All methods below were carried out in accordance with relevant guidelines and regulations.
Ethical approval was obtained from the Graduate School of Medicine and Faculty of Medicine, The University of Tokyo (10057), the research ethics committees of the Tokyo Metropolitan Institute of Medical Science (Approval number: 12-35), and SOKENDAI (The Graduate University for Advanced Studies) (2012002). Informed consent was obtained from both children and their parents, and written informed consent was obtained from the parents.
Study procedures and participants
We used the data from the Tokyo Early Adolescence Survey (T-EAS), a population-based cross-sectional survey conducted in three municipalities in Tokyo (Setagaya, Mitaka, and Chofu) . The T-EAS was designed to function as a baseline survey for the Tokyo Teen Cohort study (TTC), a longitudinal cohort study for which the fourth wave of data collection is underway (http://ttcp.umin.jp/). The T-EAS was conducted from October 2012 to January 2015, and participants were recruited by the Basic Resident Register in the municipalities. Children aged 10 years born between September 1, 2002, and August 31, 2004 and lived in the municipalities during the survey period and their parents (mothers; 98.5%) were invited to join in the T-EAS. Among 10,234 potential child-parent pairs, 5,756 pairs refused to participate in the T-EAS. Finally, 4,478 children were included, together with their parents (response rate: 43.8%) .
Trained investigators conducted the survey over the course of two visits. At the first visit, we obtained written informed consent from the parents and assent from the children to participate in the T-EAS. The child-parent pairs were required to answer questionnaires separately by the second visit. There were columns on the questionnaire for the parent to copy records from the Maternal and Child Health Handbook (MCH Handbook). The MCH Handbook is distributed to almost all (>98%)  pregnant women in Japan by local municipalities to record maternal and child health information during pregnancy, the perinatal period, and after-birth . At the second visit, the participating child-parent pairs were asked to complete the other self-reported questionnaires separately. These questionnaires pertained to sensitive contents such as mental health problems and placed in an envelope immediately after completion. During the second visit, the investigator also performed physical measurements and psychological examinations.
Children were excluded from the analyses if they needed special care for urological or neurological deficits other than NE or had taken medication for these conditions in the 2 weeks prior to participation (n = 44). Of the 4,434 participants, the data from 90 were not used for the final logistic regression analysis due to missing values. Finally, the data of 4,344 child-parent pairs were available for analysis. Demographic characteristics such as age and estimated intelligence quotient (IQ) (bilateral t-test), sex prevalence and parental educational background (chi-square test) did not significantly differ between included and excluded children (p>.05).
Delay in developmental milestones
To collect data regarding developmental delays, we created columns for copying the MCH Handbook records in the T-EAS questionnaire for the parents. The developmental milestones studied were as follows: ‘Holding their head up’ (at 3 months of age), ‘Rolling over’ and ‘sitting’ (6-7 months), ‘Crawling’ and ‘standing’ (9-10 months), ‘Walking with support’ and ‘Understanding some words’ (12 months), ‘Walking independently’ and ‘Speaking a few words’ (18 months), ‘Talking in two-word sentences’ (24 months), and ‘Drawing a circle’ and ‘Saying their name’ (36 months). We asked the parents, for instance, “Did your child hold his/her head up at 3 months old?”. In their answer, the parents would refer to the corresponding column in the MCH Handbook. Answer options were Yes (0) and No (1), and dichotomous variables for each delay were created according to the answers.
We also counted each participant’s total number of developmental milestone delays described above (range: 0-12), as an indicator of severity and consistency of children's developmental delay. Participants without any information about their developmental milestones (n = 31 (0.7%)) were excluded from this procedure.
The participating parents were asked about their children’s NE with the following question: “Does your child currently wet the bed at night?”, which was scored on a 5-point scale: (1) never, (2) sometimes, (3) once or twice a week, (4) three or four times a week, and (5) using diapers. We classified the answers into three groups. Children with answer (1) were assigned to the ‘none’ group, children with (2) to ‘infrequent wetting,’ and children with (3-5) to ‘frequent wetting’ since they could meet the diagnostic criteria for enuresis according to the DSM-5 .
Other potentially confounding factors
Sex, age, child’s intelligence quotient (IQ), low birth weight, and parents’ educational backgrounds were included in the analyses as potential confounding factors [31–36]. IQ was estimated using a short form of the Wechsler Intelligence Scale for Children (WISC-III), consisting of two subsets (Information and Picture Completion) . A formula for estimating the child’s IQ (estimated IQ = 3.3 × Information + 2.0 × Picture Completion + 45.6) was obtained from the results of the full version of the WISC-III performed by clinical psychologists in 28 children, 1 year after participation in the T-EAS. The IQ calculated by the formula explained 78% of the variance in IQ from the full version. Birth weights were collected from the MCH Handbook. We created a dichotomous variable for low birth weight (<2,500 g), in line with the 10th Version of the International Classification of Diseases . The parents’ educational backgrounds were assessed according to their highest levels of education, scored on a 6-point scale : (1) Junior high school (or lower), (2) high school (leaving before graduation), (3) high school (graduated), (4) vocational school or 2-year college, (5) 4-year college or university, and (6) postgraduate education (or higher).
Following the demographic statistics calculation, we created a cross-tabulation table for developmental milestones in infancy and NE status in early adolescence (none, infrequent wetting, and frequent wetting). Chi-square tests were conducted to examine distribution differences. Subsequently, the total number of developmental delays recorded in the MCH Handbook was tabulated for each participant. We examined whether the distribution and average number of developmental delays differed among NE groups using the chi-square test and analysis of variance with a post hoc Bonferroni procedure. Finally, multinomial logistic analyses were performed, which employed developmental delay as an independent variable and NE as a dependent variable using the None group as reference. We created two models. In the crude model, the odds ratio (OR) of each developmental delay was calculated without adjustment. In the adjusted model, we adjusted for sex, age, estimated IQ, low birth weight, and parents’ educational background. All variables were simultaneously forced into the model, and multi-collinearity not indicated (VIF<1.5). The Statistical Package for Social Science version 24 was used for all statistical analyses.