19 testing, severe pre-existing breathlessness and moderate baseline functioning difficulty. Future studies are required to detail the underlying mechanisms for symptom clusters. The concurrence of symptoms, the complex relationships with sociodemographic factors, and the vague cellular and physiological mechanisms support the necessity of further investigation and a new paradigm integrating biological, sociodemographic and psychological factors to explain long COVID.28
This study does not examine acute versus long-term symptoms directly but instead considers a pragmatic patient-reported long-term symptomology. We provide information on the association of prior COVID-19 positivity and reporting of symptoms, presenting that COVID-19 infection was a significant positive predictor for various patient-reported symptoms across different bodily systems. Incorporating the susceptibility of sociodemographic subgroups, this information may help to prepare the healthcare systems for increased demands and assist in the development of targeted approaches to the management of population health in the aftermath of the pandemic.
Our findings likely represent the symptom burden experienced by Alberta residents by (1) capturing the COVID-19-infected Alberta residents regardless of the policy changes—from the widely available PCR testing phase to the home antigen testing-focused phase during the pandemic and (2) by including populations from the communities who may be less likely or less able to seek care, especially during the surges of the pandemic. First, the study population exhibits a degree of representativeness to the adult population in Alberta with a small effect size on age and sex. Moreover, as 41% of COVID-19 infections were asymptomatic and 2.1% resulted in hospitalisation among the COVID-19 population,29 asymptomatic and/or non-hospitalised populations are commonly excluded from the large-scale long COVID research. A previous study assessing electronic medical records in the USA reported prior COVID-19 infection was a significant positive predictor for anosmia, palpitations, fatigue, chest pain and joint pain, but not for headache and cough compared with non-infected patients.30 Our study, on the other hand, identified the association of COVID-19 positivity with more symptoms. In fact, analyses on non-hospitalised UK adults during the early pandemic have identified more symptoms at ≥12 weeks after infection compared with adults testing negative.24 The discrepancy in the symptom burden between studies may be partially attributable to the population differences. Including a wider population, especially those who had no COVID-19 records in the medical files or those who were less likely to seek care, we provide a more complete picture of the health of COVID-19 survivors in Alberta. Our findings on the presenting symptoms and subgroup risks for symptom development are important for the development of more appropriate clinical guidelines for patient triage, symptom management and treatment.
Our study has limitations. First, response bias remains a major confounder in any survey studies, which limits the representativeness of the study. Examining the demographic of the population in Alberta from the 2021 Canada Census, visible minorities, the Indigenous population and males were under-represented in our study. Of note there was a similar demographic distribution of positive and negative respondents, indicating that the report of excess symptoms was likely unrelated to the responder bias. Second, the current study did not evaluate the impact of variant wave, COVID-19 reinfection and vaccination status, which warrants further studies. Third, we cannot exclude the possibility of misclassification due to false negative and false positive COVID-19 results, however, these conditions would suppress the observed effect.
Conclusion
The results of this cross-sectional analysis of a large, population-based, provincial survey suggest that prior COVID-19 positive status was associated with the reporting of more and a wider range of symptoms compared with respondents with COVID-19 negative status, including olfactory dysfunction, cardiopulmonary and neurological symptoms. Moreover, female sex, middle age, Indigeneity, unemployment, the severity of the acute illness at the test and pre-existing health conditions independently predicted more and various symptoms at follow-up. Future research should focus on evaluating the long-term clinical implications of these clusters of symptoms, establishing better care and unveiling the mechanisms underlying the symptom clusters. From a clinical perspective, resources should support post-COVID-19 recovery.
Supplementary Material
Acknowledgments
We would like to thank the Neurosciences, Rehabilitation and Vision Strategic Clinical Network (NRV SCN) and Health System Access (HSA), Alberta Health Services (AHS) for administrative and dissemination support. Moreover, we would like to acknowledge the Alberta SPOR SUPPORT Unit (AbSPORU), which contributes to dissemination support and the development of analytic strategies.
Footnotes
Contributors: XC, CN, JB and CH led the development of the study protocol. XC, JB and TW processed and analysed the survey data. XC wrote the original draft, with input from CN, JB, CH, TW and BM. JB accepted full responsibility for the work of the study, had access to the data, and controlled the decision to publish as the guarantor. All authors reviewed and approved the final version of the manuscript.
Funding: This work was funded by the Ministry of Health, Government of Alberta (grant number 012676).
Competing interests: None declared.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review: Not commissioned; externally peer reviewed.
Supplemental material: This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
Data availability statement
Data are available on reasonable request. Data protection regulations will be coordinated with the University of Alberta ethics committee, Alberta Health Services and Alberta Health. Data containing identifiable information will not be shared due to the privacy of the participants and ethical reasons. Deidentified data are available to researchers under a signed data disclosure agreement. Researchers are invited to contact the corresponding author or file a request through Alberta SPOR SUPPORT Unit (https://absporu.ca/) immediately following publication with ethical approval obtained and agreement with Alberta Health and Alberta Health Services. The study protocol was published in BMJ Open and is publicly available online (https://bmjopen.bmj.com/content/13/2/e067449.long). The aggregate data and code used in this analysis are available from the corresponding author on reasonable request.
Ethics statements
Patient consent for publication
Not applicable.
Ethics approval
This study involves human participants and was approved by the University of Alberta ethics committee (Study ID: Pro00112053). Participants gave informed consent to participate in the study before taking part.
References
- 1. Arsenault C, Gage A, Kim MK, et al. COVID-19 and resilience of healthcare systems in ten countries. Nat Med 2022;28:1314–24. 10.1038/s41591-022-01750-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Douglas M, Katikireddi SV, Taulbut M, et al. Mitigating the wider health effects of COVID-19 pandemic response. BMJ 2020;369:m1557. 10.1136/bmj.m1557 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Shiels MS, Haque AT, Berrington de González A, et al. Leading causes of death in the US during the COVID-19 pandemic, March 2020 to October 2021. JAMA Intern Med 2022;182:883–6. 10.1001/jamainternmed.2022.2476 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Nalbandian A, Sehgal K, Gupta A, et al. Post-acute COVID-19 syndrome. Nat Med 2021;27:601–15. 10.1038/s41591-021-01283-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Lopez-Leon S, Wegman-Ostrosky T, Perelman C, et al. More than 50 long-term effects of COVID-19: a systematic review and meta-analysis. Sci Rep 2021;11:16144. 10.1038/s41598-021-95565-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. National Institute for Health and Care Excellence: Clinical Guidelines . In COVID-19 rapid guideline: managing the long-term effects of COVID-19, copyright © NICE 2020. London: National Institute for Health and Care Excellence (NICE); 2020. [Google Scholar]
- 7. Davis HE, McCorkell L, Vogel JM, et al. Long COVID: major findings, mechanisms and recommendations. Nat Rev Microbiol 2023;21:133–46. 10.1038/s41579-022-00846-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Belitski M, Guenther C, Kritikos AS, et al. Economic effects of the COVID-19 pandemic on entrepreneurship and small businesses. Small Bus Econ 2022;58:593–609. 10.1007/s11187-021-00544-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Pak A, Adegboye OA, Adekunle AI, et al. Economic consequences of the COVID-19 outbreak: the need for epidemic preparedness. Front Public Health 2020;8:241. 10.3389/fpubh.2020.00241 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. McAlister FA, Parikh H, Lee DS, et al. Health care implications of the COVID-19 pandemic for the cardiovascular practitioner. Can J Cardiol 2023;39:716–25. 10.1016/j.cjca.2022.11.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Wang H, Paulson KR, Pease SA. Estimating excess mortality due to the COVID-19 pandemic: a systematic analysis of COVID-19-related mortality, 2020-21. Lancet 2022;399:1513–36. 10.1016/S0140-6736(21)02796-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Reif J, Heun-Johnson H, Tysinger B, et al. Measuring the COVID-19 mortality burden in the United States: a microsimulation study. Ann Intern Med 2021;174:1700–9. 10.7326/M21-2239 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Beesoon S, Bakal JA, Youngson E, et al. Excess deaths during the COVID-19 pandemic in Alberta, Canada. IJID Reg 2022;5:62–7. 10.1016/j.ijregi.2022.08.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Xiong J, Lipsitz O, Nasri F, et al. Impact of COVID-19 pandemic on mental health in the general population: a systematic review. J Affect Disord 2020;277:55–64. 10.1016/j.jad.2020.08.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Petersen MW, Dantoft TM, Jensen JS, et al. The impact of the COVID-19 pandemic on mental and physical health in Denmark - a longitudinal population-based study before and during the first wave. BMC Public Health 2021;21:1418. 10.1186/s12889-021-11472-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Chen X, Bakal J, Whitten T, et al. Assessing the impact of COVID-19 pandemic on the health of residents and the Healthcare system in Alberta, Canada: an observational study—the Alberta POST-COVID follow-up study. BMJ Open 2023;13:e067449. 10.1136/bmjopen-2022-067449 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Sigfrid L, Drake TM, Pauley E, et al. Long Covid in adults discharged from UK hospitals after COVID-19: a prospective, multicentre cohort study using the ISARIC WHO clinical characterisation protocol. Lancet Reg Health Eur 2021;8:100186. 10.1016/j.lanepe.2021.100186 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Michelen M, Manoharan L, Elkheir N, et al. Characterising long COVID: a living systematic review. BMJ Glob Health 2021;6:9. 10.1136/bmjgh-2021-005427 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Matta J, Wiernik E, Robineau O, et al. Association of self-reported COVID-19 infection and SARS-Cov-2 serology test results with persistent physical symptoms among French adults during the COVID-19 pandemic. JAMA Intern Med 2022;182:19–25. 10.1001/jamainternmed.2021.6454 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Xie Y, Xu E, Bowe B, et al. Long-term cardiovascular outcomes of COVID-19. Nat Med 2022;28:583–90. 10.1038/s41591-022-01689-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Tran VT, Porcher R, Pane I, et al. Course of post COVID-19 disease symptoms over time in the compare long COVID prospective E-cohort. Nat Commun 2022;13:1812. 10.1038/s41467-022-29513-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Burhamah W, AlKhayyat A, Oroszlányová M, et al. The psychological burden of the COVID-19 pandemic and associated lockdown measures: experience from 4000 participants. J Affect Disord 2020;277:977–85. 10.1016/j.jad.2020.09.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Daugherty SE, Guo Y, Heath K, et al. Risk of clinical sequelae after the acute phase of SARS-Cov-2 infection: retrospective cohort study. BMJ 2021;373:n1098. 10.1136/bmj.n1098 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Subramanian A, Nirantharakumar K, Hughes S, et al. Symptoms and risk factors for long COVID in non-hospitalized adults. Nat Med 2022;28:1706–14. 10.1038/s41591-022-01909-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Butowt R, Bilinska K, von Bartheld CS. Olfactory dysfunction in COVID-19: new insights into the underlying mechanisms. Trends Neurosci 2023;46:75–90. 10.1016/j.tins.2022.11.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Tan BKJ, Han R, Zhao JJ, et al. Prognosis and persistence of smell and taste dysfunction in patients with COVID-19: meta-analysis with parametric cure modelling of recovery curves. BMJ 2022;378:e069503. 10.1136/bmj-2021-069503 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Raman B, Bluemke DA, Lüscher TF, et al. Long COVID: post-acute sequelae of COVID-19 with a cardiovascular focus. Eur Heart J 2022;43:1157–72. 10.1093/eurheartj/ehac031 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Saunders C, Sperling S, Bendstrup E. A new paradigm is needed to explain long COVID. Lancet Respir Med 2023;11:e12–3. 10.1016/S2213-2600(22)00501-X [DOI] [PubMed] [Google Scholar]
- 29. Ma Q, Liu J, Liu Q, et al. Global percentage of asymptomatic SARS-Cov-2 infections among the tested population and individuals with confirmed COVID-19 diagnosis: a systematic review and meta-analysis. JAMA Netw Open 2021;4:e2137257. 10.1001/jamanetworkopen.2021.37257 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Baskett WI, Qureshi AI, Shyu D, et al. COVID-specific long-term sequelae in comparison to common viral respiratory infections: an analysis of 17 487 infected adult patients. Open Forum Infect Dis 2023;10:ofac683. 10.1093/ofid/ofac683 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
bmjopen-2023-078119supp002.pdf (9MB, pdf)
bmjopen-2023-078119supp001.pdf (1.7MB, pdf)
Data Availability Statement
Data are available on reasonable request. Data protection regulations will be coordinated with the University of Alberta ethics committee, Alberta Health Services and Alberta Health. Data containing identifiable information will not be shared due to the privacy of the participants and ethical reasons. Deidentified data are available to researchers under a signed data disclosure agreement. Researchers are invited to contact the corresponding author or file a request through Alberta SPOR SUPPORT Unit (https://absporu.ca/) immediately following publication with ethical approval obtained and agreement with Alberta Health and Alberta Health Services. The study protocol was published in BMJ Open and is publicly available online (https://bmjopen.bmj.com/content/13/2/e067449.long). The aggregate data and code used in this analysis are available from the corresponding author on reasonable request.