A quantitative cross-sectional study design with both descriptive and analytical components was employed, using a researcher-administered questionnaire to describe maternal health knowledge and examine its association with selected socio-demographic variables.

Three antenatal clinics were included in the study, all located in the Northern Tygerberg sub-district, Western Cape. The Northern Tygerberg sub-district has a population of fewer than 424 945 residents, comprising diverse racial and socio-economic groups with an average annual growth rate of 5.31%.¹⁹ The antenatal clinics are sections or segments of the Midwifery Obstetric Units (MOUs) located on the community health centre (CHC) premises. The antenatal clinics were purposely selected because they offer basic antenatal care and serve a large catchment area of pregnant women from lower socio-economic backgrounds in the Northern Tygerberg sub-district, Western Cape. The antenatal clinics provide an 8-h antenatal care service that includes regular check-ups, health education and promotion, screening tests, monitoring foetal development and supportive services.

The target population was 900 pregnant women attending antenatal clinics during the data collection period. The inclusion criteria for the study were pregnant women (18 years and over) who attended antenatal clinics for antenatal services. The calculator.net sample size calculator was used to calculate the sample size, with the margin of error of 6% and confidence level of z = 95% and population proportion of p = 50%, suggesting a sample size of 267.²⁰ A 6% margin of error was selected to balance statistical precision with feasibility while maintaining acceptable estimate accuracy for this cross-sectional study. Furthermore, the sample size calculated for the descriptive objective was also adequate for the analytical component, as post hoc power analysis demonstrated sufficient power to detect moderate effect sizes at the 5% significance level. Selecting p = 50% is statistically conservative and prevents underestimation when the true proportion is unknown. The sample size was calculated as 267 to account for an expected non-response rate, ensuring that the study would still have enough participants for reliable results even if some did not respond. A systematic sampling method was used by selecting every third attendee to be approached for participation in the study. The decision to select every third patient was undertaken to ensure an even distribution of participants while maintaining logistical feasibility for data collection. When the selected third patient declined to participate, a systematic substitution procedure was applied. The study population comprised approximately 900 pregnant women, from which a sample of 267 was required. Using systematic sampling, every third patient was selected. The interval was derived by dividing the total population (n = 900) by the sample size (n =267), yielding approximately three.

The questionnaire included the valid and reliable Maternal Health Literacy Inventory in Pregnancy (MHELIP) scale with 48 statements, of which 21 statements measure maternal health knowledge (21 items).²¹ The maternal health knowledge scale includes statements rated on a 5-point Likert scale from ‘I don’t know at all’ (1), ‘I know a little’ (2), ‘I know something’ (3), ‘I know a lot’ (4) and ‘I know fully’ (5). To calculate the total score, the recommended formula was used (Score = [raw score – minimum possible raw score]/[maximum possible raw score – minimum possible raw score] × 100) and then classified into four categories: inadequate (0 to 50), problematic (50.1 to 66), sufficient (66.1 to 84) and excellent (84.1 to 100). The inadequate and problematic maternal health knowledge categories were defined as limited health literacy (n < 66.1/100), while the sufficient and excellent maternal health knowledge categories were defined as desired health literacy (n > 66/100). The internal consistency of the maternal health knowledge scale in this sample was calculated for scale reliability and was acceptable (Cronbach’s α = 0.78) although lower than that reported in a previous study by Taheri et al.,²² which found a reliability coefficient of α = 0.94. The original questionnaire was adapted into English and subsequently back-translated into isiXhosa to enhance comprehension. The isiXhosa is the dominant language in the study setting; therefore, this language was used for data collection.

Prior to data collection, the questionnaire was pretested at three antenatal clinics within the Northern Tygerberg sub-district. The pretesting of the questionnaire was carried out a month before the main study (July 2022) at the three selected antenatal clinics on 15 pregnant women who met the established inclusion criteria. The pretesting of the questionnaire was conducted to assess the feasibility of the study procedures, refine data collection instruments, evaluate recruitment and logistical processes and identify potential methodological challenges prior to the main study. The respondents on the retesting of the questionnaire were excluded from the main study.

Data were collected at the three antenatal clinics located in the Northern Tygerberg sub-district between August 2022 and November 2022 after obtaining ethics approval from the university and permission from the Department of Health (DOH) and managers of the antenatal clinics. Data were collected in clinic waiting rooms over approximately 30–40 days per clinic, with data collection at each site completed before proceeding to the next clinic. Two trained research assistants with master’s degrees in nursing collected the data between 8:00 and 16:00 each working day during the 4-month period. Research assistants received face-to-face training on the administration of the questionnaire, the provision of study information that included the study’s objectives, potential risks, respondents’ expectations and the procedures for obtaining written informed consent. Standardised data collection procedures, verification of completed questionnaires, data entry checks and strict adherence to predefined inclusion and exclusion criteria were employed to ensure the quality and integrity of data collection. The researcher addressed potential biases by employing a structured questionnaire with a standardised administration protocol, emphasising anonymity and confidentiality to encourage honest responses and utilising systematic sampling to minimise the influence of convenience sampling. The respondents were presented with the study information and provided written informed consent. The researcher and assistants administered the questionnaire to all pregnant women who met the inclusion criteria. Data collection was conducted in the morning during tea time and in the afternoon during lunch time, when the participants were not busy with nurses, to avoid interrupting maternal services. The questionnaire took approximately 25 min to complete.

The study data were captured and cleaned in the Statistical Package for the Social Sciences (SPSS) version 29.0 for analysis. Age was categorised into two groups: ≤ 35 years and > 36 years. This classification is based on existing literature, which defines advanced maternal age as 35 years and older, while individuals under 35 years are considered to be within the normal reproductive age range. Advanced maternal age has been associated with increased risks during pregnancy and childbirth, including a higher incidence of maternal and neonatal complications such as gestational hypertension and preeclampsia. Categorising participants in this manner facilitates a meaningful analysis of age-related differences in health knowledge, behaviours and outcomes.²³

The researchers categorised 21 maternal health knowledge statements into four thematic groups to facilitate understanding of the phenomenon: pregnancy and well-being (11 items), foetal development and monitoring (6 items), preparation for labour (2 items) and postnatal care (2 items). Socio-demographic characteristics were presented using frequencies and percentages. In addition, the Likert scale options were recoded as ‘I don’t know’ (0) and ‘I know a lot’ (1). Maternal health knowledge statements were analysed using mean, standard deviation and 95% confidence interval (CI). The Mann–Whitney test was used to test significant associations between maternal health knowledge statements and socio-demographic characteristics to test the hypotheses. The Mann–Whitney (U) test was employed to examine differences between two independent groups because the outcome variable was not normally distributed. The maternal health knowledge was ranked into three categories, namely inadequate (0–50/100), problematic (50.1 to 66/100), sufficient (66.1–84/100) and excellent (85 and above). Combining sufficient and excellent simplifies the presentation of results without losing meaningful distinctions relevant to the research objectives. The ‘sufficient’ and ‘excellent’ categories were combined into the ‘sufficient to excellent’ (66.1 – 100) category (called desired level of maternal health knowledge) because of the small number (n = 6, 2.4%) of respondents in the ‘excellent’ category. There were no missing data in the population. Therefore, there was no need for the researchers to address the bias.

Ethics approval was obtained from the University of the Western Cape Humanities and Social Sciences Research Ethics Committee (HSSREC), reference number REC: BM21/02/05. Permission to conduct the study was sought from the Western Cape Department of Health (DOH), sub-district manager and antenatal clinic managers. Participation in the survey was voluntary, and no personal information was collected from respondents. Participants were informed about their voluntary participation, that their responses were anonymous and that they could easily withdraw from the survey at any given time.

The results show the socio-demographic characteristics according to parity of 248 pregnant women (response rate: n = 248/267, 92.8%) who provided informed consent to participate in this study (Table 1). The mean age of the respondents was 27.4 years (standard deviation [s.d.] = 7.53), with most of the respondents (n = 204, 82.3%) being less than 35 years old. More than half (n = 133, 53.6%) of the respondents did not have a partner, and the majority of the respondents (n = 222, 89.5%) had attained secondary education and higher. More than half (n = 126, 50.8%) of the respondents did not know the gestational age of their pregnancy. Over half (n = 127, 52.2%) of the respondents were employed, and three-quarters (n = 186, 75.0%) of the respondents lived with other people, such as family and the baby’s father (Table 1). Significant differences were observed between maternal health knowledge and level of education (Table 1). Respondents who had secondary education and higher were more likely to have low maternal health knowledge than their counterparts who had primary education (89.4 vs. 10.6, χ² = 3.570, p < 0.001). Furthermore, a significant association between maternal health knowledge and living status, where the respondents who live alone were more likely to have low maternal health knowledge than their counterparts who live with others (75.8 vs. 24.2, χ² = 4.568, p < 0.028).

The maternal health knowledge subscale was measured using 21 statements, with higher mean scores indicating higher levels of self-rated maternal health knowledge. The overall score for maternal health knowledge was: 58.50 out of 100, s.d. = 13.79, range: 13–100. More than half of the respondents (n = 127; 51.2%, 95% CI: 45.0–57.4) were categorised as possessing problematic maternal health knowledge, with scores ranging from 50.1 to 66 (Table 2).

An additional 55 participants (22.2%, 95% CI: 17.0–27.4) demonstrated inadequate knowledge (scores between 0 and 50), while 66 respondents (26.6%, 95% CI: 21.1–32.1) exhibited sufficient to excellent knowledge, scoring between 66.1 and 100 (see Table 2). A higher proportion of the respondents with problematic maternal health knowledge were in their second or subsequent pregnancy (n = 74, 58.2%) compared to their first pregnancy (n = 53, 41.7%) (Figure 1), but the differences were not statistically significant (Table 3). Similarly, no significant differences were found between maternal health knowledge and level of education and gestational age.

Respondents rated their knowledge on foetal development and monitoring domain the highest 3.4 (s.d. = 0.77), (95% CI: 3.30–3.49); followed by pregnancy and well-being 3.33 (s.d. = 0.65), (95% CI: 3.25–3.41); and preparing for labour 3.22 (s.d. = 1.06), (95% CI: 3.19–3.46). The lowest rated is the postnatal care 3.26 (s.d. = 1.17), (95% CI: 3.11–3.40). There were no significant differences between the average knowledge scores for the domains of knowledge (Figure 2). The comparison is limited to these domains, and the overlap of the 95% CIs indicates the relative similarity in maternal health knowledge levels across domains, rather than statistically tested differences between them.

Pregnancy health and well-being knowledge was rated with 11 statements, with an overall mean of 3.33, (s.d. = 0.65), (95% CI: 3.25–3.41). The highest rated maternal health knowledge statement was ‘I know natural physical changes during pregnancy’ 3.75 (s.d. = 1.15), (95% CI: 3.60–3.89). This was followed by the statement ‘I know natural psychological changes during pregnancy’ 3.60 (s.d. = 1.23), (95% CI: 3.44–3.75). The lowest-rated statements were ‘I know proper activity and status in pregnancy’ (M = 3.15, s.d. = 1.33), (95% CI: 2.98–3.32) and ‘proper exercise during pregnancy’ (M = 3.15, s.d. = 1.33), (95% CI: 2.88–3.30).

Respondents with more than one pregnancy (M = 3.54, s.d. = 1.16) rated their knowledge of the acceptable and normal amount of weight gain during pregnancy significantly higher than first-time pregnant respondents (M = 3.07, s.d. = 1.23), U = 89.46, p = 0.003). Similarly, they rated knowledge of common pregnancy-related problems such as nausea, vomiting and lower back pain, significantly higher (M = 3.31, s.d. = 1.19) than those in their first pregnancy (M = 2.89, s.d. = 1.23) (U = 94.22, p = 0.001) and knowledge of the safety and appropriateness of vaccines during pregnancy (M = 3.35, s.d. = 1.17 vs. M = 3.00, s.d. = 1.13), respectively (U = 86.10, p = 0.019) (Table 3). There was a relationship between the pregnancy health and well-being items and number of pregnancies.

Foetal development and monitoring were rated the most knowledgeable area (M = 3.40, s.d. = 0.77), (95% CI: 3.30–3.49). The highest rated statement was ‘I know the appropriate referral timing for pregnancy examinations (visits)’ (M = 3.69, s.d. = 1.18), (95% CI: 3.54–3.84), and the lowest-rated statement was ‘I know the normal number of foetal movements’ (M = 3.14, s.d. = 1.34), (95% CI: 2.97–3.31) (Table 3). Respondents with more than one pregnancy rated knowledge of diagnostic examinations (e.g. ultrasound and laboratory tests) significantly higher than those with a first pregnancy (M = 3.79, s.d. = 1.07 vs. M = 3.47, s.d. = 1.12), respectively (U = 85.14, p = 0.025). However, respondents with a first pregnancy rated their knowledge of warning or risk signs during pregnancy higher than those with more than one pregnancy (M = 3.53, s.d. = 1.30 vs. M = 3.12, s.d. = 1.35), respectively (U = 60.59, p = 0.017) (Table 3). There were no significant differences were found for level of education and gestational age.

This study provides valuable insights into the existing body of knowledge on maternal health literacy. The primary outcome of this study was maternal health knowledge, measured using 21 items of the MHELIP scale. There are few studies conducted in Africa that specifically used maternal health knowledge of MHELIP scale. One study that employed the MHELIP scale conducted among pregnant women in Eswatini found that higher levels of maternal health literacy were significantly associated with more appropriate utilisation of antenatal care services, compared to women with lower health literacy.²⁴ A second study that used the MHELIP scale, conducted in Nigeria, discovered that about 41.6% of respondents had inadequate maternal health literacy.²⁵ The two studies demonstrated socio-demographic characteristics comparable to those of this study, particularly with respect to the pregnant women’s levels of education, employment status and age.

In this study, respondents exhibited a problematic level of maternal health knowledge, with an overall mean score of 58.50 out of 100 (s.d. = 13.79; range: 13–100), indicating inadequate to problematic health knowledge among 182 participants (73.4%). This finding contrasts with a study conducted in Turkey, which reported that pregnant women possessed a sufficient level of maternal health knowledge, with an average score of 68.44.²⁶ This is also supported by the finding that over a half (50.8%) of the respondents did not know the gestational age of the current pregnancy. Not knowing gestational age is inconsistent with studies conducted in Ethiopia and United States, which reported that most pregnant women were aware of the gestational age of pregnancy and were accurate in estimating their pregnancy duration.^9,27 This finding suggests that variations in overall maternal health literacy and that knowledge of gestational age may be an indicator of overall maternal health literacy in some settings.

Most pregnant women in our study rated their knowledge regarding natural physical changes during pregnancy as the highest 3.75 (s.d. = 1.15). A study conducted in Saudi Arabia concurs with this study, approximately 82% of the women recognised mood swings as a normal symptom of pregnancy, 80.1% were aware that nausea commonly occurs, 75.9% identified fatigue as a typical symptom and 68.9% knew that fainting or dizziness can also be experienced during pregnancy.²⁸ This suggests that most pregnant women in this study had a reasonably good level of knowledge about the natural physical changes that occur during pregnancy. The result not only reflects a positive trend in maternal health knowledge but also suggests that education on pregnancy-related body changes should continue to ensure all pregnant women are well informed, especially those who may fall below the average knowledge level.

Our findings revealed that pregnant women rated knowledge of physical activity and exercise in pregnancy as the lowest 3.15 (s.d. = 1.33). In agreement with this study is a study conducted in Iraq that reported nearly two-thirds (60%) of women had poor knowledge levels about physical exercise during pregnancy.²⁹ This suggests that pregnant women in this study have a poor understanding of the normal physiological adaptations of pregnancy and physical activity, although some variation remains, indicating the potential benefit of reinforcing educational support during antenatal care. The lack of knowledge on proper activity during pregnancy might lead to risks of problems such as excessive weight gain, pregnancy conditions (gestational diabetes and hypertension disorders) and decreased stamina for labour and delivery. This low knowledge in both studies may be attributed to limited antenatal health education, inadequate counselling by healthcare providers and cultural beliefs that discourage physical activity during pregnancy.

Knowledge regarding the appropriate timing for referral during pregnancy examinations was rated relatively high, 3.69 (s.d. = 1.18). In contrast with the study conducted in Ethiopia, 48% of the respondents reported that they did not know the right time and purpose of making the antenatal booking.³⁰ Knowledge of appropriate referral timing for pregnancy examinations is instrumental to prevent complications for maternal and neonatal death. We found that pregnant women rated knowledge on the normal number of foetal movements as the lowest 3.14 (s.d. = 1.34). Similarly, another study conducted in Indonesia reported that more than three-quarters (n = 136, 79.4%) of the respondents had limited or poor knowledge about the normal number of foetal movements.³¹ The finding of this study is of concern as limited/diminished foetal movement is a danger sign of pregnancy and missing it might result in foetal death. The difference could be because of the differences in the timing of antenatal check-up. There is a need to empower pregnant women about the normal number of foetal movements and accelerate efforts to prevent complications of pregnancy that lead to death. The empowerment will assist in achieving the Sustainable Development Goal number 3, which seeks to ensure health and well-being for all.

The study’s results indicated that pregnant women rated their knowledge of pain relief methods during vaginal delivery as the highest 3.33 (s.d. = 1.27). Similarly, a South African study found that more than half (n = 200, 52.1%) of pregnant women lack knowledge regarding non-pharmacological pain relief methods.³² The finding in this study indicates that pregnant women might be able to request pain relief methods to help them cope during labour, which could make the delivery process easy and without complications.

Knowledge of postpartum care of the mother was rated among the lowest 3.30 (s.d. = 1.29). The findings of this study are inconsistent with those of a study conducted in Ethiopia, which revealed that postpartum mothers were knowledgeable about postpartum topics such as heavy vaginal bleeding and healthy, balanced foods.³³ The current results show that the most of pregnant women demonstrated lower understanding of the knowledge required for postpartum care of the mother. This suggests that while many women are aware of essential aspects of pregnancy and well-being, there remains a need for strengthened education and support to ensure comprehensive preparedness for the postpartum period.

Another contrasting finding comes from a study conducted in Nigeria among pregnant women attending specialist hospitals, which reported that participants demonstrated a significant level of knowledge regarding postnatal care.³⁴ Although the Nigerian study was conducted among pregnant women attending specialist hospitals, its reported level of postpartum care knowledge is comparable to that observed in this study, which indicates that pregnant women remain largely unaware of key aspects of postpartum care, including physical recovery, mental health and the importance of follow-up medical visits.

Lastly, the lowest-scoring item on the maternal health literacy scale in the postpartum period pertained to neonatal and infant care (3.14, s.d. = 1.38). A study conducted in Ghana revealed that more than a half (50.7%) of the mothers had good knowledge about the essential new-born care.³⁵

The midwives/nurses working at antenatal clinics in the Western Cape should provide pregnant women with information on different aspects of maternal health knowledge, such as pregnancy and wellness, foetal development and monitoring, preparation for labour and postnatal care. Future studies should endeavour to include pregnant women attending antenatal clinics in other sub-districts. The antenatal units should be teaching pregnant women on the identification of normal foetal movements during pregnancy, as well as on factors influencing foetal health, such as medication use and pregnancy-related disease symptoms. According to the National Integrated Maternal and Perinatal Care Guidelines, these symptoms are recognised as danger signs of pregnancy, indicating potential complications that require timely intervention.³⁶ Further research is warranted to explore the perspectives of pregnant women, and maternal health knowledge should be systematically incorporated into nursing practice.

This is one of the few studies conducted in South Africa to assess maternal health knowledge of pregnant women using the MHELIP scale, with 48 statements, of which 21 statements measure maternal health knowledge (21 items). A key strength of this study lies in its focus on maternal health knowledge among pregnant women, an area in which limited research has been conducted in the region, particularly using this measurement scale.

The study was limited to pregnant women attending three antenatal clinics in the Northern Tygerberg Sub-district, which may not represent the broader population of pregnant women in other regions or healthcare settings leading to selection bias. Furthermore, pregnant women who did not attend antenatal care were excluded from the study. Future studies should endeavour to include pregnant women attending antenatal clinics in other sub-districts. There could have been potential bias observed in the results. The respondents were recruited from clinic attendees, potentially excluding women who do not access antenatal care services or those receiving care from non-facility providers. This may introduce selection bias and over-represent women who are more health-conscious or have better access to care.