It has been reported that psychological symptoms such as depressive symptoms and anxiety symptoms during the postpartum period affect physical health and child rearing and that these effects are important social issues in Japan [1] [2] and in the world [3] [4]. Regarding the transition of depressive symptoms in postpartum women, it was shown in a longitudinal study that the proportion of women with depressive symptoms peaked at 2 weeks postpartum and was decreased at 3 months postpartum [5]. Another study conducted for a 6-month postpartum period showed that the proportion of women with depressive symptoms peaked at one month postpartum [6]. On the other hand, a cohort study showed that the proportion of women with anxiety symptoms increased until one month postpartum and was decreased at 4 months postpartum [7]. Changes in ovarian steroid hormones and hypoactivation of the hypothalamic-pituitary-adrenal axis have been suggested to be biological risk factors for postpartum psychological symptoms [8] [9] [10]. Given that ovulation was found at about 6 months after delivery in lactating women and at 45 days after delivery in non-lactating women [11], the changes in psychological symptoms during a 6-month postpartum period need to be clarified in a prospective study.

Regarding the possibility of differences depending on parity, it has been reported that there were no differences in the prevalence of depressive symptoms and anxiety symptoms between primiparous women and multiparous women [7]. However, during a one-month postpartum period, the proportion of women with depressive symptoms was high in primiparous women [6]. In addition, primiparity was shown to be a risk factor for postpartum depression [12] [13]. The results of those studies are controversial. On the other hand, it has been reported that psychological status of postpartum women was closely associated with positive feelings of postpartum women towards newborn infants such as feelings that taking care of the baby is fun or the baby is cute [14]. However, regarding the association of psychological status of postpartum women with feelings for involvement with newborn infants, a difference between primiparous women and multiparous women has not been clarified.

The aims of this prospective study were to clarify the transition of depressive symptoms and anxiety symptoms during a 6-month postpartum period and to clarify the differences in associations of depressive symptoms and anxiety symptoms with feelings for involvement with newborn infants in primiparous women and multiparous women.

This study was conducted from February 2017 to July 2018. The necessary sample size was calculated to be 150 by using permissible errors (5%), reliability (95%) and a ratio of a population (50%), and the appropriate sample size was determined to be 229 considering the probable number of uncollected questionnaire forms for a prospective study. Postpartum women who had given birth through vaginal delivery or Cesarean section were recruited at two birth centers in Japan in which the number of births is approximately 750 per year. The periods of admission for delivery were about 5 days for primiparous women and about 4 days for multiparous women, and breastfeeding alone for the infants was not promoted in either of the hospitals. On the day of delivery and 3 days postpartum, we asked postpartum women to participate in our study and we obtained agreement by signature on a consent form. We distributed five self-administered questionnaires at 3 days postpartum, 2 weeks postpartum, one month postpartum, 3 months postpartum and 6 months postpartum. We collected the questionnaires at 3 days postpartum through a collection box and we collected the questionnaires at 2 weeks, one month, 3 months and 6 months postpartum by mail.

Women who were taking medication for psychological diseases and women who in the past had experienced death of a newborn infant or emergent hospitalization of a newborn infant were excluded. Women who were having difficulties in answering the questionnaire because of insufficient recovery mentally and physically and women who could not provide agreement and who could not answer the questionnaire by their own will were also excluded.

The contents of the questionnaires were as follows. The questionnaire given at 3 days postpartum had questions on background characteristics including age, parity, marital status, treatment for infertility and delivery method. The questionnaire given at one month postpartum had questions on abnormal results of medical examinations for the women and their infants. The questionnaire given at 6 months postpartum had questions on reoccurrence of menstruation until 6 months postpartum. The questionnaires given at all time points had questions on presence of an advisor, status of the mother’s health, status of the infant’s health, anxiety regarding family finances, relationship between the couple, sleeping hours and lactation method and also included the Japanese version of the Edinburgh Postnatal Depression Scale (EPDS). This version of EPDS was revised as a Japanese version for Japanese women, and a value of 9 points or more is assessed as a high score in Japan [15]. EPDS is a 10-item self-administered questionnaire that was originally developed to screen for postpartum depression [16].

Two scales were used in the study. First, we used psychological symptoms in a standard climacteric scale [17], which is a questionnaire for degrees of menopausal symptoms. Estrogen deficiency symptoms including psychological symptoms have been found in postpartum women as well as postmenopausal women. Thus, in Greene’s climacteric scale, we asked about depressive symptoms (feeling tired or lacking in energy, loss of interest in most things, feeling unhappy or depressed, crying spells and irritability) and anxiety symptoms (heart beating quickly or strongly, feeling tense or nervous, difficulty in sleeping, excitable, pain attacks and difficulty in concentrating) with points from 0 (no symptom) to 3 (severe symptom). The points for depressive symptoms ranged from 0 to 15 and the points for anxiety symptoms ranged from 0 to 18 with a higher score indicating a stronger degree of a symptom. The reliability and validity of the scale were confirmed in a previous study [17].

We also used a questionnaire on involvement with infants that was developed in Japan in 2005 by Matsumura [18] since depressive symptoms and anxiety symptoms in postpartum women have relationships with involvement with newborn infants. The questionnaire included 7 questions on positive feelings and negative feelings for involvement with newborn infants. Two questions about wanting to hug newborn infants because they are so cute and about wanting to dearly protect newborn infants were given points from 1 (definitely no) to 5 (definitely yes). Five questions about finding it hard to take a break from the newborn infant and getting refreshed, not wanting to be near the newborn infant because the infant cries a lot, being tired from dealing with the newborn infant, putting up with a great deal around the newborn infant, and not wanting to change diapers and other messy things were given points from 1 (definitely yes) to 5 (definitely no)．Thus, high total points mean positive feelings for involvement with newborn infants and low total points mean negative feelings for involvement with newborn infants. With respect to reliability for the scale, Cronbach’s coefficient alpha was 0.733. The validity of the scale was confirmed by confirmatory factor analysis and correlation between the subscales [18].

We analyzed data for postpartum women who responded to all of the 5 questionnaires during the 6-month postpartum period. Differences in characteristics between primiparous women and multiparous women during the 6-month postpartum period were analyzed by the t-test and the chi-square test. We used Fisher's exact test when the number of subjects was five or less. The scores for depressive symptoms and anxiety symptoms are shown as means ± standard deviations. Differences in the scores for depressive symptoms and anxiety symptoms were analyzed by the paired t-test. Differences between factors in an increased group in which anxiety symptoms increased from one month postpartum to 3 months postpartum and an unchanged group in which anxiety symptoms did not increase during the same period were analyzed by using the chi-square test. Differences between factors in an increased group in which depressive symptoms increased from 3 months postpartum to 6 months postpartum and an unchanged group in which depressive symptoms did not increase during the same period were also analyzed by using the chi-square test. Correlations of depressive symptoms and anxiety symptoms with feelings for involvement with newborn infants during the 6-month postpartum period were analyzed by using Spearman’s rank correlation coefficient test. Correlations of scores for depressive symptoms and anxiety symptoms with EPDS scores at all 5 time points during the 6-month postpartum period were also analyzed by using Spearman’s rank correlation coefficient test. All statistical analyses were conducted with SPSS statistics ver. 24 (IBM Corp.).

This study was approved by the Research Ethics Committee of Tokushima University Hospital (approval number 2702). We first distributed an explanation sheet and a consent form to the subjects, and we received written informed consent for participation after explanation of the study including the aim of the study and methods used in the study, right of free participation, personal privacy protection and disadvantage by non-participation. We recruited participants who provided written informed consent.

We distributed 229 questionnaires to two hospitals. Responses to all of the questionnaires at 3 days postpartum, 2 weeks postpartum, one month postpartum, 3 months postpartum and 6 months postpartum were obtained from 121 (52.8%) of the 229 women. The primiparous women were 56 and multiparous women were 65. The mean ages of the primiparous women were 30.8 ± 4.9 years and the multiparous women were 32.7 ± 4.3 years. There was a significant difference between the mean ages of the primiparous women and multiparous women at 3 days postpartum (p = 0.028). Also, the proportions of the women who had received treatment for infertility were significantly different in primiparous women and multiparous women (p = 0.043) (Table 1).

There was a significant difference in status of the infant’s health at 3 days postpartum between primiparous women and multiparous women (p = 0.007). There were also significant differences in the lactation method between primiparous women and multiparous women at 2 weeks postpartum (p = 0.001) and at one month postpartum (p = 0.007). There were no significant differences in any of the items between primiparous women and multiparous women at 3 months and 6 months postpartum. The proportions of women with anxiety

^aAge is shown as mean ± standard deviation are shown in age. Statistical analysis was used by t-test. ^bStatistical analysis was used by chi-square test (5 or less are Fisher’s test). *p < 0.05.

about family finances ranged from 37.5% to 50.0% in primiparous women and from 32.3% to 41.5% in multiparous women during the 6-month postpartum period (Table 2).

We compared scores for depressive symptoms and anxiety symptoms with EPDS scores in order to confirm whether scores for depressive symptoms and anxiety symptoms are appropriate for evaluation of postpartum psychological symptoms. The scores for depressive symptoms and anxiety symptoms showed strong positive correlations with EPDS scores (ρ = 0.566 and p < 0.001 at 3 days postpartum, ρ = 0.755 and p < 0.001 at 2 weeks postpartum, ρ = 0.709 and p < 0.001 at 1 month postpartum, ρ = 0.598 and p < 0.001 at 3 months postpartum, and ρ = 0.602 and p < 0.001 at 6 months postpartum).

Transition of depressive symptoms and anxiety symptoms according to parity (Figure 1(a), Figure 1(b))

Regarding the transition of depressive symptoms and anxiety symptoms according to parity during the 6-month postpartum period，the score for depression at 3 days postpartum in primiparous women was significantly higher than that in multiparous women (p = 0.025). However, there were no significant differences in the scores for depressive symptoms and anxiety symptoms between primiparous women and multiparous women at the other four time points during the postpartum period.

Also, scores for both depressive symptoms and anxiety symptoms showed peaks at 2 weeks postpartum and decreased thereafter. The scores for depressive