Prevalence and associated factors of postpartum depression among postpartum mothers in central region, Eritrea: a health facility based survey

This study used analytical cross-sectional study design to evaluate the magnitude of and factors associated with postpartum depression.

This research was part of a study which was conducted to validate the Edinburgh Postnatal Depression Scale in Eritrea. The study was conducted in four primary health care facilities of Zoba Maekel (Central Region), Eritrea. All of the health facilities in the Region were categorized in to two groups, based on rural and urban setting, from which two health facilities were selected from each setting using simple random sampling technique. Accordingly, Tsaeda-kristian and Seregeka Health Centers were selected from the rural health facilities, and Semenawi Asmara Health Center and Freselam Health Station were selected from health facilities within the city of Asmara. These health facilities are meant to provide primary health care services like immunization, family planning, antenatal care and other basic health services, which make them ideal places to encounter postpartum women in our setting. Hence, the study was purposively conducted in primary health care facilities. The target population for this study was recently delivered mothers within 2 to 14 weeks postpartum. Mothers confirmed to be free from any known chronic illness and mental illnesses were included in the study.

The sample size for this study was determined by using the single population proportion formula [n = (z² (P *(1-p))/d²]. Where: z-score value = 1.96, p (the proportion of postpartum depression) was taken as 50% (0.5), and margin of error (d) was set to be at 5% (0.05). Based on these parameters the initial sample size (n₁) was 384. We used a correction factor [n₂ = (n₁*N)/ (N + n₁))] to adjust the sample size to the actual population size (N), which was estimated to be 17,369. Based on this n₂ was found to be 375. Finally, 5% of the sample was added for the expected non-response. Therefore, the final sample size was 393. The sample size was allocated to the selected health facilities based on probability proportional to the size, size being the number of mothers attending the health facilities for child immunization program. Then the researchers developed a sampling frame of eligible mothers, those who were present during the data collection period, in each health facility and study participants were selected by using simple random sampling technique.

A structured closed-ended questionnaire was developed to be used as data collection tool to capture the socio-demographic and maternity-related information of the study participants. The standard Diagnostic and Statistical Manual of Mental Disorders Fifth Edition (DSM-5) [2] was used to assess depression in the postpartum women. DSM-5 is an international diagnostic manual for mental disorders developed by the American Psychiatric Association and WHO in collaboration with other institutions and professionals working in the field of mental health. This manual has ten-item diagnostic criteria for Major Depressive Disorders (MDD). Accordingly the presence of five or more of the following symptoms during 2-weeks period, which represent a change from previous functioning, qualifies for the diagnosis of major depressive symptoms. This manual is definitive criteria for the diagnosis of MDD in the study setting and it requires a qualified professional to administer it.

First, the participants were interviewed, in a private room within the health facility, for the socio-demographic and other maternity related questions. Then DSM-5 interview for MDD was conducted by experienced qualified psychiatric-nurse in a separate room. Based on the diagnostic interview the mothers were categorized as depressed and not-depressed. Mothers diagnosed as having depression during the MDD interview were offered counseling on how to proceed with their condition.

The dependent variable for this study was status of the mother with regard to postpartum depression, as diagnosed by the DSM-5’s criteria. The socio-demographic and maternity related variables presumed to influence the likelihood of developing PPD, were the independent variables. The main maternity-related data included; gravidity, gestational age at delivery (term/preterm), mode of delivery, place of delivery, and postpartum days of the mother.

On bivariate logistic regression analysis nine socio-demographic and maternity related variables of the mothers were found to predict PPD. However, multivariate analysis revealed only six of them to be independent predictors of postpartum depression. The employment status of the mothers was categorized into employed, unemployed and housewives. Unemployed mothers are those who previously been employed or are potentially employable but currently have no job. Whereas, housewives represent mothers who are married, have no job and are not seeking for employment. The housewives were 0.24 times less likely to develop PPD, as compared to the employed mothers (AOR = 0.24, 95% CI: 0.06–0.97; p = 0.046), those mothers who perceived their economic status as low were 13 times more likely to develop PPD as compared to the mothers who had good economic status (AOR = 13.33, 95% CI: 2.66–66.78; p = 0.002). Mothers who did not have support from their partner or the father of the baby had higher odds (AOR = 5.8, 95% CI: 1.33–25.29; p = 0.019) of experiencing PPD. Participants whose last pregnancy was unplanned were 3.39 times more likely to have postpartum depression (AOR = 3.39, 95% CI: 1.24–9.28; p = 0.017). Mothers who experienced illness after delivery were more likely to develop PPD as compared to their counterparts (AOR = 7.42, 95% CI: 1.44–34.2; p = 0.016). Finally, mothers from the Southwest of Asmara were found to have higher odds (AOR = 6.35, 95% CI: 1.73–23.23; p = 0.05) of being affected by postpartum depression (Table 3).

This study indentified several socio-demographic and maternity-related factors as being associated with PPD. Perceived economic status was a significant predictor of PPD; mothers who perceived their economic status as low were 13.3 times more likely to develop PPD, as compared to those who perceived their economic status as good. This finding is supported by studies conducted in different settings, like Poland [18], Singapore [35], Kenya [36] and Northern California [37]. The impact of poverty on the household in general and women in particular is a well known phenomenon [1, 38]. In the current study setting, childbirth is a well revered and celebrated event. In such cases mothers who have few economic resources may find it stressful to prepare the necessary funds to celebrate the event, which may lead them to PPD.

Studies have consistently reported employment status as a significant predictor of PPD; unemployed mothers being at higher risk than the employed [21, 39, 40]. Similarly in this study employment status of the mother was significantly associated with PPD. On bivariate analysis unemployed mothers were 4.5 times more likely to have depression, compared to the employed mothers. However, on multivariate analysis the housewives had a 74% lower risk of experiencing PPD, compared to the employed mothers. A study conducted in Iran reported contrary findings; housewives were at higher risk of PPD than were employed mothers [41]. The concern of employed mothers about securing their job, on returning to work after delivery, which is not a concern for housewives [42‐44], can be a reason for the lower risk of PPD among this group. A study on pregnant Egyptian women found similar findings, with employed women reporting higher levels of depression and anxiety, compared to housewives [22].

Mothers who had no husband/partner support after delivery were 5.8 times more likely to develop postpartum depression than those who had partner support. The effect of partner support or nature of the couple’s relationship has been reported as significant predictors of PPD [36, 42, 45, 46]. Partner support can be viewed from different perspectives. First, partner support can be economic, l such as contributing or covering expenses towards child rearing. Another aspect of the support can be the father’s accepting the baby as his child, particularly in the case of single mothers. This indicates that the above-mentioned factors are most likely to affect single mothers. Finally, mothers may need their partners to share chores, as the additional task of infant care can overwhelm them [47]. In any case, the role of the male partner’s support in preventing and alleviating postpartum depression is commendable [41, 48].

Unplanned pregnancy was another significant factor associated with PPD. In this study mothers whose last pregnancy was unplanned were found to be at higher odds (three times) of developing PPD. Similarly, Barton et al. [49], Sunder et al. [50] and Timothy et al. [33] reported the association between unplanned pregnancy and subsequent PPD in women. However, a study from Hanoi, Vietnam, found no association between unplanned pregnancy and PPD [51]. In this study 42.9% of unintended pregnancies were reported by single mothers as compared to 17.1% of married mothers. This makes interpreting the pathway by which pregnancy intention affects PPD somehow complex, as it could be related to social stigma, inability to bear the responsibility, financial difficulty, or affirming fatherhood.

In this study mothers who became ill after delivery were seven times more likely to experience postpartum depression. Mother’s illness was assed by asking the mother if medication has been prescribed by a physician for the reported problem, to avoid confusion with psychosomatic symptoms. Studies from Uganda [31], and Pakistan [52] have affirmed the link between maternal illness immediately following delivery and postpartum depression. This may be related to the known effect of physical health problems on mental wellbeing [53, 54].

Finally, residence of the mothers was found to be a significant predictor of PPD in this setting. Mothers from Southwest-Asmara were 6.3 times more likely to have PPD as compared to mothers from Northeast- Asmara. The former is an area where relatively affluent people reside, while the latter is a more traditional and slummy residential area. Logically we expected the effect to go the in the opposite direction, with mothers from the slummy area to be more affected by PPD. One possible explanation for this paradox is the culturally determined social support a mother receives during and after delivery in this traditional residence area. Normally the expectant mother is comforted by her neighbors, family members and significant others during and after delivery. This social support ranges from accompanying the pregnant mother to health facility for delivery to washing her and her infant’s clothes. Such social support networks are common in more traditional residential areas where people live a communal life. Viewing the two settings in this way, mothers from Southwest-Asmara would be less likely to receive this kind of social support, than would mothers from Northeast- Asmara. The role of a strong social network in reducing the risk of postpartum depression has been reported in many studies [55‐58]. In addition, study findings further indicated that social and cultural norms influence the link between postpartum mental problems and social relationships, by defining what is preferred, expected and exchanged in interpersonal interactions [33]. However, further study is necessary to understand the effects of residence and social interaction in PPD in the current study setting.

This study included only mothers who delivered live-births and who have a healthy child. Mothers with chronic disease, stillbirth, whose infant died after delivery, and mothers with infants in neonatal intensive care units were not included. This limits the ability of this research to healthy mothers with healthy babies. Furthermore, the fact that this was a facility based study the results may not enable us to generalize about the mothers in the general population of the Central Region in which the study was conducted.