Risk factors of post-COVID-19 conditions attributed to COVID-19 disease in people aged ≥50 years in Europe and Israel

Objectives High age, male sex and pre-existing comorbidities are risk factors for a more severe development of COVID-19, and individuals surviving COVID-19 may experience persistent symptoms afterwards referred to as ‘post-COVID-19 condition’, which represents a range of symptoms after recovering from COVID-19. This study aims at identifying risk factors of post-COVID-19 conditions among people aged ≥50 years. Study design We conducted a cross-sectional study based on data from the Survey of Health, Ageing and Retirement in Europe. Methods A multiple logistic regression model was used to investigate age, sex, education, comorbidities, smoking, body mass index, and COVID-19 hospitalisation as risk factors of post-COVID-19 condition. Results Participants aged ≥70 years (odds ratio [OR] 1.61) with medium (OR 2.38) and lower (OR 2.14) educational levels have a higher risk of post-COVID-19 conditions. In addition, when considering the severity of the COVID-19 disease, those who were hospitalised due to COVID-19 had a 26 times higher risk of post-COVID-19 conditions compared with those who were only tested positive (OR 25.9). Conclusions This study supports that health inequalities exist across educational levels with respect to post-COVID-19 conditions, although misclassification may be more common among lower educated participants. The results suggest that policy makers should increase educational interventions towards increasing health literacy.


Introduction
By fall 2022, the COVID-19 pandemic has been present for more than 2 years, and repercussions of COVID-19 may well be observed. Although COVID-19 strikes at all ages, older people are at increased risk of a critical course of disease, as are people with pre-existing comorbidities. 1 Following COVID-19 disease, lingering symptoms lasting from weeks to months may be experienced. The World Health Organisation has recently coined this as 'post-COVID-19 condition', defined as a condition that "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." 2 Studies describing the risk factors of post-COVID-19 health conditions are limited and show conflicting results, but increasing age, female gender, multimorbidity, high body mass index (BMI), lower educational level, having several symptoms during COVID-19 infection, and a more severe acute phase of COVID-19 disease, have all been associated with post-COVID-19 condition. 3,4 Full recovery from COVID-19 is important to keep people's work capacity, productivity, and ability to return to normal everyday activities, and thus, risk factors of post-COVID-19 must be explored. To our knowledge, no previous studies have explored the possible risk factors in a large population-based sample of middle-aged and older adults comparing several countries. Thus, the aim of the study was to investigate risk factors of post-COVID-19 conditions among people aged 50 years within 27 European countries and Israel.

Methods
We used data from the second wave of the Survey of Health, Ageing and Retirement in Europe (SHARE) COVID-19 survey (SCS2) conducted from June to August 2021. 5 Participants, who had tested positive with COVID-19 (Supplementary figure F1), were asked, "Have you experienced any long-term or lingering effects that you attribute to your Covid Illness?" with the following answer options: 'fatigue', 'cough, congestion, shortness of breath', 'loss of taste or smell', 'headache', 'body aches, joint pain', 'chest or abdominal pain', 'diarrhoea, nausea', 'confusion', and 'other'. Based on this question, an outcome variable was created and dichotomised, indicating whether the respondents had 'no symptoms' or 'one or more'. The following variables were selected as the primary potential risk factors for post-COVID-19 conditions: age (50e69 and 70 years), sex (male, female), educational level (according to the International Standardized Classification of Education [ISCED] classified into lower [ISCED groups 0e2], medium [ISCED groups 3e4] and higher [ISCED groups 5e6]), comorbidity ('no diseases', 'one or two diseases', and 'three or more diseases'), smoking ('not smoking', 'smoking now', 'previous smoker'), BMI ('normal weight' (BMI 18.5 and <25 kg/m 2 ), 'underweight' (BMI <18.5 kg/m 2 ), 'overweight' (BMI 25 and <30 kg/m 2 ) and 'obese' (BMI 30 kg/ m 2 ) and a COVID-19 hospitalisation variable indicating if respondents have been hospitalised due to COVID-19 ('yes' or 'no') and serve as a proxy of the severity of the COVID-19 disease. Moreover, the variable 'country', representing the 28 SHARE countries, was included (Germany, The Netherlands, Belgium, Luxembourg, Switzerland, Austria, France, Spain, Portugal, Malta, Italy, Greece, Cyprus, Estonia, Latvia, Lithuania, Poland, Czech Republic, Slovakia, Slovenia, Hungary, Croatia, Romania, Bulgaria, Denmark, Sweden, Finland, and Israel; Supplementary Table S2). Age, sex, diseases, and COVID-19 hospitalisation were drawn from SCS2, and all other variables were drawn from SHARE wave 8 (conducted in 2020) 6 or the latest wave possible.
Multiple logistic regression models were used to study the risk factors of post-COVID-19 conditions, and the analyses were performed in three steps. First, we included age, gender, education, comorbidity, smoking and BMI (model 1); second, we added a 'country' variable (model 2); and finally, we included the 'COVID-19-hospitalisation' variable (model 3). In all analyses, the crosssectional individual weights supplied by SHARE were applied to yield a representative sample.

Results
Of all SCS2 participants aged 50 years (n ¼ 49,044), 6.5% (n ¼ 3156) reported having tested positive for COVID-19 and were thus included in the final analysis (supplementary figure F1). In the final sample, 42.4% were aged 70 years, and 61.6% were females. The proportion of people with a high educational level was 19.2%, 45.1% with a medium level, and 30.4% had a lower educational level (10% were missing). Among the respondents, 19.4% had three or more comorbidities and 24.1% had none, 39.1% were overweight, and 28.4% had a normal BMI. Only 7.9% reported to be smokers, and 16.4% were hospitalised due to COVID-19 disease (Supplementary  Table S3).
Of those who tested positive, almost one-quarter (23.5%) reported at least one post-COVID-19 condition, the most common being 'fatigue' (18.8%), followed by 'cough, congestion or shortness of breath' (13.5%) and 'body aches or joint pain' (13.2%; Supplementary Table S4). In the crude analysis, respondents with higher age (70 years), medium or low education, multimorbidity, and obesity were at higher risk of post-COVID-19 conditions. However, in the adjusted model ( Table 1). Sex, comorbidity, BMI, and smoking did not show any association with post-COVID-19 conditions.

Discussion
Post-COVID-19 conditions are common among COVID-19einfected people aged 50 years living in the 28 SHARE countries and affects, in particular, older people and those with lower educational level. The association to lower educational level has not been described previously in a European population. However, educational attainment is a social determinant of health, as well as a predictor of the severity of COVID-19 disease. 7 Low educational level is also associated with low health literacy, 8 that is, the ability to reflect upon one's illness and understand how to distinguish between symptoms from chronic disease and symptoms related to COVID-19, and poorer health literacy may explain an overreporting of symptoms.
Not only is increasing age associated with COVID-19 disease severity, 1 but it is also associated with higher risk of post-COVID-19 conditions, 4 and in line with our findings. This may be explained by an age-associated adverse immunological response, as specific cell changes persist longer in older compared with younger individuals, 9 and therefore, the symptoms may be more persistent.
The most salient predictor of post-COVID-19 conditions in our study was the COVID-19 hospitalisation, which may represent the severity of the COVID-19 disease and is in keeping with other studies. 4 However, as intensive treatment during hospital stay can cause similar symptoms, 10 it may be difficult to determine if the symptoms are attributed to the in-hospital treatment, COVID-19, or a combination.
The strength of this study was the large and representative sample of 28 countries using standardised methods for data collection. Also, this study comprised both hospitalised and nonhospitalised participants. However, it is a weakness that SHARE did not collect information on the duration of the post-COVID-19 symptoms to account for the extent of the post-COVID-19 condition. A time frame for long-term or lingering symptoms would have improved the comparability across individuals. In general, we cannot exclude residual confounding; for instance, both hospitalisations and post-COVID-19 conditions may be correlated with vaccination status. Finally, as a self-reported questionnaire, the post-COVID-19 conditions may be over-or under-reported due to information bias.

Conclusion
The results indicate that lower educational level, higher age, and prior hospitalisation for COVID-19 disease increase the risk of post-COVID-19 conditions. Although biological mechanisms may explain the adverse effects of higher age and disease severity, the effect of a lower educational level for post-COVID-19 conditions is more likely explained by lower health literacy. In conclusion, these findings have implications for public policies in Europe and Israel, as we shed light on the social inequalities in health, which still exist at large. We encourage policy makers to increase their focus on educational interventions and implement programmes and policies to increase health literacy and hereby reduce inequalities in health.

Ethical approval
None required.