Prevalence and risk factors associated with common mental disorders among pregnant women in Kumasi, Ghana: a facility-based survey in selected primary care settings

Abstract

Background

Common mental disorders (CMDs) during pregnancy are linked to adverse maternal and neonatal outcomes, particularly in LMICs. However, risk factors for CMDs during pregnancy have received limited attention in preventive strategies. Therefore the study aimed to examine the prevalence and risk factors associated with common mental disorders among pregnant women in selected primary healthcare settings in Kumasi, Ghana.

Methods

A facility-based cross-sectional study was conducted among pregnant women in three selected public primary care facilities within Kumasi in the Ashanti Region of Ghana. A random sampling technique was used to select 232 pregnant women in their second and third trimesters who responded to the Self-Reported Questionnaire (SRQ-20). Descriptive and inferential analyses, including chi-square tests and firth logit regression models, were conducted using Stata (version 17) to identify the prevalence and factors associated with common mental disorders. Odds ratios were calculated with a 95% confidence interval to determine the association.

Results

This study found a CMD prevalence of 12.0% (95% CI: 8.3–16.9%) among the studied population. Overweight (adjusted Odds Ratio [aOR] = 0.17; 95% CI:0.03–0.96) and obese (aOR = 0.11; 95% CI:0.02–0.66) individuals exhibited a lower risk of CMD compared to those with normal BMI, while those with diabetes demonstrated a significantly increased risk (aOR = 8.59; 95% CI:1.41–52.24).

Conclusion

The significant link between diabetes and increased CMD risk underscores the necessity for comprehensive care strategies that address both physical and mental health needs concurrently by integrating care pathways into primary obstetrical care.

Background

Common mental disorders (CMDs) are a group of mental health conditions that manifest through depression and anxiety with non-specific physical complaints that disrupt day-to-day functioning [1].They are usually prevalent among pregnant women in primary care settings. CMDs may include clinically diagnosed disorders or signs and symptoms-based reports amongst pregnant women [2]. A systematic review on the prevalence and associated factors of CMDs in women found that common mental disorder prevalence in women ranged from 9.6 to 69.3% [3] and was high in pregnancy. CMDs during pregnancy have gained significant global public health attention, necessitating the integration of mental health services into primary obstetric care settings [4–6]. Maternal common mental disorders and their adverse outcomes are interrelated, and complex and involve biological, psychosocial, and environmental interactions [7]. CMDs are prevalent in pregnancy and are associated with non-psychotic symptoms like difficulty in concentration, fatigue, irritability, low self-esteem, difficulty in sleeping, frequent substance misuse, and memory loss [8–10]. As opposed to popular erroneous belief, research has found that pregnancy is not protective against the development or relapse of mental disorders [11].

Common mental disorders remain understudied especially in Lower-middle-income countries (LMICs) despite the high morbidity and mortality rate associated with pregnancy [12]. Research has found that common mental disorders in pregnancy are prevalent globally, however, it is 19.7% higher amongst women who are less resourced and with reduced access to good health services [1]. In high-income countries, the prevalence of common mental disorders among pregnant women was found to range from 7 to 15% [7, 8]. However, common mental disorders accounted for 15.6% of pregnant women in Lower-middle-income countries as well as 19.8% in the postnatal period [13]. Research suggests that common mental disorders have negative effects on pregnant women’s psychological well-being [14]. These effects lead to neglect of personal hygiene, lower quality of life, low work productivity, reduced life expectancy, substance abuse, suicide and infanticide ideation, family dysfunction [14] and disruptions in activities of daily living [3].

Poor social support during pregnancy, genetic predisposition, maternal stress, and a history of an underlying mental condition are among the multiple factors that increase maternal and child health outcomes negatively [8, 15]. Moreover, common mental disorders in pregnancy can negatively affect child development up to 16 years of age [16, 17]. Pregnant women who suffer from common mental disorders have reduced or no antenatal clinic attendance and poor nutrition which may lead to severe weight loss. Common mental disorders in pregnancy may also lead to poor obstetric complications [15], complicated delivery, and worse neonatal outcomes such as an increased risk of preterm birth with low birth weight. Various studies have reported that pregnant women with untreated mental disorders like bipolar disorder, post-traumatic stress disorder, and obsessive-compulsive disorders may experience impaired psychological, emotional, cognitive, and behavioral development in children [8, 18, 19]. They also experience physical effects [20] such as malnutrition and decreased immunological function in children [21, 22]. Bipolar disorder particularly can lead to maternal functional incapacities and frequent hospitalizations [23]. Although maternal and child mortality has been linked with negative mental health outcomes during pregnancy among Ghanaian women [24], the majority of studies have focused on pregnant women in the capital city, Accra [25–27].

Despite the burden of common mental disorders and their deleterious effects on maternal and child health, the associated risk factors have not been focused on prevention [28] especially in LMICs. Ghana introduced a comprehensive Mental Health Bill in 2012 that aimed to protect the rights of persons with mental health challenges following international human rights standards however, it excludes maternal mental health care. There is a rapid traction of research on common mental disorders during pregnancy. Still, information on their corresponding preventive interventions is scanty, especially within the context of primary antenatal settings in Ghana. Consequently, this study formed part of a broader study; it was a necessary preliminary exercise aimed to provide data that (1) would inform the broader study aim to develop appropriate systems and procedures for incorporation of screening, diagnosis and treatment of pregnant women with common mental disorders (2) the study would have a beneficial impact for healthcare providers to prioritize maternal mental healthcare delivery. This study aimed to examine the prevalence of common mental disorders and their associated risk factors among pregnant women attending antenatal care in primary care settings.

Methods

Study design and data source

This is a facility-based cross-sectional study that was conducted among pregnant women in their second and third trimesters who had set-time appointments in three selected state-managed primary antenatal care facilities within the Kumasi metropolis. Kumasi is the second-largest city in Ghana, situated 300 km away from the capital city, Accra. Kumasi spans an area of 223.1 square kilometers and is divided into ten administrative districts. Interviewer-administered questionnaires (programmed on a Tablet) were used to collect the data between February and March 2024.

Study setting

For this study, three hospitals were strategically chosen due to their reputation as state-managed health facilities offering comprehensive primary-level antenatal and obstetric care designated as sites A, B, and C. For example, site A, runs the largest antenatal care services at the primary care level. The maternity clinic attends to 1053 pregnant women for antenatal care annually, according to their bio-statistical unit. Site B affiliated with the Christian Health Association of Ghana (CHAG), serves a population of 154,526 [29]. It offers general, surgical, and maternal care services for both out-patients and in-patients, attracting pregnant women from Kwadaso Municipal and surrounding areas, as well as patients from the Ahafo and Bono East regions. Similarly, site C, a peri-urban state-owned hospital within Kumasi under the Atwima Nwabiagya North Municipal, serves a population of 155,025. It provides general, obstetric/gynecological, and surgical services.

Sampling technique

In the three selected health facilities, the number of registered pregnant women in their second and third trimesters was 500. Of these, 210, 190, and 100 were found in sites A, B, and C respectively. Initially, with the assistance of a statistician, a sample size of 244 was calculated, including a 10% attrition rate to cater for potential non-response due to the nature of the sampled population. The determination of this sample size was conducted through a mathematical formula, [30], in which n is the sample size, N is the total number of registered pregnant women who were in their second and third trimesters in the three (3) selected health facilities, and e is the margin of error. The calculated sample size was proportionally allocated based on the number of registered pregnant women in each facility. Therefore, 103, 93 and 48 from sites A, B and C respectively. Participants were selected using systematic random sampling. The sampling interval (Kth) was calculated as the ratio of registered pregnant women (N_i) to the allocated sample size (n_i) for each facility.

• Site A: N_A = 210, n_A = 103, K_A = 210/103 ≈ 2.

• Site B: N_B = 190, n_B = 93, K_B = 190/93 ≈ 2.

• Site C: N_C = 100, n_C = 48, K_C = 100/48 ≈ 2.

A uniform interval of every 2nd woman was used for simplicity. A random starting point (1 or 2) was chosen for each facility using a random number generator. Starting from this point, every 2nd eligible pregnant woman with a scheduled antenatal care visit (February–March 2024) was selected until the allocated sample size was reached.

Inclusion/exclusion criteria

Pregnant women in their second/third trimesters who consented to participate in the study were selected. Pregnant women in their first trimester, those who did not consent to participate in the study or did not visit the facilities for ANC at the time of the study, and those who had known mental problems were excluded from the study. Ultimately, the actual sample size used in the final analysis was 226 after dropping six missing values from the data from a response rate of 95.1%.

Data collection tools and procedures

The data were collected using computer-assisted personal interviewing (CAPI) addressing study respondents’ socio-demographic and obstetric characteristics. Common mental disorders (outcome of interest) among the respondents were assessed using the standardized World Health Organization’s (WHO) Self-Reporting Questionnaire (SRQ-20) widely used to screen for mental health disorders in primary care and community settings. The SRQ-20 is a 20-item questionnaire comprising yes/no inquiries concerning the occurrence of depressive, anxiety, and somatic symptoms within the past 30 days. The questionnaire was translated from English to the local language (Twi) using language experts. Interviewer-administered questionnaire downloaded on a Tablet that covered the questionnaire’s purpose, item-by-item explanations, culturally sensitive communication, and techniques to minimize bias during interviews. A pilot test with 20 pregnant women at a separate facility (site D) in Kumasi confirmed the translation’s reliability (Cronbach’s alpha = 0.71) and comprehensibility, whilst assessing completion time and the need for adjustments. Pilot-tested questionnaires were exempted from the final analysis. Most of the sampled population were interviewed during their scheduled antennal care (ANC) visits at the health facilities.

Variables and measures

Outcome variable

Common mental disorder status was the outcome variable. This was assessed using the WHO’s standardized Self-Reporting Questionnaire (SRQ-20). The SRQ-20 is a screening tool that has been validated and used for common mental disorders in several countries [31–33] including Ghana [34] and was designed to enhance and improve early detection of CMDs in primary health settings in LMICs [35]. The SRQ is composed of twenty yes/no items asking about depressive, anxiety, panic, and somatic symptoms during the preceding 30 days. Each 20 item is scored 0 or 1. A score of 1 indicates that the symptom was present during the past month and a score of ‘’0’’ signifying that the symptom was absent. The SRQ has been proven to be more effective and requires no trained personnel as a lay health worker can administer it [36]. A cut-off point of 7/8 (7 ‘yes’s’ a non-case and 8 ‘yes’s’ a case) is the most commonly used cut-off point in developing countries and has been used extensively in the community and primary care settings [32, 37]. Multiple research shows it has good internal reliability in primary care settings in LMICs, especially in sub-Saharan Africa [35, 36, 38]. The reliability of the SRQ 20 was assessed by calculating the scale’s Cronbach’s alpha to measure internal consistency. Internal consistency was good with a Cronbach’s alpha of 0.71 [36]. Therefore the SRQ seemed to be a suitable tool for this study. Respondents who answered “Yes” to six or more of the twenty questions were categorized as common mental disorder cases or otherwise if less. Existing studies validated this criterion [39, 40].

Exposure variables

Based on extant literature [3, 7, 39–41] thirteen explanatory variables were selected for the analyses. These variables include age (< 25 years, 25–29, 30–34, 35–39, and 40 or above), level of education (no formal education, basic, secondary, and tertiary), place of residence (rural, and urban), employment status [unemployed, employed (formal), employed (informal)], ethnicity (Akan, Dagomba, Ewe, and other), parity (0–1, 2, 3, 4 and above), gestational weeks (2nd trimester, and 3rd trimester), body mass index [BMI: Underweight (< 18.5 kg/m²); Normal (18.5–24.99 kg/m2); Overweight (25.0–29.99 kg/m2); Obese (> = 30 kg/m2)], frequency of fruit consumption (daily, weekly, and monthly), engage in exercise (yes or no), hypertension status [Normal: (systolic pressure < 140mmHg or diastolic < 90mmHg); hypertensive: systolic pressure > 140mmHg or diastolic > 90mmHg), diabetes status (yes or no), experience of motion sickness (yes or no), and experience of caesarean section in the past pregnancy (yes or no).

Statistical procedure and analysis

Descriptive analyses summarized the distribution of background characteristics and Common Mental Disorder (CMD) prevalence among pregnant women. CMD prevalence was calculated as a proportion with its 95% confidence interval (CI) using the normal approximation. Chi-square tests assessed associations between explanatory variables and CMD, with proportions reported across selected characteristics.

Furthermore, bivariate and multivariate Firth logistic regression models evaluated associations between explanatory variables and CMD, adjusting for confounders selected based on literature and bivariate significance. A multicollinearity test confirmed no multicollinearity (mean VIF = 9.46), ensuring robust model estimates. In our dataset, small cell counts in certain categories, such as the low prevalence of common mental disorders (CMD) among specific subgroups (e.g., only 3 CMD cases among 7 women with diabetes), resulted in zero odds or sparse data issues when using standard logistic regression. This led to unstable or undefined odds ratio estimates, as the model struggled to converge. In such cases, regrouping or recategorizing variables often fails to eliminate convergence problems [42]. To address this, Firth logistic regression (firthlogit) was employed, as it applies a penalized likelihood approach to produce unbiased parameter estimates in the presence of sparse data or small cell counts. Two models were fitted. In the first model, we looked at the association between each explanatory variable and the outcome. The second model was fitted to account for the effects of each explanatory on the outcome. A multicollinearity test was performed on each explanatory variable, and the results showed a mean-variance inflation factor (VIF) of 9.46, showing the nonexistence of multicollinearity. Using a 95% confidence interval, the odds ratios for each variable were determined. The data were analyzed using Stata (Version 17) (Corp, College Park, TX, USA). We were guided by the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline in presenting the results [43].

Results

Descriptive results

Prevalence of Common Mental Disorder (CMD) among pregnant women by background characteristics

Of the 226 respondents, 27 (12.0%, 95%CI: 8.30–16.90) were identified as having a common mental disorder (CMD). Among age groups, those aged 35–39 exhibited the highest proportion of CMD at 17.8%, followed closely by those under 30–34 at 14.5%. Regarding educational level, the highest proportion of CMD was observed among those with basic education (20.0%), while the lowest proportion was among those with tertiary education (5.7%). In terms of marital status, the highest proportion of CMD was found among never-married individuals (18.0%). Women with a normal body weight had a significantly higher prevalence of CMD (22.2%) than those who were obese (4.4%) (p = 0.002). Similarly, women living with diabetes had a significantly higher prevalence of CMD (42.9%) (p = 0.010), as did individuals with motion sickness (28.6%) (p = 0.014) (Table 1).

Table 1.

Distribution of common mental disorder (CMD) among pregnant women by background characteristics and other key independent variables

Variables	Sample (n(%)	CMD cases	Chi-square(X2): p-values
		Frequency[n(%)]
Age (years)			3.9958; 0.407
< 25	21 (9.3)	1(4.8)
25–29	73 (32.3)	6(8.2)
30–34	76 (33.6)	11(14.5)
35–39	45 (19.9)	8(17.8)
40 and above	11 (4.9)	1(9.1)
Highest educational level			5.7372; 0.125
No formal education	5 (2.2)	14(18.4)
Basic	76 (33.6)	1 (20.0)
Secondary	92 (40.7)	9 (9.8)
Tertiary	53 (23.5)	3(5.7)
Employment status			5.1476; 0.076
Unemployed	27 (12.0)	2 (3.6)
Employed (Formal)	55 (24.3)	22 (15.3)
Employed (Informal)	144 (63.7)	3 (11.1)
Parity			4.7202; 0.094
0–1	91 (40.3)	6 (6.6)
2	67 (29.6)	9 (13.4)
3 or more	68 (30.1)	12 (17.7)
Body mass index			14.6510; 0.002
Underweight	73 (32.3)	16 (21.9)
Normal	18 (8.0)	4 (22.2)
Overweight	67 (29.6)	4 (6.0)
Obese	68 (30.1)	3 (4.4)
Engage in exercise			1.7216; 0.189
No	83 (36.7)	13 (15.7)
Yes	143 (63.3)	14 (9.8)
Diabetes status			6.5610; 0.010
No	219 (96.9)	24 (11.0)
Yes	7 (3.1)	3 (42.9)
Experience motion sickness			6.0824; 0.014
No	205 (90.7)	21 (10.2)
Yes	21 (9.3)	6 (28.6)

*p<0.05, **p<0.01, n: number of observations, All p-values are two-sided

Bivariate and multivariate firthlogit regression analysis of risk factors associated with CMD

The binary logistic regression analysis presented in Table 2 examines the factors associated with Common Mental Disorder (CMD) among pregnant women. Model I displays the crude odds ratios (COR) along with 95% confidence intervals (CI), while Model II demonstrates the adjusted odds ratios (aOR) after controlling for various demographic and health-related variables. Parity, body mass index, diabetes and motion sickness emerged as significant predictors of CMD among pregnant women. Notably, women with parity levels of 3 or more children exhibit higher odds of CMD [COR = 2.91; 95%CI:1.06–7.95] compared to those with less than two children, however, this significant association was attenuated after adjusting for other factors. Additionally, women who were overweight and obese had a significantly lower risk of CMD [overweight: aOR = 0.17; 95%CI:10.03–0.96, obese: aOR = 0.11, 95%CI:0.02–0.66]. Moreover, women who were living with diabetes were at 8.59 times greater risk of CMD, highlighting the importance of considering comorbidities such as diabetes in mental health assessments. Lastly, motion sickness showed a significant positive association with CMD [COR = 3.60; 95%CI:1.30–9.97], however, the association receded after accounting for other factors in the model (Table 2).

Table 2.

Bivariate and multivariate Firthlogit regression analysis of associated factors of common mental disorder (CMD) among pregnant women

Variables	Model I		Model II
	COR (95% CI)	P-value	aOR (95% CI)	p-value
Age (years)
< 25	0.51[0.05–5.55]	0.582	0.66[0.03–14.11]	0.792
25–29	0.67[0.10–4.47]	0.683	1.82[0.18–18.61]	0.613
30–34	1.23[0.20–7.60]	0.824	1.33[0.15–11.61]	0.794
35–39	1.59[0.24–10.25]	0.628	2.47[0.32–18.98]	0.383
40 and above	1.00		1.00
Parity
< 2	1.00		1.00
2	2.14[0.75–6.11]	0.157	1.96[0.52–7.43]	0.323
3 or more	2.91[1.06–7.95]	0.037	2.07[0.43–9.99]	0.365
Body mass index
Underweight	0.92[0.28–3.04]	0.897	0.56[0.11–2.89]	0.488
Normal	1.00		1.00
Overweight	0.23[0.05–3.04]	0.042	0.17[0.03–0.96]	0.044
Obese	0.17[0.04–0.78]	0.022	0.11[0.02–0.66]	0.016
Monthly	1.00		1.00
Yes	6.21[1.44–26.69]	0.014	8.59[1.41–52.24]	0.020
Diabetes status
No	1.00		1.00
Yes	6.21[1.44–26.69]	0.014	8.59[1.41–52.24]	0.020
Experience motion sickness
No	1.00		1.00
Yes	3.60[1.30–9.97]	0.014	3.03[0.88–10.36]	0.078

*p < 0.05, **p < 0.01; n: number of observations; All p-values are two-sided; COR: crude odds ratios; aOR: adjusted odds ratios; n: sample size

Discussion

Common mental disorder during pregnancy is associated with poorer maternal health, increased odds of obstetric complications, preterm birth, and neonatal complications, which are of major public health concern that substantially affect women’s and children’s quality of life in low and middle-income countries [44–46]. Therefore, examining the nuances of common mental disorders and their associated factors is of great value to public health and clinical practice to improve obstetrical outcomes and fetal well-being. Our study investigated the prevalence and risk factors of common mental disorders (CMD) among 226 pregnant women in primary antenatal care settings in Kumasi, Ghana. We found a CMD prevalence of 12.0% and firth logistic regression identified two significant risk factors: body mass index (BMI) and diabetes.

This study found a CMD prevalence of 12.0% (95% CI: 8.3–16.9%) among 226 pregnant women attending antenatal care in primary care settings in Kumasi, Ghana. Compared to other studies, our prevalence is lower than the 18.1% (95% CI: 15.5–21.0%) reported in Northeast Ethiopia [39] and the 35.8% (95% CI: 34–38%) in Southeast Ethiopia [40].This similarity may be attributed to shared characteristics such as predominantly urban populations. In contrast, the non-overlapping CI with the Southeast Ethiopia study likely reflects substantive differences in study populations and methodologies. For instance, the Southeast Ethiopia study reported a higher CMD prevalence, which has been linked to a higher proportion of rural residents, greater food insecurity, and limited access to mental health services [40].

Our findings indicate a striking association between diabetes and CMD, with women living with diabetes being 8.59 times more likely to experience CMD. This aligns with studies indicating that chronic illness can be a significant stressor, leading to higher rates of CMD [40, 47] among pregnant women. A meta-analysis encompassing 16 observational studies from LMICs found that women with GDM had nearly twice the odds of experiencing perinatal depression compared to those without GDM (pooled OR 1.92; 95% CI 1.24–2.97) [48]. Similarly, research from Malaysia indicated that nearly two-fifths of women with GDM experienced anxiety symptoms, and one-tenth had depressive and stress symptoms [49]. Managing diabetes demands rigorous, ongoing self-care routines, such as regular blood sugar monitoring, adherence to dietary restrictions, and consistent medication management. These demands can become particularly burdensome during pregnancy, which can be overwhelming and exacerbate mental health issues during pregnancy [50]. Also, the hormonal fluctuations associated with pregnancy can complicate glycemic control, introducing additional stress for women with diabetes [51].

Interestingly, our results indicate that overweight and obese women had a lower risk of CMD, contrary to the general trend where higher BMI is often associated with greater psychological distress among pregnant women compared to those with normal body weight [52, 53]. This divergence might suggest pregnancy-specific dynamics, where increased BMI does not carry the same mental health burden, possibly due to different societal attitudes or physiological factors during pregnancy. This aligned with study in Brazil [54], which linked lower BMI to depression and anxiety. Similarly, a previous study of over 50,000 Danish adults also showed that obesity-related genetic variants (FTO rs9939609, MC4R rs17782313) were linked to reduced psychological distress [55]. Despite obesity’s link to diabetes, they affect CMD through distinct pathways. Obesity may be protective in Ghana, particularly Kumasi, due to cultural valuation of larger body sizes and pregnancy-related factors like normalized weight gain [56]. Conversely, diabetes heightens CMD risk via chronic stress from management and metabolic effects like hyperglycemia [50]. Our research has demonstrated a need to improve both physical/chronic and mental health aspects of all Ghanaian pregnant women accessing antenatal care. Given the observed association of CMD with maternal health conditions such as diabetes, the inclusion of standardised psychiatric assessments in the ANC protocols could facilitate early detection and treatment of psychiatric distress. This would allow timely intervention, reduce adverse outcomes in mothers and newborns and align with global recommendations for comprehensive maternal care.

Implications for practice and further research

These findings suggest several practical implications for the management and support of pregnant women. The strong association between diabetes and CMD highlights the need for an integrated care pathway that simultaneously addresses physical and mental health to improve obstetrical outcomes. Considering the mitigated risk of CMD with higher BMI, healthcare providers should be cautious not to assume increased psychological distress based solely on BMI during pregnancy. Finally, the nuanced role of parity and transient health conditions like motion sickness in predicting CMD calls for personalized approaches to mental health screening and intervention during pregnancy. In terms of research, future studies should explore the mechanisms underlying the relationships observed in this study, particularly how sociodemographic and systemic health factors interact to influence CMD risk. Further research might explore whether those findings reflect protective psychosocial factors or different stress responses in overweight or obese pregnant women. Longitudinal designs would help clarify the directionality of these associations, improving our understanding of CMD’s development during pregnancy and informing targeted interventions to support at-risk groups.

Strengths and limitations

This study highlights the urgency to acknowledge the prevalence of CMD and its associated factors among pregnant women in the primary antenatal healthcare setting in Kumasi, which has several implications for public health and clinical practice. Despite providing valuable insights, our study has limitations that must be considered by readers in the of course interpreting the findings. This study may be subject to selection bias, as participants were recruited from public primary care facilities and may not represent all pregnant women in the region. Additionally, despite adjusting for known confounders, the possibility of residual confounding from unmeasured factors remains. These limitations should be considered when interpreting the findings. The cross-sectional design restricts our ability to infer causality. Longitudinal studies are needed to determine the temporal relationship between these factors and common mental disorders. Additionally, measuring common mental disorders based on a 20-item questionnaire which was self-reported data might introduce recall and social desirability bias, particularly in the reporting of psychological symptoms and health conditions. Furthermore, wide confidence intervals for some estimates, such as the association between diabetes and CMD (aOR = 8.59; 95% CI: 1.41–52.24), reflect uncertainty in the effect size, likely due to the small number of participants with diabetes (n = 7). This limits the precision of these findings and suggests caution in generalizing the strength of such associations.

Conclusion

Modern obstetric care providers in primary healthcare settings should be aware of the association between the risk of chronic physical conditions and mental disorders in pregnancy and the increasing trend of each diagnosis. This study elucidates several important factors associated with the occurrence of common mental disorders among pregnant women, emphasizing the complexity of these associations. The significant link between diabetes and increased common mental disorder risk underscores the necessity for comprehensive care strategies that address both physical and mental health needs concurrently, by integrating chronic conditions like diabetes care into obstetrical care. The unexpected association between higher BMI and lower common mental disorder risk during pregnancy calls for a nuanced understanding of mental health dynamics in this unique population. While some factors like parity and motion sickness lost their significance after adjustment, their initial associations suggest that multiple interacting variables contribute to common mental disorder outcomes. Overall, this study contributes valuable insights into the multifactorial nature of common mental disorders during pregnancy, indicating specific areas for intervention and further research to optimize health outcomes for pregnant women. Future research should focus on intervention-based approaches to develop and evaluate culturally sensitive mental health services tailored to the needs of pregnant women in Ghana and similar settings.

Abbreviations

ANC

Antenatal care

aOR

Adjusted odd ratio

BMI

Body mass index

CHAG

Christian Health Association of Ghana

CMD

Common mental disorder

COR

Crude odds ratios

GSS

Ghana Statistical Service

SRQ-20

Self-Reporting Questionnaire

WHO

World Health Organization

Authors’ contributions

S.F.A. conceptualized the idea for the topic, collected the data, and drafted the original manuscript and subsequent revisions. S.A. and P.B. (research supervisors) guided the conceptualization of the research project and data collection and reviewed the manuscript and subsequent revisions. All authors read and approved the final manuscript.

Funding

This research was conducted as part of a doctoral thesis in Nursing and was made possible through a HEARD PhD scholarship to SFA at the University of KwaZulu-Natal (UKZN), funded by the Swedish International Development Agency (SIDA). Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of HEARD, UKZN or SIDA.

Data availability

Data is provided within the manuscript.

Declarations

Ethics approval and consent to participate

This study was approved by the University of KwaZulu-Natal’s Biomedical Research Ethics Committee, South Africa (BREC/00006040/2023), and the Ghana Health Service Ethics Review Committee (GHS-ERC:001/11/23). Permission was also granted by the administrative committees of the three selected health facilities. The purpose of the study was explained to pregnant women to inform their decision to participate or otherwise and questions that emerged were clarified. Pregnant women who agreed to participate in the study signed an informed consent form (English or Twi - a local language, depending on their preference). Interviewer-administered questionnaires (uploaded on a Tablet) were used to collect the data as the principal investigator is fluent in both languages. To ensure privacy and confidentiality, interviews were conducted in a separate room in the antenatal department after respondents had consented to participate in the study. Data were deidentified to avoid tracing information back to the respondents.

Consent for publication

The consent for data collection also included permission for the publication of the respondents’ anonymized data.

Competing interests

The authors declare no competing interests.