Association of Acute Headache of COVID-19 and Anxiety/Depression Symptoms in Adults Undergoing Post-COVID-19 Rehabilitation

As a common non-respiratory symptom of COVID-19, headache should not be overlooked, and its characteristics should be recorded with scrutiny. Identifying risk factors associated with post-COVID headache will ensure immediate action and counseling for this population of patients. Therefore, the study aimed to investigate the relationship between headache and psychological state (stress level, depression, and anxiety symptoms) in adults undergoing post-COVID-19 rehabilitation. In addition, we used mediation analysis to evaluate the mediation effect of psychological variables in the relationship between headache and quality of life. This cross-sectional study included 147 patients undergoing post-COVID-19 rehabilitation at the Public Hospital in Poland (64 males, 83 females, with mean age of 56.97 years). Psychological parameters were evaluated using the Hospital Anxiety and Depression Scale (HADS), the Perceived Stress Scale (PSS-10), and the brief World Health Organization Quality of Life Scale (WHOQOL-BRIEF). Additionally, all participants completed a questionnaire related to COVID-19 symptoms and their severity, the place of COVID-19 treatment, and the need for oxygen therapy during hospitalization. Of all participants, 65% experienced headache during COVID-19. Of the participants with headache, there were significantly more females in this group (69% vs. 31%), and they were significantly younger (mean age 55.47 vs. 59.78 years). Participants with headache had a 27% higher HADS-D score, a 21% higher HADS-A score, and a 13% higher PSS-10 score. Moreover, gender and headache were found to be important predictor variables for total HADS and HADS-D, accounting for 11% and 7%, respectively. Mediation analysis has shown that the tested psychological variables mediated 39–68% of the total effect of headache influence on WHOQOL domains. In conclusion, our study demonstrated several relationships between headache that occurred during COVID-19 and symptoms of depression, anxiety, and perceived stress level during post-COVID rehabilitation also in the context of quality of life. Our results show that patients who experienced headaches during COVID-19 are at high risk of developing anxiety-depressive symptoms later. Female gender is associated with a higher prevalence of headache during COVID-19.


Introduction
Patients with coronavirus disease 2019 (COVID- 19) can experience a wide range of clinical manifestations, from no symptoms to critical illness. Initially, COVID-19 was described as a respiratory disease; however, when the number of cases began to increase, there were reports that other organs and systems were also affected by the disease as well [1].
Headache attributed to COVID-19 can be persistent and disabling in certain cases, even in patients with no prior history of headache [21]. In the first 180 days, the time course of post-COVID headache appears to be stable, but longitudinal studies are necessary [22]. The predictors of an increase in the intensity of headache [7] attributed to COVID-19 include fever, dehydration, and gender. The severity of COVID-19-related headache is greater in patients with primary headache group disorders [23].
As a common non-respiratory symptom of COVID-19, headache should not be overlooked, and its characteristics should be recorded with scrutiny. Identifying risk factors associated with post-COVID headache will ensure immediate action and counseling for this population of patients. Therefore, the study aimed to investigate the relationship between headache and psychological state (stress level, depression, and anxiety symptoms) in adults undergoing post-COVID-19 rehabilitation.

Study Settings and Participants
This cross-sectional study included 147 patients who survived COVID-19 (64 females, 83 males, with a mean age of 56.97, SD 9.86), the average time elapsed between onset and commencement of post-COVID rehabilitation was 19.13 (SD 5.61) weeks. Included patients were undergoing post-COVID-19 rehabilitation at the Public Hospital in Glucholazy (Glucholazy, Poland). Dyspnea and fatigue, which hindered everyday functioning, was the main reason for admitting the patient to the ward and undergoing pulmonary and cardiovascular rehabilitation. Exclusion criteria were limited to patients under 18 years of age, a cognitive impairment that prevents psychological evaluation, any serious psychiatric disorders (e.g., bipolar disorder, major depression, and schizophrenia), and initiation of psychiatric or psychological treatment during the course of the study. Data were collected from patients admitted to the rehabilitation ward between July and September 2021. The research was in accordance with all relevant national regulations, institutional policies, and tenets of the Declaration of Helsinki. All study procedures were approved by the Institutional Review Board of the Wroclaw University of Health and Sport Sciences (Wroclaw, Poland) (nr 13/2021). No compensation was offered to the participants. Written informed consent was obtained from all individuals included in the study.

Survey
All participants completed a questionnaire related to their sociodemographics, such as age, sex, education, marital status, and employment. They also responded to questions regarding their current health and health before COVID-19 (good, neither good nor bad, bad), physical activity before COVID-19 (yes/no), and what the symptoms of COVID-19 were and their severity. Additionally, questions were asked about comorbidities, the place of COVID-19 treatment, and the need for oxygen therapy during hospitalization.

Depression and Anxiety Symptoms
To assess the severity of symptoms of depression and anxiety, the Hospital Anxiety and Depression Scale (HADS) was used. It is a 14-item scale scoring from 0 to 3 for each item. Seven items related to anxiety (HADS-A), while the remaining seven related to depression (HADS-D). The global score ranged from 0 to 42, with a cut-off point of 8/21 for anxiety and 8/21 for depression. The higher the score, the greater the anxiety or depression symptoms. According to the authors, the Cronbach alpha ranges from 0.78 to 0.93 for HADS-A and from 0.82 to 0.90 for HADS-D, and the test/retest correlations were r = 0.80 [24].

Assessment of Stress Level
The Perceived Stress Scale (PSS-10) was used to measure psychological distress. It contains 10 questions on a five-point scale from 0 to 4 [25]. Participants are asked to rate their stress levels during the past month. The higher the score, the greater the feeling of stress. The scale demonstrated a satisfactory internal consistency (Cronbach's alpha = 0.69).

Quality of Life Assessment
Quality of life was assessed using the short version of the World Health Organization Quality of Life (WHOQOL-BREF). This questionnaire assesses quality of life in four domains: physical health, psychological, social relationships, and environmental. It contains 26 questions with answers arranged on a five-point Likert scale. WHOQOL-BREF is a cross-culturally valid assessment of quality of life, with an internal consistency alpha varying in each domain from 0.68 to 0.82 [26].

Data Analysis
Data were analyzed using SPSS 25.0 software (IBM Corp, Washington, DC, USA) and JASP 0.16.1 software (University of Amsterdam, Amsterdam, The Netherlands). Continuous variables are presented as means and standard deviations (SDs), and the categorical responses are presented as frequencies and percentages. Multiple linear regression (stepwise) was used to identify the association between headache during COVID-19 and depression and anxiety symptoms and stress levels. Prior to the multiple linear regression analysis, the assumption of a linear relationship (using the point biserial correlation coefficient) between the outcome variable and the independent variables was tested (variable encoding: 0-headache, 1-without headache). Three variables (stress level, depression, and anxiety symptoms) were tested as possible mediators of the relations between headache and quality of life outcomes. Mediators were tested by calculating bias-corrected 95% CI using bootstrapping (5000) with the JASP software. The variables were constructed as independent single mediator models. These models were based on different domains of quality of life, that is, physical health, psychological, social relationships, and environmental health. The results were presented as the effect size of the total, direct, and indirect effects. A significance level of α < 0.05 was established.

Between-Group Comparison
Taking into account the results of the intergroup comparison, significant statistical differences were found in all variables examined, except for WHOQOL social relationships domain (p = 0.051). Participants with headache had a 27% higher level of HADS-D (p = 0.001), a 21% higher level of HADS-A (p < 0.001), and a 13% higher level of PSS-10 (p = 0.28). Participants with headache were also characterized by statistically significantly lower scores for the quality-of-life domains ( Table 2). They obtained scores of 12, 6, and 8% lower for the domains of physical health (p = 0.016), psychological (p = 0.018), and environmental (p = 0.002), respectively. All statistically significant differences were within the moderate effect size range (Cohens d ranged from 0.43 to 0.61).

Predictors
In order to examine how headache experienced during COVID-19 can explain a statistically significant amount of variance in post-COVID depression, anxiety, and stress level, stepwise multiple regression was used. Headache, gender, and age were included as a predictor. For total HADS and HADS-D, gender and headache were found to be important predictor variables, accounting for 11% and 7% of the variance in these models (p = 0.01 and p = 0.002, respectively) ( Table 3). Among the variables included in the anxiety model (HADS-A), headache was revealed to be a significant predictor. This model accounted for 4% of the anxiety variance (p = 0.02). Taking into account the stress level (PSS-10), it was found that gender was a significant predictor (p = 0.001).
terms of quality of life, headache during COVID was associated with worse outcomes of post-COVID physical health (r = 0.19), social relationships (r = 0.19), and environmental (r = 0.25) in WHOQOL-BREF domains.

Predictors
In order to examine how headache experienced during COVID-19 can explain a statistically significant amount of variance in post-COVID depression, anxiety, and stress level, stepwise multiple regression was used. Headache, gender, and age were included as a predictor. For total HADS and HADS-D, gender and headache were found to be important predictor variables, accounting for 11% and 7% of the variance in these models (p = 0.01 and p = 0.002, respectively) ( Table 3). Among the variables included in the anxiety model (HADS-A), headache was revealed to be a significant predictor. This model accounted for 4% of the anxiety variance (p = 0.02). Taking into account the stress level (PSS-10), it was found that gender was a significant predictor (p = 0.001).

Mediation Analysis
Three variables (stress level, depression and anxiety symptoms) were tested as possible mediators of the relations between headache during COVID-19 and post-COVID quality of life outcomes (Figure 2). In all the models tested, the direct effect was not statistically significant. In turn, the indirect effect was significant, which resulted in the significance of the total effect. The tested variables mediated 39%, 60%, 68%, and 43% of the total effect of physical health, psychological, social relationships, and environmental WHOQOL domains, respectively (Table 4).

Discussion
COVID-19 is a systemic inflammation that affects all age groups, with a high mortality rate and severe adverse outcomes. It involves the nervous system, blood vessels, lung, heart, liver, gastrointestinal system, kidney, eyes, and other organs [27]. It is important to keep in mind that as the COVID-19 pandemic rapidly sweeps across the world, it induces a considerable degree of negative economic and psychosocial consequences that can contribute to poor mental health.
COVID-19-related headache was a commonly reported symptom in many studies, but there was a great diversity in its frequency, severity, character, and duration, with a prevalence of headache ranging from 3.5 to 34% [4]. There is no consensus on how COVID-19 affects the nervous system, and the mechanisms involved in COVID-19 headache are also unclear. Psychological and neurological symptoms after COVID-19 may be explained by different pathophysiological bases such as direct neuroinvasion with damage to the neuronal pathway, indirect effects mediated by hypoxia, hypertension, coagulopathy, and cytokine storm in the CNS, worsening of preexisting brain diseases or new disorder [28].
In the present study, 65.3% of the participants reported headache during COVID-19. The frequency of headache observed in this study was much higher than that reported in most studies. We have to remember that the study included people who suffered long-term complications after COVID and therefore entered rehabilitation.
Previously published surveys showed increased symptoms of depression, anxiety, and stress-related to COVID-19, as a possible result of psychosocial stressors such as fear of the disease, loss of life, and economic issues [29][30][31]. Numerous studies in the general population have consistently shown that headache is more prevalent in women than in men. The most important risk factors for headache include overuse of acute migraine medications, ineffective acute treatment, obesity, depression, stressful life events, age, and low education level [32,33].
None of the results of the published surveys evaluated the relationship between headache as a somatic symptom related to the COVID-19 pandemic and psychological parameters or quality of life for long-COVID observation. It seems to be important to find a simple method to identify the group at high risk for developing psychiatric disorders and to provide early preventive measures or treatment to avoid further consequences. To our knowledge, this is the first study to investigate whether COVID-19 with a coexisting headache has a significant negative impact on patient psychological parameters and quality of life for 6 months after an acute COVID-19 infection.
Our findings present a comparison of the results of sociodemographic and clinical data between two groups of people undergoing post-COVID rehabilitation: 1/with acute comorbid headache during the disease and 2/without this symptom. All examined psychological parameters differed significantly to the disadvantage of the headache group, which means that people who reported headache during COVID-19 infection had worse post-COVID emotional state: a higher level of anxiety and depression, a higher level of stress and, at the same time, lower quality of life. The largest difference was observed for depression (r = −0.27).
In order to answer the question: which variable may predispose to lowering or increasing the level of depression, i.e., to determine the direction of the described changes, the authors of the study performed a regression in which depression was the dependent variable. We have observed that 11% of the variability in the HADS-D model could be explained by gender and headache, and these variables predisposed them to higher depression scores. Furthermore, headache caused the average HADS-D score to increase by almost 2 points while belonging to the female gender increased this result by an average of 1.69 points. Other authors have reported similar results regarding the higher prevalence of headache in women [34][35][36].
The headache during COVID-19 itself did not significantly affect post-COVID quality of life; however, by influencing psychological parameters, it ultimately indirectly influenced its domains as well. The greatest indirect impact on quality of life was recorded for its areas: social and psychological. Our results are consistent with those of Perlis et al. [15] The authors found that individuals with prior COVID-19, who reported headache during acute infection, appeared to have an elevated risk of depressive symptoms. However, as the authors underline, as a cross-sectional study, the possibility that individuals with current depression are more likely to recall or report headache cannot be excluded.

Limitation
Despite its novelty and many strengths, this study is not without some limitations. This study was carried out in a single center and involved only inpatients, this decreases the ability to generalize the study and, therefore, extrapolates the results to patients with mild forms of the disease, who do not require post-COVID-19 rehabilitation. Additionally, we did not have information on the mental state of the enrolled participants before they contracted COVID-19. Moreover, we did not explore whether the patients suffered of headache or migraine before infection.

Open Questions and Future Research Directions
Although COVID-19 affects primarily physical health, the secondary influence of issues related to the pandemic on mental health should also be considered. Further studies with larger samples, prospective nature with multiple points of clinical data collection and analysis, taking into account inflammatory markers, could help clarify the mechanisms of COVID-19 headache and its association with severity of acute phase of the disease and its long-term consequences. When analyzing the results of this study, numerous questions arose that remain open, such as (1) Did subjects with COVID-19 headache have anxietydepressive symptoms before COVID-19? (2) Does the presence of anxiety-depressive symptoms prior to COVID-19 predict a more severe course of infection with a severe headache? (3) Is COVID-19 headache and the subsequent presence of anxiety-depressive symptoms related to inflammatory causes? (4) Could the COVID-19 headache be the cause of a new depressive episode? (5) Why does COVID-19 headache occur more frequently in women? Perhaps answering these questions will provide a clearer understanding of the relationship between headaches and the mental state of COVID-19 patients.

Conclusions
Our study demonstrated several relationships between headaches that occurred during COVID-19 and the emotional state of a patient who underwent post-COVID rehabilitation. Headache during COVID-19 was strongly associated with an increased intercity of depressive-anxiety symptoms and higher stress levels. In turn, psychological parameters were found to be significant mediators of the impact of headache on the quality of life of COVID-19 survivors. Female gender was significantly associated with headache during COVID-19 infection and post-COVID psychological disturbances.