This study aimed to examine the relationship between the difficulties experienced by children with neurodevelopmental disorders in implementing infection prevention measures and the deterioration of their relationships with their caregivers and friends.

Study participants were recruited at Osaka University Hospital’s Center for Developmental Medicine and Child Psychiatry (Pediatric Development Outpatient Clinic) and the Kanazawa University Research Center for Child Mental Development. We recruited 150 caregivers of children with neurodevelopmental disorders attending elementary or middle school as of October 2020 (aged 6–15 years). The survey period was October–December 2020. The study was approved by the Osaka University Hospital’s international review board, and written consent was obtained from the parents.

With the University of Malaya’s cooperation, we created an original questionnaire focusing on the participants’ concerns and their families’ needs during the COVID-19 pandemic [21] [see Additional file 1]. In addition to sociodemographic data (including age, sex, diagnosis, school, and family members) and the circumstances of the child and family when schools were closed (parents’ work status, limitations imposed during the state of emergency declaration, presence or absence of infections, presence or absence of the child’s involvement with their friends, presence or absence of caregivers’ mental health concerns, and presence or absence of social support offered), the child’s relationships, behavioral problems, and difficulties implementing infection control measures were also evaluated.

The participants were asked about their child’s relationship with others at three time points: before the spread of infection (January 2020), when schools were closed (April 17‒May 7, 2020), and after schools reopened (June 2020).

The following questions were asked about their child’s relationships with friends, parents, and siblings: “How was your child’s relationship with parents and friends before the pandemic, during the state of emergency declaration, and after schools reopened?” The answers were rated on a five-point scale: “Good,” “Fairly good,” “Neither good nor poor,” “Fairly poor,” or “Poor.”

Questions about the child’s behavioral problems at three time points were evaluated to reveal the symptoms of neurodevelopmental disorders and the deterioration of secondary disorders. Fourteen problem-related items were established: hyperactivity, carelessness, social withdrawal, stereotyped speaking patterns, obsession, repetitive behaviors, sleep problems, rebellious behavior, food refusal, irritability, aggression, tantrums, self-injury, and school non-attendance. For each item, they were asked whether the behavior occurred before the pandemic and whether there was any worsening or new onset of these behavioral problems during the school closure period or after schools reopened. The participants were asked to choose items applicable to their children.

Questions about infection control were measured on the basis of infection countermeasures proposed in Japan as the “new normal.” These comprised six items: “Social distancing,” “Washing hands thoroughly after arriving home,” “Wearing a mask when going out,” “Taking one’s temperature and checking one’s health status every morning,” “Frequently airing out living spaces,” and “Voluntarily complying with stay-at-home requests.” The participants were asked to mention infection control measures that the child found challenging to implement since the spread of the infection. Multiple answers were accepted.

Statistical analysis was conducted using JMP Pro 15.2.0. We calculated the frequency and percentages of the child’s behavioral problems, deterioration in their relationships, and difficulty implementing infection control measures. The Wilcoxon signed-rank test was used to analyze changes in relationships with others from before the pandemic to during school closure, and from school closure to after school reopening.

To investigate the factors related to deterioration in the child’s relationship with parents or friends, the objective variable of changes in scores in the deterioration of relationships with parents or friends was used. The scores for relationships with parents or friends before the pandemic were considered as the baseline, from which the difference in the scores when schools were closed was calculated to determine “scores for deterioration in relationships with parents or friends while schools were closed.” Thereafter, the scores for relationship with parents or friends while schools were closed were considered the baseline, and the difference in these scores after schools had reopened was calculated to determine the deterioration in relationships after schools reopened. The more the relationship deteriorated, the higher the deterioration scores. The deterioration scores of relationships with others were considered as the objective variable, and for each of the six types, independent and multiple regression analyses were performed for the presence or absence of difficulty implementing infection control measures, presence or absence of caregivers’ mental health concerns, and the presence or absence of deterioration of one or more problematic behaviors. In the multiple regression analysis, adjustments were made for age and sex. All continuous variables used for analysis were centralized and analyzed.

We also analyzed using a sample without children who do not attend school to examine whether school non-attendance may affect the results of the relationships, because the amount of time spent with parents or friends would be affected by school non-attendance.