An Overall Trend of Headache Remission during COVID-19 Pandemic for Chinese Patients with Preexisting Headache Highlights the Role of Family Supports

Background: Impact of COVID-19 pandemic on Chinese patients with pre-existing headache remains unclear. The present study investigated the impact of COVID-19 pandemic on headache patients in China. Methods: A survey was conducted through an online survey platform from June 6, 2020. Demographic characteristics, PHQ-9 scores, ISI scores, COVID-19 questionnaires and headache prole section were included in the online questionnaires. Results: Eventually a total of 15,000 participants from China completed the online questionnaire. Among them, 2806 participants (18.71%) had preexisting headache disorders. Our analysis showed a reality of those participants with remission of headache duration (3.414±6.859 vs 4.033±7.325, P=0.000), headache days per month (1.788±2.989 vs 2.092±3.694, P=0.000), and headache intensity (4.110±1.609 vs 4.290±1.680, P=0.000), during COVID-19. Smoking (OR=1.397, 95% CI 1.090 to 1.790, P=0.008) and getting support from family members during the social isolation (OR=1.656, 95% CI 1.075 to 2.550, P=0.022) were the independent factors of remission in headache duration. Education level (OR=1.478, 95% CI 1.103 to 1.980, P=0.009) and having relative or acquaintance got COVID-19 (OR=0.643, 95% CI 0.458 to 0.902, P=0.011) were the independent factors of remission in headache severity. And participants living in Wuhan area, having COVID-19 symptoms or COVID-19 diagnosis and having relative or acquaintance got COVID19 are more likely to get headache deteriorated. Conclusions: Participants presented an overall trend of headache remission during COVID-19. Family supports played an important role in headache remission. coronavirus disease 2019; PPE: personal protective equipment; RAS: renin-angiotensin system; non-steroid drug; PHQ-9: Patient Questionnaire-9; ISI: Insomnia Severity Index; IES-R: Impact Event Scale Revised; HIS: International Headache TTH: tension-type headache.

Insomnia Severity Index (ISI), a brief self-report instrument, was applied to assess subjective sleep quality.
The ISI comprises seven items. Each item is rated on a 4-point Likert-type scale, ranging from 0 to 3, and the total score ranges from 0 to 28. A higher score suggests more severe insomnia 13 .

COVID-19 questionnaires
COVID-19 questionnaires included having COVID-19 symptoms or COVID-19 diagnosis, having relative or acquaintance got COVID-19, getting support from family members, getting support from the outside world during the social isolation.
The Impact of Event Scale Revised (IES-R), a self-administered questionnaire, was applied to assess the psychological impact of COVID-19. The IES-R comprises 22 items, ranging from 0 to 4, and the total score ranges from 0 to 88. A higher score suggests more severe symptoms 14 .
The headache pro le section The headache pro le section included items on history of preexisting headache disorders, characteristics of accompanying symptoms (nausea, vomiting, photophobia, phonophobia and osmophobia, aggravation by or causing avoidance of routine physical activity, lacrimation, rhinorrhea, conjunctival injection, and eye bilges, remission by massage), headache triggers (increased stress, sleep too short, sleep too long, menstruation, bright light, noise or special smell, special foods and alcohol), headache intensity/headache days per month/headache duration at baseline or during COVID-19, factors associated with headache during COVID-19(stress, sleep quality, changes in the law of life and decreased social interaction).

Statistical Analysis
Statistical analysis was performed with SPSS V. 25.0 software (IBM, West Grove, Pennsylvania, USA). Categorical data were reported using proportions, while continuous data with means and standard deviations. The chi-square test was used to nd out whether differences in proportions are signi cant. Mann-Whitney U test or the Wilcoxon Signed Rank Test was used to compare non-normally distributed data. Binary logistic regression was rst performed to explore variables associated with headache relief. Then multivariate logistic regression analyses were performed to verify the independent in uence factor associated with headache relief. A p-value < 0.05 was considered to be statistically signi cant.
Post hoc comparisons using Bonferroni test evidenced that the aggravating group presented signi cantly higher proportion of having relative or acquaintance got COVID-19, compared with the remission group (P = 0.000) and unchanged group (P = 0.000). The aggravating group also demonstrated signi cantly higher proportion of smoking (P = 0.000), compared with the remission group, and signi cantly higher proportion of living in Wuhan area (P = 0.006) and getting less support from family members during the social isolation (P = 0.001), and lower ISI score (P = 0.014), compared with the unchanged group. Figure 5 shows that participants living in Wuhan area, having COVID-19 symptoms or COVID-19 diagnosis and having relative or acquaintance got COVID-19 are more likely to get headache deteriorated.
Participants living in Wuhan area took a higher exacerbation rate of headache duration (13.54% vs 6.31%, P = 0.005), headache days (19.79% vs 9.59%, P = 0.001) and headache intensity (8.33% vs 3.28%, P = 0.008). And participants having COVID-19 symptoms or COVID-19 diagnosis took a higher exacerbation rate of headache duration (13.25% vs 6.18%, P = 0.001), headache days (23.18% vs 9.19%, P = 0.000) and headache intensity (6.62% vs 3.28%, P = 0.029). Participants having relative or acquaintance got COVID-19 also took a higher exacerbation rate of headache duration (10.61% vs 5.96%, P = 0.001), headache days (18.44% vs 8.70%, P = 0.000) and headache intensity (9.78% vs 2.53%, P = 0.000).  3, Getting Unchanged or more support from the outside world vs Getting less support from family members.   16 . Similar to the previous study, our analysis showed a higher prevalence of headache in females (20.57%), compared with that in males (16.22%), while the headache prevalence in the present study was 18.71%, slightly lower than in the previous study. Besides, prevalence of headache peaked during adolescence. As shown in previous studies, headache prevalence among childhood or adolescence varied widely, from 5.9 to 82%, bringing high costs to individual and society [17][18][19] . It presented a gradual upward trend throughout childhood, peaking at about 11 to 13 years old [17][18] . And headache in children or adolescents can result in impaired psychosocial function, as a major factor associated with school absenteeism and poorer quality of life 18 . A consensus was reached that the natural pattern of headache may change through childhood to adulthood 18 .

Predictors of remission
Hormonal changes, different levels of stress exposure, and the life events and circumstances behind each stress exposure were the possible explanations for the evolution of headaches over time 18 . Moreover, those who had headaches in childhood or adolescence may be at a higher risk of getting additional physical or psychological problems in adulthood 19 . Given that headache in childhood or adolescents had a high prevalence, and caused various negative effects not only on childhood but also on adulthood, it is vital for us to understand the clinical feature and evolution of headache from childhood to adulthood.
The main target of this research was to investigate the impact of COVID-19 pandemic on headache patients in China. Interestingly, but not surprisingly, our analysis showed a reality of those headache participants with a tendency to improve or remit in headache duration, headache frequency and headache intensity during COVID-19 outbreak. Meanwhile, we inquire some headache patients, as well as friends and colleagues. In line with the present study, they reported a tendency to improve or remit in headache attacks during the social isolation. We also observed that, migraine was more likely to have remission in headache duration, compared with TTH and CH. Further work is required to demonstrated the underlying causes.
Previous studies demonstrated evidences that the risk of headache attacks decreased in holidays or days off 20 . Due to the war against COVID-19, we got a special long holiday. Then, why does the holiday give participants a remission in headache attacks? Several factors are particularly noteworthy to report. For instance, time off is important for workers or students to recover from load effects accumulated at work or study 21 . Less stress from work or study could be a factor related to headache remission during the social isolation. Additionally, previous studies reported not eating on time as a precipitating factor of headache attacks [22][23] . The underlying mechanism could be explained by causing relative hypoglycemia, further activating the adrenal medulla 21 . Eat regularly during the special holiday could be another factor related to headache remission. Then, sleep quality is an acknowledged factor affecting headache attacks. Our analysis showed that ISI scores were signi cantly lower in the remission group than in the aggravating group. Also, it should be particularly emphasized that getting support from family members during the social isolation were the independent factors of remission in headache duration. A previous study demonstrated that patients from functional families appeared to show relatively low levels of distress regardless of the severity of their headache 24 . COVID-19 outbroke when Spring Festival was approaching, the period for a reunion with family members in China. Families got together during the special holiday, and have more time to communicate and listen to each other. Similarly, with regard to headache participants, they got more opportunities to communicate their headache-related problems with family members and obtained more effective support, thus reduced the headache-related or mental burden. Given that, we call on more family supports involved to help headache patients improve or remit in headache attacks.
Besides getting support from family members, education level and smoking were also independent factors of remission in headache attacks. A previous study showed that remission was positively associated with increased educational level 25 . On the contrary, another study demonstrated that participants with higher education tended to having more headaches, especially migraine 26 . The present study found that participants with a junior colleague or lower degree were more likely to have remission in headache severity during COVID-19, compared with participants with a university degree. Smoking was a precipitating factor of headache attacks. Participants not smoking were more likely to have remission in headache duration.
Rather curiously multivariate logistic regression analyses showed that having relative or acquaintance got COVID-19 was a protective factor factors of remission in headache severity. By synthesizing the data, we found that the aggravating group presented signi cantly higher proportion of participants having relative or acquaintance got COVID-19, compared with the remission group and unchanged group, while the unchanged group presented a signi cantly lower proportion, compared with the remission group. And participants in the unchanged group greatly outnumber participants in the aggravating group ( Third, sampling error is unavoidable by using this online questionnaire, for example, illiterate, the elderly, children and those who cannot afford to buy a smartphone may have a less participation in the survey.
Fourth, some factors such as BMI, marriage status and regular diet were not assessed.

Conclusion
The present study provided the rst data about the impact of COVID-19 pandemic on headache patients in China. Adverse impacts of COVID-19-related events on headache patients, especially those living in highrisk area, cannot be overlooked, though participants presented an overall trend of headache remission during COVID-19. Smoking, education level and getting support from family members during the social isolation were the independent factors of remission in headache. We call on more family supports involved to help headache patients improve or remit in headache attacks. 2, Getting Unchanged or more support from family members vs Getting less support from family members; 3, Getting Unchanged or more support from the outside world vs Getting less support from family members.