Psychiatric morbidity among pregnant and non pregnant women in Ibadan, Nigeria

Abstract A comparative cross-sectional study was conducted among 991 pregnant and 674 non-pregnant women of reproductive age attending healthcare facilities in Ibadan, Nigeria using the General Health Questionnaire-12 (GHQ), and WHO self-reporting questionnaire (SRQ). Logistic regression analysis was conducted to identify predictors of psychiatric morbidity at p < 0.05. A significantly higher proportion of pregnant women experienced psychological distress on the GHQ (51.8%) and psychiatric morbidity on SRQ (33.3%) compared with 28.6% and 18.2% of non-pregnant women, respectively. Predictors of psychiatric morbidity among pregnant women were the type of facility, poor satisfaction and communication with partners, the experience of violence in the home, previous abortions, and previous history of depression. Psychiatric morbidity among non-pregnant women was predicted by younger age, previous history of depression, poor satisfaction and communication with partners. There is a need for early identification of psychiatric morbidity among women of reproductive age, to ensure early interventions and prevent long-term disability. Impact statement What is already known on this subject? Psychiatric morbidity has immense effects on a woman’s quality of life, social functioning, obstetric outcome, and economic productivity. What do the results of this study add? Psychiatric morbidity among women of reproductive age is high. Pregnant women when compared to non-pregnant women had significantly higher rates of psychiatric morbidity. This high prevalence of psychiatric morbidity in both groups was predicted by poor satisfaction and communication with partners, and a previous history of depression. What are the implications of these findings for clinical practice and/or further research? Simple screening for women of reproductive age attending healthcare facilities may help with the early identification of psychiatric morbidity leading to prompt interventions, and preventing long-term disability.


Introduction
Mental disorders are a leading cause of morbidity in the world today (WHO 2019).These disorders especially depression, anxiety and stress-related disorders are common among women in the reproductive age group (WHO 2008, WHO 2018).There is increasing attention to the mental health of women in low-and-middle-income -countries (LAMICS) who are within the perinatal period but less attention has been given to their non-pregnant peers.A recent study in rural south west Nigeria found significant psychiatric morbidity among non-pregnant women with anxiety disorder prevalence rates of 28.8% and anxiety and depression prevalence rates of 19.8% (Soyannwo et al. 2020).Studies have shown that the incidence of mental health problems rises substantially in child-rearing years (Thapar et al. 2012).This has been attributed to the fact that females carry the physical burden of child bearing, and in many cultures, take up the greater burden of child-rearing.The transition to motherhood often represents a major life change, with substantial impact on financial, social-emotional and physical daily functioning.Career changes and possible employment breaks, may also occur in the lives of women to facilitate the caregiver role, all of which may contribute to the increased female vulnerability to mental health disorders (Goncalves et al. 2016).
One out of three to one out of five women in LAMICs, and about one in ten in developed countries, have significant mental health problems during pregnancy and after childbirth (WHO 2008).Additionally, the incidence of mental health problems among women in the perinatal period increases by 3 to 5 times for those exposed to intimate partner violence (WHO 2001).Pregnancy has been recognised as a time of elevated risk for psychiatric morbidity, predominantly depressive and anxiety disorders (Soares and Zitek 2008).Risk factors for psychiatric morbidity in pregnancy has been reported to include maternal age, stressful life events, stressful events during the pregnancy and delivery (Fatoye et al. 2004, Adewuya et al. 2006, Wisner et al. 2006, Prince et al. 2007, Fisher et al. 2012, Sulyman et al. 2016).Pregnancy hormones-oestrogen and progesterone which steadily increase throughout pregnancy until just before birth also affect the brain and subsequent mental state of women (Steiner et al. 2003).These hormones are known to have complex interactions with brain neurotransmitters leading to the mood fluctuations, irritability, and depression commonly experienced during pregnancy.Stress during pregnancy has also been found to be associated with pregnancy related complications, hospital admissions, domestic violence, lack of or low social support, poverty, previous abortion or still birth, poor marital relationship, instrumental vaginal delivery, emergency caesarean section, manual removal of placenta, and poor maternal experience of control during childbirth (Andersson 2004, Adewuya et al. 2006), all which further increase the risk of mental health problems among affected women.
During pregnancy, mild symptoms of stress and anxiety may result from hormonal changes as well as the impending transition into parenthood.Thus there may be a need for psychosocial adjustments to role changes especially in nulliparous women (Adewuya et al. 2006, Chan 2013).Most women are able to cope with these changes using their own internal resources as well as the social support available to them (Adewuya et al. 2006, Chan et al. 2013).However, a significant number of pregnant women go on to develop anxiety disorders with prevalence rates ranging from 17.7% to 59.5% (Adewuya et al. 2006, Lee et al. 2007, Chan et al. 2013, Soyannwo et al. 2020).Some studies have also shown a higher prevalence of psychiatric disorders in the antenatal period compared to the postnatal period (Adewuya et al. 2006, Faisal-Cury and Rossi 2007, Lee et al. 2007, Grant et al. 2008, Chan et al. 2013).
There are more studies worldwide on the mental wellbeing of women in the perinatal period (Gadanya et al. 2018, Agbaje et al. 2019, Ezeme et al. 2020) but there is a paucity of studies on psychiatric morbidity among women of reproductive age who are not within the perinatal period, especially in LAMICs like Nigeria.This study therefore aimed to assess psychiatric morbidity among a group of pregnant versus non-pregnant women in the reproductive age group using validated research tools.

Materials and methods
This was a comparative cross-sectional study among consenting pregnant and non-pregnant women in Ibadan, Southwest Nigeria conducted between September 2017 and February 2018 using an interviewer-administered questionnaire.The study was conducted among women aged 18 to 45 years attending the three tiers of public health care service delivery in Nigeria -primary, secondary and tertiary.In Nigeria, basic health care is provided at primary health care facilities which are usually manned by nurses and community health care workers who are trained to manage and treat only minor ailments.More serious health conditions are referred to secondary health care facilities which are an intermediate level of care.The most serious ailments are referred to tertiary healthcare facilities which provide specialist care.Because payment for health care services in Nigeria is mainly out of pocket, primary health care services are the preferred option for all types of health problems by people from lower socioeconomic classes who cannot afford more expensive specialised health care.Two of the 11 districts or local government areas (LGA's) in Ibadan were purposively selected based on the availability of facilities in the three tiers of health care services.One primary and one secondary healthcare facility were randomly selected from the thirteen primary healthcare, and two secondary healthcare facilities were available in these districts.The only tertiary health facility in the city of Ibadan was also selected.
The sample size for the study was determined using the formula for comparing two proportions with the sample size calculator in WinPepi version 16.The minimum sample size (n) for each group was calculated using a power of 80%, a statistically significant level of 5% and adjusting for the cluster effect of 2 and a 10% non-response rate (Abramson 2011) giving an estimated sample size of 1,096 (540 pregnant and 546 non-pregnant).Systematic sampling was used to select respondents in these health facilities.Pregnant women presenting for antenatal care or booking in the second trimester (14 weeks and above) at the antenatal clinic and nonpregnant women attending the out-patient clinics for nongynaecological conditions in all the selected health facilities were recruited for the study while those who presented with emergency conditions or were too ill to give consent were excluded.Women who met the inclusion criteria were informed about the study by their health professionals, and written informed consent was obtained before the administration of the questionnaire by the trained research assistants.The mental health status of 991 antenatal clinic attendees at the gestational age of �14 weeks was compared with 674 non-pregnant women of reproductive age attending the outpatient clinics of the same facilities.Women who showed signs of psychological distress during interviewing were encouraged to speak with their healthcare professionals for possible referrals as needed.
The WHO Self Reporting Questionnaire (SRQ), General Health Questionnaire 12 (GHQ À 12) and adapted Respondent and Partners section of the WHO women's health and life events survey was used to collect data from study participants.In this study occupational group was classified into six categories using the Registrar General Occupational Classification (Park 2009) and the income was based on the country's monthly minimum wage of 30,000 Naira (equivalent to $73.2).
The validated questionnaires were translated into Yoruba (the local language) and back-translated to English to ensure meaning was retained.Data on demographic and socioeconomic characteristics; obstetrics, medical history, and living conditions were also collected.Data collected were entered, cleaned and analysed using the Statistical Package for the Social Sciences (SPSS) version 23.Data were reported in frequencies and proportions for categorical variables and means and standard deviations for quantitative continuous variables.Bivariate analysis was conducted to explore the association between independent and dependent variables using Chisquare test.Variables on bivariate analysis were significant at 10% and were subjected to binary logistic regression analysis to detect predictors of psychiatric morbidity among pregnant and non-pregnant women.Ethical approval was obtained from the Oyo State Ethics Review Committee (AD 13/479/432).

Results
A total of 1,665 women of which 991(59.5%) were pregnant and 674 (40.5%) were non-pregnant participated in the study.Their mean ages were 30.1 ± 5.4 and 29.6 ± 8.4 years (t ¼ 1.472, p ¼ 0.141), respectively.A significantly higher proportion 808 (81.5%) of the pregnant women compared with 346 (51.3%) were in the 25-39 age category (p < 0.001).A higher proportion of the pregnant women were married 953(96.2%)compared with the non-pregnant women 401(59.5%)which was statistically significant (p < 0.001).Furthermore, more women in both groups were from monogamous family settings and currently living with their spouses (p < 0.001).A significantly higher proportion of pregnant women experienced psychological distress on the GHQ, 513 (51.8%), and psychiatric morbidity on the WHO SRQ 330 (33.3%) compared with 28.6% and 18.2% respectively of nonpregnant women (p < 0.001), (Table 1).
We explored the association between psychiatric morbidity and sociodemographic characteristics (Table 2), family characteristics and obstetric and medical history (Table 3).Among pregnant women, sociodemographic factors significantly associated with psychiatric morbidity were the type of facility (p < 0.001) family characteristics, partners' communication (p < 0.001), the experience of arguments (p ¼ 0.001, the experience of quarrels (p ¼ 0.034), the experience of violence in the home (p < 0.001), the experience of family-related stress (p < 0.001), and for obstetric and medical history, previous abortion (p < 0.001), number of living children (p ¼ 0.033), satisfaction with living children's' gender (p < 0.001), medical conditions in current pregnancy (p < 0.001), previous history of depression (p < 0.001) and family history of depression (p < 0.001).The highest proportion of psychiatric morbidity was found in women attending tertiary facilities (39.4%), followed by secondary facilities (34.0%), and the least was in women from primary care facilities (19.3%).Pregnant women who had experiences of frequent quarrels with their spouses, experience of domestic violence, family-related stress, previous abortions, medical conditions in the current pregnancy, a past history of depression, and a family history of depression had higher rates of psychiatric morbidity compared to those who did not have these characteristics.
In non-pregnant women, correlates of psychiatric morbidity were mainly family characteristics which included satisfaction with marriage (p ¼ 0.001), partners' gainfully employed (p ¼ 0.031), partners' occupation (p ¼ 0.026), financial support from other sources apart from spouses (p ¼ 0.028), poor partner communication (p ¼ 0.021), the experience of arguments in the marriage (p ¼ 0.023), the experience of violence in the home (p < 0.007), and obstetric and medical history; previous abortion (p < 0.006), previous history of depression (p < 0.001) and family history of depression (p < 0.001).Women who were unsatisfied with their marriage had employed partners, experience domestic violence, had previous abortions, previous history of depression, and family history of depression had higher rates of psychiatric morbidity.
Predictors of psychiatric morbidity among pregnant women included the type of facility, dissatisfaction with marriage, poor partner communication, experience of violence, previous abortion, and previous history of depression.Pregnant women who attended primary health care facilities were significantly less likely to have psychiatric morbidity compared with those who were attending tertiary health facilities (95% CI: 0.083 À 0.441; OR ¼ 0.192).Pregnant women who reported partial spousal communication were less likely to experience psychiatric morbidity than those who did not communicate at all (95% CI: 0.064 À 0.835; OR ¼ 0.231).Those who did not experience violence were significantly less likely to have psychiatric morbidity compared with those who did (95% CI: 0.187 À 0.518; OR ¼ 0.311).Pregnant women with a previous history of abortion were 2 times more likely to have psychiatric morbidity than those without previous history of abortion (95% CI: 1.203 À 2.944; OR ¼ 1.882).Those with a previous history of depression were 3 times more likely to have psychiatric morbidity than those without previous history of depression (95% CI: 1.782 À 5.429; OR ¼ 3.111).
Predictors of psychiatric morbidity among non-pregnant women included age, satisfaction with marriage, spousal communication and previous history of depression.Younger nonpregnant women aged 25-39 years were 2 times more likely to have psychiatric morbidity than their counterparts aged 40 years and above (95% CI: 1.175 À 4.571; OR ¼ 1.822).Non-pregnant women who reported they were not satisfied with their marriage were 3.5 times more likely to have psychiatric morbidity than those who reported satisfaction (95% CI: 1.165 À 10.491; OR ¼ 3.496).Non-pregnant women who reported partial spousal communication were less likely to experience psychiatric morbidity than those whose spouses did not communicate

Discussion
This study sought to determine the levels of psychiatric morbidity among a group of pregnant versus non-pregnant women in the reproductive age group at primary, secondary and tertiary levels of health care in Ibadan, Nigeria.The sociodemographic characteristics of our participants were similar to those from previous studies in this environment (Oladimeji et al. 2019).We found a mean age of 30.1 ± 5.4 years vs 29.6 ± 8.4 years for pregnant and non-pregnant women respectively, which is consistent with the 28.9 years (SD ¼ 5.21) and 30 years (SD ¼ 5.14) found by Oladimeji et al. (2019).This study also found that the majority of women were married, Christians and had at least a secondary level of education which is consistent with our findings.
The pregnant women in this study had significantly higher rates of psychiatric morbidity compared to the non-pregnant women.While there are not many studies comparing psychiatric morbidity between pregnant and non-pregnant women, an Australian study found that among pregnant and nonpregnant women who were matched for age and education, levels of psychiatric morbidity were similar (Barber and Steadman 2018).However, a Nigerian study (Adewuya et al. 2006) found that pregnant women had significantly higher rates of anxiety disorders when compared to their non-pregnant counterparts, especially in the presence of a pre-existing medical condition.As reported in previous literature, hormonal changes and changes in family and work dynamics may contribute to the added stress of women who are pregnant thus increasing their risks of psychiatric morbidity.The socioeconomic issues in this context mandate that more and more women have to work, and social support from the extended family is on the decline (Akinwaare et al. 2019).
The factors associated with psychiatric morbidity among both groups of respondents were poor communication with partners, frequent arguments in the intimate relationship, violence in the home, previous abortions, a previous history of depression and a family history of depression.Previous studies have shown a protective effect of marital satisfaction on psychological distress among women and poor satisfaction, poor communication, increased domestic violence, and poor financial and social support are all associated with psychological distress among pregnant women and non-pregnant women worldwide (Qadir et al. 2013, Jonsdottir et al. 2017).Though there are varying reports about the effects of foetal loss on women's mental health, studies have shown an overall increase in psychiatric morbidity associated with induced abortions (Heikinheimo et al. 2017).
Predictors of psychiatric morbidity among pregnant women were a type of facility, poor satisfaction with the marital relationship, poor communication with partners, the experience of violence in the marriage, previous abortions and previous history of depression.On the other hand, psychiatric morbidity among non-pregnant women was predicted by younger age (25 -40 years), poor satisfaction with marriage and communication in the home, and a previous history of depression.The marital relationship is one of the strongest relationships a woman can have and can either reduce or increase the risk of stress and its associated emotional problems.Studies from similar contexts to our study have all reported higher prevalences of psychiatric morbidity among women who had worse marital satisfaction (Odinka et al. 2018, Omidvar et al. 2018, Ezeme et al. 2020).In Ethiopia, there are reports that antenatal women in unsatisfactory marital relationships were about eight times more likely to be depressed than those in cordial marital relationships (Mengistu et al. 2020).
It is not clear why pregnant women who attended primary healthcare facilities had reduced risks of psychiatric morbidity in this study.Many of these women are usually from lower socioeconomic backgrounds and would have been expected to be more vulnerable to psychiatric morbidity.However, they are likely to have greater support from extended family members such as mothers and mothers-in-law who like them do not work in high pressured jobs, and are therefore available to help with child care and other household chores.It is well established that women who enjoy greater social support in the perinatal period experience better mental health (Maharlouei, 2016).In addition, because these women do not work in high pressured jobs, they may also be able to work more flexibly during pregnancy or stop work totally thereby reducing their overall stress.
It is also noteworthy that younger age (25-40 years) was a significant predictor of psychiatric morbidity among nonpregnant women in this study.In the study context, age 25 -40 is the age bracket where most women are married and would start a family (National Population Commission, 2019).The stress associated with building up a new marriage, becoming a mother, child rearing and finding a balance between family duties and work may increase women's vulnerability to psychiatric morbidity during this time.By the time women are 40 years and above, children have grown and some of the burdens of child-rearing reduce (Craig and Sawrika 2009).Work may be also more stabilised with many of them being promoted to senior positions, leading to greater work flexibility.
The limitations of the study include its cross-sectional design which precludes the ability to infer causality.Data was by self-report which may be clouded by the participant's emotional states, and screening instruments were used to measure psychiatric morbidity.On the other hand, the strengths of this study are its large sample size, with respondents across the three tiers of healthcare services available in the country.This study made use of validated questionnaires, and several demographic and medical factors were also explored.

Study implications
The high rates of psychiatric morbidity among women of childbearing age in Nigeria call for interventions to aid early detection and treatment.Healthcare workers should be trained to ask simple screening questions which could include questions about the marital relationship at women's first point of contact with health services.This will aid early identification and help to reduce long-term adverse consequences.

Table 1 .
Study participants' characteristics and prevalence of psychiatric morbidity.
Bold values indicate significance at p < 0.05.

Table 2 .
Sociodemographic correlates of psychiatric morbidity among pregnant and non-pregnant women.

Table 3 .
Association between partners' characteristics, and obstetric and past medical history and psychiatric morbidity among pregnant and non-pregnant women.