Association between maternal psychological distress and children's neurodevelopment in offspring aged 4 years in Japan: The Tohoku Medical Megabank Project Birth and Three‐Generation Cohort Study

An association between maternal psychological distress and children's development has been reported, but reports from Japan are limited. This study aimed to examine the association of maternal psychological distress with children's neurodevelopment in Japan.

Regarding maternal psychiatric symptoms such as depression and anxiety, previous studies have reported an association between maternal psychiatric symptoms and children's development. [5][6][7][8][9][10][11][12] Two studies have found that maternal prenatal and post-natal depression were associated with children's development. 5,7 Another study shows an association between ongoing anxiety during the pre-natal and post-natal periods and children's development. 6 One study shows an association between maternal pre-natal and post-natal depression and children's cognitive and social-emotional development in the low-income population. 8 Furthermore, systematic review and meta-analysis have been conducted on this topic. 11,12 Although the association between maternal psychiatric symptoms in pre-natal and post-natal periods and children's development has been studied world-wide, there has been limited published research in Japan. To the best of our knowledge, only two studies have assessed this topic in Japan. 9,10 One study examined the association between maternal depression in the perinatal period and children's behavioural problems, 9 and the other examined the association between maternal depression in the post-natal period and children's language development. 10 The pre-natal and postnatal environment may differ among countries owing to cultural differences, medical settings and social settings. Thus, we considered that further studies on the relationship between maternal psychological distress and children's development were needed, focusing on the Japanese population.
Considering the above circumstances, this study focused on children's neurodevelopment. This study aimed to examine the association of maternal psychological distress in pre-natal and post-natal periods with children's neurodevelopment in the Japanese population.

Study design and population
This study is based on the Tohoku Medical Megabank Project Birth and Three-Generation Cohort Study (TMM BirThree Cohort Study). The details of the TMM BirThree Cohort Study are described elsewhere. [13][14][15][16] Between July 2013 and March 2017, pregnant women were recruited into the TMM BirThree Cohort Study at 50 obstetric clinics and hospitals in Miyagi and Iwate prefectures, Japan. All participants obtained explanations from trained genome medical research coordinators and signed consent. The exclusion criteria applied to 23 130 pairs of mothers and children were as follows: withdrawn informed consent (n = 505), multiple participation (n = 875), missing information on maternal psychological distress in early pregnancy (n = 720), missing information on maternal psychological distress in the 2 years postpartum (n = 8880), missing information on children's neurodevelopment at age 4 (n = 4344) and taking psychotropic drugs between pre-to early pregnancy (n = 160). In total, 7646 mother-child pairs were included in the analysis (Fig. 1).
The TMM BirThree Cohort Study protocol was reviewed and approved by the Ethics Committee of Tohoku University Tohoku Medical Megabank Organization (2013-1-103-1).

Maternal psychological distress
The Japanese version of the Kessler Psychological Distress Scale (K6) was used to evaluate maternal psychological distress in early pregnancy and 2 years postpartum. 17,18 The K6 is a short scale comprising six questions, developed by Kessler et al., 17 and the Japanese version was developed and validated by Furukawa et al. 18 A K6 score of 5 or higher was defined as psychological distress in this study. This cut-off score has been suggested to maximise the sum of specificity and sensitivity in the Japanese version of K6 in screening for mood and anxiety disorders. 19 Furthermore, we classified mothers and children pairs based on psychological distress into four categories in no psychological distress in both pre-natal and post-natal periods (none), only the pre-natal period (pre-natal only), only the post-natal periods (post-natal only) and both the pre-natal and post-natal periods (both).

Neurodevelopmental assessments
The Ages & Stages Questionnaires Third Edition (ASQ-3) was used to assess neurodevelopment in children. 20,21 The ASQ-3 can assess children from 1 month to 66 months of age. In this study, the parents responded to the Japanese translation of ASQ-3 at 4 years of age. 21 The ASQ-3 consists of six questions divided into five domains of communication: gross motor, fine motor, problem solving and personal and social skills. The questions are answered with 'yes' (=10), 'sometimes' (=5) or 'not yet' (=0), and each domain is scored on a range of 0-60. 20 In the case of missing one or two questions out of six, the remaining total score was adjusted from 0 to 60 by multiplying it by a correction coefficient of 1.2 or 1.5, respectively. 20 A score less than À2 standard deviation relative to the mean in reference indicates developmental delay, and further assessment is needed. 20

Confounders
We have selected covariates that may influence the association between maternal psychological distress and the children's neurodevelopment, referring to previous studies. [5][6][7][8][9][10][11][12] Information on a children's sex was garnered from birth records. Information about maternal age at delivery (years) and parity (never, one or more) was gathered from the medical record. We divided maternal age into four categories: <25, 25-29, 30-34 and ≥35 years. Information on maternal alcohol drinking (never, former and current), maternal cigarette smoking (never, stopped before pregnancy, stopped after pregnancy and current) and paternal smoking (never, stopped before pregnancy, stopped after pregnancy and current) were collected from the questionnaire during early pregnancy. Information on household income (<4 000 000, 4 000 000-5 999 999 and ≥6 000 000 Japanese yen/year) was gathered from the mid-pregnancy questionnaire. Maternal educational attainment data (high school graduate or less, junior college or vocational college graduate, university graduate or above and others) were collected from questionnaire in 1 year postpartum.

Statistical analysis
The participant characteristics according to the four categories of maternal psychological distress were examined. Since all variables were categorical variables, the data were presented as frequencies and percentages. The association between maternal psychological distress of four categories and five domains of developmental delay in children was evaluated using multivariable logistic regression analysis to estimate odds ratios (ORs) and 95% confidence intervals (CIs), and the category of none (no psychological distress in both pre-natal and post-natal periods) was used as a reference. Multiple imputations imputed incomplete confounders by chained equation. 20 Plausible synthetic values were generated from given exposure, outcome and the other confounders in the data. Twenty sets of quasi-complete data were analysed in the multivariate analyses independently and integrated the estimates. 22 In addition, as of the mothers and children pairs who agreed to participate in the TMM BirThree cohort study, only about 35% were included in our analysis, we performed a dropout analysis, that is, a comparison of the characteristics of the participants in the analysis with those of the excluded participants to examine external validity. The characteristics of the participants according to the two groups of participants included (n = 7646) and participants not included (n = 14 104) were described, and P values were obtained from χ 2 tests comparing participant characteristics. The variables used in the dropout analysis were as follows: maternal age at delivery, parity, household income, educational attainment, maternal alcohol drinking, maternal cigarette smoking, paternal cigarette smoking and children's sex.

Characteristics of the study population
The participant characteristics according to the four categories of maternal psychological distress are shown in Table 1. Among the 7646 mothers, 4256 (55.7%) were assessed to have no psychological distress in both pre-natal and post-natal periods, 1302 (17.0%) in pre-natal only, 863 (11.3%) in post-natal only and 1225 (16.0%) in both pre-natal and post-natal periods. The participants who had psychological distress in pre-natal or postpartum periods were characterised as being younger, having never given birth and having a lower household income. Among 7646 children, 303 (4.0%) had a developmental delay of ASQ-3 domain of communication, 328 (4.3%) of gross motor, 378 (4.9%) of fine motor, 292 (3.8%) of problem solving and 348 (4.6%) of personal-social.

Multivariable logistic regression analysis for maternal psychological distress and children's developmental delay
The association of maternal psychological distress in pre-natal and post-natal periods with children's neurodevelopment by multivariate logistic regression were shown in Table 2

Comparison of characteristics of included and not included participants
Results of dropout analysis comparing included and not included participant's characteristics are shown in Table 3. Mothers excluded from the analysis were younger, had lower incomes, were less educated, were more likely to be smokers and had higher rates of paternal smoking.

Discussion
This study examined the association between maternal pre-natal and post-natal psychological distress and children's neurodevelopment in the Japanese population. We found that maternal psychological distress in only the postpartum period and both pre-natal and postpartum periods were associated with developmental delays in communication, gross motor, fine motor, problem solving and personal-social. Only pre-natal maternal psychological distress was associated with the developmental delay of communication.
In this study, the OR values of children's developmental delay were highest in the maternal psychological distress in both prenatal and post-natal periods. This result is consistent with previous research showing that persistent depression symptom and anxiety in the pre-natal and post-natal periods are associated with children's developmental delay. [5][6][7] The possible mechanism by which the pre-natal psychological distress adversely affects the children's neurodevelopment is downregulation of 11β-HSD2 in the placenta. 23 Typically, placental 11β-hydroxysteroid dehydrogenase type 2 (11β-HSD2) protects the fetus from exposure to cortisol by inactivating cortisol. However, it has been reported that 11β-HSD2 in the placenta is downregulated by maternal anxiety. 23 The possible cause of the negative impact of postpartum maternal psychological distress on child development is a decreased quality of parenting. 24 Maternal pre-natal psychological distress was not associated with developmental delay of gross motor, fine motor, problem solving and personal-social in the children. The mechanisms are unclear, but the association of each domain may vary depending on whether the mother experienced psychological distress in prenatal period, post-natal period, or both pre-natal and post-natal periods. In other words, maternal psychological distress in prenatal or post-natal periods may be specifically associated with a particular phenotype of child developmental delay. Further research is needed to examine in detail which phenotypes of children's developmental delay are associated with pre-natal or postnatal psychological distress.
This study's results indicate the importance of continuous screening and early intervention for maternal psychological distress, and further research focusing on how to ameliorate maternal psychological distress. Because psychological distress in both the pre-natal and postpartum periods was most associated with the risk of developmental delay, it is important not to allow pre-natal psychological distress to continue into the postpartum period. The environments surrounding women and women's psychological states change drastically during pre-natal and post-natal periods. So, screening for psychological distress and care might be important regardless of whether postpartum or prepartum. However, considering the risk for the children's developmental delay, prenatal care for the mothers may be especially important.
This study had limitations. First, the TMM BirThree Cohort Study was conducted mainly in Miyagi Prefecture, limiting the generalisability of the findings. However, it has been shown that the prevalence of maternal postpartum depression, average children's birthweight, length and prevalence of low birthweight in the TMM BirThree Cohort Study are nearly in line with national averages, 14 making it unlikely that the characteristics of Miyagi Prefecture residents deviate significantly from the national characteristics of Japan. Second, of the mothers and children pairs who agreed to participate in the TMM BirThree cohort study, only about 35% were included in our analysis, which affect the external validity of our results. In fact, the dropout analysis confirmed differences in the characteristics of included and not included participants ( Table 3). The characteristics of the excluded mothers, younger and having lower income, were similar to mothers who experienced psychological distress in pre-natal and post-natal periods (Tables 1,3). Given these circumstances, the results of this study may underestimate the association between maternal psychological distress and children's neurodevelopment. Third, genetic and epigenetic factors were not accounted in this study. If developmental delay were observed in the children, it is possible that the genes causing the maternal psychological distress were inherited on to the children, leading to the developmental delay. Furthermore, pre-natal stress and anxiety have been reported to increase methylation of the child's glucocorticoid receptors, 24 therefore epigenetic pathways may be an intermediate factor in the association between maternal psychological distress and children's development. Fourth, the psychological distress screened for in the K6 is different from specific symptoms such as depression or anxiety, which limits the comparison between the present study and previous studies because most previous studies on this topic have focused on depression or anxiety. Fifth, the present study could not consider the efficacy of medication for psychological distress. Further studies are needed to examine how treatment of psychological distress might change the association with children's neurodevelopment. Sixth, since the ASQ-3 is a parent-completed questionnaire; if the mother completed who has psychological distress, the children's neurodevelopment may not be assessed correctly. Seventh, the possibility cannot be ruled out that the symptoms of children's developmental delay are apparent until around the age of 2 and that this is causing maternal post-natal psychological distress through parenting. Finally, we used ASQ-3 only at 4 years old. So, it is needed to study the children at later ages as the followup study progresses. The strength of this study was that the analysis was conducted in one of the largest sample size populations of any study examining the relationship between maternal psychological distress and child development. It was possible to adequately evaluate the association between maternal psychological distress and children's neurodevelopment.

Conclusions
In conclusion, maternal pre-natal and post-natal psychological distress was associated with a higher risk of children's developmental delay in the Japanese population. Furthermore, continuous psychological distress from the pre-natal to post-natal period was most strongly associated with a higher risk of children's neurodevelopmental delay.