This prospective survey aimed to investigate the extent of symptoms and daily restrictions in the course of COVID-19 in 60 non-hospitalized patient whereas we have put a major focus on mental illness and the effects on daily life and society.
Accordingly, 48.3% of the participants developed PC (WHO definition) (4). It appeared that PC, which is purely defined by persisting physical symptoms, cannot be limited to these disorders, additionally it directly or indirectly affects the quality of life of the individual. After six months 20% still reported restrictions in everyday activities such as sports, education or work. In addition 20% continued to suffer from anxiety and more than 40% continued to be concerned about their health.
Post-COVID prevalence in outpatient care
A single term for symptom persistence after acute SARS-CoV-2 infection has yet to be found. Currently, there are many different terms such as post-acute COVID, chronic COVID syndrome, or the more commonly used Long-COVID (LC) and PC (12, 13). The definition criteria here are often different and inconsistent between studies. This makes it fundamentally difficult to put data from different studies together.
According to WHO; fatigue, shortness of breath or breathing difficulties, memory-, concentration-, or sleep-problems are the predominant PC symptoms (14). This is largely consistent with our data. A systematic review of 37 included studies on PC reported that common long-lasting symptoms also include loss of smell, taste, and cough, which tended to play a minor role in our cohort (15). The PC prevalence reported in the literature shows an enormous range, which can be explained by methodological differences as well as differences in patient demographics and comorbidities (16). Standardized syndrome and symptom definitions as well as operationalizations are needed for more accurate data in the future. Another factor that may lead to differences in prevalence is the viral variant. Studies have shown that LC occurs less frequent in the Omicron variant than in the Alpha variant (21% vs. 50.5%) with the latter being predominant in our study population (17, 18).
Predictive factors for PC
Several risk factors for PC and LC have already been described. These include, for example, female gender, obesity, smoking and a wide range of comorbidities (6, 19, 20). However, little has been reported on early symptoms that have an effect on the long-term course of the disease, although this information may be useful for primary care physicians to provide early preventive or counseling measures. We were able to show that the presence of fatigue, reduced concentration, headache and deteriorated sleep quality at week 4 is predictive for later PC. This finding is not very surprising, since these symptoms are those that are predominantly still present at week 12 or persist until then, whereas other symptoms show a significant decrease between weeks 4 and 12 (see Fig. 2). The fact that persistent fatigue and/or cognitive impairment are predominant symptoms on PC should give reason to further characterize the neurobiological background in order to create adequate treatment strategies (21).
Effect on daily activities
Relevant influences on daily activities such as negative influence on sports, education or work, which were present in 20% in our group, were confirmed in a study by Jacobson et al. in 2021 on functional impairment, which showed health-related disabilities in work and general activities in 35% resp. 46% of the non-hospitalized patients after 3 to 4 months (22). One reason for the significantly higher prevalence in this study compared to ours could be a different SARS-CoV variant. In addition, the WPAI (Work Productivity and Activity Impairment Questionnaire) questionnaire used by Jacobson et al. has a different question structure (23). However, it seems obvious that limitations in daily life have a significant impact on mental health in particular.
Underreported mental symptoms
A meta-analysis showed that the most common mental symptoms after COVID-19 are anxiety, sleep disorders and depression (24). This can be confirmed partly by this study. We showed that six months after infection, 17% still suffered from mental impairments such as anxiety or depressive symptoms, while up to 40% were concerned that their health would deteriorate again or not completely recover. Many studies do not survey mental disorders by using common diagnostic criteria. We decided to use the GAD 7 as a reliable test in the final survey and were able to show that 10% of the subjects suffered from moderate or severe anxiety symptoms. In a meta-analysis from 2021, Schou et al. were able to detect the severity of the disease, the duration of the symptoms, and the female gender as risk factors for a mental disorder. Fortunately, in the same study, as well as in our study, it could be shown that there is a tendency for symptoms to improve over time (25).
As an explanatory model for the development of PC the biopsychosocial model is discussed (26). Here, biological factors (e.g., pre-existing conditions), health behaviors, psychological factors (e.g., social isolation, perceived threat), and social resp. situational factors (economic situation, availability of medical care) interact to define individual exposure and vulnerability. This determines the clinical and psychosocial (anxiety, depression, sleep disturbances) course of the disease which in turn influences the long-term physical and psychological consequences. It shows that COVID-19 and a resulting PC Syndrome is not a purely physical condition, but is based on a multifactorial disease development in which, in addition to physical aspects of the disease, social factors such as limited daily activities, isolation, and economic threat play a significant role (26, 27).
Increased economic burden and impact on health care system
We were able to show that PC poses a financial burden for the German social systems in two aspects. On the one hand, the prolonged illness leads to a longer incapacity to work (MD 11–15 days), during which the employer is obliged to continue paying wages, on the other hand, PC patients showed a significantly higher need for outpatient medical services. A representative data collection of a large German health insurance showed that PC patients with a mild course were on sick leave for an average of 90 days per year. In comparison, the average sick leave of the insured was only 15 days (28). In addition, it can be assumed that even after returning to work, the performance capacity may still be significantly limited and in some cases an occupation may no longer be possible due to the persistent symptoms and thus, an occupational disability is imminent or at least an occupational retraining is necessary. This is partially demonstrated by a large American survey which showed that PC was associated with a greater likelihood of unemployment and a lower likelihood of full-time work (29). Overall, due to the small number of cases in this study and the fact that there is not enough data on this topic, further research is needed to obtain reliable conclusions.
Strengths and limitations
In this prospective data collection, we were able to generate an almost gapless data set with the help of structured questionnaires, as well as personal regular long-term follow-up. Possible bias of the collected data due to different interviewers was excluded, as the standardized interviews were conducted exclusively by one person. Recruitment was conducted through the public health department, and every last individual who tested positive for SARS-CoV-2 during the recruitment period were invited to participate in this study without any exceptions. However, the need for active contact by patients does not eliminate the possibility of bias, as patients who actively choose to participate in a COVID-19 study may be more engaged with their disease than other patients.
Recruitment occurred when the alpha variant (B.1.1.7) was predominantly active (7, 8). Recent studies show that the subsequent Omicron variant has less severe courses and also leads to less LC cases (17, 30, 31). In addition, our cohort was predominantly unvaccinated (80% unvaccinated). In vaccinated collectives, the expression of LC is significantly lower. An initial dose of vaccine decreases the risk of LC by 13%, a second vaccination lowers it by an extra of 9% (32). Thus, one must assume that collectives currently affected by COVID-19 face less severe consequences than reported in our study. In general, it would be informative to conduct a survey of a similar cohort presently to gain a deeper comprehension of the disease expression and personal coping mechanisms.
Furthermore, only a relatively small study population was available in this study and we had no control group. Data on symptoms were based on subjective assessments and were not collected by rational parameters, e.g., respiratory rate or physical fitness performance tests. The final application of the GAD-7 at the time of six months does not allow a comparison to the situation before the infection or at the beginning of the disease.