Maternal health literacy among pregnant women attending antenatal care in Ghana: a cross-sectional study

Abstract

Introduction

Maternal health literacy (MHL) plays a critical role in ensuring positive maternal and neonatal health outcomes; however, limited evidence exists on the functional aspect of MHL among pregnant women in Ghana. This study assessed the level of functional MHL, associated demographic factors, and perceived barriers among pregnant women attending antenatal care (ANC) at a university hospital in Ghana.

Methods

A hospital-based cross-sectional study was conducted among 235 pregnant women attending ANC from November 2024 to February 2025. Participants were selected using simple random sampling. Data were collected through a structured questionnaire adapted from validated health literacy tools and analyzed using Stata/SE version 17.0. Descriptive statistics were used to summarize the data. Associations were assessed using chi-square and Wilcoxon rank-sum tests, while logistic regression was used to identify associated factors.

Results

The median age of participants was 29 years (IQR: 24–33), and 55.3% had tertiary education. Overall, 54.9% of the women demonstrated high MHL. Educational level, age, employment status, and parity were associated with MHL at the bivariate level; however, only education remained independently associated with MHL in the adjusted model. Women with primary [aOR = 0.02, 95% CI: 0.01–0.11] and secondary education [aOR = 0.27, 95% CI: 0.14–0.54] were less likely to have high MHL compared to those with tertiary education. The most reported barriers were poor communication from healthcare providers (47.2%) and lack of resources (34%).

Conclusion

Nearly half of pregnant women demonstrated low functional MHL, with educational attainment identified as a factor associated with MHL. Strengthening health-literacy-responsive communication and integrating group antenatal care may enhance comprehension and informed decision-making.

Introduction

Health literacy (HL) has emerged as a significant public health concern, with the World Health Organization (WHO) underscoring its relevance in fulfilling the goals of the 2030 Agenda for Sustainable Development [1]. HL is described as the capacity to obtain, comprehend, and utilize knowledge to sustain optimal health for oneself and the community [2]. In the context of pregnancy, this is known as maternal health literacy (MHL) and is essential for managing personal and fetal health throughout gestation [3]. MHL refers to a woman's ability to access, understand, critically appraise, and apply maternal and child health information to make informed decisions during pregnancy, childbirth, and the postpartum period [4], [5]. MHL is significantly associated with enhanced health outcomes. Studies indicate that higher MHL is associated with improved maternal and neonatal outcomes, such as earlier and more frequent antenatal care (ANC), reduced complications, and better infant Apgar scores [6], [7], [8]. Consequently, enhancing MHL is crucial for improving health outcomes and reducing maternal health disparities.

Sub-Saharan Africa (SSA) faces significant challenges regarding maternal and neonatal mortality, with an estimated maternal mortality ratio for the region standing at 454 per 100,000 live births in 2023 according to the 2025 WHO report [9], placing a heavy burden on healthcare systems and communities across the region. Restricted access to health information, inadequate literacy levels, and socio-cultural barriers consistently hinder MHL, contributing to adverse maternal and neonatal outcomes [4], [10], [11]. Evidence from studies conducted in countries such as Nigeria and Ethiopia has explored the impact of MHL in influencing health service utilization, including increased ANC attendance, improved nutritional practices, and more timely healthcare-seeking behaviors [12], [13]. However, disparities persist, with evidence from recent demographic and health surveys indicating that ANC utilization remains suboptimal across SSA despite the WHO's updated recommendation of eight ANC contacts [14], [15]. These persistent gaps necessitate a deeper exploration of context-specific factors, including those that influence MHL in specific local settings.

Despite its importance, research on HL among pregnant women in Ghana has yielded inconclusive results, with some studies indicating sufficient literacy levels and others highlighting shortcomings [16], [17], [18]. However, a key limitation identified in previous studies conducted in Ghana is the over-reliance on general HL measurement tools that do not adequately capture the functional dimension of MHL, a particularly underexplored area [16], [19], [20]. In this study, functional health literacy (FHL) refers to the practical ability of pregnant women to access, understand, and use maternal health information in making appropriate health decisions during pregnancy [21], [22]. It was operationalized using self-reported measures assessing women's ability to read and engage with health-related materials, understand and follow healthcare providers' instructions, and independently complete medical tasks such as filling out forms. It also captured access to and use of mobile health information sources, frequency of interaction with health information within their social environment, and the influence of cultural beliefs and social support on health-related decision-making. For example, Lori et al. (2014) found that although pregnant Ghanaian women could identify danger signs such as sleeplessness and loss of appetite, many were unable to translate this knowledge into appropriate health-seeking behaviors [17], revealing a clear gap between antenatal health education received and its real-world application. This issue is especially concerning in a context where maternal mortality remains high (234 per 100,000 live births in Ghana, compared to the global average of 197 per 100,000 live births) [9], highlighting the need for effective strategies to improve maternal outcomes. Addressing low FHL is therefore essential, particularly among women with limited formal education and those in underserved communities. Although emerging interventions such as Group Antenatal Care (G-ANC) have shown potential in increasing MHL [16]. However, many pregnant women continue to face challenges in accessing, understanding, and utilizing essential health information. These challenges are compounded by a significant gap in the literature on FHL within many healthcare settings in Ghana, as most existing studies focused on general rather than functional dimensions of HL [19], [20].

Building on this identified gap, existing literature has documented several sociodemographic and contextual factors that may influence MHL, including educational attainment, age, parity, communication with healthcare providers, and access to health-related resources. Women with higher levels of formal education are consistently shown to be better able to comprehend, evaluate, and apply health information [23], [24]. Age and parity have also been associated with MHL, as older and multiparous women may benefit from cumulative exposure to ANC services and prior pregnancy-related experiences [25]. Effective communication with healthcare providers further enhances women's understanding of medical guidance and strengthens trust in health services [26]. In addition, access to mobile health applications and printed educational materials can diversify and strengthen the channels through which women obtain and process health information [27], [28], [29]. Nevertheless, much of the existing evidence is derived from studies conducted outside Ghana or within broader sub-Saharan African contexts. This underscores the need for context-specific study to empirically examine these associations within the Ghanaian setting.

To address this critical gap, the present study assessed the functional health literacy levels of pregnant women attending the ANC clinic at a university hospital in Ghana, examined the demographic factors associated with these levels, and explored the reported barriers to MHL. By providing a comprehensive assessment, this research contributes evidence for the development of targeted health education programs and policies. The findings provide crucial insights for targeted interventions that address specific health literacy-related barriers, ultimately ensuring that all pregnant women, irrespective of educational background, can access optimal healthcare services and achieve better maternal and child health outcomes.

Methods

Study design and population

We conducted a hospital-based cross-sectional study to assess functional health literacy, barriers, and their associated factors among pregnant women. The study population comprised all pregnant women attending ANC clinic at the Kwame Nkrumah University of Science and Technology (KNUST) Hospital during the study period. Eligible participants were pregnant women aged 18 years and above, attending ANC clinic at the study site, and in their second trimester to near full-term (12–40 weeks), who were willing to provide informed consent. Pregnant women with cognitive impairments or language barriers that could hinder meaningful participation, as well as those who had attended fewer than two antenatal clinic visits at the hospital, were excluded from the study. Pregnant women less than 18 years were excluded due to ethical considerations and hospital policy requirements for parental consent, which could complicate recruitment and compromise data reliability. In addition, participants with fewer than two ANC visits were excluded to ensure that respondents had sufficient exposure to maternal health information and services, allowing for a meaningful assessment of functional maternal health literacy. Data collection was carried out over four months, spanning from November 2024 to February 2025.

Study setting

The study was conducted at the KNUST Hospital, a district-level healthcare facility located in Kumasi in the Ashanti Region of Ghana. The hospital serves a diverse population drawn from both urban and peri-urban communities and provides a wide range of comprehensive maternal and child health services. Situated on the KNUST campus along the main Accra–Kumasi highway, the facility has a bed capacity of approximately 150 and offers specialized services in gynecology, obstetrics, medicine, surgery, pediatrics, and emergency care. The hospital employs about 200 permanent and 140 contract staff, and it serves as a referral center for several neighboring hospitals and health centers within the Oforikrom municipality. The study area lies within the middle belt of Ghana, characterized by a mix of urban and peri-urban settlements.

KNUST Hospital was selected for this present study due to its high ANC attendance and the heterogeneity of its client population. These attributes provided a robust platform for examining MHL within a real-world clinical setting. The hospital's routine integration of health education into ANC services further enabled an in-depth exploration of FHL in practice.

Sample size determination and sampling technique

The sample size was calculated using Yamane's formula [30] at a 95% confidence level and a 5% margin of error. This approach was considered appropriate because the study population was finite and well-defined, with an estimated monthly antenatal care (ANC) attendance of 600 pregnant women (N = 600). In addition, no reliable prior estimate of the prevalence of functional maternal health literacy in this specific population was available to inform a single population proportion calculation. Therefore, Yamane's formula was adopted to obtain a representative sample from the finite population. Substituting the values into the formula, n = N / 1 + N (e)², gave a minimum sample size of 240. Allowing for a 10% non-response rate, the final sample size was 264. However, a total of 235 eligible pregnant women who met the inclusion criteria consented to participate in the study, representing a response rate of approximately 89%. This response rate was considered satisfactory for a facility-based cross-sectional study and provided sufficient observations to estimate maternal health literacy levels and assess associated factors [31], [32], [33].

A simple random sampling technique was employed to select study participants from eligible ANC attendees. The estimated monthly ANC attendance of 600 women constituted the sampling frame. During each ANC clinic session, all pregnant women who met the inclusion criteria and none of the exclusion criteria were identified. Following an explanation of the study objectives and procedures, written informed consent was obtained from willing participants. A lottery method was then used to randomly select participants. The number of “Yes” papers was determined based on the number of participants required from each clinic session, while the remaining papers were labeled “No.” All papers were identical in size, shape, and appearance, folded uniformly, and mixed thoroughly before selection. Each eligible woman selected one folded paper, and those who picked a “Yes” paper were enrolled in the study. This process was repeated at successive ANC clinic sessions until the required number of participants was obtained. By ensuring that all eligible women attending ANC during the study period had an equal probability of selection, the sampling procedure enhanced the representativeness of the study sample and minimized selection bias.

Data collection instrument and techniques

An adapted health literacy questionnaire was developed based on validated instruments, including the Brief Health Literacy Screen (BHLS) [34], the eHealth Literacy Scale (eHEALS) [35], and the Health Literacy Questionnaire (HLQ) [36]. Items were selected based on their relevance to the functional, digital, and sociocultural dimensions of health literacy among pregnant women and adapted to our local context. The drafted questionnaire was reviewed for content relevance and clarity, and pretested with 20 pregnant women attending ANC at Ejisu Government Hospital, a district hospital, prior to data collection to assess comprehensibility, cultural appropriateness, and face validity. Necessary refinements were made based on pretest feedback to improve clarity and ensure suitability for the study population. Data were collected through face-to-face interviews using the structured questionnaire, which comprised three main sections: demographic information; functional, digital, and sociocultural aspects of health literacy; and barriers to health literacy (Supplementary File 1).

Study variables

Dependent variables

Maternal health literacy was assessed using a structured 9-item questionnaire. Four items were adapted from the BHLS [34] to evaluate how often participants read health-related materials (e.g., brochures, pamphlets), how often they required assistance when reading hospital materials (never, rarely, sometimes, or often), their confidence in understanding medical instructions provided by healthcare professionals, and their confidence in independently completing medical forms (not confident, slightly confident, moderately confident, or very confident). Two items were adapted from the eHEALS [35] to assess access to health information through mobile health applications (yes/no) and the frequency of using such applications to obtain health information (never, rarely, sometimes, or often). The remaining three items were adapted from the HLQ [36] to assess whether cultural beliefs and practices influence health decisions, how often participants discuss health information with family or community members (never, rarely, sometimes, or often), and whether they receive adequate social support from family or community regarding health decisions (yes/no). One item was negatively worded and reverse-scored. Higher total scores reflected higher levels of MHL. Based on the median score of 21, participants scoring below 21 were classified as having low MHL, while those scoring 21 or above were classified as having high MHL. The internal consistency of the 9-item scale, assessed using McDonald's omega, was 0.7, indicating acceptable reliability.

Independent variables

The independent variables comprised the demographic characteristics of the participants, including age (in years), educational level (no formal education, primary, secondary, or tertiary), employment status (employed or unemployed), and parity (0, 1, 2, or ≥ 3 children). In addition, potential barriers to maternal health literacy were examined. These included limited accessibility to health services and information within the hospital, cultural beliefs and practices influencing health decisions, language barriers in understanding health information, inadequate resources, and poor communication from healthcare providers.

Data analysis

Data were entered into Microsoft Excel, cleaned, coded, and analyzed using Stata/SE Version 17.0 [37]. Continuous variables were assessed for normality using the Shapiro–Wilk test and visually inspected with histograms. Non-normally distributed variables were summarized using medians with interquartile ranges (IQRs), while categorical variables were presented as frequencies and percentages. Associations between demographic characteristics and MHL were examined using Pearson's chi-square test, with Fisher's exact test applied where expected cell counts were less than five to ensure statistical accuracy. The Wilcoxon rank-sum test was used to compare differences in skewed continuous variables between groups. Additionally, a bar chart was used to illustrate the reported barriers to maternal health literacy among participants.

All demographic variables were first examined using bivariate logistic regression to assess their independent association with MHL. Variables were then entered into a multivariable logistic regression model to adjust for potential confounding, based on prior evidence of their association with MHL [12], [20], [24], [25]. Multicollinearity was assessed using the Variance Inflation Factor (VIF), and only variables with VIF values below 5 were retained in the multivariable model. The overall diagnostics (Mean VIF = 1.98; Minimum VIF = 1.29; Maximum VIF = 4.17) indicated no evidence of substantial collinearity [38], [39].

Results

Demographic and maternal health literacy distribution among study participants

The median age of participants was 29 years (IQR: 24–33), with a difference observed among women in the low literacy group, having a median age of 30 years (IQR: 26–34) compared with 28 years (IQR: 23–32) among those in the high literacy group. Educational attainment varied across literacy groups: most women with tertiary education were in the high literacy group (74.4%), whereas women with primary education were more represented in the low literacy group (27.3%). Employment status differed between groups, with 31.8% of unemployed women in the high literacy group compared with 18.9% in the low literacy group. Parity also varied across literacy levels, with 39.5% of nulliparous women in the high literacy group, whereas 30.2% of women with three or more births were more common in the low literacy group (Table 1).

Table 1.

Sociodemographic characteristics of study participants stratified by literacy level.

Characteristics	Frequency (N = 235)1	Low literacy (n = 106)1	High literacy (n = 129)1	p-value2
Age (years)				0.003W
Median (IQR)	29 (24, 33)	30 (26, 34)	28 (23,32)
Educational level				<0.0001F
No formal education	9 (3.8)	9 (8.5)	0 (0.0)
Primary	31 (13.2)	29 (27.3)	2 (1.6)
Secondary	65 (27.7)	34 (32.1)	31 (24.0)
Tertiary	130 (55.3)	34 (32.1)	96 (74.4)
Employment status				0.03
Unemployed	61 (26.0)	20 (18.9)	41 (31.8)
Employed	174 (74.0)	86 (81.1)	88 (68.2)
Parity				0.003
0	81 (34.5)	30 (28.3)	51 (39.5)
1	58 (24.7)	20 (18.9)	38 (29.5)
2	48 (20.4)	24 (22.6)	24 (18.6)
≥3	48 (20.4)	32 (30.2)	16 (12.4)

IQR = Interquartile range; 1 = Median (IQR), frequency (%); 2 = Pearson's chi-squared test; W = Wilcoxon rank-sum test; F = Fisher's exact test.

Distribution of maternal health literacy among study participants

Overall, more than half of the women (54.9%) demonstrated high health literacy, while 45.1% had low literacy levels (Fig. 1). In terms of specific literacy behaviors, more than half of the participants (53.2%) reported sometimes reading health-related materials such as brochures or pamphlets. Likewise, 53.2% indicated being very confident in understanding medical instructions given by healthcare providers. Almost half (46.8%) sometimes needed help reading hospital materials, while 45.1% were very confident in filling out medical forms by themselves. Regarding digital health access, 62.1% of respondents had access to mobile health applications, with 39.6% reporting that they sometimes used these apps to obtain health information. Additionally, 55.3% stated that their cultural beliefs and practices never influenced their health decisions. More than half (56.2%) sometimes discussed health information with family or community members, and a majority (77.0%) received adequate social support from family or community regarding health decisions. The overall median health literacy score was 21 (IQR: 17–23), with minimum and maximum scores of 7 and 28, respectively, out of a possible score of 32. The scale demonstrated acceptable internal consistency, with a McDonald's omega coefficient of 0.7 (Table 2).

Fig. 1.

Distribution of maternal health literacy.

Table 2.

Distribution of maternal health literacy among study participants.

Maternal health literacy question items	Response (N = 235)	Percentage (%)
How often do you read health-related materials (e.g., brochures, pamphlets)?
Never	39	16.6
Rarely	22	9.4
Sometimes	125	53.2
Often	49	20.8
How confident are you in understanding medical instructions given by healthcare providers?
Not confident	12	5.1
Slightly confident	26	11.1
Moderately confident	72	30.6
Very confident	125	53.2
How often do you need help reading hospital materials?
Never	51	21.7
Rarely	46	19.6
Sometimes	110	46.8
Often	28	11.9
How confident are you in filling out medical forms by yourself?
Not confident	33	14.1
Slightly confident	40	17.0
Moderately confident	56	23.8
Very confident	106	45.1
Do you have access to health information through mobile health apps?
No	89	37.9
Yes	146	62.1
How often do you use mobile health apps to get health information?
Never	69	29.4
Rarely	18	7.6
Sometimes	93	39.6
Often	55	23.4
Do your cultural beliefs and practices influence your health decisions?
Never	130	55.3
Rarely	32	13.6
Sometimes	49	20.9
Often	24	10.2
How often do you discuss health information with family or community members?
Never	42	17.9
Rarely	29	12.3
Sometimes	132	56.2
Often	32	13.6
Do you receive adequate social support from family or community regarding health decisions?
No	54	23.0
Yes	181	77.0
Overall maternal health literacy score [Median, (IQR)]	21 (17, 23) [Out of 32]	Min (7), Max (28)

McDonald's omega coefficient of 9 items = 0.7; Min = Minimum maternal health literacy score; Max = Maximum maternal health literacy score.

Demographic factors associated with maternal health literacy

In the bivariate logistic regression analysis, age, educational level, employment status, and parity were associated with maternal health literacy. After adjusting for potential confounders, only educational level remained independently associated with maternal health literacy. Compared with women with tertiary education, those with primary education [aOR = 0.02, 95% CI: 0.01–0.11] and secondary education [aOR = 0.27, 95% CI: 0.14–0.54] had lower odds of literacy (Table 3).

Table 3.

Logistic regression for the demographic factors associated with maternal health literacy.

	Crude		Adjusted
Characteristics	OR1	95% CI3	OR2	95% CI3
Age (years)
	0.93	0.88–0.97	0.93	0.86–1.00
Educational level
Tertiary	Ref		Ref
No formal education	–	–	–	–
Primary	0.02	0.01–0.11	0.02	0.01–0.11
Secondary	0.32	0.17–0.60	0.27	0.14–0.54
Employment status
Employed	Ref		Ref
Unemployed	2.00	1.09–3.69	0.88	0.37–2.05
Parity
0	Ref		Ref
1	1.12	0.55–2.26	2.00	0.86–4.66
2	0.59	0.29–1.21	1.36	0.55–3.38
≥3	0.29	0.14–0.62	0.81	0.30–2.23

1 = Crude odds ratio; 2 = Adjusted odds ratio; 3 = 95% confidence interval; − = Stata couldn't compute the estimates due to low outcome observation; Ref = Reference category.

Distribution of barriers to maternal health literacy

Fig. 2 shows the distribution of reported barriers to maternal health literacy. Poor communication from healthcare providers was the most reported barrier (47.2%), followed by lack of resources (34%) and language difficulties in understanding health information (27.2%). Cultural beliefs and practices accounted for 13.2%, while limited accessibility to health services and information at the hospital was the least reported barrier (6%).

Fig. 2.

Distribution of barriers to maternal health literacy.

Discussion

This study assessed functional maternal health literacy (MHL), its demographic associated factors, and perceived barriers among pregnant women attending antenatal care at a university hospital in Kumasi, Ghana. Although more than half of the participants demonstrated high MHL, a striking 45% still exhibited low literacy despite the university-hospital setting, where exposure to health information is relatively high. The strong association between educational attainment and MHL underscores the enduring importance of formal education in shaping women's ability to read, interpret, and use health information effectively. Women with only primary or secondary education were less likely to have high MHL, consistent with a large body of Ghanaian and international evidence showing that education is a primary driver of health literacy and maternal health knowledge [17], [40]. The persistence of low literacy even among women with some secondary education suggests that functional literacy skills, distinct from basic schooling, remain uneven and may require targeted, context-specific interventions.

Our findings resonate with prior Ghanaian studies that have highlighted deficits in comprehension of antenatal instructions, danger signs, and medication labels, even when information is routinely provided in clinical settings. Lori et al. (2014) reported that although women could recall many health messages, they often struggled to apply them [17], a pattern reflected in our sample where nearly half required assistance reading hospital materials or completing medical forms. This reinforces the idea that information delivery alone does not guarantee understanding or actionable knowledge. However, our results also diverge from recent intervention-focused research. The cluster randomized trial on Group Antenatal Care (G-ANC) by Lori et al. (2024) demonstrated substantial improvements in MHL regardless of educational background [16], suggesting that participatory, discussion-based ANC formats may help reduce literacy-related disparities. In contrast, our cross-sectional findings reflect outcomes under the conventional individual ANC model, where disparities by educational level persist. This contrast highlights the potential transformative effect of structured group-based models on maternal comprehension and engagement.

The barriers identified—poor communication from healthcare providers, language difficulties, and resource constraints—shed light on systemic and structural issues that impede MHL. Almost half of our participants reported poor provider communication, aligning with some Ghanaian studies showing that time pressure, limited interpersonal counselling, and inconsistent use of local languages weaken ANC message delivery. Language challenges, even in an urban facility, indicate that clinical communication often relies heavily on English or medical jargon, which may not align with women's linguistic comfort or literacy levels. Furthermore, more than a third of participants cited resource limitations, echoing findings from studies reporting that lack of digital access, instructional materials, and simplified tools hinder women's ability to reinforce health messages at home [41]. Social support was relatively high in this sample, but the fact that a sizable proportion struggled with digital health tools underscores a digital literacy gap that could widen inequities as mHealth platforms expand nationally.

Taken together, these findings suggest that maternal health literacy in Ghana is shaped by a complex interplay of individual education, communication quality within healthcare encounters, language accessibility, and structural resource constraints. Although the study setting is a university hospital, where one might expect higher literacy levels, the notable proportion of low MHL indicates that the challenge is broader than individual capability. It reflects system-level communication gaps and the absence of health-literacy–responsive care models that actively assess and address women's understanding. Addressing these gaps is essential because MHL influences care-seeking, adherence, birth preparedness, and maternal–newborn outcomes. Without intentional system-level improvements, low literacy will continue to undermine otherwise well-designed antenatal interventions.

Implications for policy, practice, and research

The findings of this study have several implications for policy, clinical practice, and future research. The findings of the present study strongly support integrating and scaling group antenatal care (G-ANC) within Ghana's maternal health system, as evidence consistently demonstrates its effectiveness in improving maternal health literacy, strengthening peer support, and increasing adherence to recommended ANC visits. The Ministry of Health, through the Ghana Health Service, should incorporate G-ANC into national ANC guidelines, prioritizing its rollout in public hospitals and primary care settings, accompanied by targeted investments in training of midwives and ANC providers, infrastructural adjustments, and scheduling systems that facilitate structured group sessions. In addition, policy efforts by the Ghana Health Services should prioritize standardization of G-ANC delivery to ensure consistency, quality, and equitable access across both urban and rural health facilities. For clinical practice, the results underscore the need for strengthened communication competencies among ANC providers, given that nearly half of the women cited provider communication challenges as a barrier to understanding and applying health information; consequently, ANC staff should be trained in patient-centred communication approaches such as the teach-back method, use of plain language, inclusive counselling styles, and culturally sensitive explanation techniques, supported by visual tools, demonstrations, and peer-facilitated discussions to enhance comprehension. Health facilities must also ensure that maternal health education is routinely delivered in predominant local languages and supported by interpreters or multilingual staff where necessary. For research, future work should adopt longitudinal or experimental designs to determine whether improvements in maternal health literacy translate into clinically meaningful outcomes such as reduced maternal complications, improved fetal growth indicators, and enhanced neonatal health, while qualitative studies should further explore the mechanisms underlying communication breakdowns and sociocultural influences that limit knowledge application. Implementation science research is also essential to identify effective strategies for scaling G-ANC in low-resource contexts, including assessments of cost-effectiveness, workforce requirements, operational feasibility, and fidelity of delivery to ensure sustainable integration into Ghana's health system.

Strengths and limitations

This study has several strengths. First, by employing a structured, adapted instrument combining items from validated scales (BHLS, eHEALS, HLQ), we were able to measure multiple dimensions of MHL (functional, digital, sociocultural) in a way that reflects real-world behavior. Second, the sample was drawn from a busy university hospital in a peri-urban Ghanaian setting, which enhances the relevance of our findings to similar health facilities and populations.

However, there are limitations. As a cross-sectional study, we cannot infer causality between literacy levels and health behaviors or outcomes. The nine-item scale we used, though adapted from validated tools, may not capture all nuances of functional health literacy in this cultural and clinical context; its internal reliability (McDonald's omega of 0.7), though acceptable, was modest, which may affect the precision of measurement. Our classification of “low” vs. “high” literacy based on a median split, while practical, may oversimplify the continuum of health literacy. Confirmatory factor analysis (CFA) was not conducted to further validate the factor structure of the adapted instrument, limiting full confirmation of construct validity in this population. In addition, the study was conducted at a single university hospital in Kumasi, which may limit the generalizability of the findings to pregnant women in other settings, particularly in rural health facilities where socioeconomic conditions, literacy levels, and healthcare access may differ. Lastly, women who declined to participate or were absent during clinic visits may have differed systematically from those enrolled, introducing the possibility of selection and non-response bias. Nevertheless, the relatively high response rate (89%) and probability-based sampling approach help to reduce these concerns. Future studies should therefore adopt multi-site designs across diverse healthcare settings to improve external validity and provide a more comprehensive understanding of maternal health literacy across different population groups in Ghana. Self-reported barriers may also be subject to social desirability bias, and interviewer-administered data collection may have introduced minimal interviewer effects.

Conclusion

This study reveals that nearly half of pregnant women attending ANC at a university hospital in Ghana demonstrate low functional maternal health literacy. Educational attainment emerged as a powerful determinant, and poor communication from healthcare providers was the most frequently reported barrier. These findings underscore the critical need to strengthen health communication strategies, especially through models such as group antenatal care, tailored to women with lower education and greater childcare responsibilities. Policymakers and clinicians should prioritize interventions that promote clear, culturally appropriate, and accessible information delivery. Future research should rigorously test whether improving MHL translates into better maternal and neonatal health outcomes and investigate scalable, sustainable ways to embed health literacy promotion into routine ANC.

Ethical considerations and consent to participate

Ethical clearance for the study was obtained from the Committee on Human Research, Publications and Ethics (CHRPE) of the School of Medical Sciences, Kwame Nkrumah University of Science and Technology (Approval ID: CHRPE/AP/1128/24), in compliance with the principles of the Declaration of Helsinki. Administrative approval was subsequently secured from the management of KNUST Hospital before data collection. The objectives and procedures of the study were thoroughly explained to all eligible participants, after which a signed informed consent was obtained. Participants were assured of strict confidentiality of the information provided and anonymity throughout the research process.

Availability of data

The anonymized datasets generated and/or analyzed during the current study are not publicly available but can be obtained from the corresponding author upon reasonable request.

CRediT authorship contribution statement

Felicia Dankwah: Writing – review & editing, Writing – original draft, Resources, Methodology, Formal analysis, Data curation, Conceptualization. Julius Kwabena Karikari: Writing – review & editing, Methodology, Formal analysis. Paul Okyere: Writing – review & editing, Supervision, Conceptualization.

Consent for publication

Not applicable.

Funding

The authors had no support or funding for this current study.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Questionnaire