Sociocultural and health system determinants of maternal mortality in Ghana: findings from a systematic review

Abstract

Background

Maternal mortality remains a significant public health concern that has received sustained global attention and substantial investment over the past few decades. However, this problem is high in Ghana, with most deaths attributed to direct or indirect causes and very little to sociocultural and system issues. This systematic review sought to map the multifaceted or related risk factors and causes of maternal mortality among women in Ghana.

Methods

We conducted a comprehensive systematic search of PubMed, CINAHL, and EMBASE to identify studies published between 2000 and 2023, using the WHO definition of maternal mortality as a guide. The review adhered to the Joanna Briggs Institute’s (JBI) guidelines for systematic reviews and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) framework. The methodological quality of included studies was appraised using the Mixed Methods Appraisal Tool (MMAT).

Results

Nineteen (19) studies met the inclusion criteria. The synthesis revealed that maternal mortality in Ghana is primarily driven by direct causes such as hemorrhage and sepsis, and indirect causes including anemia and sickle cell disease. Sociocultural determinants, such as place of delivery, geographical location, cultural beliefs, lack of support, and poverty, further shaped maternal outcomes. Health system factors, including inadequate antenatal care, shortages of blood and essential medications, poor supervision and documentation, weak referral systems, and delays in care, were also critical contributors.

Conclusion

This review underscores that maternal mortality in Ghana is driven not only by direct medical causes but also by systemic, socioeconomic, and cultural barriers. Addressing these requires strengthening emergency obstetric care, improving referral and data systems, and ensuring consistent provider training in managing obstetric emergencies. Equally important are policies that reduce sociocultural and financial barriers and promote community-based maternal care. These actions are critical to accelerating progress toward achieving national and global maternal health targets.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-025-25134-5.

Background

Maternal mortality remains a significant public health concern that has received sustained global attention and substantial investment over the past few decades [1]. Although the relentless quest to mitigate the problem has yielded a consistently dramatic decline, low and middle-income countries continue to record the highest rates of maternal mortality worldwide. In 2020, the World Health Organization (WHO) indicated that sub-Saharan Africa (SSA) and Southern Asia accounted for almost 87% (253,000) of the total estimated maternal fatalities worldwide. SSA alone accounted for around 70% of global maternal deaths (202,000), while Southern Asia accounted for 16% (47,000) [2]. At the same time, Eastern Europe and Southern Asia experienced the highest overall reduction in the Maternal Mortality Ratio (MMR) between 2000 and 2020, with decreases of 70% (from 38 to 11) and 67% (from 408 to 134), respectively.

The high number of maternal deaths in some parts of the world reflects the inequalities in accessing quality health services. It highlights the continuous existing gap between developed and developing countries [2].

Despite recording a very high MMR in 2020, SSA also achieved a 33% reduction in MMR between 2000 and 2020 [2]. Maternal mortality, as defined by the World Health Organization (WHO) in the International Statistical Classification of Diseases and Related Health Problems, 11th edition (ICD-11), refers to “the death of a woman while pregnant or within 42 days of the end of pregnancy, irrespective of the duration and site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management but not from accidental or incidental causes” [3]. It is against this milieu that the United Nations reaffirmed its commitment to achieving a global maternal mortality ratio of less than 70 per 100,000 live births by 2030 in Sustainable Development Goal 3.1 [4].

In Ghana, the MMR has gradually declined due to several factors, such as the robust systems put in place to ensure safe pregnancy and delivery [5]. The Ghana Health Service (GHS) reported that a total of 875 maternal deaths occurred in 2018 and 838 in 2019. The number decreased to 776 in 2020, despite an increase in total births and a decrease in the institutional MMR from 117 in 2019 to 106 in 2020 during the COVID-19 pandemic and its accompanying effects [5]. Also, at the 2022 National Health Sector Annual Summit, the Ghana Ministry of Health (MOH) reported that Ghana’s MMR stood at 310 per 100,000 live births. Furthermore, the institutional maternal mortality rate had risen from 109.2 per 100,000 live births to 119.5 in 2021, reflecting a 9.4% increase [6, 7]. Additionally, the Central Intelligence Agency (CIA) in 2023 ranked Ghana 34th out of 186 countries with the highest incidence of maternal mortality, recording about 263 deaths per 100,000 live births [8].

Maternal death trends in Ghana have been consistent over the past two decades, with most deaths attributed to direct or indirect causes. Rural areas are more affected than urban areas [1], largely due to the low number of skilled birth attendants in these communities [1]. Direct factors contributing to maternal mortality were identified as obstetric complications (such as obstetric hemorrhage or hypertensive disorders in pregnancy, sepsis, obstructed labor, complications of anesthesia or caesarean section), medical interventions, omissions, and incorrect treatment [1]. Postpartum hemorrhage (PPH) is the primary cause of maternal death resulting directly from childbirth, with eclampsia being the second most common cause [1]. Besides, research in 2012 listed the five direct leading causes of maternal mortality as postpartum hemorrhage (22.5%), abortion-related causes (19.3%), hypertensive disorders in pregnancy (17.8%), puerperal sepsis (8.9%), and anemia (7.3%), constituting about 75% of total maternal deaths [9]. Indirect causes include malaria, sickle cell anemia, HIV/AIDS, and Tuberculosis (TB) [1, 10]. Inasmuch as direct and indirect causes of maternal mortality increase the chances of maternal and infant mortality, research shows sociocultural factors serve as facilitators and barriers to maternal healthcare utilization, consequently leading to maternal and infant deaths [11]. Socio-cultural determinants refer to social, cultural, environmental, and community factors, as well as living and working conditions, that can either support or hinder maternal health [12]. Evidence further revealed that sociocultural factors such as competition for more children, food taboos, blood transfusions, antenatal attendance, poverty, and poor transport systems are contributing factors to maternal deaths [1, 10]. This is backed by Thaddeus and Maine’s framework, which suggests that factors like these often result in delays in hospital arrival, decision-making, and treatment, leading to maternal mortality [13].

While there is a growing body of literature that has explored the trends, prevalence, causes, and risk factors of maternal mortality in Ghana, the gaps in understanding its persistence remain. A comprehensive review of how these causes are distributed across different populations and regions within Ghana is lacking. Additionally, existing studies often focus on either medical causes [1, 9] or sociocultural factors [1, 10–12] in isolation without fully exploring how these elements interact and compound each other in specific local contexts. There is also limited insight into the barriers to the effective implementation of maternal health interventions, particularly in rural areas where access to skilled birth attendants and healthcare services is limited [12, 14–16].

Moreover, although direct and indirect causes of maternal mortality have been well-documented, the role of sociocultural determinants such as community beliefs, transportation issues, and access to antenatal care requires deeper exploration [17, 18]. Research on the persistence of maternal mortality in the face of government policies and healthcare improvements remains insufficient, particularly regarding the impact of health system deficiencies [10, 14, 18]. Understanding the distribution and interaction of these risk factors across different demographic and geographic groups in Ghana is essential for addressing the persistent maternal mortality rate and achieving the country’s health goals. Therefore, this review aims to systematically map the causes and risk factors contributing to maternal mortality among women in Ghana.

Methods

This systematic review was conducted following the Joanna Briggs Institute’s (JBI) guidelines for systematic reviews [19] and is reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) checklist [20]. The review question for this study was guided by the Population, Exposure and Outcome (PEO) mnemonic, a framework for systematic reviews of aetiology and risk [19]. The population comprised deaths occurring among women during pregnancy or within 42 days post-pregnancy in Ghana. The exposure was the causes and risk factors, and the outcome was maternal mortality. This culminated in the question, “What are the causes and risk factors contributing to maternal mortality among women in Ghana?”

Eligibility criteria

This review included articles that met the following criteria listed in Table 1.

Table 1.

Inclusion and exclusion criteria

Criterion	Inclusion	Exclusion
Publication type/Study Design	Primary research studies with findings on maternal mortality as per the WHO definition. Studies that employ quantitative descriptive cross-sectional, qualitative, mixed methods, or case studies reporting or addressing the causes or risk factors of maternal mortality.	All secondary studies, opinions and conference papers, abstracts, and editorials. Studies pertaining to developing theories, conceptual frameworks, or study protocols. Studies that lacked abstracts or full texts.
Language and Date	All studies published in English between 2000 and 2023.	Studies published in other languages other than- English were excluded.

Search strategies

A comprehensive systematic search was conducted in PubMed, CINAHL, and EMBASE to identify studies on the causes and risk factors of maternal mortality among women in Ghana. The search strategy was developed in consultation with an informatician from the University of Health and Allied Sciences (UHAS), Ho, Volta Region, Ghana. Keywords included “maternal mortality,” “maternal death,” “sociocultural determinants,” “health system factors,” and “Ghana.” The initial search was first conducted in PubMed. To identify keywords and index terms, we analyzed the words in the titles and abstracts of the studies. Secondly, using the identified keywords and index terms, a general search across the databases was performed. Thereafter, an individual search strategy was developed for each database, considering the differences in Thesaurus terminology and indexing. A detailed search strategy for PubMed is attached as an additional file 1. Lastly, to ensure comprehensiveness, the reference lists of all included studies were manually screened to identify additional relevant literature not captured in the database search.

Study selection

The search results were uploaded to EndNote 20, a reference manager, for screening and duplicate removal. After duplicate removal, two independent reviewers (DBA and FK) screened the titles and abstracts of the studies. For studies that met the inclusion criteria, these independent reviewers (DBA and FK) retrieved and screened the full-text articles. Whenever disagreements arose between the two reviewers, a third reviewer (AA) resolved them through further adjudication.

Assessment of methodological quality

The quality of the included studies were evaluated using the Mixed Methods Appraisal Tool (MMAT) [21]. Two separate reviewers (DBA and FK) conducted the critical assessment; disagreements were discussed and settled by another review author (AA). The MMAT includes five questions that are based on the study design, and two screening questions concerning the suitability of the data collection technique and the clarity of the research questions. The standards for a non-randomized, quantitative study design were used. The requirements for a study include representative participants, suitable measurements, comprehensive outcome data, accounting for confounders, and proper exposure. The reviewers assessed each criterion as ‘yes’, ‘no’, or ‘can’t tell’, and the total quality score was not computed as recommended by Hung et al. [21].

Data extraction

Using the standardized JBI data extraction method, two reviewers (DBA and FK) extracted the data independently, and a third reviewer (AA) validated the results [19]. The extracted information included the author, year, setting, the study’s objectives, design, sample, etiology, and risk factors of maternal mortality. Two reviewers (DBA and FK) examined each article independently using the data extraction template, discussed the charted data, and updated the data extraction sheet as necessary. A third reviewer (AA) led the team to resolve discrepancies and disagreements through further adjudication.

Data synthesis

Meta-analysis was not possible since the included studies differed significantly regarding variables examined, analytic technique, and comparison groups. As a result, the authors synthesized the quantitative data narratively, combining related risk variables and presenting them in tables. A convergent integrated method was utilized to convert the data into a narrative format, as the extracted data were derived from both quantitative and qualitative studies [22]. Following JBI’s recommendation, we qualitized quantitative data for data transformation since this is less likely to be wrong when codified than when qualitative data is given numerical values [22, 23]. Qualitizing involves extracting data from quantitative investigations, transforming it into textual descriptions for integration with qualitative data, and offering a narrative interpretation of the quantitative findings. Following the convergent synthesis of the modified data, the reviewers conducted a thorough assessment of the gathered data to select groups based on semantic similarity.

Results

Search results and study characteristics

The initial search through the electronic databases yielded 953 records. Duplicate records (105) were removed, and then the titles and abstracts were screened. Following the title and abstract screening, 40 articles were screened for eligibility. After full-text screening, 19 articles were included in this review. Of the 19 studies included in this review [1, 9, 10, 12, 14–18, 24–33], 14 used quantitative study designs [1, 9, 14, 15, 17, 24–32], four used mixed method designs, and only one was a case study (Fig. 1). The cumulative total sample size of maternal mortalities reported for the studies reviewed was 3672. Three were nationwide studies [16, 25, 32], four from the Northern Region [12, 15, 17, 24], and two were conducted in the Upper West, Eastern, and Greater Accra regions [1, 9, 18, 26, 27, 29], respectively. Again, a study each from the Upper East, Volta, Western, and Brong Ahafo regions [10, 28, 31, 33] was included in this review. Seven of these studies were conducted in the Northern/Savannah and coastal/southern belt [1, 10, 12, 15, 17, 24, 26–29, 31, 33], respectively. Three were multizonal studies [16, 25, 32], and two were from the middle belt [9, 14, 18]. While most of the deaths [11] reported for the review studies were from urban areas [1, 9, 10, 14, 15, 17, 18, 26, 27, 30, 31], five were from rural communities [12, 24, 28, 29, 33] (see Table 2).

Fig. 1.

Prisma flow diagram

Table 2.

Characteristics of included studies

Author (year)	Study location	Study type	Population	Percentage (MMR)	Key findings
Nsor-Anabiah, (2019) [17]	Tamale Teaching Hospital	Quantitative study	A total of 272 women died out of 61,368 who came on admission to the facility.	439.7	Direct or medical causes: • Eclampsia • Sepsis • Severe anaemia • Pneumonia • Unsafe abortion Indirect causes • Malaria • HIV/AIDS • CSM • Heart failure • Severe anaemia • Sepsis • Pregnant multiple times Socio-cultural factors • Multiple pregnancies as proof of fertility • Belief that real women give birth at home and only go to a hospital in emergencies. • Resistance to Caesarean Section • Poor road network • Poverty • Food taboos • Reluctance to accept blood transfusion
Galaa et al. (2016) [1]	Wa Regional Hospital	Quantitative study	73 maternal mortality cases at the Wa Regional Hospital from January 1, 2005, to December 31, 2010	520.4	Direct causes: (60% of the deaths) • Haemorrhage (19%) • Eclampsia (15%) • Sepsis (11%) • Abortion related difficulties (8%) • Obstructed labour (7%) Indirect cause • Already existing ailments • Malaria • Hepatitis • CSM • Anaemia • HIV/AIDS and TB • Cardiac arrest • Hypovolaemic shock • Thyroid toxicosis • Organ failure • Heart attack Socio-cultural factors • Pressure in polygamous homes to have more children • Indiscriminate use of herbal intoxication • Belief in spiritual interference • Seeking deity’s permission before seeking healthcare • Prestige in having many children • Belief that menopause is the only morally accepted contraceptive • Home delivery to prove fidelity • Food taboos • Poverty • Opposition to blood transfusion
Apanga (2018) [10]	Upper East Region of Ghana	Case study	One case	Not stated	Major contributory factors to maternal deaths in Ghana • Delay in seeking medical care • Poorly resourced nature of health facilities makes it difficult to carry out medical and laboratory tests • Lack of logistics and equipment • Using oral quinine as the first drug of choice for uncomplicated malaria in the third trimester of pregnancy
Jakperik et al. (2023) [16]	Nationwide	Quantitative study	A sample of 1,052 women was selected from all ten administrative regions of Ghana in which 188 maternal deaths occurred.	(17.87%).	• Age at death • Marital status • Season(rainy/dry) • Region, • Place of death • Place of residence • Religion • Ethnicity • Level of education
Ganyaglo et al. (2012) [9]	Eastern Regional Hospital	Quantitative study	A total of 191 maternal deaths over the period out of the total of 19,965 live births	957	Direct Causes (constituting about 75% of total maternal deaths). • Postpartum hemorrhage (22.5%) • Abortion-related causes (19.3%) • Hypertensive disorders in pregnancy (17.8%) • Puerperal sepsis (8.9%) • Anemia (7.3%)
Asamoah et al. (2011) [25]	Nationwide	Quantitative study	A total of 605 maternal deaths	Not stated	Key findings • Haemorrhage was the highest cause of maternal mortality (22.8%). • Married women had a significantly higher risk of dying from haemorrhage, compared with single women. • Women aged 35–39 years had a significantly higher risk of dying from haemorrhage. • The risk of dying from abortion decreases as maternal age increases. • The risk of dying from infectious diseases was high in younger women. • Malaria is the single highest cause of maternal deaths • Risk of maternal deaths due to miscellaneous causes increased with age.
Issah, (2011) [29]	Upper West Region	Quantitative study	47 maternal deaths were reported and audited		Medical causes of death • Ruptured uterus • Sepsis • Hemorrhage • Eclampsia • Meningitis • Anesthetic complications • Chronic liver disease • Unsafe abortion • HIV/AIDS • Thyrotoxicosis • Gastric perforations • Fetal retention syndrome • Severe pneumonia • Sickle cell crisis • Acute asthmatic attack Other causes of maternal mortality • Inadequate quality of care and emergency response systems • Non-adherence to treatment protocols • Poor referral systems in the hospitals
Lee et al. (2012) [30]	Ahanti Region, Komfo Anokye Teaching Hospital,	Quantitative study	322 with 30,269 livebirths in the period 1 January 2008 to 30 June 2010	1004 per 100,000 livebirths	Major causes • Hypertensive states of pregnancy (26.4%) • Hemorrhage (16.8%) • Genital tract sepsis (10.6%) • Early pregnancy deaths (8.4%) • Infections (8.7%)
Geelhoed et al. (2003) [28]	Brong-Ahafo Region	Quantitative study	229 maternal deaths were identified.	1077 per 100,000 live births.	Direct causes (156/229 deaths, 68.1%) • Abortion complications (18.8% of all deaths). • Sepsis (34/229 deaths, 14.8%) • Hemorrhage (27/229 deaths, 11.8%) Indirect (61/229 deaths, 26.6%) • Hepatitis (15/229 deaths, 6.6%) • Anemia (12/229 deaths, 5.2%) • Unknown causes (5.2%)
Gumanga et al. (2011) [15]	Northern region, Tamale Teaching Hospital	Quantitative study	280 maternal deaths from 1 st January 2006 to 31 st December 2010, 139 audited maternal deaths that occurred between 1 st January 2008 and 31 st December 2010	1018.7	Direct maternal deaths contributed 99 deaths (71.2%) Main causes: • Sepsis (19.8%) • Hypertensive disorders (18.6%) • Hemorrhage (15.8%) • Unsafe abortion (11.5%) • Obstructed labour Indirect causes contributed 40 deaths (29%) • Anemia (8.7%) • Sickle cell disease (5.7%) • Malaria (5.0%) • HIV • Viral Hepatitis & its complications • Meningitis Renal failure
Adusi-Poku et al. (2015) [14]	Ahanti Region, Komfo Anokye Teaching Hospital	Mixed method	33 maternal deaths were reviewed.	Not stated	Key contributors to maternal deaths • Poor documentation • Non-involvement of senior clinicians in the management of cases • Laboratory-related issues, particularly in relation to blood and blood products Promptness of care and adequacy of intensive care facilities and specialists in the hospital.
Boafor et al. (2021) [26]	Greater Accra Region	Quantitative study	276 maternal deaths, giving a maternal mortality ratio of 604 per 100,000 live births.	604	The leading causes of maternal death • Hypertensive disorders (37.3%), • Hemorrhage (20.6%), • Sickle cell disease (8.3%), • Sepsis (8.3%), • Pulmonary embolism (8.0%). • Women with no formal education, • Women who had less than four antenatal visits • Emergency cesarean sections
Owusu-Sarpong et al. (2017) [18]	Eastern Regional Hospital	Mixed method	A total of 43 maternal deaths occurred, out of which 37 (86%) were audited	Not stated	Leading causes of maternal death • Haemorrhage • Eclampsia and • Unsafe abortion Other causes of maternal deaths • Delays in reaching the regional hospital • Non-involvement of senior clinicians in the management of cases • Inadequate equipment • Lack of intensive care facility • Operational challenges related to laboratory and blood bank services
Okoh-Owusu et al. (2023) [31]	Western Region	Quantitative study	67 out of the 93 (73.1%)	Not stated	Causes of maternal deaths • Hypertensive diseases (35.8%) • Hemorrhage (31.3%) • Sepsis (7.5%) • Ruptured uterus • Sickle cell disease Other findings • Hypertensive diseases in pregnancy, hemorrhage, and sepsis account for about 75% of institutional maternal deaths in the Western Region of Ghana • 80% of deaths occurred among women who attended ANC at least three times during pregnancy.
Der et al. (2013) [27]	Greater Accra Region, Korle-Bu Teaching Hospital	Quantitative study	Of 5,247 deaths among women aged 15–49, 12.1% (634) were pregnancy-related.	Not stated	Direct obstetric causes contributing 79.5% of deaths (504) include • Hemorrhage (21.8%) • Abortion (20.8%) • Hypertensive disorders (19.4%) • Ectopic gestation (8.7%) • Uterine rupture (4.3%) • Genital tract sepsis (2.5%) Indirect obstetric causes contributing 20.5% of deaths (130) include • Infections outside the genital tract (9.2%) • Anemia in pregnancy (2.8%) • Sickle cell disease (2.7%) • Pulmonary embolism (1.9%) • Disseminated intravascular coagulation (1.3%)
Abdulai et al. (2014) [24]	Northern Region Tolon District	Mixed-methods study	Not stated	Not stated	Beliefs and factors that contribute to maternal mortality • ANC leads to difficulties in delivery, allergies, fears, cultural explanations of certain diseases, and cultural appropriation of women’s conduct. • Women who go to the hospital to deliver are prime suspects of infidelity. To avoid this suspicion, women refrain from going to the hospital to deliver, even if the labor is life-threatening. • Most Dagomba women believe that only their husbands have the right to see or access their private parts. • Women who eat eggs and fish are regarded as bad women by Dagomba culture. • Most rural women fear the hospital environment because it gives them bad memories. • Some rural women dread doctors and nurses because they think they possess certain powers that enable them to treat the sick. • The belief that when the hospitals/doctors carry out operations, they cut some parts of their bodies or steal these body parts for some purposes. • The belief that people suffering from diseases like anthrax, snake bites and boils will die when they come into contact with steel.
Sumankuuro et al. (2018) [32]	Nationwide	Quantitative study	1240 deaths occurred among women aged 15–49 years, out of which 164 were maternal deaths	The prevalence of maternal death was 13% (164)	• Age and marital status were significant determinants of maternal mortalities. • 98.7% of women whose death was maternal-related, completed less than four antenatal visits, 98.7%. • Women who had excessive bleeding during labour/delivery showed a significant association with maternal death. • The utilization of traditional/herbal and spiritual remedies during pregnancy was strongly associated with maternal mortalities.
Sarkodie et al. (2014) [12]	East Gonja District, Northern Region	Quantitative study	213 women in fertility age (WIFA), 15–49 years, consented to the study, 199 respondents were studied	Not stated	Medical Causes • Haemorrhage 16(35%) • Sepsis 4(9%) • Hypertensive disorders 5(11%) • Unsafe Abortion 5(11%) • Obstructed labour 4(9%) • Others like anemia, diabetes, liver diseases, malaria 12(27%) Socio-cultural Causes • Poor staff attitude (63%) • Non-attendance of ANC (60.3%) • Spouse support is lacking (47%) • Lack of transport (44.1%) • Delay in reporting to health facility for treatment (36.3%) • Refusing referral (32%) • Delivery Outside health facility (30%) • Negligence of health worker (24%) • Lack of money (22.1%) • lack of Support from relatives (16%) • Poor health service quality (13.7%) • Superstition/Customary practices around labour (12.5%) • TBA delivery (10%) • Use of local medicine in pregnancy and delivery (10%) • Inexperienced health workers 7.4%)
Cham (2017) [33]	South Tongu District Volta region	Quantitative study	A total of fifty-four (54) community members took part in the study.		Causes of Maternal Mortality • Poor nutrition • Non-attendance of ANC • Spousal neglect and witchcraft Risk Factors • Anaemia and poor nutrition • Lack of family planning (FP) education and services, • Little awareness of the need for FP, • Cultural and religious factors • Divorce-remarriage

Quality assessment outcome

Using the Mixed Method Appraisal Tool (MMAT), the studies for this review were assessed for their quality, validity, reliability, and relevance. These studies were grouped and evaluated based on their study designs. Out of the 19 studies, 15 were evaluated using the quality criteria for quantitative studies, three employed the quality criteria for mixed method studies, and one was evaluated using the quality criteria for qualitative studies. All the studies assessed passed the screening criteria of whether there were straightforward research questions and whether the data collected addressed the objectives of these studies. For the quantitative studies, the sampling strategies were suitable to address the research question and represented the target population. However, except for two studies [1, 12] that had appropriate measurements, the other 12 quantitative studies did not have appropriate measurements. Also, the criteria relating to whether or not the risk of nonresponse bias was low were only applicable to the study conducted [12]. Additionally, the statistical analysis for all quantitative studies was appropriate to answer the research question.

Considering the assessment of the case study, appropriate approaches were used to answer the research questions. The data collection procedures were also suitable. The findings derived from the study data were sufficient, and the data acquired adequately supported the interpretation of the results. This was coherent with the qualitative data sources, collection, analysis, and interpretation.

Regarding the assessment of the mixed-method studies, all the studies met the assessment criteria. They provided a sufficient rationale for using a mixed-methods design to address the research questions. The studies effectively integrated the different components to answer the research questions. Additionally, the various components of the studies adhered to the quality criteria of each methodological tradition employed. The studies, therefore, were considered suitable for this review (see additional file 2).

Causes and factors associated with maternal mortality

The causes and contributing factors associated with maternal mortality were systematically categorized into five distinct groups: [1] direct causes [2], indirect causes [3], socioeconomic/cultural factors [4], system/health system factors, and [5] other miscellaneous causes (see Table 2). This categorization emerged through a comprehensive review of existing literature on maternal mortality. This followed a critical analysis of the studies included in our review, where we identified recurring patterns and commonalities in the reported findings. The thematic grouping allowed for a clearer understanding of the multifaceted contributors to maternal mortality, facilitating the identification of key areas requiring targeted interventions and improved healthcare strategies.

Direct causes of maternal mortality

This review found eight primary factors that directly contributed to maternal mortality in Ghana. The primary causes of maternal death were hemorrhage, sepsis, eclampsia, and abortion-related complications. Among many studies, hemorrhage was identified as the most prevalent cause of maternal mortality [1, 9, 10, 12, 17, 26–32]. The succeeding highest causes of death, as shown in the studies reviewed, were eclampsia [1, 10, 12, 15, 17, 18, 26, 27, 30, 31], sepsis [1, 9, 10, 12, 17, 27, 28, 30–32], and abortion-related difficulties [1, 9, 10, 12, 15, 17, 18, 25, 27, 28], which were reported among 11 and 10 studies, respectively. Additionally, four studies revealed obstructed labor as a direct cause [1, 10, 12, 15], and early pregnancy was reported as the least among all the direct causes of obstetric deaths [30].

Indirect causes of maternal mortality

Regarding the indirect causes of maternal mortality in Ghana, which are described as existing preconditions or opportunistic ailments that have fatal consequences on pregnancies and childbirths, nine studies revealed anaemia/sickle cell anaemia [1, 9, 10, 12, 15, 17, 26, 28, 30] as indirect causes of maternal mortality. This was closely followed by malaria, reported by five studies [1, 10, 12, 15, 17]. HIV/AIDS cases were also identified in 3 studies [1, 10, 17], and two studies each showed that liver disease [12, 28] and tuberculosis [TB] [1, 10] were contributors to obstetric deaths. Other already existing conditions that were triggered during the pregnancy were diabetes [12], heart failure [17], cerebrospinal meningitis [17], and infection outside the genital tract [27]. These conditions were reported in studies conducted in the East Gonja District, Wa Regional Hospital, Korle-Bu Teaching Hospital, and nationwide, respectively. Multiple pregnancies were also shown in one study [17].

Socioeconomic/cultural factors of maternal mortality

Socioeconomic/cultural factors and practices were also identified to be linked to maternal deaths in Ghana. The decision or the factors influencing the decision of the place of delivery were shown in 4 studies [12, 15–17]. This factor was substantial in all studies that reported sociocultural factors related to maternal mortality. Traditional beliefs and the use of local remedies during pregnancy were seen in 4 studies [1, 12, 17, 32]. Also, taboos [1, 17], the marital status of the woman during pregnancy [25, 33], level of education [16, 26], poverty [1, 17], and the woman’s age [25, 32] were identified in 2 studies each. The least frequently reported sociocultural factors associated with maternal mortality were geographical location, place of residence, ethnicity, religion, and fitness [16]; lack of support and refusal of medical referrals [12]; misconceptions, suspicion, competition, certain tribes, cultural appropriation, food practices, happiness, and fear [24]; witchcraft, neglect, and poor nutrition [33]; and sanctity [17](see additional file 3).

System/health system problems

This theme involved an assessment of all the general system or health system problems that could contribute to maternal deaths. Overall, 12 studies [10, 12, 14–18, 26, 29, 31–33] identified system or health system problems that led to obstetric deaths in Ghana. Of these studies, antenatal care attendance [10, 12, 26, 31–33] emerged at the top of the identified problems. This was closely followed by the shortage of blood, which was reported in four studies [10, 14, 18, 29]. Delays in reaching the health facility and receiving emergency care, or in the referral system, were identified in five studies [10, 12, 15, 18, 29]. Three studies [12, 16, 18] reported that transportation difficulties were contributors to maternal deaths. It was observed, yet, that bad road networks [17, 18], poor communication between the healthcare team, the woman and her family [14, 29], staff attitude towards providing health services [12, 14], the referral system [18, 29], lack of Intensive Care Unit (ICU) facilities and logistics [10, 14, 18] as well as inadequate or inappropriate interventions [15, 17] by staff during the management of the women were factors leading to maternal death. Owusu-Sarpong et al. [18] revealed that poor documentation during pregnancy, especially during referral, the referral system, and poor supervision of juniors by senior staff were causes of maternal deaths. It was also shown [12] that the quality of health services provided, and inexperienced staff, contributed to obstetric deaths. In addition, an article [14] reported that staff strength and lack of intensive care facilities can lead to maternal deaths. The findings of other studies also showed that the non-availability of medicines and non-adherence to treatment [29], education [33], mode of delivery [26], and lack of logistics [10, 18] were factors associated with maternal mortality in Ghana.

Other causes

Other causes of maternal mortality in this study comprised obstetric deaths of unspecified/unclassified origins. Three studies showed embolism [25, 26, 31], two studies each also identifying rupture of the uterus [25, 27] and complications of anesthesia [25, 28] as causes of maternal mortality. It was also revealed that ectopic gestation and disseminated intravascular coagulation [27], complications of obstetric surgery [25], thyrotoxicosis, gastric perforations, acute asthmatic attack, and pneumonia were also contributors to maternal deaths [29].

Discussion

This review aimed to synthesize evidence on the causes and risk factors contributing to maternal mortality among women in Ghana. While direct and indirect medical factors remain relevant, the findings revealed that sociocultural and health system determinants were most frequently cited across studies. This pattern highlights how entrenched cultural norms, socioeconomic inequalities, and persistent weaknesses in the health system create structural and contextual barriers that amplify clinical risks and undermine maternal outcomes. Addressing these broader determinants is therefore critical to reducing preventable maternal deaths and achieving sustainable improvements in maternal health in Ghana.

Our review found that the trend of direct medical causes such as haemorrhage, sepsis, eclampsia, and abortion-related maternal deaths was consistently reported among several studies as the major causes in the past years. However, their rates were alternating within these periods. For instance, it was reported in a study published in 2003 by Geelhoed et al. [28] that the leading cause of maternal death was abortion-related complications, which were closely followed by sepsis and haemorrhage. Another study in 2023 showed the same causes, and these accounted for about 75% of institutional maternal deaths [31]. These studies reported similar causes of maternal death identified by studies conducted in sub-Saharan Africa and globally [34, 35]. The consistencies in these causes of maternal deaths suggest that the strategies developed to reduce these events are poorly adhered to, not properly implemented, or not practiced. There is, therefore, a critical need to re-evaluate some of the current management protocols for these causes of maternal mortalities to identify gaps in the management system.

Additionally, there is a need to train and retrain health staff on various guidelines and protocols for managing these cases. Considering the financial challenges most institutions face in organizing these training sessions for their staff [36], there is a need for the government and other non-governmental organizations to invest in the skills advancement and acquisition of the healthcare team to help reduce the rates of maternal mortalities in Ghana.

The review also found that anaemia/sickle cell anaemia and malaria were indirect causes of maternal deaths in Ghana. Accordingly, these factors accounted for about 65% of all the indirect maternal deaths that occurred [15]. These findings are consistent with a study by Priya et al. conducted at Tanda Hospital in India, which investigated common complications leading to maternal deaths. The study revealed that the majority of women who succumbed to hemorrhage were also affected by anemia [37], which also accounted for 25.7% of all maternal deaths. These conditions are preventable with simple and inexpensive interventions at the antenatal clinic, yet they contribute to the deaths of mothers. Although obstetric haemorrhage was strategically tackled, as mentioned in this study, the rates of nutritional anaemia and other poor general conditions failed to prevent pregnancy-related tragedies [37]. Preexisting anemia worsens as pregnancy progresses, leading to congestive cardiac failure and death. It also impedes the mother’s ability to resist infection or cope with bleeding, which increases the likelihood of her dying during childbirth by a factor of four [37, 38].

While it has been established that sociocultural and economic factors significantly influence the use of maternal health services, they often act as barriers, discouraging pregnant women from seeking obstetric care. This occurs despite existing recommendations encouraging pregnant women to seek medical care and recognize early warning signs of pregnancy complications [39, 40]. In typical Ghanaian and African societies, traditional beliefs and customs determine where the pregnant woman should deliver the baby, what should or should not be done during this stage, their perceptions about health, and how they should relate with others during this period. These actions can negatively affect the outcome of a woman’s pregnancy [41]. For instance, a study posits that the way community members adhere to their cultural norms determines how they view diseases and other related conditions [42]. Culture and ethnicity create a unique pattern of beliefs and perceptions as to what “health” or “illness” means [43]. These beliefs influence how symptoms are recognized, to what they are attributed, and how they are interpreted, hence affecting how and when health services are sought. Poverty may also create a challenge for maternal health because most of these deaths occur in marginalized and impoverished rural communities. As a result, when pregnant women are referred from these communities, they are unable to go to facilities with advanced healthcare services because they cannot pay for such services. Even though health services are made free for those with valid health insurance cards in Ghana [11, 44], it becomes extremely difficult for those who are taken ill or referred to do some laboratory tests for further investigations. This prevents them from seeking healthcare and only presents to the health facility in dire situations. This is consistent with the findings of a study that highlighted shortages of prescription drugs listed under the National Health Insurance Scheme (NHIS), which forces patients to pay out-of-pocket for essential medications and services. This further hinders or delays access to healthcare, particularly for marginalized and impoverished groups. Additionally, the overwhelming workload faced by healthcare providers can negatively impact the quality of service delivery, including maternity care [44, 45]. Similarly, the findings of a study suggest that women in low-income countries are highly vulnerable because they suffer from poverty, have inadequate knowledge about maternal healthcare benefits, have issues with transportation-related costs of referrals to higher health facilities, lack autonomy to make health decisions, and have safety concerns that prevent them from accessing and utilizing maternal healthcare facilities [46].

The major highlights of the system/health system problems, which create a situation of too little, too late, were antenatal (ANC) attendance, shortage of blood, and delays, causing pregnancy-related deaths. While it is often believed that women who attend ANC have higher chances of delivering safely without complications, our study observed, on the contrary, that many of these women are the victims of these deaths. Notably, it was observed that these maternal deaths are largely a result of several interrelated and overlapping health system causes. For instance, history taking and documentation of all procedures and findings are crucial tasks that must be carried out to inform accurate decision-making about women’s health and well-being; however, it is often overlooked by most health professionals [18]. This bridges continuity and leads to wrong diagnoses or inappropriate interventions, eventually perpetuating the death of these women.

Furthermore, poor supervision of staff by superiors and the attitude of staff towards clients can lead to inappropriate interventions. These findings align with research showing that disrespect, abuse, and mistreatment of women in health facilities often stem from gender, ethnic, or social class biases. Such biases may be perpetuated by healthcare providers, including female providers, who are themselves subject to the structural inequalities embedded within the healthcare system [47, 48]. While these various mistreatments can occur at the level of the interaction between the woman and service provider, a complex range of systemic failures at the levels of the health facility and health system contribute to their occurrence. These may include poor supervisory structures, insufficient staffing, inadequate supply chains, poor physical conditions, and policies, facility cultures, and power dynamics that systematically disempower women [48].

Implications for practice/policy

The findings from this systematic review highlight critical areas for targeted interventions to reduce maternal mortality in Ghana. Building on this, enhancing healthcare provider training and protocols is crucial. The consistent reporting of hemorrhage, sepsis, and eclampsia as leading causes of maternal deaths underscores the need for ongoing training of healthcare providers in managing these conditions. Regular workshops and simulations on obstetric emergencies, with a focus on early recognition and timely intervention, should be mandated.

To complement these efforts, improving antenatal and postnatal care is essential. While antenatal care attendance is high, the review found that many women who attend ANC still experience poor outcomes. Policymakers should focus on ensuring that ANC visits are not only frequent but also comprehensive, with a focus on early detection and management of risk factors such as anemia and hypertension. Community-based postnatal care programs, supported by trained midwives and community health workers, can bridge the gap between hospital discharge and long-term maternal health.

Addressing sociocultural barriers is another critical area. The review identified traditional beliefs, cultural practices, and misconceptions as significant barriers to maternal healthcare utilization. Community-based education programs, led by respected local leaders and healthcare providers, can help dispel myths and encourage women to seek timely care. Delays in referrals were a recurring theme in the review. Strengthening referral systems, particularly between lower-level facilities and regional hospitals, can reduce delays in accessing emergency care. This could involve improving communication systems, such as the use of mobile technology, and ensuring that referral facilities are adequately staffed and equipped to handle emergencies.

Limitations

This study has several notable limitations. A meta-analysis could not be conducted due to significant disparities among the included studies. Most studies were concentrated in urban areas, particularly in secondary and tertiary institutions such as district, regional, and teaching hospitals, with only a few focusing on rural settings where healthcare accessibility is constrained by systemic and logistical challenges. Additionally, the studies relied primarily on institutional records and maternal audits, which do not fully capture the circumstances leading to complications and subsequent deaths. Furthermore, many of the maternal deaths examined were referrals from smaller facilities, and the majority of studies were conducted in teaching hospitals, where institutional maternal mortality rates tend to be higher. Lastly, a significant portion of the studies were drawn from the northern and savannah regions, largely due to the rate of cases from rural communities. In these areas, higher poverty and illiteracy rates have a profound impact on healthcare utilization and access.

Conclusion

This systematic review presented a comprehensive overview of the determinants of maternal mortality across various regions in Ghana. Key determinants such as inadequate resources, infrastructure, and personnel, access to quality healthcare, poverty, traditional beliefs and customs, socioeconomic status, education, presence of skilled birth attendants, and geographical location played a crucial role in maternal health outcomes, including death. The consistencies in reporting these causes of maternal deaths suggest that the strategies developed to reduce these events are poorly adhered to, not correctly implemented, or not practiced. Therefore, to reduce this burden, interventions such as strengthening healthcare systems through re-evaluating some of the current obstetric care protocols to identify gaps in the management system, improving healthcare infrastructure, enhancing provider training, and promoting community-based initiatives are essential to achieving substantial and sustainable reductions in maternal mortality. Additionally, collaborative efforts involving governments, healthcare providers, and communities are imperative to ensure that all women have access to the care they need for safe pregnancies and deliveries. This review underscores the importance of continued research and policy efforts to reduce maternal mortality and improve maternal health outcomes in Ghana. Future research should focus on evaluating the effectiveness of targeted interventions and exploring innovative approaches to overcome barriers to maternal healthcare.

Supplementary Information

Below is the link to the electronic supplementary material.

Authors’ contributions

D.B.A., and A.A., conceptualized the study and wrote the manuscript. FK., S.S.D., J.J.O., E.A.A., R.A.A., K.N.R., P.A., V.B.B. contributed to writing and reviewing the manuscript. J.J.O., K.N.R, P.A., and V.B.B. supervised the study. All the authors reviewed and approved the final manuscript.

Funding

This study did not receive any funding.

Data availability

Not applicable.

Declarations

Ethical approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.