Cumulative exposure to maternal psychological distress in the prenatal and postnatal periods and atopic dermatitis in children: findings from the TMM BirThree Cohort Study

Atopic dermatitis (AD) is a common chronic inflammatory skin disorder characterized by clinical features with remission and relapse of itch and eczematous lesions [1]. AD is likely to have negative effects on sleep, discrimination [1], and complexity of family relationships [2, 3]. Moreover, AD is often comorbid with other atopic diseases such as asthma, allergic rhinitis, wheeze, and food allergy [1]. AD is often observed in children under the age of 5 years [4], and the prevalence tends to be higher than at older ages [5]. Up to approximately 20% of children are affected by AD worldwide [1], and the national cohort study in Japan found that 15.3% of 2-year-old children had AD [5]. Considering these circumstances, reducing the risk of developing AD in children is a public health issue.

Maternal mental health problems have been examined as possible risk factors for AD in children [4]. Twelve studies have examined the association between prenatal maternal mental health problems alone and AD in children [6‐17], and 10 of them showed that a maternal mental health problem was associated with an increased risk of AD in children [7‐11, 13‐17]. Recent studies have shown that maternal cortisol is transmitted to the fetus through the placenta. This leads to changes in immune functions in the child, which subsequently result in AD [4]. Additionally, three studies showed that a postnatal maternal mental health problem alone was associated with an increased risk of AD in children [16‐18].

Considering maternal mental health through the perinatal period, some mothers experience mental health problems through both the prenatal and postnatal periods [19]. However, no study has examined the cumulative impacts of prenatal and postnatal maternal psychological distress on AD in children as far as we know, whereas only the cumulative impacts of prenatal and postnatal maternal mental health problems on wheeze [20] and asthma [21] were examined. We hypothesized that cumulative exposure to maternal psychological distress through the prenatal and postnatal periods is associated with an increased risk of AD in children rather than examining either prenatal or postnatal maternal psychological distress. Examining this association may emphasize the importance of maternal mental health support through both prenatal and postnatal periods.

The purpose of this study was therefore to examine the cumulative impacts of prenatal and postnatal maternal psychological distress on the development of AD in children.

The characteristics of the participants are shown in Table 1. Psychological distress was experienced by approximately half of all women in the prenatal or postnatal period. This included 14.3% of women in only the prenatal period, 13.4% in only the postnatal period, and 19.0% in both periods. A total of 1169 (14.0%) children developed AD between the ages of 1 and 2 years.

Table 2 shows the RRs and 95% CIs for the development of AD in children. Maternal psychological distress in both prenatal and postnatal periods was associated with an increased risk of AD in children compared to no psychological distress in both periods (crude RR, 95% CI: 1.34, 1.20–1.47). This association remained after adjusting for potential confounders (adjusted RR, 95% CI: 1.34, 1.20–1.47). Maternal psychological distress in only the postnatal period was associated with an increased risk of AD in children compared to no psychological distress in both periods (crude RR, 95% CI: 1.22, 1.06–1.38) that remained after adjusting for potential confounders (adjusted RR, 95% CI: 1.23, 1.07–1.39). Maternal psychological distress in only the prenatal period was not associated with an increased risk of AD in children compared to no psychological distress in both prenatal and postnatal periods (crude RR, 95% CI: 1.13, 0.97–1.29) that remained after adjusting for potential confounders (adjusted RR, 95% CI: 1.14, 0.98–1.30). Additional file 2 shows that maternal psychological distress in both prenatal and postnatal periods, and in only the postnatal period was associated with an increased risk of AD, whereas maternal psychological distress in only the prenatal period was not associated.

This study examined the association between cumulative exposure to maternal psychological distress in the prenatal and postnatal periods and the development of AD in children. Mothers with psychological distress in both the prenatal and postnatal periods were the most likely to report AD in their children. Psychological distress in only the postnatal period was also associated with AD, whereas psychological distress in only the prenatal period was not associated with AD.

Cumulative exposure to maternal psychological distress in the prenatal and postnatal periods was associated with an increased risk of AD in children. Chiu et al. (2012) reported that the combined impact of high prenatal and postnatal maternal stress was associated with an increased risk of wheeze in children compared to low maternal stress in both periods (adjusted odds ratio, 95% CI: 3.04, 1.67–5.53) [20]. Brew et al. (2018) also found that cumulative exposure to maternal anxiety or depression in the prenatal and postnatal periods was associated with an increased risk of asthma in children compared to no anxiety or depression in both periods (adjusted odds ratio, 95% CI: 1.50, 1.08–2.09) [21]. To the best of our knowledge, this is the first study to examine the association between cumulative exposure to maternal psychological distress in the prenatal and postnatal periods and AD in children. Considering that the adjusted RRs for maternal psychological distress in only the prenatal period, only the postnatal period, and both periods were 1.14, 1.23, and 1.34, respectively, there may be an additive effect.

Possible biological mechanisms underlying the association between each of prenatal and postnatal maternal mental health problems on AD can be suggested. In utero, maternal stress releases corticotropin-releasing hormone (CRH) via activation of the hypothalamic-pituitary-adrenal (HPA) axis [32], and CRH is transported to the fetus through the placenta, which stimulates the fetal HPA axis to secrete glucocorticoids [33]. This may lead to a modification of the infant’s immune system [32]. It was also suggested that oxidative stress potentially contributes to this mechanism [13]. In the postnatal period, high maternal stress promotes immunoglobulin E expression and the allergen-specific proliferative response in the child. This changes immune functions and enhances the inflammatory response in the child [34]. Additionally, a distressed mother tends to lack cognitive function [35] and to express rejection of the child [36, 37]. This may lead to low responsiveness to the child’s needs or signals [35, 36, 38]. Such a poor quality mother-infant interaction may increase the risk of AD [37]. An animal study also demonstrated that low maternal responsiveness was significantly associated with a higher inflammatory stress response in the infant [39].

In regard to maternal psychological distress in only the postnatal period, the results of the present study also supported previous studies that found an association between a postnatal maternal mental health problem and AD in children [16‐18]. Therefore, a caregiver’s psychological distress after delivery may be a trigger for AD to manifest in children.

On the other hand, prenatal maternal psychological distress was not associated with an increased risk of AD in children. This is inconsistent with most previous studies that demonstrated an association between a prenatal maternal mental health problem and the life or point prevalence of AD in children [7‐11, 13‐17], whereas the development of AD was examined in the present study. Moreover, anxiety or depression was mostly examined as an exposure in previous studies [4], and the impacts of maternal psychological distress in the prenatal period may be more likely to appear at an early age, for example, before the age of 1 year. A nationwide study in Japan reported that 17.0% of children manifested AD before 1 year of age, and the prevalence of AD was higher at the age of 1 year than at the age of 2 and 3 years [5]. In the present study, children who had AD at the age of 1 year were excluded. Therefore, the association between prenatal maternal psychological distress and AD might not have been evident.

The present findings have some implications for reducing the risk of AD in children. Since the cumulative impacts on AD in children were identified, it is important to enhance continuous support for mothers through the prenatal period to the postnatal period. A qualitative study conducted in the UK showed that mothers experienced a lack of continuous care from the prenatal period to the postnatal period [40]. Another qualitative study conducted in China reported that some mothers were willing to receive education about pregnancy and the postpartum period from health care professionals, whereas others were unwilling to seek help unless their condition became serious [41]. Those mothers thought it was irresponsible to disclose their own negative feelings and preferred to talk to a trustable person, such as their own mothers, rather than healthcare professionals [41]. Apart from healthcare professionals, mothers’ partners may play a significant role. A study reported that it is important that men be involved in the care before childbirth to enhance the couple’s relationship and women’s autonomy [42]. Moreover, another study reported that support especially from partners may reduce the risk of AD in children [37]. In Japan, the government published a guideline from the prenatal and postnatal support project in 2017 and passed legislation requiring municipalities to make efforts to provide support for postpartum women in 2019 [43]. These projects are expected to cover potential high-risk mothers and play a role in reducing distress in every mother throughout the perinatal period.

The present study has several limitations. First, this study was conducted in one of the 47 prefectures in Japan. Therefore, the results of this study cannot be generalized. However, the TMM BirThree Cohort Study was able to include approximately half of the newborns in Miyagi Prefecture [22]. Additionally, since the participants in this cohort may have been interested in health and had a willingness to participate, there may have been a selection bias. Second, 35.3% of the participants in the TMM BirThree Cohort Study were included in this study. Variables that were collected at 1 or 2 years after delivery tended to have missing data because of a decrease in the follow-up rate, as reported in another cohort study in Japan [44]. Moreover, the participants included in the analysis were less psychologically distressed than participants who were not included in the analysis (Additional file 1). The cohort study also reported that less psychologically distressed mothers were more likely to continue participation [44]. Although there was no difference in the development of AD between the participants included in the analysis and those who were not included in the analysis, the development of AD may be underestimated in this study, since there have been less psychologically distressed mothers and less diagnosed as AD in parents (Additional file 1). Finally, maternal psychological distress and AD in children were assessed by self-reported questionnaires that may have led to misclassification. However, the Japanese version of the K6 scale was validated [25], and the sensitivity and specificity of a cut-off value of 4/5 were 100 and 68.7%, respectively [26]. A systematic review demonstrated that self-reported AD in children might have been overestimated [45]. Moreover, mothers with psychological distress may be likely to report that their children have poor health. A study reported that mothers with anxiety were more likely to report asthma in children. However, no association was observed between maternal anxiety and a diagnosis of or medication for asthma in children [46]. This suggests that the health perception may be different between psychologically distressed and non-psychologically distressed mothers. Therefore, the reports of AD may have been overestimated in this study.

The present study found that cumulative exposure to maternal psychological distress in the prenatal and postnatal periods was associated with an increased risk of the development of AD in Japanese children at the age of 2 years. Continuous support from partners, family members, and society through the prenatal period to the postnatal period may be important to improve maternal psychological distress, which could potentially reduce the development of AD in children.