The present study identified the patterns of long COVID and its association with chronic diseases in adults and older adults living in a municipality in southern Brazil. Four patterns considerably different from each other were identified: respiratory pattern, neuromusculoskeletal, neurosensory and cognitive, with the latter having the highest prevalence. Chronic diseases increased the probability of developing long COVID patterns by up to 121%, with emphasis on anxiety, heart and spinal problems, which were associated with all patterns. Only diabetes mellitus was not associated with any of the long COVID patterns identified.
Attempts to relate preexisting underlying diseases to the long duration of COVID-19 symptoms are recent, as it is known that the severity and mortality of infected patients is directly related to the presence of comorbidities such as hypertension, diabetes, CVDs, acute cardiac injuries and chronic respiratory diseases [5]. The investigation of symptom patterns of long COVID and their association with chronic diseases is still incipient. Thus, the comparison of the findings of the present study with the others must be carried out with caution, especially due to the differences between the analysis methods.
Regarding the relationship between diabetes and long COVID, there seems to be a bidirectional relationship between these variables [13, 14]. Harding et al. [14] showed that COVID-19 survivors may be at increased risk of new-onset diabetes and that preexisting diabetes is also a risk factor for the development of postacute sequelae of COVID-19. In turn, Ssentongo et al. [15] demonstrated that COVID-19 was a risk factor for the development of diabetes, causing an increase of 66% among survivors. Furthermore, our findings on the lack of association between diabetes and long COVID patterns corroborate the studies by Thompson et al. [16], Peghin et al. [17] and Austin et al. [18]. Finally, it should be mentioned that some of the participants may have developed diabetes after the infection, but this was not evaluated in our study.
The presence of heart problems increased the probability of the occurrence of the four identified patterns from 41–73%. A study carried out by Xie et al. [19] showed increased risks for heart disease in long COVID patients, with these results being observed in patients without heart disease prior to COVID-19 infection and in those who did not require hospitalization during the acute phase of COVID-19. In addition, there also seems to be a bidirectional relationship between COVID-19 and diseases of the cardiac system [20]. Persistent symptoms, such as dyspnea associated with tachycardia and postexertion fatigue, were described in a cohort of individuals who had mild symptoms in the acute phase and who did not have preexisting diseases of cardiac or respiratory origin before the infection; however, they developed cardiac symptoms in the long phase [21]. This suggests that cardiac symptoms may be related to chronic inflammatory cardiac involvement when there is no structural heart disease or when there is an increase in cardiac biomarker levels [21]. In addition, CVDs affect immune function and are related to the prognosis of COVID-19 [22]. Diabetes and obesity are among the mechanisms for developing CVD, [23] which would explain the association of heart problems with all patterns, given that most of our sample (73.3%) was overweight.
In our findings, the presence of breathing problems increased the probability of presenting the breathing pattern by 70%. Chronic diseases such as asthma and history of myocardial infarction were related to long COVID in nonhospitalized patients in a Polish study [24]. This suggests that preexisting diseases may influence adaptive immune responses and that they may also be linked to the development of long COVID symptoms [25].
With the exception of diabetes, all chronic diseases investigated in our research were associated with the neuromusculoskeletal pattern. It is likely that a multifactorial process triggers the symptoms of long COVID [25]. One of the pathophysiological mechanisms of long COVID is directly associated with chronic hyperinflammation, where damage to the blood‒brain barrier occurs, increasing permeability to neurotoxic substances, such as elevated interleukin-6 (IL-6), which impairs muscle metabolic homeostasis and stimulates muscle loss [26]. The musculoskeletal system can be directly affected by SARS-CoV-2 in myocytes or indirectly affected through the systemic release of cytokines that cause changes in muscle homeostasis [27]. It is known that patients with chronic diseases are the most affected by COVID-19, and it seems reasonable to consider that this may have contributed to the development of the neuromusculoskeletal pattern.
Lifestyle interferes with health and contributes to unfavorable outcomes, such as the development of diseases and mortality, or it plays a protective factor for those with healthy habits [28]. In this study, 7 out of 10 participants were overweight, and 6 out of 10 did not practice physical activity, factors that are directly linked to back problems and musculoskeletal and cardiometabolic diseases, which may justify the association with the patterns of long COVID found.
The occurrence of depression and anxiety has been recurrent in the long COVID in population-based studies and systematic reviews. Our results identified an association between anxiety and all patterns evaluated [29, 30]. One-quarter of the sample had this morbidity, slightly below that observed in a Brazilian cohort of out-of-hospital adults, which was approximately one-third [31]. That study also observed that the main neuropsychiatric manifestations of long COVID were memory loss, sleep problems and depressive feelings and that patients with depression were more likely to have memory problems [31]. The association of depression with the neuromusculoskeletal pattern can be attributed to prolonged resting time, lack of motivation caused by the disease, and physical inactivity, which very likely directly impacts muscle function [32].
Our findings showed that hypertension was related to the neuromusculoskeletal pattern. A previous diagnosis of hypertension increased the prevalence of this pattern by 33%, diverging from Thompson et al. [16] who found no association between high blood pressure and long COVID. Although the relationship between hypertension and severe cases of COVID-19 has been evidenced in the literature, [33] the relationship with long-term symptoms still seems uncertain.
Some limitations must be noted. The cross-sectional design is especially subject to potential reverse causality, since there is the possibility of diseases resulting from infection with the coronavirus, especially hypertension, diabetes mellitus and cardiovascular diseases. However, to minimize this bias, the questions about symptoms considered the acute phase and the time of the interview (long COVID). In addition, we mention the use of self-reported information. However, such an approach is considered valid and widely used in surveys on health conditions due to the simplicity and low cost in large population samples.
One of the strengths of this study is the use of a representative population-based sample from a middle-income municipality. Another point that deserves to be highlighted is the high response rate, using a promising method for research in the health area, especially with regard to the acquisition of knowledge during outbreaks of infectious diseases in rapid evolution, such as COVID-19. Furthermore, to the best of our knowledge, this is the first study to investigate long COVID symptom patterns and their association with chronic noncommunicable diseases.
In conclusion, we detected four patterns of long COVID, which shows that a number of persistent symptoms after COVID-19 infection can constitute a pattern of behavior. We observed that among the patterns discovered, the cognitive pattern was the most prevalent, followed in order by the respiratory, neuromusculoskeletal and neurosensory patterns. In addition, associations of these patterns with chronic diseases have been demonstrated, which reveals that individuals affected by these patterns are more vulnerable to developing symptoms of long COVID. These results point to the need to monitor people with chronic diseases and to intensify research on the patterns of symptoms of long COVID, in addition to the elaboration of public policy proposals to serve this population group.