Epidemiological knowledge of long-term outcomes following a COVID-19 episode remains limited, although it is becoming increasingly clear that some patients who have had an acute COVID-19 experience, reports not being fully recovered several months after onset of COVID-19 symptoms accompanied by long-term persistent symptoms. It is known as long-COVID, post-COVID syndrome or Post-Acute Sequelae of SARS-CoV-2 infection (PASC).1–3 It appears to occur in both hospital and community settings. Among the long-term persistent symptoms, the following would stand out: fatigue, headache, chest pain, dyspnoea, neurological, psychological, and cardiovascular symptoms, with an impact on quality of life as assessed by the EuroQol tool.3–5

At this point of knowledge, the World Health Organisation (WHO) has developed a clinical case definition of post COVID-19 condition in adults by a Delphi consensus, stating that “Post COVID-19 condition occurs in individuals with a history of probable or confirmed SARS-CoV-2 infection, usually 3 months from the onset of COVID-19 with symptoms that last for at least 2 months and cannot be explained by an alternative diagnosis”. As common symptoms, WHO include fatigue, shortness of breath, cognitive dysfunction but also others, and these symptoms may be new onset following initial recovery from an acute COVID-19 episode or persist from the initial illness. Symptoms may also fluctuate or relapse over time.6

Due to the huge number of people affected by COVID-19 after the pandemic years and the growing evidence of long-term sequelae, a complete epidemiological understanding of long-term effects of COVID-19 seems to be necessary for policy makers and healthcare systems, and ultimately for the society that must understand the challenges faced by long-term COVID patients.7

The International Severe Acute Respiratory and emerging Infections Consortium (ISARIC) WHO Clinical Characterisation Protocol (CCP) was designed by international consensus in 2012 for any severe or potentially severe acute infection of public health interest.8 In the COVID-19 pandemic context, the ISARIC Global COVID-19 follow-up working group, has conducted an ongoing COVID-19 long term follow up study9 and an ISARIC survey (as a screening tool for persisting symptoms) has been developed, in collaboration with the WHO and a range of experts.10 The use of this “ISARIC COVID-19 follow-up survey foradults” allows a same standardized data collection among the different international studies in order to obtain valid comparisons of their results among the different countries and settings such as hospitalised and community settings. Its use therefore provides further insight into the burden of disease, and who is at greatest risk of developing long-term complications.

The objective of this study was to characterize long-term patient reported outcomes in adults after acute COVID-19 in Spain, by using the standardized data collection form mentioned above (the ISARIC survey as a screening tool for persisting symptoms).

More than 15% of patients, reported NFFR of their COVID-19 illness. This % is lower than that reported by other studies, but the mixed characteristics of patients (with and without hospital admission for COVID-19) has to be taken into account when comparing with other studies focusing only on patients with more severe disease requiring admission. In this regard, the meta-analysis performed by O’Mahoney et al. (2022)13 includes data from hospitalised (n=122 studies), non-hospitalized (n=11 studies), and mixed as our study (n=36 studies). Among studies with non-hospitalised patients, the pooled prevalence of COVID-19 survivors experiencing at least one symptom during follow-up was 34.5%. Among mixed studies including both groups, the pooled prevalence was 37.8% and lastly, in studies with patients requiring admission, the pooled prevalence was 52.6%.13 The meta-analysis published by Chen et al. (2022)7 would also support this dose–response pattern, with prevalences of PASC of 34% and 33% for non-hospitalized or mixed studies, and 54% for studies with hospitalised patients.

These meta-analyses also show that the prevalence of post-acute symptoms decreases as more time has elapsed since the COVID-19 illness. In this sense, our study would be very long-term compared to those of the studies included in the aforementioned meta-analyses, with means of follow-up of 126 days13 or being <120 days in most of published studies reported by Chen et al. (2022).7

The high vaccination rate in Cantabria and Spain must also be considered in the context of a COVID-19 complete immunization as a protective factor for long- COVID-19 symptoms. In the case of the meta-analysis with the highest prevalence of long-term symptoms (prevalence of at least one symptom of 80%), it is based on studies prior to January 2021, which implies that vaccination rates would be very low or non-existent. It would be also based on studies that in no case exceed an average of 110 days of follow-up.14 Nevertheless, it should be noted that the strength of the current evidence on vaccination as a protective factor for long-COVID-19 symptoms is limited and our results do not support a clinically relevant protective effect either.15–17 Finally, it is plausible that the earlier more virulent variants would be associated with more permanent symptoms.18 According to this hypothesis, studies with patients infected earlier in the pandemic (and thus with more virulent variants) could also report a higher prevalence of long-term symptoms.

In terms of the strengths of our study, the methodology used in the form of simple random sampling of the total number of cases in a whole autonomous community denotes that we have real-life conditions, providing a population-based view of real-life conditions at the community level. If we restrict to our patients with hospital or ICU admission, our percentage increase to 23.4% and 33.3% respectively. Percentages that would be more similar to those reported in other studies such as Sigrid et al. (2021)4 although lower than the pooled prevalences of the mentioned meta-analyses.7,13

Among our predictors of NFFR, having been admitted to ICU showed the strongest magnitude of association (adjusted OR=2.11). However, due to our small sample of patients who required ICU admission (n=18), this association did not yield statistical significance. Female sex and older age were independent predictors of NFFR, yielding statistical significance. Our results for the female sex are supported by other reviews and meta-analyses such as that of Aiyegbusy et al. (2021)5 or Chen et al. (2022)7 reporting an OR of 1.57 (95%CI 1.09–2.26) very similar to ours, and also with primary studies such as the Sigfrid et al. (2021)4 based on the same ISARIC survey as ours. Interestingly, for age, our results contrast with those of Sigfrid et al. in their prospective multicentre cohort study in adults discharged from UK hospitals after COVID-19, because in this study, women under 50 years old and those with more severe acute disease in-hospital, had the worst long-term outcomes. In terms of meta-analyses, those published by Aiyegbusy et al. (2021)5 and Chen et al. (2022)7 support the association between older age and higher prevalence of long-term symptoms, whereas that of O’Mahoney et al. (2022)13 show inconclusive results. The 70+ year olds in our study were less likely to report NFFR than the reference category (<50 years) (OR=0.90). It is possible that this result may be explained by a survival bias due to the fact that the most severe cases have died. Thus, if those >70 years of age susceptible to long-term sequelae have not survived, they have not had a chance to have them.

Main symptoms in our group NFFR (fatigue. headache. chest pain. shortness of breath & dyspnoea. cognitive dysfunction. neurological. psychological) coincide with those reported in primary studies, also listed in Annex 3 and Table 2 of the WHO report6 and as it is logical, coincide with those reported in most of the published systematic reviews and meta-analyses, with the percentages of each symptom being in very similar ranges.5,7,13,14,19–21 Regarding cardiovascular symptoms, in our study the prevalence was low, but it must be taken into account that we only included Chest Pain and Palpitations. If we compare these two specific cardiovascular symptoms, we find similar prevalences reported in meta-analyses: i.e. 9.3% and 11%.14,19

In terms of quality of life, no dose-response pattern was observed because higher punctuations were observed in COVID-19 patients feeling “Recovered or unsure” than in close contact without COVID-19. However, large differences in quality of life were found between COVID-19 patients. Restricting to COVID-19 patients, those NFFR scored on average 9.63 points less in Euroquol than those patients feeling “recovered or unsure”, with independence of sex, age, immunization status and severity of COVID-19. They were 2.6 times more likely to have low Euroquol scores (≤the median of 80 points). In this sense, Aiyegbusy et al. (2021) meta-analysis, includes five studies that specifically use Euroquol to analyze quality of life.1,22–25 One of these studies reports a difference of 10 points, very similar to ours.1

Our study includes the group of close contacts without evidence of infection as a comparative group, allowing a dose–response pattern between the three groups. The dose–response pattern is one of Bradford Hill's classic causality criteria and would therefore support the association between a higher prevalence of these symptoms and NFFR. Using the same tool based on the ISARIC WHO clinical characterization protocol, would allow valid comparisons with other studies. The similarity of symptoms found through this standardized survey, coinciding with other primary studies in other countries, and the mixed nature of our setting (with and without hospital admission for COVID-19) would support both the internal and external validity of our results, supporting the existence of a pattern after COVID-19 compatible with a complex clinical picture known as long-COVID, post-COVID syndrome or PASC. In terms of limitations, it should be noted that we only have a single cut-off point (the questionnaire was administered only once after infection). In this sense, prospective studies that take more than one visit, following the ISARIC protocol more completely, would support the persistence of these symptoms over time. Nevertheless, minimum time from diagnosis to completing our survey was 143 days (4.7 months) with a median time of 385 days (12.7 months), supporting the long-term post COVID-19 context of our results.

More than 15% of patients in our real-life population-based study, reported NFFR of their COVID-19 illness after a minimum time of 4.7 month from diagnosis. Female sex and older age were in our study statistically significant independent predictors of NFFR. Most prevalence symptoms in our patients were in accordance with WHO definition of post COVID-19 condition in adults and were in order of frequency: Fatigue; Loss of smell; Problems speaking or communicating; Loss of taste; Confusion/lack of concentration; Persistent muscle pain and Shortness of breath/breathlessness. When comparing the three ordinal groups (Close contacts without infection, COVID-19 feeling recovered, and COVID-19 NFFR) the prevalence of these symptoms was increasingly higher among each ordinal group with a statistically significant dose-response pattern (p<0.001). Large differences in quality of life were found between patients NFFR and those patients feeling “recovered or unsure” reporting 9.63 points less in Euroquol with independence of sex, age, immunization status and severity of COVID-19.

There was no funding associated with this study.

O.P. conceptualization, methodology, writing-review & editing. M.S. methodology, analysis and interpretation of data, drafted the work. L.R. analysis and interpretation of data, writing-review & editing. J.M.C. methodology, analysis and interpretation of data, writing-review & editing. A.H.L. conceptualization, methodology, supervision, writing-review & editing.

There are no conflicts of interest associated with this study.