株式会社ブレインスリープ (Brain Sleep Co., Ltd., Headquarters: Chiyoda-ku, Tokyo; Representative Director: Atsushi Hirota; hereinafter "Brain Sleep") and a research team led by Dr. Seiji Nishino (Chief Research Advisor at Brain Sleep), Professor of Psychiatry and Director of the Sleep and Circadian Neurobiology Laboratory at Stanford University School of Medicine, analyzed the results of an online survey conducted in January 2021 targeting approximately 10,000 workers across all 47 prefectures in Japan (allocated by gender, age, and prefecture). They investigated the relationship between sleep conditions and infection with the novel coronavirus disease (hereinafter "COVID-19"). The results revealed that the prevalence of sleep apnea syndrome (hereinafter "SAS") was 13.1 times higher in COVID-19 infected individuals compared to non-infected individuals. Conversely, individuals with a history of SAS had a 16.6 times higher risk of COVID-19 infection. We are pleased to announce that a research paper summarizing the findings of this study (lead authors: Masahiro Nakajima, Ryota Amano; corresponding author: Seiji Nishino) has been published in the international scientific journal "Scientific Reports" (published by Nature Research, UK). With the COVID-19 pandemic since 2020, the importance of sleep in boosting immunity and enhancing vaccine effectiveness has been re-recognized. While it is undeniable that improving sleep quality is crucial for enhancing immunity, suffering from sleep disorders also has a significant impact, leading to renewed attention on the importance of appropriate treatment for sleep disorders. Among sleep disorders, SAS is said to have a particularly high number of cases. Since the beginning of the COVID-19 pandemic, there have been international reports suggesting that SAS increases the risk of COVID-19 infection. For example, a study of 10 hospitals in the Chicago metropolitan area indicated that individuals with SAS as a comorbidity had approximately eight times higher risk of COVID-19 infection (References <1-3>). While the relationship between SAS and COVID-19 has been recognized since the outset, large-scale surveys targeting the general public have been rare, with most conventional reports focusing on hospitalized patients or data analysis from medical record systems. In this study, we conducted a large-scale online survey via the internet, targeting 10,339 members of the general public in Japan, particularly working individuals, to investigate the relationship between the presence or absence of SAS, sleep habits, and lifestyle habits, and the COVID-19 infection rate. The survey participants were allocated from a major online survey company's sample members to ensure even distribution by gender and age, and in some prefectures, considering population composition (see Figure 1). The survey content included: ① Questions related to individual attributes such as family composition and occupation ② Questions related to sleep habits and other sleep-related matters ③ Questions related to smoking habits ④ Questions related to stress and productivity, etc. ⑤ Questions related to health conditions since 2020 ⑥ Questions related to COVID-19 infection prevention The response data, consisting of a total of 140 variables from these question groups, were analyzed. Figure 1: Distribution of survey participants by gender, age, and prefecture In this study, among the 10,339 responses collected, 10,323 responses were considered valid after excluding 16 respondents whose answers regarding the number of cigarettes smoked per day or height and weight were judged as outliers. For determining the presence or absence of COVID-19 infection, respondents were asked to choose from three options ("yes," "no," "prefer not to answer") for the following six questions. Those who answered "yes" to either question ② or ③ were classified as COVID-19 positive (hereinafter "COVID-19+"), and those who answered "no" to all questions from ① to ⑥ were classified as COVID-19 negative (hereinafter "COVID-19-"). This determination method aimed to reduce the possibility of infected individuals being mixed in with non-infected individuals, providing a conservative estimate. The survey was conducted in February 2021, and vaccinations had not yet begun. 1 Suspected of having COVID-19 and had a medical consultation 2 Suspected of having COVID-19 and was hospitalized 3 Suspected of having COVID-19 and underwent hotel recuperation 4 Suspected of having COVID-19 and underwent home recuperation 5 Had self-isolation at home as a close contact of a COVID-19 patient 6 Family, relatives, friends, or colleagues were diagnosed with COVID-19 Among the 10,323 valid respondents, 144 were COVID-19+ and 8,693 were COVID-19-, making a total of 8,837 individuals the subjects of this study's analysis. For the presence or absence of SAS, individuals who answered "yes" to the question "Have you been diagnosed with SAS at a medical institution and received treatment?" were identified as SAS patients. Among the analyzed subjects, there were 282 SAS patients (hereinafter "SAS+") and 8,555 non-SAS patients (hereinafter "SAS-"). Table 1 shows the distribution of analyzed subjects by gender and age. It was confirmed that many COVID-19 infected individuals were in their 20s (younger age group) and more often male than female. SAS patients were prevalent not only among those aged 60 and above but also among those in their 20s. Among the total of 8,837 analyzed subjects, there were 282 SAS patients, resulting in a prevalence rate of 3.2%, which is consistent with previously reported SAS prevalence rates in Japan (Reference <4>). Furthermore, the higher proportion of COVID-19 infected individuals in their 20s (Reference <5>) and the higher proportion of SAS patients also in their 20s (Reference <6>) are consistent with other reports. Table 1: Distribution of 8,837 analyzed subjects by gender, age, COVID-19 infection status, and SAS prevalence status Cross-tabulation of COVID-19 and SAS revealed a strong association between COVID-19 infection status and SAS prevalence status. Among the total 8,837 analyzed subjects, 51 out of 144 COVID-19+ individuals (35.4%) were SAS+, while only 231 out of 8,693 COVID-19- individuals (2.7%) were SAS+. A chi-squared test showed a p-value well below 0.01, statistically confirming a significant relationship between the two. Similar results were obtained even when limiting the analysis to individuals in their 20s, who had a high number of COVID-19 cases (see Table 2). Table 2: Relationship between COVID-19 infection status and SAS prevalence status for all analyzed subjects and for individuals in their 20s *The proportion of COVID+ among SAS+ is 18.1% (=51/282), and the proportion of COVID-19+ among SAS- is 1.1% (=93/8,555). Therefore, individuals with a history of SAS infection were found to have a 16.6 times higher risk of COVID-19 infection (18.1/1.1=16.6 times). In this survey, a large number of questions totaling 140 variables were asked. To identify variables that significantly contribute to the presence or absence of COVID-19 infection among them, multivariate statistical analysis was also performed. The dependent variable was either the presence or absence of COVID-19 infection or the presence or absence of SAS. The analysis subjects were considered in two cases: all age groups and only individuals in their 20s. A total of four patterns of multivariate logistic regression analysis were conducted: 1 Dependent variable: presence or absence of COVID-19 infection; Analysis subjects: all age groups 2 Dependent variable: presence or absence of COVID-19 infection; Analysis subjects: only individuals in their 20s 3 Dependent variable: presence or absence of SAS; Analysis subjects: all age groups 4 Dependent variable: presence or absence of SAS; Analysis subjects: only individuals in their 20s For explanatory variables, after organizing the initial 140 variables by removing questions with redundant meanings and open-ended items, reducing them to 67 variables, approximately 10 variables were selected for each pattern based on statistically significant variables from univariate analysis and clinical judgment. The results for each pattern are shown in Tables 3-4. In Pattern ①, where the dependent variable was the presence or absence of COVID-19 infection, "not wearing a mask when going out," "influenza infection," and "SAS prevalence" were identified as explanatory variables with high odds ratios (hereinafter "OR"). Similarly, in Pattern ③, where SAS prevalence was the dependent variable, "COVID-19 infection" was identified as an explanatory variable with a high OR. Table 3: Each explanatory variable and statistical analysis results for Pattern ① (dependent variable: presence or absence of COVID-19 infection, analysis subjects: all age groups) (variables with OR of 5 or more are shown in orange, variables with p-value of 0.05 or more are shown in gray) Table 4: Each explanatory variable and statistical analysis results for Pattern ③ (dependent variable: presence or absence of SAS, analysis subjects: all age groups) (variables with OR of 5 or more are shown in orange, variables with p-value of 0.05 or more are shown in gray) Regarding the explanatory variables with high ORs, comparing the proportion of COVID-19 infected and non-infected individuals between those to whom each explanatory variable applies and those to whom it does not, it can be confirmed that the proportion of COVID-19 infected individuals is remarkably higher among the former. Among these, as mentioned above, the proportion of COVID-19 infection among SAS patients is 18.1%, which is high even when compared to other explanatory variables, indicating that COVID-19 infection is more widespread among SAS patients. Nearly identical results were obtained in the analysis limited to individuals in their 20s. Figure 2: Proportion of responses for major explanatory variables identified in Pattern ① for COVID-19 infected and non-infected individuals, respectively This study confirmed that SAS prevalence significantly increases the risk of COVID-19 (and influenza) infection at a statistically significant level. While the exact cause is not fully understood, previous research suggests the involvement of Angiotensin Converting Enzyme 2 (ACE-2) (Reference <7>), and it has also been pointed out that mouth breathing due to SAS may increase the risk of viral infection in the upper respiratory tract. With the continued prevalence of COVID-19 anticipated this winter, further research is expected to identify the cause, also from the perspective of preventing COVID-19 or influenza infection. ◆About the journal of publication: International scientific journal 'Scientific Reports' Title: Influences of sleep and lifestyle factors on the risk for covid-19 infections, from internet survey of 10,000 Japanese business workers URL: https://www.nature.com/articles/s41598-022-22105-3 Authors and Affiliations: Masahiro Nakajima1, Ryota Amano2, Naoya Nishino3, Yasutaka Osada1,4, Yuriko Watanabe1, Akifumi Miyake1, Shintaro Chiba5,6, Seiji Nishino1,3 1: Brain Sleep Research Institute (Brain Sleep Co., Ltd.) 2: Graduate School of Information Sciences, Applied Information Sciences, Tohoku University 3: Department of Psychiatry, Stanford University School of Medicine 4: Anti-Aging and Preventive Medicine, Nippon Medical School 5: Ota Sleep Science Center, Ota General Hospital Memorial Research Institute 6: Department of Otorhinolaryngology, Jikei University School of Medicine ◆References <1> Miller, M. A. & Cappuccio, F. P. A systematic review of COVID-19 and obstructive sleep apnoea. Sleep Med Rev 55, 101382 (2021) <2> A Maas, M. B., Kim, M., Malkani, R. G.,bbott, S. M. & Zee, P. C. Obstructive Sleep Apnea and Risk of COVID-19 Infection, Hospitalization and Respiratory Failure. Sleep Breath 25, 1155-1157 (2021) <3> Hariyanto, T. I. & Kurniawan, A. Obstructive sleep apnea (OSA) and outcomes from coronavirus disease 2019 (COVID-19)pneumonia: a systematic review and meta-analysis. Sleep Med 82, 47-53 (2021) <4> Okura, M. et al. Polysomnographic analysis of respiratory events during sleep in young nonobese Japanese adults without clinical complaints of sleep apnea. J Clin Sleep Med 16, 1303-1310 (2020) <5> Ministry of Health, Labour and Welfare "COVID-19 Treatment Guidelines, 3rd Edition" https://www.mhlw.go.jp/content/000668291.pdf <6> Nishijima, T. et al. Prevalence of sleep-disordered breathing in Japanese medical students based on type-3 out-of-center sleep test. Sleep Med 41, 9-14 (2018) <7> Barcelo, A. et al. Angiotensin converting enzyme in patients with sleep apnoea syndrome: plasma activity and gene polymorphisms. Eur Respir J 17, 728-732 (2001) Comment by Dr. Kentaro Iwata, Professor, Division of Infectious Diseases, Graduate School of Medicine, Kobe University This is a study using an online survey to examine the relationship between sleep apnea syndrome (SAS) patients and COVID-19 infection. The research topic is academically interesting and clinically valuable. The finding that COVID-19 infected groups had significantly more SAS patients is consistent with findings from previous research and is clinically understandable. Even after adjusting for gender, BMI, and other factors in multivariate logistic regression analysis, SAS was associated with COVID-19 infection (OR 7.34, 95% CI 3.44-2.36). Other factors included influenza infection, which should be interpreted not as influenza infection causing COVID-19, but rather as both infections sharing similar risk factors. Additionally, younger age and not wearing a mask were correlated with infection risk, which also aligns with findings from previous research. The 35.4% of COVID-19 infected individuals having SAS seems somewhat high, which might indicate a potential bias in the information collection method, as this was an online survey conducted by Brain Sleep. Similarly, it is possible that the COVID-19 negative group included individuals who were asymptomatic or mildly infected and unaware of their infection, and such individuals might have a lower rate of underlying medical conditions, which could have influenced the odds ratio. This possibility cannot be ruled out. It was also interesting that the BMI of the participants in this online survey was surprisingly low, suggesting that individuals with higher health awareness might have been more likely to respond to the survey. This impression differed slightly from the COVID-19 patients with SAS we see in the hospital ward. Previous research has already examined the impact of SAS on severe illness and mortality, and further research in this area is expected (JAMA Network Open. 2021 Nov10;4(11):e2134241.). ＜Dr. Kentaro Iwata: Profile＞ Graduated from Shimane Medical University (now Shimane University) in 1997. After serving as a resident physician at Okinawa Prefectural Chubu Hospital and an internist at St. Luke's-Roosevelt Hospital, Columbia University, he became an Infectious Disease Fellow at Beth Israel Medical Center, Albert Einstein College of Medicine. In 2003, he moved to China and worked at Beijing International SOS Clinic. Returned to Japan in 2004, served as Head of Infectious Diseases and Head of General Internal Medicine and Infectious Diseases at Kameda Medical Center (Chiba Prefecture). Has been in his current position since 2008.
Log in
