Impact of COVID-19 pandemic on the socioeconomic inequality of health behavior among Japanese adolescents: a two-year-repeated cross-sectional survey ===================================================================================================================================================== * Akira Kyan * Minoru Takakura ## Abstract **Background** Although disparities in socioeconomic status in health behaviors have been highlighted globally, they are not well understood in Japanese adolescents. The purpose of this study was to clarify the changes in socioeconomic disparities in adolescents’ fundamental health behaviors, such as physical activity, screen time (ST), sleep, breakfast intake, and bowel movement before and during COVID-19. **Methods** This was a repeated cross-sectional study which used data from the 2019 and 2021 National Sports-Life Survey of Children and Young in Japan. Data of 766 and 725 participants in 2019 and 2021, respectively, were analyzed. Favorable health behaviors were defined as daily moderate-to-vigorous physical activity (MVPA) of at least 60 minutes, ST of less than 2 hours, sleep of 8 to 10 hours, daily breakfast intake, and bowel movement frequency of at least once in every 3 days. We calculated the slope index of inequality (SII) and relative index of inequality (RII) in each health behavior for equivalent household income levels for assessing absolute and relative economic inequalities. **Results** Compliance with MVPA and ST recommendation significantly declined from 20.1% and 23.0% in 2019 to 11.7% and 14.9% in 2021, respectively. The SII and RII increased in MVPA for income levels, but decreased in daily breakfast in 2019 to 2021. Although the widening and narrowing of the disparity was inconclusive for ST, it exacerbated for the higher income groups. **Conclusions** Our study revealed widening of economic disparities in the achievement of recommended MVPA and narrowing of it in breakfast intake among adolescents before and during COVID-19. **What is already known on this subject** * Although disparities in the economic status of the family and neighborhood have been observed in several fundamental health behaviors among western countries’ adolescents, they have not been well understood in Japan. It has been pointed out that the economic disparities caused by the COVID-19 pandemic threaten to increase disparities in health and health behaviors. **What this study adds** * This is the first study to examine the change in socioeconomic inequalities in various health behavior among Japanese adolescents before and during the COVID-19 pandemic. * We found widening of socioeconomic disparities in the achievement of recommended PA * We found narrowing of socioeconomic disparities in breakfast intake. **How this study might affect research, practice or policy** * Continuous monitoring of how this affects health in the short and long term is needed. Keywords * Epidemiology * Health inequities * Household income * Physical Activity * Public health ## Introduction The issue of socioeconomic disparities of health in Japan has been pointed out only in the last 20 years, after the period of rapid economic growth. A systematic review of health disparities in Japan reports that the gradient of disparities is modest in the first decades since the 1990s compared with western countries [1]. In the last decade, as it worsened, the situation has been recognized as a public health concern [1]. Since 2013, correcting the situation has been one of the goals of Japan’s health policies [2]. In western countries, disparities by the economic status of the family and neighborhood have been observed in several fundamental health behaviors, such as physical activity (PA) [3], sedentary behavior (screen time (ST)) [4], sleep behavior [5], bowel movement frequency (constipation) [6], and food intake [7], among adolescents. In Japan, on the other hand, they are not well understood. This may be because it has only been a few decades since the health inequalities in Japan were first called to attention. In relatively recent very few examples of the inequality of health behavior, daily breakfast intake is higher in the high-income group [8], while socioeconomic disparities in food and nutrient intake have been restrained owing to serving of school lunches on weekdays in schools [9]. There are no studies on the relationship between socioeconomic status (SES) and exercise in Japanese adolescents, except for one inconclusive study in a local Japanese journal [10]. During the COVID-19 pandemic, the fundamental health behaviors have deteriorated among adolescents around the world [11,12]. Likewise, national surveys among Japanese adolescents reported that exercise, including active play, declined significantly, screen time of 2 hours or more per day increased significantly, and daily breakfast intake, adequate sleep of 8-10 hours, and the frequency of bowel movement remained almost unchanged [13,14]. The COVID-19 pandemic has not only disrupted healthcare, but also the lifestyle and social and economic activities of people. In Japan, measures to prevent the spread of the pandemic included policies to restrict people’s activities, such as closure of restaurants and restriction on large-scale events [15]. Thereby, the service industry, including the restaurants, and tourism industry, experienced a significant drop in sales, and many businesses were forced into unemployment and bankruptcy [15]. In contrast, the information and communication industry, including online businesses, saw an increase in sales [15]. Thereby, the accompanying widening of household income inequality became a concern [16]. The public health concern associated with it is that this may increase disparities in people’s health or health behaviors [17]. The possibility of increasing disparities in mental well-being [18] and recreational PA [19] among Japanese adults has been reported recently. Regarding dietary intake among Japanese adolescents, a recent retrospective study found that meat, fish, eggs, and vegetables intake worsened, especially in low-income households, during school closure (from 16 April to 6 May, 2020), while it mended after school lunch restarted [20]. However, no study has been conducted on other health behaviors in either Japan or the rest of the world [21]. Adolescent health behaviors have been documented to not only determine the physical and mental growth and development and health during adolescence, but also how these are carried over into the future [22]. It is also a right for all children to have a healthy life [22]. If they are suffering from inequality because of the home environment, which is beyond their control, they are victims. While some health behaviors, which are affected by the COVID-19 pandemic, should be recovered for adolescents, understanding how these behaviors vary with household income is important to form intervention strategies (providing necessary support to those who need it) and policies. Thus, this study aimed to clarify the trends in socioeconomic disparities in adolescents’ health behaviors, especially the fundamental health behaviors that constituted daily life, such as PA, ST, breakfast intake frequency, and bowel movement frequency before and during COVID-19. ## Methods ### Data source This study used data from the 2019 and 2021 National Sports-Life Survey of Children and Young conducted by the Sasakawa Sports Foundation (SSF) [14]. The survey is a nationwide cross-sectional survey that has been conducted every two years since 2001. From the 2017 survey, the target age groups were categorized into ages 4-11 and 12-21. The survey mainly measures exercise and sports participation of children and youth during after-school periods and holidays, and sports environments as well as health behaviors including sleep time, media usage time, and bowel movement frequency. Data are collected using self-administered questionnaires from adolescents and parents/guardians between June and July of each survey year. The study sample of each year was selected using the two-staged stratified random sampling method from 225 locations (2019: 61 in metro areas, 94 in cities with a population of over 100,000, 50 in cities with a population of less than 100,000, and 20 in towns and villages; 2021: 61 in metro areas, 96 in cities with a population of over 100,000, 50 in cities with a population of less than 100,000, and 18 in towns and villages), which were proportionally distributed from the strata by district/city size based on the population of the Basic Resident Register as of January 1st in the prior year of each survey year. The study area established at the time of the census in 2015 was used as the primary sampling unit at the time of the survey. The target sample size allocated to each location was 10 to 19. Survey points were extracted by the probability proportional sampling method, using the following formula to calculate the sampling interval for strata in which two or more survey points were assigned. #### Total number of people aged 4 to 21 years in the study area in the strata / Number of survey sites calculated in strata A total of 3,000 individuals participated in the survey. Detailed information on the survey methods can be found on the SSF website [14]. In this study, we used data from the surveys of young people aged 12 to 21 in 2019 and 2021, as all variables of interest were available. The number of responses for 2019 and 2021 were 1,675 (response rate 55.8%) and 1,663 (response rate 55.4%), respectively. The analysis included 12- to 18-year-olds, excluding 18-year-olds who were not attending high school. In the Japanese educational system, most students enrolled in high school are 15- to 18-year-olds. Of these, most 17- and 18-year-olds are enrolled in the third grade of high school and share school routines. This was the reason for including 18-year-olds attending high school in the analysis, while the age categories for the analysis of PA, ST, and sleep recommendations were up to age 17. After excluding individuals with missing variables of interest, data from 766 participants (359 boys and 407 girls) in 2019 and 725 participants (365 boys and 360 girls) in 2021 were used for analysis. Ethical approval was not required, as this study was a secondary analysis conducted using public datasets from the SSF that did not include identifiable personal information. ### Measures #### Physical activity PA was measured using the Japanese version of the Health Behaviour in School-aged Children survey (HBSC) questionnaire [23]. The HBSC questionnaire is widely used globally as the basis of moderate-to-vigorous PA (MVPA) surveillance of adolescents [24]. In the questionnaire, PA referred to any activity that increased the heart rate and made an individual feel out of breath for some time. PA can include sports, school activities, playing with friends, or walking to school. Examples of PA include running, brisk walking, rollerblading, biking, dancing, skateboarding, swimming, playing soccer, basketball, football, etc., and surfing. The question item was as follows: ‘Over the past 7 days, on how many days were you physically active for a total of at least 60 minutes per day?’. Participants were dichotomized into either active or inactive based on whether they achieved seven days per week of 60 minutes MVPA, according to the World Health Organization (WHO) guidelines on PA and sedentary behavior [25]. #### Screen time ST was assessed by asking about recreational TV/DVD viewing time and computer/game/smartphone usage on weekdays and weekends separately. These questions were formulated as follows: ‘For how many hours do you watch TV or DVD or use computer, video games (including TV, computer, and cellular device games, etc.), or smartphone per day outside of school and/or work?’. Possible answers to the question were ‘less than half-hour/day,’ ‘half-hour to 1 h/day,’ ‘1–2 h/day,’ ‘2–3 h/day,’ ‘3–4 h/day,’ ‘4–5 h/day,’ ‘more than 5 h/day,’ and ‘I don’t know’. The answer categories were recorded in minutes using the midpoint method. The ‘I don’t know’ option was deemed a missing case. Average minutes of daily ST were weighted by weekdays and weekends using the following formula: ![Formula][1] We then categorized the participants based on the cut-off of 2 h of recreational ST [25]. #### Sleep duration Sleep duration was assessed by questioning on bedtime and awakening hours on weekdays and weekends separately. Participants were asked to report the times they typically went to sleep at night and woke up in the morning on weekdays and weekends. Sleep duration was calculated by subtracting wake-up time from bedtime. Average minutes of daily sleep were weighted by weekdays and weekends using the following formula: ![Formula][2] The participants were grouped based on whether they had a sleep duration that was within the range (8.0–10.0 h/night) recommended by the National Sleep Foundation [26]. #### Breakfast Breakfast frequency was determined by asking the question ‘How often do you have breakfast per week?’; the possible responses were ‘almost every day’, ‘eating 4-5 days per week’, ‘eating 2-3 days per week’, and ‘very few’. Participants were dichotomized into almost every day or others based on the recommendation of the Ministry of Agriculture, Forestry and Fisheries, Japan [27]. #### Bowel movement frequency Bowel movement frequency was determined by asking the question ‘How often do you have bowel movement?’; the possible responses were ‘almost every day’, ‘once per two days’, ‘once per three days’, ‘less than once per three days’, and ‘irregularly’. We dichotomized participants into two groups: ‘almost every day to once every three days’ and ‘less than once every three days and irregular’, with reference to the definition of constipation [6]. While the definition of constipation ranges from self-reported constipation to clinical criteria regarding frequency of bowel movements and symptoms [6], one of the symptoms included in the Rome IV criteria is ‘two or fewer defecations in the toilet per week [6]. #### Household Income The parents/guardians of the participants were asked about their annual pre-tax household income using 11 options ranging from ‘no income’ to ‘10 million yen or more’. The ‘I don’t know’ option was deemed a missing case. The midpoint of each option was substituted for the household income. The equivalent household income was calculated by dividing household income by the square root of the number of members in the household [28]. Subsequently, the equivalent household income was categorized based on one-half of the median equivalent household income as a cut-off point: ‘less than 1.375 million yen’, ‘1.375 to less than 2.75 million yen’, and ‘2.75 million yen or more’[29]. One million yen was roughly ten thousand US dollars at the time of the survey. #### Confounding factors Place of residence, family structure, sex, age, sports participation, self-rated health, and preference of PA were used as covariates, and were considered as potential confounders [3,30]. The place of residence was categorized into two groups based on the population of the cities. Family structure was assessed using information regarding the people living in the home, coded as ‘living with both parents’ or ‘other.’ Sports participation was assessed by asking whether the participants were involved in extra-curricular exercise activity in school, sports clubs in local community, and private. Self-rated health was dichotomized into ‘good’ and ‘poor’. Preference for PA was dichotomized into ‘like’ and ‘dislike’. ### Statistical analysis Initially, the percentage of compliance with health behavior recommendation was estimated by each income level and each survey year. Cochran-Armitage test for trend was used to assess time trends in the percentage of compliance with each recommendation of health behavior and prevalence of constipation over time by each income level. Fisher’s exact test was performed to evaluate differences in compliance with each recommendation of health behavior and prevalence of constipation among different categories for each income level by each survey year. Next, socioeconomic inequalities in health behavior between low- and high-income groups were assessed by absolute and relative measures. Both absolute and relative measures were estimated with 95% confidence intervals (CIs) for income levels in each survey year. For the absolute measures, the slope index of inequality (SII) [31] was calculated using generalized linear models with binomial distribution and identity link function. The coefficient yielded an estimate of the absolute inequality. When this binomial model failed to converge, a generalized linear model with normal distribution and identity link function was used [32]. For the relative measures, the relative index of inequality (RII) [31] was calculated using generalized linear models with binomial distribution and log link function. The exponentiated coefficient yielded an estimate of the relative inequality. The SII and RII were estimated using ridit score for income levels as an independent variable. The SII and RII were summary measures of inequality as the changes in health behavior between the bottom and top points in the income hierarchy while accounting for the cumulative distribution in each income level [33]. Time trends of absolute and relative measures were assessed by the inclusion of the interaction term between income levels (ridit score for the SII and RII) and survey year [33]. In the model, survey year was treated as a continuous variable coded as 1 for 2019 and 2 for 2021 [33]. Wald test was used to test if the interaction term was statistically significant. Finally, these models were adjusted for confounding factors, such as place of residence, family structure, sex, age, sports participation, self-rated health, and preference of PA. To explore the association between the outcomes, when each health behavior was considered in the dependent variable as an outcome, other health behaviors were included in the independent variables. ## Results Table 1 shows the distribution of study participants by sociodemographic characteristics and survey years. There were significant relationships between survey year and MVPA and ST (P < 0.05). The number and percentage of missing each variable were shown in the e-Table (appendix). The missing percentages for each variable were similar in both surveys. Household income was missing by approximately 22% in each survey. This is because the “I don’t know” option was treated as a missing value. View this table: [Table 1.](http://medrxiv.org/content/early/2022/08/17/2022.08.11.22278499/T1) Table 1. Distribution of study participants by sociodemographic characteristics Table 2 shows the prevalence in compliance with each health behavior recommendation by income levels and survey years. Overall, although the compliance with PA and ST recommendation significantly declined from 20.1% and 23.0% in 2019 to 11.7% and 14.9% in 2021, respectively, the compliance with sleep and breakfast frequency recommendation and constipation prevalence remained unchanged between 2019 (63.2%, 83.6%, and 10.1%, respectively) and 2021 (64.7%, 83.3%, and 10.5%, respectively). While declining trends in PA were observed for all income levels, declining trends in ST were observed for only high-income level. Regarding breakfast consumption, no overall trend of change was observed between 2019 and 2021, but a declining trend was observed only at the high-income level, although it was not statistically significant. Table 2 also shows the results of Fisher’s exact test. A clear difference between 2019 and 2021 emerged in breakfast intake: the percentage of breakfast intake was lower in the low-income group and higher in the highest-income group in 2019 (P < 0.001), but the differences between the groups obscured in 2021 (P = 0.151). A similar trend was observed for ST, although it was not statistically significant (P in 2019 and 2021 were 0.097 and 0.863, respectively). On the other hand, PA showed differences by income levels in 2021, although it was not statistically significant (P = 0.063), but not in 2019 (P = 0.907). These differences by socioeconomic levels between the study periods were not found in compliance with sleep duration (P in 2019 = 0.916; P in 2021 = 0.356) and bowel frequency (P in 2019 = 0.503; P in 2021 = 0.265). View this table: [Table 2.](http://medrxiv.org/content/early/2022/08/17/2022.08.11.22278499/T2) Table 2. Prevalence of health behavior in 2019 and 2021 according to the equivalent household income Table 3 shows the SII and RII in each health behavior for the income levels by survey years. Significant increase and decrease trends of disparities were shown in the compliance of MVPA and daily breakfast intake. The SII in MVPA increased from −2.37% (95% CI 13.62– 8.87) in 2019 to 11.2% (95% CI 2.90– 19.50) in 2021. Although it was not statistically significant in the crude model (P for trends = 0.057), the adjusted model showed significant increase trends (P for trends = 0.032). The RII in MVPA increased from 0.86 (95% CI 0.43– 1.73) in 2019 to 3.03 (95% CI 1.16– 7.93) in 2021 (P for trends = 0.038). In contrast, the SII in breakfast intake decreased from 20.22% (95% CI 9.4– 31.04) in 2019 to 1.35% (95% CI −9.87– 12.57) in 2021 (P for trends = 0.018). The RII also decreased from 4.34 (95% CI 2.05– 9.18) in 2019 to 1.11 (95% CI 0.5–2.47) in 2021 (P for trends = 0.015). The SII in ST decreased from 20.72% (95% CI 7.71– 33.72) in 2019 to 4.74% (95% CI − 7.97– 17.45) in 2021, but it was not statistically significant (P for trends = 0.085). Similarly, for RII, the disparity decreased numerically but was not statistically significant. For sleep duration and bowel movement, no disparities or increase and decrease trends were observed for the survey years. Except the SII in PA, all adjusted models showed almost the same results as the crude models. View this table: [Table 3.](http://medrxiv.org/content/early/2022/08/17/2022.08.11.22278499/T3) Table 3. Slope index of inequality (SII) and relative index of inequality (RII) in health behavior according to the income levels by survey year ## Discussion This is the first study to examine the time trend in socioeconomic inequalities in various health behaviors among the Japanese adolescents before and during the COVID-19 pandemic. We found widening of socioeconomic disparities in the achievement of recommended PA and narrowing of socioeconomic disparities in breakfast intake among adolescents. For MVPA particularly, while all income groups showed deterioration during the COVID-19 pandemic, the trend was more prominent in the lowest-income groups. Breakfast intake frequency showed an improving trend among lower income groups, while it was worsening among higher-income groups. For ST, the widening and narrowing of the disparity was inconclusive. However, it exacerbated in the higher income groups. Sleep duration and bowel movement frequency did not change from pre-COVID-19 time to the present, and no socioeconomic disparities were observed. The percentage of the present study participants achieving the recommended sleep duration was poor at approximately 35%, while other countries achieved roughly 65% [34]. The prevalence of constipation at about 10% was not relatively high [6]. The association between SES and PA observed in other countries [3] has not been reported in Japan [10]. This was probably due to having extracurricular activities as a social system in Japan that allowed many adolescents to participate in sports without spending money. Japan has a system of extracurricular activities in which teachers are engaged in sports coaching from junior high school through high school [35]. In most cases, the cost of participation in extracurricular activities is small, with minimal expenses for the essential equipment. The course of study stipulate a policy for extracurricular activities [36] and almost all schools have it. This is a unique initiative worldwide [35]. According to national comprehensive survey data, as of 2019, the percentage of second-year middle school students enrolled in sports/exercise programs was as high as 66.5% [13], which meant that it played an important role as an equitable opportunity for PA for the Japanese adolescents. The decline in PA among Japanese adolescents during the COVID-19 pandemic was probably the result of the prohibition of that activity due to the pandemic. The widening socioeconomic disparity in PA may be due, in part, to the dependency of accessibility of sports activities on the SES of the family. The local government took measures, such as banning extracurricular activities, at the municipal level to prevent the spread of COVID-19, and the decision to exercise was up to the individuals. Meanwhile, paid sports organizations continued to provide certain services, such as online lectures. The association between paid sports activities and household income has been pointed out before the pandemic [37]. That is, it can be presumed that a larger proportion of adolescents from families with sufficient household income were engaged in paid sports and benefited from them, while a larger proportion of those from low-income families, who were not engaged in paid sports, were inactive. Further research is needed to clarify these details. Socioeconomic disparities in compliance with breakfast intake recommendation have narrowed: the percentage of lower-income groups has improved, while that of higher-income groups has declined. The reason for the decrease in the percentage of higher income groups is not clear, but it is possible that the overall increase in people’s health awareness due to the pandemic [38] has led to an improvement in the percentage of low-income groups consuming breakfast. Assuming that the acquisition of knowledge about diet causes behavioral change [39], the change to have regular breakfast, irrespective of its content, may be easier than changing other health behaviors. This is because food resources are also abundant in Japan. A meta-analysis in high-income Western countries found that young people from lower socioeconomic backgrounds have higher levels of ST compared with those from higher socioeconomic backgrounds [4]. According to our results, in 2019, before the COVID-19 pandemic, the trend in Japan was similar to that in other countries. It can be seen that ST has also worsened for higher income groups due to the pandemic. Although the economic disparity can be interpreted to have been eliminated, the overall deterioration is an urgent public health issue. It is plausible to assume that the lifestyles of the higher income groups are changing since the restrictions on going out have led people, at all income levels, to spend more time at home. As mentioned above, higher income groups are more likely to participate in paid private tutoring and sports activities [37]. It is assumed that adolescents from higher-income families had more ST at home due to the COVID-19 pandemic. Similar changes may be occurring around the world, but no research has yet elucidated it, except for our study. It should be urgently verified. ### Limitations This study had some limitations. Because this is not a longitudinal study, individual trajectory is unknown. Although the sampling method of the present study is appropriate to ensure the representativeness of the data, eligible samples included in the analysis were 45.7% and 43.6% of all respondents in 2019 and 2021, respectively. However, the proportion of population in each district/city based on their sizes was similar to that of the nation as a whole. Therefore, the representativeness of the data would not deviate considerably. Another limitation is that all data are from self-reported questionnaires, which cannot deny the possibility of including biases such as social desirability. However, the question items used in this study were almost identical to those used in large international or national surveys. For example, PA, ST, sleep, and breakfast frequency were similar to those used in the HBSC [24] and Japanese comprehensive surveys [13]. Thus, these items have sufficient validity as questionnaire to capture the health behaviors of adolescents. The number of the missing value of income was relatively large, but the pattern of missing values was similar in each survey year. This may not be a critical issue in discussing trends in socioeconomic disparities. It also has been pointed out that disparities in these health behaviors have been recognized for other SES indicators, such as parental education level, jobs, and psychosocial deprivation. Impacts to health behaviors may vary by SES indicators [3]. Future studies need to examine disparities in other SES indicators to identify needs for support. Although a number of factors related to each health behavior were taken into account, dietary intake related to bowel movement frequency was only taken into account for breakfast intake frequency due to a lack of data. No socioeconomic disparities in bowel movement frequency have been detected previously in Japan; our results are in line with previous studies [30]. Given that bowel movement frequency is related to dietary intake, school lunches may also contribute to reducing the socioeconomic inequality in bowel movement. To clarify this, future studies need to examine dietary intake and bowel movement simultaneously. ## Conclusion In conclusion, our study revealed that economic disparities in the achievement of recommended PA widened, while it narrowed for breakfast intake among adolescents before and during COVID-19. The fact that we found socioeconomic disparities that were insignificant before the pandemic confronted us with the need for urgent supports and medium- and long-term monitoring. Possible supports include financial assistance to low-income populations, interventions that encourage more effective health behavior change among them, etc. Because of the relatively mild restrictions on behavior in Japan compared to other countries (e.g. no lockdowns are imposed) [12], a severer situation is feared in other countries. Indeed, a recent ecological study among Scottish adolescents reported on the cautionary on a possible widening socioeconomic inequality in PA and ST [40]. Further individual-level research is necessary to verify this. ## Supporting information e-Table [[supplements/278499_file02.docx]](pending:yes) ## Data Availability All data produced are available online at [https://www.ssf.or.jp/thinktank/sports\_life/application/index.html](https://www.ssf.or.jp/thinktank/sports_life/application/index.html) ## Author statements ## Ethical approval Ethical approval was not required, as this study was a secondary analysis conducted using public datasets from the SSF that did not include identifiable personal information. ## Funding This study was supported by the Grant-in-Aid for Scientific Research (JSPS KAKENHI Grant Number 20K10473 and 20K13966) from the Japan Society for the Promotion of Science. ## Competing interests The authors declare that they have no competing interests. ## Authors contribution All authors contributed to the concept or design of the study and the acquisition, analysis, or interpretation of data for the work. A.K. drafted the manuscript. M.T. critically revised the manuscript. All authors gave final approval and agreed to be accountable for all aspects of work, thus ensuring integrity and accuracy. ## Acknowledgements We would like to thank Editage ([www.editage.com](http://www.editage.com)) for English language editing. ## Footnotes * **Telephone numbers** A.K.: +81-78-803-7721 M.T.: +81-98-895-1255 * **Email addresses** A.K.: a-kyan{at}pony.kobe-u.ac.jp M.T.: minoru{at}med.u-ryukyu.ac.jp * Received August 11, 2022. * Revision received August 11, 2022. * Accepted August 17, 2022. * © 2022, Posted by Cold Spring Harbor Laboratory This pre-print is available under a Creative Commons License (Attribution 4.0 International), CC BY 4.0, as described at [http://creativecommons.org/licenses/by/4.0/](http://creativecommons.org/licenses/by/4.0/) ## References 1. Ikeda N, Saito E, Kondo N, et al. What has made the population of Japan healthy? Lancet 2011;378:1094–105. doi:10.1016/S0140-6736(11)61055-6 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S0140-6736(11)61055-6&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21885105&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.11.22278499.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000295548900026&link_type=ISI) 2. Ministry of Health L and W. Healthy Japan 21 (the second term). 2018.[https://www.nibiohn.go.jp/eiken/kenkounippon21/en/](https://www.nibiohn.go.jp/eiken/kenkounippon21/en/) (accessed 23 Jun 2022). 3. O’Donoghue G, Kennedy A, Puggina A, et al. Socio-economic determinants of physical activity across the life course: a ‘DEterminants of DIet and Physical ACtivity’ (DEDIPAC) umbrella literature review. PLoS One 2018;13:e0190737. doi:10.1371/JOURNAL.PONE.0190737 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/JOURNAL.PONE.0190737&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.11.22278499.atom) 4. Mielke GI, Brown WJ, Nunes BP, et al. Socioeconomic Correlates of Sedentary Behavior in Adolescents: Systematic Review and Meta-Analysis. Sport Med 2017;47:61–75. doi:10.1007/S40279-016-0555-4/FIGURES/5 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/S40279-016-0555-4/FIGURES/5&link_type=DOI) 5. Sosso FE, Khoury T. Socioeconomic status and sleep disturbances among pediatric population: a continental systematic review of empirical research. Sleep Sci 2021;14:245. doi:10.5935/1984-0063.20200082 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.5935/1984-0063.20200082&link_type=DOI) 6. Mugie SM, Benninga MA, Di Lorenzo C. Epidemiology of constipation in children and adults: A systematic review. Best Pract Res Clin Gastroenterol 2011;25:3–18. doi:10.1016/j.bpg.2010.12.010 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.bpg.2010.12.010&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21382575&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.11.22278499.atom) 7. Cislak A, Safron M, Pratt M, et al. Family-related predictors of body weight and weight-related behaviours among children and adolescents: a systematic umbrella review. Child Care Health Dev 2012;38:321–31. doi:10.1111/J.1365-2214.2011.01285.X [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/j.1365-2214.2011.01285.x&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21752064&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.11.22278499.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000302352300003&link_type=ISI) 8. Okamoto S. Parental socioeconomic status and adolescent health in Japan. Sci Rep 2021;11:12089. doi:10.1038/s41598-021-91715-0 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41598-021-91715-0&link_type=DOI) 9. Murayama N, Ishida H, Yamamoto T, et al. Household income is associated with food and nutrient intake in Japanese schoolchildren, especially on days without school lunch. Public Health Nutr 2017;20:2946–58. doi:10.1017/S1368980017001100 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1017/S1368980017001100&link_type=DOI) 10. Yamakita M, Sato M, Ando D, et al. Associations between parent’s educational levels and children’s time spent in exercise. Res Exerc Epidemiol 2017;19:36–43. doi:10.24804/ree.19.36 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.24804/ree.19.36&link_type=DOI) 11. Farello G, Di Lucia A, Fioravanti B, et al. Analysis of the impact of COVID-19 pandemic on functional gastrointestinal disorders among paediatric population. Eur Rev Med Pharmacol Sci 2021;25:5836–42. doi:10.26355/EURREV\_202109\_26802 [CrossRef](http://medrxiv.org/lookup/external-ref?access\_num=10.26355/EURREV_202109_26802&link_type=DOI) 12. Kharel M, Sakamoto JL, Carandang RR, et al. Impact of COVID-19 pandemic lockdown on movement behaviours of children and adolescents: a systematic review. BMJ Glob Heal 2022;7:e007190. doi:10.1136/BMJGH-2021-007190 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1136/BMJGH-2021-007190&link_type=DOI) 13. Japan sports agency. National survey of the physical strength, exercise ability and exercise habits. 2021.[https://www.mext.go.jp/sports/b\_menu/toukei/kodomo/zencyo/1368222.htm](https://www.mext.go.jp/sports/b_menu/toukei/kodomo/zencyo/1368222.htm) (accessed 24 Jun 2022). 14. Sasakawa Sports Foundation. Publications. 2021.[https://www.ssf.or.jp/en/publications/](https://www.ssf.or.jp/en/publications/) (accessed 21 Jul 2022). 15. Statistics Bureau of Japan. Summary of the Results of Monthly Survey on Service Industries. 2018.[https://www.stat.go.jp/english/data/mssi/kekka/index.html](https://www.stat.go.jp/english/data/mssi/kekka/index.html) (accessed 22 Sep 2021). 16. Cabinet office. 2021 Analysis of the Survey of Children’s Living Conditions Report. 2022.[https://www8.cao.go.jp/kodomonohinkon/chousa/r03/pdf-index.html](https://www8.cao.go.jp/kodomonohinkon/chousa/r03/pdf-index.html) (accessed 23 Jun 2022). 17. Ahmed F, Ahmed N, Pissarides C, et al. Why inequality could spread COVID-19. Lancet Public Heal 2020;5:e240. doi:10.1016/S2468-2667(20)30085-2 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S2468-2667(20)30085-2&link_type=DOI) 18. Kanbayashi H, Hommerich C, Sudo N. Impact of COVID-19 Pandemic on Household Income and Mental Well-Being: Sociol Theory Methods 2021;36:259–77. doi:10.11218/ojjams.36.259 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.11218/ojjams.36.259&link_type=DOI) 19. Kyan A, Takakura M. Socio-economic inequalities in physical activity among Japanese adults during the COVID-19 pandemic. Public Health 2022;207:7–13. doi:10.1016/j.puhe.2022.03.006 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.puhe.2022.03.006&link_type=DOI) 20. Horikawa C, Murayama N, Kojima Y, et al. Changes in Selected Food Groups Consumption and Quality of Meals in Japanese School Children during the COVID-19 Pandemic. Nutrients 2021;13:2743. doi:10.3390/nu13082743 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3390/nu13082743&link_type=DOI) 21. López-Bueno R, López-Sánchez GF, Casajús JA, et al. Potential health-related behaviors for pre-school and school-aged children during COVID-19 lockdown: A narrative review. Prev Med (Baltim) 2021;143:106349. doi:10.1016/j.ypmed.2020.106349 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.ypmed.2020.106349&link_type=DOI) 22. Patton GC, Sawyer SM, Santelli JS, et al. Our future: a Lancet commission on adolescent health and wellbeing. Lancet 2016;387:2423–78. doi:10.1016/S0140-6736(16)00579-1 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S0140-6736(16)00579-1&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=27174304&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.11.22278499.atom) 23. Tanaka C, Kyan A, Takakura M, et al. Validation of the Physical Activity Questions in the World Health Organization Health Behavior in School-Aged Children Survey Using Accelerometer Data in Japanese Children and Adolescents. J Phys Act Heal 2021;18:151–6. doi:10.1123/jpah.2019-0671 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1123/jpah.2019-0671&link_type=DOI) 24. Inchley J, Currie D, Cosma A, et al. Health Behaviour in School-aged Children (HBSC) study protocol: background, methodology and mandatory items for the 2017/18 survey. 2018. 25. World Health Organization. WHO guidelines on physical activity and sedentary behaviour. 2020.[https://www.who.int/publications/i/item/9789240015128](https://www.who.int/publications/i/item/9789240015128) (accessed 10 Sep 2021). 26. Hirshkowitz M, Whiton K, Albert SM, et al. National Sleep Foundation’s sleep time duration recommendations: methodology and results summary. 2015. doi:10.1016/j.sleh.2014.12.010 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.sleh.2014.12.010&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=29073412&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.11.22278499.atom) 27. Ministry of Agriculture F and F. What Are the Benefits of Shokuiku (Food and Nutrition Education)? 2018.[https://www.maff.go.jp/j/syokuiku/evidence/index.html](https://www.maff.go.jp/j/syokuiku/evidence/index.html) (accessed 9 Jul 2022). 28. OECD. What are equivalence scales? 2011.[http://www.oecd.org/els/soc/OECD-Note-EquivalenceScales.pdf](http://www.oecd.org/els/soc/OECD-Note-EquivalenceScales.pdf) (accessed 10 Sep 2021). 29. Keppel K, Pamuk E, Lynch J, et al. Methodological Issues in Measuring Health Disparities. Vital Health Stat 2 2005;:1–16./pmc/articles/PMC3681823/ (accessed 10 Sep 2021). 30. Yamada M, Sekine M, Tatsuse T. Psychological Stress, Family Environment, and Constipation in Japanese Children: The Toyama Birth Cohort Study. J Epidemiol 2019;29:220–6. doi:10.2188/jea.JE20180016 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.2188/jea.JE20180016&link_type=DOI) 31. Mackenbach JP, Kunst AE. Measuring the magnitude of socio-economic inequalities in health: An overview of available measures illustrated with two examples from Europe. Soc Sci Med 1997;44:757–71. doi:10.1016/S0277-9536(96)00073-1 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S0277-9536(96)00073-1&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=9080560&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.11.22278499.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=A1997WM16300004&link_type=ISI) 32. Naimi AI, Whitcomb BW. Estimating Risk Ratios and Risk Differences Using Regression. Am J Epidemiol 2020;189:508–10. doi:10.1093/aje/kwaa044 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/aje/kwaa044&link_type=DOI) 33. Wachtler B, Hoebel J, Lampert T. Trends in socioeconomic inequalities in self-rated health in Germany: a time-trend analysis of repeated cross-sectional health surveys between 2003 and 2012. BMJ Open 2019;9:e030216. doi:10.1136/bmjopen-2019-030216 [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiYm1qb3BlbiI7czo1OiJyZXNpZCI7czoxMToiOS85L2UwMzAyMTYiO3M6NDoiYXRvbSI7czo1MDoiL21lZHJ4aXYvZWFybHkvMjAyMi8wOC8xNy8yMDIyLjA4LjExLjIyMjc4NDk5LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 34. Rollo S, Antsygina O, Tremblay MS. The whole day matters: Understanding 24-hour movement guideline adherence and relationships with health indicators across the lifespan. J Sport Heal Sci 2020;9:493–510. doi:10.1016/j.jshs.2020.07.004 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.jshs.2020.07.004&link_type=DOI) 35. Nakazawa A. Seeing sports as educational activities: a postwar history of extracurricular sports activities in Japan. Hitotsubashi J Soc Stud 2014;45:1–14. 36. Japan Sports Agency. Course of Study. 2018.[https://www.mext.go.jp/sports/b\_menu/sports/mcatetop04/list/1398875.htm](https://www.mext.go.jp/sports/b_menu/sports/mcatetop04/list/1398875.htm) (accessed 11 Jul 2022). 37. Benesse Education Research Institute. Third Annual Survey of Out-of-School Educational Activities 2017. 2017.[https://berd.benesse.jp/shotouchutou/research/detail1.php?id=5210](https://berd.benesse.jp/shotouchutou/research/detail1.php?id=5210) (accessed 11 Jul 2022). 38. Japan Sports Agency. Survey research on the public’s participation in sports, changes in awareness, and health status due to the outbreak of new coronavirus infection. 2019.[https://www.mext.go.jp/sports/b\_menu/sports/mcatetop01/list/detail/jsa\_00003.html](https://www.mext.go.jp/sports/b_menu/sports/mcatetop01/list/detail/jsa_00003.html) (accessed 11 Jul 2022). 39. Asakura K, Todoriki H, Sasaki S. Relationship between nutrition knowledge and dietary intake among primary school children in Japan: Combined effect of children’s and their guardians’ knowledge. J Epidemiol 2017;27:483–91. doi:10.1016/J.JE.2016.09.014 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/J.JE.2016.09.014&link_type=DOI) 40. Bardid F, Tomaz SA, Johnstone A, et al. Results from Scotland’s 2021 report card on physical activity and Health for children and youth: Grades, secular trends, and socio-economic inequalities. J Exerc Sci Fit Published Online First: 19 July 2022. doi:10.1016/j.jesf.2022.07.002 [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.jesf.2022.07.002&link_type=DOI) [1]: /embed/graphic-1.gif [2]: /embed/graphic-2.gif