What urban–rural differences exist in cultural norms and maternity care preferences: a cross-sectional study among women receiving antenatal, postnatal, and child under 5 years health services in Bo District, Sierra Leone

Abstract

Introduction

To assess rural–urban differences in cultural norms, traditional practices and preferences for culturally responsive maternity care in Bo District, Sierra Leone. Exploring urban–rural differences is essential because women living in these settings often experience distinct cultural influences and healthcare barriers that shape maternity care preferences and may require context-specific interventions.

Methods

Facility-based descriptive cross-sectional study at Bo Government Hospital (urban) and Tikonko Maternity Center (rural). Participants were 500 pregnant women, postpartum mothers and mothers of children under 5 years of age who attended maternal/child health services and resided in Bo for ≥6 months. Data were collected through interviewer-administered structured questionnaires using an Open Data Kit system called ONA—a mobile platform. Trained research assistants conducted face-to-face interviews in participants’ preferred language (Krio or Mende), recorded responses electronically in real time and obtained informed consent before data collection.

Outcomes included religion, family type, marital status and 10 culturally responsive strategies (1866 multiple-response selections). χ² tests, proportion differences and 95% CIs assessed rural–urban differences.

Results

Residency was not associated with religion (p=0.901), family type (p=0.235) or marital status (p=0.541). Cultural competence training was most endorsed (26.6%). Urban women preferred integrating safe traditional practices into facility care (χ²=9.742, p=0.0018), whereas rural women preferred language-support services (χ²=26.044, p<0.001). Other strategies showed no meaningful rural–urban differences.

Conclusions

Both groups valued culturally respectful maternity care. Cultural competence training was a priority, while preferences for traditional-practice integration and language support indicated context-specific urban and rural needs, informing culturally responsive interventions and policy in Sierra Leone.

WHAT IS ALREADY KNOWN ON THIS TOPIC.

WHAT THIS STUDY ADDS

• The study showed that rural and urban women in Sierra Leone share many preferences for culturally responsive maternity care, although important differences exist in selected culturally responsive strategies. Provider cultural competence emerged as the most strongly endorsed strategy across both settings; urban women showed greater support for integrating safe traditional practices into facility-based care, while rural women expressed stronger preferences for language support services. The study added that culturally responsive maternity care should combine universal strategies with context-specific approaches that address the differing needs and preferences of rural and urban populations.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY.

• The study provided evidence that informs the design of culturally responsive maternal health policies and programmes in Sierra Leone and other low-resource settings by supporting strategies that strengthen healthcare providers’ cultural competence, improve communication through local-language support and appropriately integrate safe traditional practices, where acceptable, to enhance the responsiveness and acceptability of maternity services.

Introduction

Maternal and newborn health (MNH) remains a major public health priority across sub-Saharan Africa (SSA), particularly in countries such as Sierra Leone, where maternal mortality continues to rank among the highest globally despite sustained investments in health-sector reforms.^1–3 Persistent maternal and neonatal deaths reflect deeply rooted structural, cultural, and social factors that influence how women access, perceive, and use healthcare services. Urban–rural inequalities further intensify these challenges: while women living in urban areas generally benefit from greater access to health facilities, skilled providers, and educational opportunities, women in rural communities continue to face barriers related to geographic isolation, poverty, and entrenched sociocultural norms. ⁴

Cultural beliefs and gender norms remain powerful determinants of maternal healthcare-seeking behaviour in Sierra Leone, particularly in rural and semi-urban communities where patriarchal systems strongly influence household decision-making.⁵ In many settings, pregnancy and childbirth are embedded within traditional beliefs, family expectations and community customs, which may discourage women from seeking skilled maternity care and reinforce reliance on traditional birth attendants and home-based delivery practices.^5,6 Socioeconomic constraints, including limited education, financial hardship and restricted decision-making autonomy, further compound these challenges, creating a complex interaction between structural disadvantage and cultural expectations that limits timely access to quality MNH.^7–9

Growing evidence from West Africa suggests that improving MNH requires interventions that extend beyond biomedical care to become culturally responsive, meaning that healthcare is delivered in ways that recognise, respect and appropriately respond to the cultural beliefs, values, languages and practices of the populations being served.¹⁰ Such interventions must also be socially grounded, that is, informed by the everyday social realities, family structures, community relationships and local contexts that shape women’s health decisions and experiences. Within this approach, cultural competence refers to the knowledge, skills, attitudes and behaviours that enable healthcare providers to communicate effectively and deliver respectful, equitable and culturally appropriate care to people from diverse backgrounds. Although respectful maternity care and cultural competence have received increasing attention, important knowledge gaps remain regarding how women themselves perceive and prioritise these approaches across different residential settings in Sierra Leone.

Existing studies have documented the influence of culture on maternal healthcare utilisation, but few have systematically compared urban and rural women’s perspectives on integrated strategies—coordinated interventions that combine clinical care with culturally appropriate communication, community engagement, respectful provider practices and the safe incorporation of acceptable traditional practices to improve service delivery. Consequently, there is limited evidence on how women from different settings prioritise strategies designed to strengthen culturally responsive MNH services, defined as MNH services that intentionally integrate high-quality clinical care with respect for patients’ cultural values, preferences, language and social context while maintaining evidence-based standards of care. This evidence gap is particularly important because persistent urban–rural disparities remain among the strongest determinants of MNH outcomes in Sierra Leone.

This study addresses this important gap by examining cultural norms, traditional practices and preferences for culturally sensitive maternal, newborn and child healthcare improvement strategies among women living in both urban and rural communities in Bo District, Sierra Leone. Specifically, the study evaluates how women perceive integrated strategies such as cultural competence training for healthcare providers, respectful communication, culturally welcoming health facility environments and the safe inclusion of culturally acceptable traditional practices within facility-based maternity care. By comparing these perspectives across residential settings, the study generates evidence to inform national policies and programmes aimed at improving equitable, person-centred MNH.

The novelty of this study lies in its comparative urban–rural perspective and its examination of integrated, culturally responsive interventions, an area of MNH research that remains underexplored in Sierra Leone and much of SSA. By demonstrating how cultural beliefs, social context and women’s preferences shape healthcare utilisation, the findings provide practical guidance for policymakers, programme implementers and healthcare providers seeking to strengthen trust in health services, improve utilisation of skilled maternity care and reduce preventable maternal and newborn deaths. Ultimately, the study highlights that sustainable improvements in MNH require healthcare systems that are both clinically effective and culturally responsive, while remaining firmly rooted in the social realities of the communities they serve.

Methodology

Study design

A descriptive cross-sectional health facility-based study design was used to examine the influence of integrated cultural strategies on maternal healthcare-seeking behaviour among pregnant women, postpartum mothers and mothers of children under 5 years of age in Bo District, Sierra Leone. This design was appropriate for capturing population characteristics, cultural norms and reported healthcare practices within a clearly defined period, allowing for meaningful comparison between urban and rural settings.

Study setting and population

The study was conducted in Bo District, located in the Southern Province of Sierra Leone, from January to May 2024. Bo District comprises both urban and rural communities with marked differences in access to healthcare, socioeconomic conditions and sociocultural practices, making it an appropriate setting for examining variations in MNH behaviours across residential contexts.

Two purposively selected health facilities representing urban and rural settings were included in the study:

• Bo Government Hospital (BGH): A tertiary-level teaching and referral hospital located in Bo City that serves a large and diverse urban and peri-urban population. The hospital provides comprehensive maternal, newborn and child healthcare services, including specialist obstetric and neonatal services, and receives referrals from lower-level health facilities across the District.

• Tikonko Maternity Center: A rural primary-level maternal health facility located in Tikonko Chiefdom that serves predominantly rural communities. The catchment is characterised by greater geographic barriers to healthcare access and stronger influence of traditional beliefs, cultural practices and community norms on maternal healthcare decisions.

These two settings were intentionally selected because they represent contrasting healthcare environments and sociocultural contexts within the same district.

Comparing pregnant women, postpartum mothers and mothers with children under 5 years of age attending an urban referral hospital with those receiving care at a rural primary health facility provided an opportunity to examine how cultural norms and beliefs, traditional practices, healthcare experiences and preferences for culturally responsive and integrated strategies to improve maternal, newborn and children under 5 years of age healthcare across diverse community settings.

This context is particularly relevant because urban–rural disparities in healthcare access and sociocultural influences remain important determinants of maternal healthcare utilisation and outcomes in Sierra Leone.

The study population comprised pregnant women, postpartum mothers and mothers of children under 5 years of age who attended antenatal, postnatal or children under 5 years of age healthcare services at either health facility during the study period.

Eligibility criteria

Inclusion criteria

Pregnant women, postpartum mothers or mothers with children under 5 years of age who had lived in the Bo District for ≥6 months before the study were eligible. A participant must have attended antenatal care (ANC), postnatal care (PNC) or received healthcare for a child under 5 years of age at one of the selected facilities. Only participants who were able and willing to provide informed consent met eligibility criteria.

Exclusion criteria

Participants were excluded if they had not accessed services at the selected facilities, did not meet the residency requirement, had medical, mental or communication limitations preventing participation or declined informed consent.

Sampling frame construction

To ensure methodological rigour, a facility-based sampling frame was constructed using ANC, PNC and children under 5 years of age attendance records from both sites: the BGH and Tikonko Maternity Center.

Extraction of Attendance Records was conducted by reviewing logbooks and electronic registers for the previous 1–2 months. This helped in the compilation of lists of all pregnant women and mothers who met the inclusion criteria.

Cleaning and verification were done by removing duplicate entries, incomplete records and non-resident cases. Residence was verified using clinic intake forms and, where necessary, cross-checked with village health worker registers.

Stratification sampling frame

The final sampling frame was stratified across facility type (urban vs rural) and maternal category (pregnant women vs postpartum mothers and mothers with children under 5 years of age).

Proportional allocation

Samples were allocated proportionally based on facility attendance volumes, maintaining representativeness across strata.

A simple random sampling technique was applied, using a computer-generated random number.

Selected participants who declined or could not be reached were replaced via a predefined replacement protocol, using the next random number on the list.

This method ensured equitable representation, minimised selection bias and produced a transparent and reproducible sampling process.

Sample size determination

Sample size was determined using Cochran’s formula for population proportions:

$n_0=\frac{Z^2\cdot p(1-p)}{e^2}$

where (Z=1.96) represented the 95% confidence level, (p=0.50) assumed maximum variability because no prior estimate of the outcome prevalence was available and (e=0.05) represented the desired margin of error. Under these calculations, the minimum required sample size was approximately 384 participants.

The calculated sample was increased by approximately 30%, yielding a final target of 500 participants (294 pregnant women and 206 postpartum mothers and mothers of children under 5 years of age). This increase was planned a priori to compensate for anticipated non-response and incomplete interviews, and to ensure adequate statistical precision for subgroup comparisons among pregnant women, postpartum mothers and mothers with children under 5 years of age in urban and rural study settings.

Although participants were recruited from two health facilities, the study did not employ a multistage cluster sampling design. Instead, the two facilities were purposively selected to represent the urban tertiary-level and rural primary-level maternal healthcare settings in Bo District. Separate sampling frames were constructed from antenatal, postnatal and mothers with children under 5 years of age attendance registers at each facility, after which participants were selected using simple random sampling within predefined strata based on facility and maternal category. Consequently, the primary sampling unit was the individual participant rather than the health facility, and a formal design effect was not incorporated into the sample size calculation.

Nevertheless, because observations obtained within the same facility may exhibit some degree of correlation, recruiting a substantially larger sample (500) than the minimum required (384) increased the precision of study estimates and strengthened comparisons across the two settings.

Outcome definitions

Clear operational definitions were established a priori to ensure consistent measurement of the study outcomes across both study sites.

The primary outcomes were participants’ cultural and sociodemographic characteristics, operationalised as place of residence (urban or rural), religious affiliation, family types and marital status. These variables were selected based on evidence from previous maternal, newborn and child healthcare literature demonstrating their relevance as sociocultural determinants of healthcare-seeking behaviour in low-resource settings and their potential contribution to urban–rural differences in maternity care preferences. Residence was classified according to the location of the health facility where participants received maternal health services. Religious affiliation was self-reported and categorised into predefined groups. Family types were classified as polygamous or monogamous, while marital status was categorised using standard demographic classifications (married, single, divorced/separated or widowed). Maternal age, parity, education level, household income and tribal affiliation were also measured.

The secondary outcomes were women’s preferences for culturally responsive maternity care strategies. These were measured using a structured questionnaire that assessed participants’ level of agreement with proposed culturally responsive approaches, including healthcare providers’ cultural competence, respectful communication, language support services, culturally welcoming health facility environments, family involvement and the safe integration of culturally acceptable traditional practices into facility-based maternity care.

The questionnaire was developed following a review of the MNH and culturally responsive healthcare literature and was refined through expert review, pretesting and community consultation before data collection. Responses were analysed descriptively and comparatively to identify differences in women’s preferences between urban and rural settings. Participants’ preferred strategies facilitated interpretation of levels of endorsement rather than representing a formally validated psychometric index.

Data collection instruments and procedure

Data were collected using a structured, interviewer-administered electronic questionnaire programmed on the ONA platform, an open-source mobile data collection application that supports secure electronic data capture, real-time data synchronisation and built-in quality-control checks. The electronic system minimised transcription errors, reduced missing data through programmed validation rules and enhanced data quality.

The questionnaire was developed from an extensive review of relevant literature, the study objectives and the conceptual framework underpinning maternal health-seeking behaviour and cultural influences on maternal healthcare utilisation. It comprised five domains: sociodemographic characteristics, cultural norms and beliefs, traditional maternal health practices, maternal healthcare utilisation and participants’ preferred approaches for integrating culturally appropriate practices into formal maternal healthcare services.

The culturally responsive approaches assessed in this study were developed through a structured process combining evidence from published literature on culturally competent care, respectful maternity care, patient-centred healthcare and community engagement, and a conceptual framework examining the influence of cultural and socioeconomic factors on maternal healthcare-seeking behaviour. The selected approaches reflected key domains of culturally responsive maternal healthcare, including provider cultural competence, respectful communication, community participation, incorporation of acceptable traditional practices where safe, language support and patient involvement in decision-making. These concepts were adapted to the Sierra Leonean maternal healthcare context and translated into participant-friendly statements. During interviews, participants selected all approaches they considered important; therefore, responses represent multiple preferences and are not mutually exclusive.^11–13

Before the conduct of the survey, the electronic questionnaire underwent a pilot test at a comparable health facility—the Njala University Medical School Campus Clinic—to evaluate question clarity, logical flow, cultural appropriateness, interviewer consistency and functionality of the electronic data collection system. Feedback obtained from participants during the pilot testing was invaluable in optimising the questionnaire. Participants’ observations on question clarity, language, cultural appropriateness and response flow informed iterative refinements to the instrument, resulting in revisions to item wording, sequencing, skip patterns and electronic programming before commencement of the main data collection.

After this pilot testing, eligible participants during the study were recruited daily from antenatal, delivery, postnatal and child welfare clinic attendance registers at each of the selected health facilities.

Each day, 10 trained enumerators reviewed the facility registers to identify women who met the study eligibility criteria. Potential participants were approached after receiving routine healthcare services; the study objectives and procedures were explained, and written informed consent was obtained before enrolment.

10 trained enumerators conducted face-to-face oral interviews and filled out all digital forms during the study data collection process.

Interviews were conducted in a private setting to ensure confidentiality while using participants’ preferred language. Standardised interviewer guidance, together with forward translation into Krio and Mende and back-translation, was used to ensure consistent interpretation of key concepts and maintain conceptual equivalence across languages. Daily supervision, Global Positioning System (GPS) verification and random call-backs (50% sample) by the enumerators’ supervisor were used to maintain data quality.

The study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement, with all methodological procedures, analyses and findings presented in line with internationally recognised guidelines for transparent and complete reporting of cross-sectional observational studies.

Data analysis

Data were analysed using IBM SPSS Statistics V.26. Before analysis, the dataset underwent quality assurance procedures through completeness checks, consistency verification, range validation and outlier assessment. Descriptive statistics were used to summarise participants’ characteristics and study variables. Categorical variables were presented as frequencies and percentages, whereas continuous variables were summarised using means and SD or medians and IQRs, where appropriate.

Comparisons between urban and rural participants were performed using Pearson’s χ² test for categorical variables. Statistical significance was assessed using two-sided tests with a significance level of p<0.05. To facilitate interpretation of the findings, 95% CIs and effect estimates were considered alongside p-values, with greater emphasis placed on the magnitude and precision of observed differences rather than statistical significance alone.

The analysis was conducted as prespecified in the study, because the study employed a descriptive cross-sectional design with predefined bivariate analyses and did not involve alternative model specifications or multiple analytical assumptions; sensitivity analyses were not undertaken.

Patient and public involvement

Women from the study communities were not involved in the initial design of the research question, outcome measures or study tools. However, to ensure the study remained grounded in local realities, we engaged community representatives, including senior midwives, Community Health Officers and women attending antenatal and postnatal services, during the pretesting of the questionnaire. Their feedback helped refine the clarity, cultural relevance and acceptability of the survey items.

During data collection, participants contributed by sharing their experiences, cultural beliefs and preferences for maternal healthcare, which directly informed the interpretation of the findings. Although patients did not participate in decisions on recruitment, data analysis or dissemination strategies, the research team plans to share the study results with the participating health facilities in Bo District through community feedback meetings and stakeholder engagement sessions. These sessions will allow women, midwives and community leaders to discuss how the findings can inform culturally responsive and respectful maternity care within their local health systems.

Result

Sociodemographic, reproductive and household characteristics of participants

Tables 1 and 2 summarise the sociodemographic, reproductive, residential, cultural and household characteristics of the 500 study participants.

Table 1. Summary of statistics for continuous sociodemographic and reproductive characteristics of study participants (N=500), in Bo District, Sierra Leone.

Bio data	Statistical summary
Age in years (pregnant women)	Mean: 24.83±4.74; SE: 0.28; range: 16–45; 95% CI 24.29 to 25.37; median: 25; mode: 25
Age in years (postpartum mothers and mothers with children under 5 years old)	Mean: 24.07±4.34; SE: 0.30; range: 17–45; 95% CI 23.47 to 24.67; median: 23; mode: 20
Monthly income (SLE)	Mean: 335.03±668.49; SE: 9.90; range: 0–10,000; 95% CI 276.29 to 393.77
Parity	Mean: 1.83±1.05; SE: 0.05; range: 1–7; 95% CI 1.74 to 1.92

Values are presented as n (%) for categorical variables and mean±SD for continuous variables.

Parity, number of pregnancies experienced; currency measure; SLE, Sierra Leonean Leone.

Table 2. Distribution of sociodemographic, residential, cultural and household characteristics among study participants (N=500), in Bo District, Sierra Leone.

Category	Subcategory	Frequency (n=500)	%
Educational level	No formal education	50	10.00
	Primary education	142	28.40
	Secondary education	268	53.60
	Tertiary education	40	8.00
Residency	Urban (Bo Government Hospital)	374	74.80
	Rural (Tikonko Maternity Center)	126	25.20
	Total	500	100
Religion	Islam	262	52.40
	Christianity	237	47.40
	Indigenous/traditional religions	1	0.20
	Total	500	100
Family type	Polygamous	66	13.20
	Monogamous	434	86.80
	Total	500	100
Marital status	Married	233	46.60
	Cohabitation	82	16.40
	Single/widow	185	37.00
	Total	500	100

Values are presented as n (%) for categorical variables and mean ± SD for continuous variables.

The mean age of pregnant women was 24.83 years (SD=4.74; 95% CI 24.29 to 25.37), with ages ranging from 16 to 45 years. Postpartum mothers and mothers with children under 5 years had a comparable mean age of 24.07 years (SD=4.34; 95% CI 23.47 to 24.67), with an age range of 17–45 years. The mean monthly household income was SLE 335.03 (SD=668.49; 95% CI 276.29 to 393.77), indicating substantial variability in income among participants. The average parity was 1.83 pregnancies (SD=1.05; 95% CI 1.74 to 1.92), with values ranging from one to seven pregnancies (table 1).

More than half of the participants had attained secondary education (53.6%, n=268), followed by primary education (28.4%, n=142), while 10.0% (n=50) had no formal education and 8.0% (n=40) had tertiary education (table 2).

Most participants were recruited from the urban study site, BGH (74.8%, n=374), whereas 25.2% (n=126) were recruited from the rural study site, Tikonko Maternity Center (table 2).

Religious affiliation was almost evenly distributed, with 52.4% (n=262) identifying as Muslim and 47.4% (n=237) as Christian. Only one participant (0.2%) reported practising an Indigenous or traditional religion (table 2).

The majority of participants belonged to monogamous households (86.8%, n=434), while 13.2% (n=66) were from polygamous households. Regarding marital status, 46.6% (n=233) were married, 37.0% (n=185) were single or widowed and 16.4% (n=82) reported cohabiting with their partners (table 2).

Tables 3 and 4 and figure 1 show that the significant predictor for skilled birth attendance was education level (AOR=0.443; p=0.006). Education level was significantly associated with skilled birth attendance. Any community support (AOR=4.405; p=0.002) was also significantly associated with skilled birth attendance. Women with community support are four times more likely to have skilled birth attendance. The following factors do not independently predict skilled birth attendance after adjustment: age (p=0.995), monthly income (p=0.056), parity (p=0.133), marital status (p=0.484), traditional practices (p=0.873) and study location (p=0.839).

Table 3. Multivariable logistic regression analysis of factors associated with skilled birth attendance among pregnant women, postpartum mothers and mothers with children under age 5 years in Bo District, Sierra Leone.

Predictor	Adjusted OR	95% CI lower	95% CI upper	P value
Age (years)	1.000	0.906	1.103	0.995
Education level (ordinal)	0.443	0.248	0.792	0.006
Household monthly income	1.183	0.996	1.405	0.056
Parity	1.397	0.903	2.161	0.133
Marital status	0.739	0.316	1.727	0.484
Any community support	4.405	1.758	11.037	0.002
Traditional practices index	0.986	0.825	1.178	0.873
Bo Government Hospital (urban site) vs Tikonko Maternity Center (rural site)	1.146	0.309	4.250	0.839

Table 4. Adjusted ORs and 95% CIs for factors associated with skilled birth attendance among pregnant women, postpartum mothers and mothers with children under age 5 years in Bo District, Sierra Leone.

Predictor	Adjusted OR	Lower error	Upper error
Age (years)	1.000	0.094	0.104
Education level (ordinal)	0.443	0.195	0.349
Household monthly income	1.183	0.187	0.222
Parity	1.397	0.494	0.764
Marital status	0.739	0.423	0.988
Any community support	4.405	2.647	6.632
Traditional practices index	0.986	0.161	0.192
Bo Government Hospital (urban site) vs Tikonko (rural site)	1.146	0.837	3.104

Figure 1. Forest Plot of Adjusted Odds Ratios and 95% Confidence Intervals for Factors Associated with Skilled Birth Attendance among Pregnant Women and Mothers with Children under Five Years Old in Bo District, Sierra Leone.

Table 5 depicts no statistically significant differences between urban and rural residents for religion, family type or marital status. Religious affiliation showed similar proportions of Islam and Christianity in urban and rural respondents (Islam: 52.9% urban vs 50.8% rural), and the χ² test did not detect a difference (χ²(2)=0.21, p=0.901; OR for Islam vs other=1.11, 95% CI 0.76 to 1.63). Family structure (monogamous vs polygamous) also did not differ significantly by residence (monogamy: 85.6% urban vs 90.5% rural; χ²(1)=1.41, p=0.235; OR=0.65, 95% CI 0.32 to 1.32). Marital status distributions (married, cohabiting, single) were likewise similar across residency (χ²(2)=1.23, p=0.541; OR for married vs others=0.84, 95% CI 0.57 to 1.24). None of the reported ORs excluded 1, and all p values >0.05, indicating no evidence of an association between urban/rural residency and these cultural/traditional variables in our sample.

Table 5. Bivariate association between place of residence and sociocultural characteristics of respondents, in Bo District, Sierra Leone.

Variable	Categories	Urban (n=374)	Rural (n=126)	χ² (df)	P value	95% CI
Religion	Islam	198 (52.9%)	64 (50.8%)	0.21 (2)	0.901	OR: 1.11 (0.76 to 1.63)
	Christianity	175 (46.8%)	62 (49.2%)	–	–	–
	Indigenous	1 (0.3%)	0 (0%)	–	–	–
Family type	Monogamous	320 (85.6%)	114 (90.5%)	1.41 (1)	0.235	OR: 0.65 (0.32 to 1.32)
	Polygamous	54 (14.4%)	12 (9.5%)	–	–	–
Marital status	Married	170 (45.5%)	63 (50.0%)	1.23 (2)	0.541	OR: 0.84 (0.57 to 1.24)
	Cohabiting	60 (16.0%)	22 (17.5%)	–	–	–
	Single	144 (38.5%)	41 (32.5%)	–	–	–

There were no significant associations between place of residence (urban vs rural) and respondents’ religion, family type (monogamous vs polygamous) or marital status; all ORs crossed unity and p values exceeded 0.05 (table 5).

Table 6 provides 1866 total responses generated through the multiple-response analysis, and three clear priority areas emerged regarding culturally responsive maternal healthcare. First, training healthcare professionals in cultural competence was overwhelmingly the most frequently endorsed strategy, accounting for 26.24% of all selections (n=487). Second, respondents demonstrated considerable support for the integration of culturally safe traditional practices, which constituted 12.77% of all selections (n=237). Notably, the vast majority of these endorsements came from urban participants (82.70%), indicating stronger urban demand for structured incorporation of traditional practices into formal maternal health services.

Table 6. Rural–urban comparison of participants’ preferred culturally responsive approaches to maternal healthcare: multiple-response frequencies and percentages (n=1866), Bo District, Sierra Leone.

Recommended approaches	Rural (frequency of yes)	Rural (% of yes)	Urban (frequency of yes)	Urban (% of yes)
Provide training for healthcare professionals on cultural competence and understanding of diverse pregnancy and childbirth practices	119	24.44	368	75.56
Incorporate traditional practices into healthcare services where possible and safe, in consultation with cultural experts	41	17.30	196	82.70
Offer patients the option to consult with or include traditional healers or midwives as part of their healthcare team	31	22.30	108	77.70
Create spaces within healthcare facilities that are welcoming and respectful of cultural diversity and practices.	61	29.00	150	71.00
Ensure that healthcare providers communicate effectively and respectfully with patients about their cultural needs and preferences	66	30.28	152	69.72
Develop and provide informational materials on pregnancy and childbirth that respect and acknowledge cultural variations	38	23.60	123	76.40
Facilitate community engagement initiatives to better understand and integrate community-specific needs and values	32	25.81	92	74.19
Implement flexible healthcare policies that allow for individualised care plans, respecting patients’ cultural backgrounds	35	26.32	98	73.68
Encourage patient and family participation in decision-making processes to ensure culturally aligned healthcare choices	29	25.43	85	74.56
Offer language support services to ensure clear communication for patients from diverse linguistic backgrounds	25	59.52	17	40.48
Total	477		1389

Participants could select more than one preferred approach; therefore, the total number of responses (n = 1,866) exceeds the study sample size (N = 500). Percentages are calculated within each response category by place of residence.

Third, the option for women to consult or include traditional healers or midwives as part of their care team accounted for 11.75% of total responses (n=139), again with urban respondents contributing the greater share (77.70%).

Other approaches, including creating culturally welcoming health-facility spaces (11.37%), respectful communication (8.67%) and culturally sensitive educational materials (7.49%), received moderate but consistent support across settings.

A notable exception to the overall urban dominance in response frequencies was the endorsement of language support services, where rural respondents accounted for almost 60% of all selections. This option represented only 2.26% of total responses.

Although some culturally responsive approaches received lower endorsement than provider cultural competence training and the integration of culturally safe traditional practices, these findings should be interpreted cautiously. Because participants could select multiple responses, lower endorsement likely reflects prioritisation rather than lack of support. Women appeared to prioritise interventions that directly improve provider interactions and acknowledge culturally meaningful practices during pregnancy and childbirth. The low endorsement of language support services likely reflects the widespread use of shared local languages between healthcare providers and women in Bo District.

In table 7, a series of χ²chi-square tests were conducted to examine whether rural and urban respondents differed significantly in the selection of culturally responsive maternity-care approaches. Proportion estimates were calculated using each group’s total number of ‘yes’ selections (rural=477; urban=1389), followed by 95% CIs for the difference in proportions. Because the data derive from a multiple-response design, the analysis assesses the relative emphasis placed on each approach across settings rather than individual-level prevalence.

Table 7. Rural–urban differences in women’s preferences for culturally responsive maternal healthcare approaches in Bo District, Sierra Leone.

Item	Rural_yes	Urban_yes	χ2	P value	Rural_prop	Urban_prop	Diff_prop	95% CI lower	95% CI upper
Provide training for healthcare professionals on cultural competence and understanding of diverse pregnancy and childbirth practices	119	368	0.44	0.507	0.2495	0.2649	−0.0155	−0.0607	0.0298
Incorporate traditional practices into healthcare services where possible and safe, in consultation with cultural experts	41	196	9.742	0.0018	0.086	0.1411	−0.0552	−0.0863	−0.024
Offer patients the option to consult with or include traditional healers or midwives as part of their healthcare team	31	108	0.839	0.3597	0.065	0.0778	−0.0128	−0.039	0.0135
Create spaces within healthcare facilities that are welcoming and respectful of cultural diversity and practices	61	150	1.401	0.2366	0.1279	0.108	0.0199	−0.0142	0.054
Ensure that healthcare providers communicate effectively and respectfully with patients about their cultural needs and preferences	66	152	2.881	0.0896	0.1384	0.1094	0.0289	−0.0061	0.064
Develop and provide informational materials on pregnancy and childbirth that respect and acknowledge cultural variations	38	123	0.356	0.5508	0.0797	0.0886	−0.0089	−0.0374	0.0196
Facilitate community engagement initiatives to better understand and integrate community-specific needs and values	32	92	0.004	0.9487	0.0671	0.0662	0.0009	−0.0251	0.0268
Implement flexible healthcare policies that allow for individualised care plans, respecting patients’ cultural backgrounds	35	98	0.043	0.8363	0.0734	0.0706	0.0028	−0.0242	0.0298
Encourage patient and family participation in decision-making processes to ensure culturally aligned healthcare choices	29	85	0.001	0.975	0.0608	0.0612	−0.0004	−0.0253	0.0245
Offer language support services to ensure clear communication for patients from diverse linguistic backgrounds	25	17	26.044	0	0.0524	0.0122	0.0402	0.0194	0.061

Values represent the number and proportion of participants endorsing each culturally responsive maternal healthcare approach. Rural–urban differences were assessed using Pearson’s χ2 tests. Differences in proportions are presented as rural minus urban proportions with corresponding 95% confidence intervals. A two-sided p<0.05 was considered statistically significant. This table evaluated approaches for each cultural responsive maternal healthcare preference separately and does not model the correlation among multiple responses provided by the same participant. Therefore, the analyses must be interpreted as comparisons of endorsement frequencies for individual approaches rather than respondent-level multivariate preferences.

Training and knowledge enhancement approaches

Training healthcare professionals in cultural competence was the most frequently endorsed approach by women in both rural and urban settings (rural=119; urban=368). The distribution did not differ statistically between groups (χ²=0.44, p=0.507). Rural respondents allocated 24.95% of their total selections to this item compared with 26.49% among urban respondents (difference: −0.0155; 95% CI −0.0607 to 0.0298).

Integration of traditional practices and providers

A strong and statistically significant rural–urban contrast emerged for incorporating safe traditional practices in consultation with cultural experts (rural=41; urban=196). Urban respondents were more likely to emphasise this approach (χ²=9.742, p=0.0018). This difference is reflected in the proportion of group-specific selections (urban: 14.11% vs rural: 8.60%), with a substantive negative difference in proportions (−0.0552; 95% CI −0.0863 to −0.024), indicating a consistently higher urban prioritisation of integrating traditional practices into facility care.

For patient options to consult or include traditional healers or midwives, no significant rural–urban difference was observed (χ²=0.839, p=0.3597). Proportions remained similar (rural: 6.50%; urban: 7.78%), with a small, nonsignificant difference (−0.0128; 95% CI −0.0390 to 0.0135).

Culturally supportive healthcare environments

The recommendation to create culturally welcoming and respectful spaces within facilities showed no statistical difference between settings (χ²=1.401, p=0.2366). Rural selections constituted 12.79% of their total, compared with 10.80% in urban areas (difference: 0.0199; 95% CI −0.0142 to 0.054), suggesting comparable prioritisation.

Similarly, the approach to ensure effective and respectful communication about patients’ cultural needs also showed no statistically significant difference (χ²=2.881, p=0.0896), though rural respondents allocated slightly more selection weight (13.84% vs 10.94%; difference: 0.0289; 95% CI −0.0061 to 0.064).

Information, community engagement and policy flexibility

Developing culturally sensitive informational materials yielded comparable selection patterns (χ²=0.356, p=0.5508), with nearly identical proportions (rural: 7.97%; urban: 8.86%).

Community engagement initiatives also showed no difference (χ²=0.004, p=0.9487), with nearly identical proportional emphasis (difference: 0.0009; 95% CI −0.0251 to 0.0268).

Similarly, flexible healthcare policies that respect individualised cultural needs demonstrated no meaningful variation (χ²=0.043, p=0.8363), and the difference in proportions was negligible (0.0028; 95% CI −0.0242 to 0.0298).

Encouraging family participation in decision-making also showed no rural–urban contrast (χ²=0.001, p=0.975), with an extremely small difference in proportions (−0.0004; 95% CI −0.0253 to 0.0245).

Language support services

A pronounced rural–urban disparity was found for language support services (rural=25; urban=17). This approach was proportionally far more important in rural responses (5.24%) compared with urban selections (1.22%). The χ² statistic indicated strong evidence of a difference (χ²=26.044, p<0.001). The difference in proportions was substantial (0.0402; 95% CI 0.0194 to 0.0610), highlighting linguistic barriers as an especially critical concern in rural settings.

Table 8 presents participants’ preferred strategies for integrating cultural considerations into maternal healthcare services. Because respondents were permitted to select more than one option, the percentages represent the proportion of total responses rather than mutually exclusive categories.

Table 8 Participants'. preferred strategies for integrating cultural considerations into maternal healthcare services (multiple responses, n=1866), Bo District, Sierra Leone.

Recommended approaches	Multiple response frequency (n=1866)	%
Provide cultural competence training for healthcare professionals	497	26.63
Incorporate safe traditional practices through consultation with cultural experts	237	12.70
Ensure respectful communication regarding cultural needs	218	11.68
Create culturally welcoming healthcare environments	211	11.31
Offer informational materials sensitive to cultural variations	161	8.63
Include traditional healers or midwives	139	7.45
Facilitate community engagement	124	6.65
Implement flexible, individualised healthcare policies	123	6.59
Encourage patient and family involvement in healthcare decisions	114	6.11
Provide language support services	42	2.25
	1866	100

The most frequently identified strategy was providing cultural competence training for healthcare professionals, accounting for 497 responses (26.63%).

The second most frequently reported strategy was incorporating safe traditional practices through consultation with cultural experts (237 responses; 12.70%), followed closely by ensuring respectful communication regarding cultural needs (218 responses; 11.68%) and creating culturally welcoming healthcare environments (211 responses; 11.31%).

Additional approaches identified by participants included offering informational materials that are sensitive to cultural variations (161 responses; 8.63%), traditional healers or traditional midwives in appropriate aspects of maternal healthcare (139 responses; 7.45%), facilitating community engagement in maternal health initiatives (124 responses; 6.65%), implementing flexible and individualised healthcare policies (123 responses; 6.59%), and encouraging patient and family involvement in healthcare decision-making (114 responses; 6.11%).

The least frequently selected strategy was providing language support services, which accounted for 42 responses (2.25%). Overall, the distribution of responses demonstrates that participants expressed preferences across a broad range of culturally responsive strategies, with particular emphasis on improving healthcare providers’ cultural competence, respectfully integrating culturally acceptable practices, promoting respectful communication and creating culturally inclusive maternal healthcare environments.

Discussion

This study examined how residency and cultural–traditional characteristics shape preferences for culturally responsive maternal healthcare in Bo District, Sierra Leone, and revealed several important findings that advance current knowledge in maternal health research in SSA.^14,15

The following supporting evidence is noteworthy. The study population was predominantly composed of young women of reproductive age in their twenties, with most having attained secondary education and residing in urban areas. This profile is consistent with recent evidence from Sierra Leone and other SSA countries, where maternal healthcare utilisation is shaped primarily by educational attainment, social support and health system factors rather than maternal age alone.^16–18 Consistent with these observations, maternal age was not independently associated with skilled birth attendance after adjustment for other variables. Education remained a significant predictor of skilled birth attendance, reinforcing its role as a key social determinant of maternal health. Women with higher educational attainment are generally better equipped to recognise obstetric risks, understand the benefits of skilled maternity care and participate actively in healthcare decisions. Within the context of culturally responsive maternity care, education may also enhance women’s ability to identify and seek respectful, culturally acceptable services that align with their values and preferences.^17,19,20 Community support was the strongest independent predictor of skilled birth attendance, with women receiving community support substantially more likely to deliver with skilled birth attendants. This finding highlights the importance of community engagement, family involvement and community health workers in strengthening trust in maternity services and promoting culturally responsive care. Recent evidence indicates that community-centred maternal health interventions improve care-seeking behaviours, increase facility-based delivery and reduce inequities in maternal health service utilisation, particularly in low-resource settings.^19,21,22 In contrast, household income, parity, marital status, traditional practices and urban–rural study location were not independently associated with skilled birth attendance after adjustment. Although these factors may influence women’s experiences and maternity care preferences, their effects appear to be mediated by broader social and health system determinants. This finding suggests that improving culturally responsive maternity care in Bo District requires strategies that strengthen women’s education, community participation and respectful, person-centred service delivery across both urban and rural settings, rather than focusing solely on demographic or geographic differences.^17,19,23 Overall, these findings support current global recommendations advocating culturally responsive, equitable and woman-centred maternity care. Integrating educational empowerment, community engagement and respectful maternity care into routine maternal health services may contribute to narrowing urban–rural disparities in maternal healthcare experiences and improving skilled birth attendance in Sierra Leone.

The lack of statistically significant rural–urban differences across these sociocultural variables suggests that cultural norms within this region may be more uniform than previously assumed, which contrasts with studies from northern and eastern Sierra Leone that have reported more pronounced cultural heterogeneity between rural and urban populations.^24,25

A key contribution of this study lies in its exploration of culturally responsive maternal-health strategies preferred by women from both settings. Across all respondents, training healthcare providers in cultural competence was the most widely endorsed strategy. This aligns with global evidence demonstrating that culturally competent care strengthens trust, promotes dignity, and improves patient–provider relationships—drivers strongly associated with increased facility births and reduced maternal mortality.^26,27 Importantly, the uniformity of support across rural and urban women underscores that cultural competence training is not merely an urban priority but a cross-cutting need in Sierra Leonean maternal healthcare.

Although several culturally responsive approaches received comparatively lower endorsement than provider cultural competence training and the integration of culturally safe traditional practices, these findings should be interpreted cautiously because participants were allowed to select multiple preferred approaches, meaning the observed pattern likely reflects prioritisation rather than rejection. Respondents appeared to prioritise interventions that addressed the most immediate and visible barriers to quality maternal healthcare, particularly provider interactions and the recognition of culturally meaningful practices. Moreover, strategies such as culturally tailored educational materials, individualised care plans, structured community engagement and family participation in decision-making may have received fewer endorsements because they are not yet routinely integrated into maternal healthcare delivery in Sierra Leone and may therefore be less familiar to women. Likewise, the relatively low endorsement of language support services likely reflects the linguistic context of Bo District, where shared local languages between healthcare providers and service users reduce the perceived need for formal interpretation services, except in some rural communities. Overall, these findings suggest that participants prioritised interventions with immediate practical benefits while underscoring the need for health systems to strengthen awareness and implementation of broader patient-centred and community-oriented approaches to culturally responsive maternal healthcare.

A particularly novel finding is the pronounced urban prioritisation of integrating safe traditional practices into formal healthcare, a result that remained statistically significant in the χ² analysis. While traditional practices are often assumed to be more deeply valued in rural settings, recent research indicates that urban populations in West Africa increasingly seek hybrid care models that combine biomedical services with culturally familiar rituals.²⁸ Urban women often navigate diverse cultural identities and may therefore request structured, safe and health system-endorsed pathways for integrating traditional beliefs. This aligns with findings from Nigeria, where urban mothers expressed the need for respectful accommodation of cultural rituals during childbirth, particularly when hospital policies are perceived to marginalise cultural identity.^29,30 Our study adds new evidence from Sierra Leone, demonstrating that urban women—not rural women—are the drivers of demand for formalised integration of cultural practices into maternal care.

Language support services emerged as a uniquely rural priority, demonstrating the highest statistically significant rural–urban difference. Rural linguistic diversity, coupled with the dominance of Krio and English in formal health settings, has been widely documented as a barrier to communication and respectful care in Sierra Leone.^31,32 Similar findings from Liberia, The Gambia and northern Ghana show that inadequate language support compromises patient comprehension, reduces satisfaction and contributes to delayed care seeking.³³ The strong rural endorsement in our study, therefore, highlights a critical gap: existing maternal-health policies insufficiently address language barriers, despite their pivotal role in patient safety and informed decision-making.

Interestingly, for most culturally responsive strategies, including respectful communication, creation of culturally welcoming facilities, provision of culturally sensitive information, family involvement and flexible policies, rural and urban respondents expressed similar priorities. This mirrors recent African evidence showing that women across geographic settings place high value on respect, empathy, privacy and agency during childbirth. The absence of rural–urban differences suggests that, while context-specific barriers exist, foundational expectations of respectful care are universally shared.

The novelty of this study lies in its integration of cultural and preference-based analysis with statistical comparisons of rural and urban maternal populations; a dimension rarely investigated in Sierra Leone. Previous studies have documented barriers such as disrespectful maternity care, limited male involvement and reliance on traditional birth attendants^34,35; however, few have empirically examined which culturally responsive strategies women themselves prioritise—and whether these priorities differ by setting. By identifying only two statistically significant differences (integration of traditional practices in urban settings and language support in rural settings), our study provides evidence that cultural responsiveness is widely desired, but the pathways through which women expect it to be operationalised can differ by geography.

Scaling up cultural competence training for healthcare providers has emerged as a priority to address diverse patient needs effectively. Emphasising cultural sensitivity in training programmes can significantly improve provider-patient relationships and service utilisation.^36–39 Safely incorporating traditional practices into healthcare was identified as essential. Encouraging consultations with cultural experts to validate and integrate traditional customs can foster community trust, acceptance and increased engagement with maternal health services.^29,33 Effective provider-patient communication and culturally welcoming healthcare environments strongly influence maternal service utilisation. Investments in communication skills training and facility design that reflect cultural values and comfort significantly improve patient experiences and outcomes.^38–40 Community-driven initiatives and active family participation in healthcare decisions were highly recommended. Policies and practices that engage families and communities align healthcare delivery with cultural expectations, strengthening local support and improving maternal health-seeking behaviours.^26,37,41 Strengthening language interpretation and translation services was emphasised as critical for overcoming linguistic barriers. Enhanced language support ensures equitable access to healthcare and better comprehension of services for linguistically diverse populations, thereby improving the overall quality of maternal care. These findings hold important implications for maternal health policy and service delivery in many countries in SSA.^26,36–40

Conclusion

This study provides new and contextually grounded evidence on how cultural and traditional characteristics intersect with maternal preferences for culturally responsive healthcare in Bo District, Sierra Leone. While residency did not significantly influence religious affiliation, family structure or marital status, rural and urban women demonstrated clear and unified expectations regarding culturally respectful maternal care. Training for cultural competence emerged as the strongest shared priority, underscoring widespread recognition that respectful and empathetic maternity care begins with provider competence. Importantly, two meaningful rural–urban contrasts were identified: urban women’s stronger preference for incorporating safe traditional practices into clinical settings and rural women’s heightened demand for language support services. These differences highlight how geography shapes the pathways through which women expect cultural responsiveness to be expressed.

The findings suggest that national maternal-health strategies in Sierra Leone should adopt a dual-focus approach: strengthening foundational cultural competence across all facilities while tailoring specific interventions—such as structured integration of traditional practices in urban hospitals and enhanced language support in rural clinics. By centring women’s own preferences, this study advances a culturally informed framework for improving maternity care quality, equity and acceptability. Future research should examine how these preferred strategies influence maternal health outcomes when incorporated into routine care and explore implementation models that build trust, promote dignity and bridge the cultural gaps that often limit women’s engagement with formal health services.

Strengths and limitations

Strengths

This study has several notable strengths. First, it employed a comparative rural–urban design that provides empirical evidence on differences and similarities in cultural norms, traditional practices and preferences for culturally responsive maternity-care approaches among women receiving MNH services in Bo District, Sierra Leone. This design offers valuable insights into contextual variations that can inform geographically responsive maternal health interventions.

Second, the study applied robust analytical methods, including Pearson’s χ² tests, differences in proportions with 95% CIs and multivariable logistic regression to examine factors associated with skilled birth attendance. These complementary analyses enabled both descriptive and inferential assessment of maternal healthcare preferences and utilisation.

Third, data quality was strengthened through the use of a pretested, interviewer-administered electronic questionnaire implemented using the Organizational Network Analysis (ONA) digital data collection platform, incorporating skip logic, real-time validation, GPS verification and routine field supervision to enhance completeness, consistency and accuracy.

Finally, the study was grounded in women’s lived experiences by directly assessing their preferences for culturally responsive maternity-care approaches. The findings therefore provide contextually relevant evidence that can inform respectful maternity-care policies, provider training and community-based maternal health interventions in Sierra Leone and similar settings.

Limitations

A few limitations should be considered when interpreting the findings. First, the cross-sectional design precludes causal inference; therefore, the observed associations cannot establish temporal or causal relationships.

Second, the study was facility-based and included women attending one urban referral hospital and one rural maternity centre. Consequently, women who did not access formal maternal healthcare services were not represented, limiting the generalisability of the findings. Furthermore, because residential setting and health facility were inherently linked, the study could not distinguish independent urban–rural effects from facility-level influences such as staffing, service organisation, patient case mix or previous exposure to culturally responsive care. The findings should therefore be interpreted as comparisons between the two study settings rather than as estimates of the independent effect of residence.

Third, participants could select multiple culturally responsive maternity-care approaches. Each approach was analysed separately using Pearson’s χ² test; therefore, the analyses did not account for within-participant correlation across multiple responses. The results should consequently be interpreted as comparisons of endorsement frequencies for individual approaches rather than overall respondent-level preference patterns. Future studies should employ analytical methods such as generalised estimating equations or mixed-effects models to account for correlated multiple-response data.

Fourth, the relatively smaller rural sample and very small subgroup sizes for some characteristics (eg, Indigenous religion) limited statistical precision and warrant cautious interpretation of subgroup findings.

Finally, the reliance on self-reported information introduces the possibility of recall and social desirability bias, particularly regarding culturally sensitive practices. These potential biases were minimised through the use of trained interviewers, standardised questionnaires, neutral interviewing techniques, confidentiality assurances and electronic data quality controls.

Potential biases and mitigation strategies

Potential sources of bias were considered during the study design and implementation. Selection bias was reduced through proportional sampling and standardised participant recruitment procedures across both study sites. Recall bias was minimised by focusing on recent pregnancy and childbirth experiences and, where feasible, cross-checking responses against maternal health records. Social desirability bias was addressed through confidential interviews conducted using neutral, non-judgemental language. Non-response bias was mitigated by incorporating a 30% oversampling allowance into the sample size calculation. Measurement bias was minimised through questionnaire pretesting, refinement of culturally responsive maternity-care items, comprehensive interviewer training and the use of electronic data collection with built-in validation checks.

Data availability statement

Data are available upon reasonable request. All data relevant to the study are included in the article or uploaded as supplementary information.