The study population was women at postpartum in the MIFI health district. The health structures were selected using the number of maternal deaths received this 2023. Out of the 20 health areas they were ranged and 6 selected taking note of the fact that the urban areas be equal to the rural areas. In the health areas the health structures were selected by balloting. The mothers were women at postpartum who have come for IWC and those at immediate postpartum in the centers pre-selected.

The women at postpartum in the health structures selected for the study who are present and accept voluntarily to take part in the study.The criteria for women who were considered to have had complications, were those women who declared to have been hospitalized and/or treated after delivery from causes linked to pregnancy and delivery.

The women of the health structures selected in the MIFI Health District who could not be available at the time of study and those who gave up on the way and those who did not give in their consent to participate.

2.4.1. Sample Size

Our study used the formula below to get the sample size for the study population of women was calculated using the Cochrane formula.

Z = 1.96.

Z = 1.96 (standard normal deviate at 95% confidence level).p = 0.408, p = 0.408, p = 0.408 (estimated prevalence of maternal mortality from prior district records).e = 0.05, e = 0.05, e = 0.05 (margin of error).

Substituting these values:

Thus, the minimum sample size for women was 372 respondents.

2.4.2. Sampling Technique

In this study stratified random sampling was used for the choice of the health district. The twenty health areas of the district were divided into four clusters. In each cluster a purposive random sampling of health structures was done from each considering the classification of health districts into health centers, medicalized centers and district hospitals. To reduce bias, the structures of the public and the private structures were included. Therefore, a convenient sampling method was used to admit participants. Only centers with Maternity services were selected. The number of participants per facility and per category of personnel was based on the Ministry of Public Health prescription of staff standards by level of antenatal care, delivery, and postpartum ([2]).

As to what concerns the women, a convenient sampling was used in each center selected. The sample size was divided equally to all the centers. Although convenience sampling is often critiqued for its limitations in representativeness, its adoption in this study is contextually justified. The sensitive nature of maternal health outcomes, combined with the ethical requirement of voluntary participation, restricts the feasibility of purely random selection. Furthermore, postpartum women are often available only briefly in health facilities, making convenience sampling the most practical approach. This strategy allowed the study to reach women directly experiencing the phenomenon under investigation, ensuring depth of data while still complementing findings with stratified facility selection to reduce bias. This choice demonstrates a balance between methodological ideals and field realities.

2.5.1. Instruments for Data Collection

Data collection instruments were designed to capture both quantitative and qualitative information. A structured questionnaire, divided into four sections, was administered to postpartum women. This tool included both closed- and open-ended questions.

2.5.2. Data Collection Procedure

Before collection of data the questionnaires were translated into French so the work may be adapted for both French and English speaking Cameroonians. The data was collected using surveys by a well-structured questionnaire for the mothers. While in the field we helped to fill the forms for those who could not read. For those who could fill we allowed them to do so.

2.5.3. Validity of the Instruments

The validity of the instruments used was established through content validation by academic supervisors from the Department of Nursing and Midwifery, Faculty of Health Sciences, University of Bamenda. Each component of the instruments was critically reviewed for relevance, clarity, fluency, and simplicity in relation to the study objectives. Based on expert feedback, necessary modifications were made, and redundant or irrelevant questions were removed to produce the final, valid version of the data collection tools.

2.5.4. Reliability of the Instruments

To ensure the reliability of the data collection instruments designed for women a pretest was conducted with 62 participants, representing 13.7% of the target study population within private health facilities excluded from the main research. The reliability of the instruments was measured using Cronbach’s alpha, complemented by a test-retest method with a two-week interval. This approach allowed comparison of responses at two different points in time, from which the correlation coefficients were calculated to assess the stability of the instruments.

Cronbach’s alpha was applied to determine the internal consistency of the items covering sociodemographic characteristics, knowledge, perceptions, and practices related to maternal mortality across the three strata of respondents (women). The statistical computation was carried out using SPSS version 25. The formula for Cronbach’s alpha is expressed as:

where:

α alphaα is Cronbach’s alpha.k is the number of items. σ i 2 is the variance of each individual item. σ t 2 is the variance of the total score for all items.

Table 1. Reliability verification of the study instruments (n = 62).

Source: Computed from SPSSv25.

The reliability analysis as shown in Table 1 of the research instruments demonstrated strong internal consistency across all sections of the tool, with Cronbach’s alpha values ranging from 0.876 to 0.892 for the individual instruments administered to women, and an overall coefficient of 0.895 for the combined 47 items. According to conventional benchmarks, coefficients above 0.70 are considered acceptable, while those above 0.80 indicate good reliability, and values approaching 0.90 reflect excellent consistency. These results therefore confirm that the questionnaire was well structured, with the items within each section measuring related constructs consistently and coherently across respondents. The similarity between the raw and standardized Cronbach’s alpha values (0.888 overall) further indicates that the scales performed reliably even when differences in item variances were accounted for. This high level of reliability implies that the instrument is stable, minimizes measurement error, and is appropriate for assessing sociodemographic factors, knowledge, perceptions, and strategies related to maternal mortality in the Mifi Health District.

Descriptive statistics were used to summarize the characteristics of respondents and maternal health indicators across the Mifi Health District from 2021 to 2023. Variables such as maternal age, education level, parity, antenatal care attendance, and place of delivery were analyzed using frequencies, percentages, means, and standard deviations. This descriptive stage provided an overview of the data, highlighted key patterns, and set the foundation for subsequent inferential analysis.

Ethical approval was obtained from the University of Bamenda Institutional Review Board, followed by authorization from the West Regional Delegation of Public Health. Local authorities were consulted, and facility-level approval was secured before data collection commenced. Participants received an information sheet outlining the study’s objectives, procedures, voluntary participation, and withdrawal rights. Written informed consent was obtained prior to participation. Confidentiality and anonymity were strictly maintained, with data coded and securely stored by the researcher. The study posed minimal risks beyond the potential discomfort of answering sensitive question.

Table 3 showed that, most (80.6%) women were attended to within 10 minutes on arrival, but 18.5% waited an hour or more, showing some service delivery delays. A large majority (88%) were accompanied to health facilities, reflecting strong family/community involvement. Nearly half (48.4%) delayed going to the hospital because they were told to wait; 44.4% went immediately, while 3.4% first tried delivering at home. This highlights systemic and cultural delays in seeking care (the “three delays” model). The most common means was motorbike (53.8%), followed by private car (20.8%), taxi (20.5%), and walking (4.6%). This suggests that unsafe or unreliable transport modes dominate. The main reasons for poor timing were lack of awareness (29.3%), financial constraints (17.4%), and other personal reasons (39.6%). About 70% recognized diseases of mother/child and fetal growth complications, while 29.1% associated pregnancy with maternal/child death. Knowledge levels are moderately high but still leave gaps. Place of birth: Almost all delivered in facilities (health center 50.4%, hospital 47.3%), with only 1.4% at home, showing high institutional delivery rates.

Table 3. Challenges of women with respect to experienced maternal complications.

Source: Computed from r analysis of results.

Regarding challenges, the time to reception of care was crucial (p = 0.0009), with those waiting one hour facing more complications (39.5%) compared to those attended within 10 minutes (13.8%). Lack of accompaniment increased complications (39% vs. 13.9%, p = 0.0003). Decision-making also played a role (p = 0.0172), with women instructed to delay arrival at hospitals having higher complication rates (23.5%). Transport means were significant (p = 0.0319), with taxi users experiencing the highest complication rate (29.2%). Financial and awareness barriers to ANC did not show significant associations, nor did knowledge of pregnancy dangers or place of birth. After a startistic regression test the highest predictor was educational level which showed that Higher Education vs Secondary: aOR = 4.12 (95% CI: 2.18 - 7.79, p < 0.001). showing that Women with higher education have 4.12 times the odds of complications compared to those with secondary education, after adjusting for age, distance, and profession. In the case of Primary vs Secondary: aOR = 5.84 (95% CI: 1.18 - 28.93, p = 0.030) Women with primary education have 5.84 times the odds of complications.

For the roles of women as shown in Table 4, women identified several roles in curbing maternal mortality. Nearly half (46.7%) proposed forming support groups, while others emphasized community events (24.2%) or connecting mothers to resources (23.9%). Women also reported providing secondary (49.9%) or primary (27.4%) emotional and practical support, while 21.9% were occasional supporters. Education of family members (41.9%) and advocacy for healthcare access (28.2%) were highlighted as key ways women influence families.

Finally, roles and perceptions revealed strong links with maternal complications. Women who engaged in support groups (p = 0.0084) and primary support provision (p = 0.0032) reported higher complication risks compared to occasional supporters. Perceptions of community support did not show significant differences, but personal stories were strongly significant (p < 0.001), with women who considered them “very effective” reporting higher complication experiences (29.5%).

Table 4. Roles of women with experienced complications.

Source: Computed from r analysis of results.

Concerning the perception of women in Table 5, perceptions revealed mixed opinions: while 51.9% believed community support had no impact on maternal outcomes, 32.2% acknowledged some effect, and only 16% considered it greatly beneficial. Most women (83.2%) reported no direct challenges related to maternal health, though 16.8% did. Regarding advocacy, 51.6% believed personal stories were “somewhat effective” in shaping health policies, while 42.5% considered them “very effective”.

Table 5. Perception of women with experienced complications.

Source: Computed from r analysis of results.

The findings highlight important socio-demographic and behavioral determinants of maternal health. The predominance of women aged 25 - 34 years aligns with the reproductive peak age in sub-Saharan Africa ([10]). Although most women lived within one hour of a health facility, complications were still prevalent, suggesting that geographical proximity alone does not guarantee timely or quality care ([15]). The significant role of education was notable, with women of higher education surprisingly showing more complications. This may reflect higher health awareness and reporting among educated women, as seen in similar studies in Ethiopia and Nigeria ([9]; [14]). The adjusted odds ratios (aOR) indicate a statistically significant association between lower levels of education and higher odds of maternal complications when compared to secondary education. This suggests that even though these women are more educated, they may be older at pregnancy, have higher career demands, or face health conditions associated with delayed childbearing all of which could increase risk. Conversely, women with primary education exhibited an even higher risk, with 5.84 times the odds of complications compared to those with secondary education. This may reflect reduced health literacy, poorer utilization of antenatal care, delayed health-seeking, or socio-economic disadvantages that limit access to timely obstetric care.

Profession and socioeconomic status also influenced maternal outcomes. Salaried workers and large-scale traders showed higher complication rates, possibly due to older maternal age or higher-risk pregnancies. Transport means, particularly reliance on taxis, was linked with complications, underscoring the importance of efficient referral and emergency transport systems, a major challenge in many African settings ([8]).

Delay in care-seeking and hospital instructions to wait before coming emerged as critical contributors to complications. This aligns with the “three delays model” ([18]), particularly delays in deciding to seek and receive appropriate care. Similarly, accompaniment to health facilities significantly reduced risks, highlighting the role of social and family support in maternal survival ([7]).

Women also acknowledged their potential role in reducing maternal mortality, particularly through support groups, community mobilization, and educating families. However, perceptions were divided: many underestimated the value of community support, despite evidence showing that women’s groups and collective action can significantly reduce maternal and neonatal deaths ([13]). The recognition of personal stories as a powerful advocacy tool reflects growing awareness of participatory approaches in shaping health policy, as seen in community-based maternal health interventions worldwide ([7]).

Overall, the study reinforces the need for multifaceted interventions addressing not just facility accessibility but also social, cultural, and behavioral barriers. Empowering women as advocates, strengthening emergency referral systems, and fostering supportive networks could significantly reduce maternal complications and mortality in Cameroon.

4.2.1. Access to Health Facilities

Distance to health facilities was significant (p = 0.004). Women living farther away (5 hours) experienced more challenges—confirming the “Three Delays Model” ([17]). Delay in deciding to seek care, Delay in reaching care, Delay in receiving care.

After arrival time of reception by the health personnel is a factor. Waiting more than one hour after arrival was associated with more maternal challenges (p = 0.0003). Delays at facilities have been globally documented as a major maternal mortality risk ([5]). The finding matches those from WHO’s 2019 report, which emphasizes reducing health facility delays to improve outcomes.

4.2.2. Social and Family Support

Accompanying person during labor is a major factor too. Being accompanied significantly reduced challenges (p = 0.0001). Companionship is protective, as emotional and logistical support improves outcomes ([4]).

Interestingly, even when husbands accompanied women, there were still high challenges. This shows that mere accompaniment without quality of care or financial empowerment is not enough ([6]) show that male involvement improves outcomes only when combined with decision-making support and financial preparedness.

4.2.3. Cultural and Economic Factors

Cultural barriers (8.83% of women) significantly impacted access to care (p = 0.0078).

This confirms that harmful traditional practices delay care ([6]). Financial constraints were the leading barrier (34.19%), confirming the role of poverty in maternal health inequalities ([16]).

Many women believed in “mystical” causes of complications (e.g., witchcraft). This shows persistent traditional beliefs which need to be addressed through culturally sensitive education. These findings are consistent with studies from rural Tanzania ([11]) where supernatural causes were commonly cited.

4.2.4. Antenatal Care (ANC) and Timing

96.29% attended ANC, but many started late (after 12 weeks). Late ANC attendance was significantly associated with maternal challenges (p = 0.033). WHO recommends that ANC should begin before 12 weeks Studies from Ethiopia and Nigeria ([26]; [23]) also report that late ANC booking contributes to maternal mortality.

4.2.5. Community Participation and Women’s Roles

Most women (87.74%) believed women can help curb maternal mortality. Their proposed roles included attending ANC, educating others, and listening to counseling. Education and awareness were cited as key strategies. Emotional support and community group formation were the main community strategies to help pregnant women. [20] stresses community-based interventions and women empowerment as crucial to curb maternal mortality.

4.2.6. Policy and System-Level Issues

Lack of healthcare facilities, poor road infrastructure, and financial barriers were highlighted as systemic issues. Most respondents (92%) emphasized the need for the government to prioritize maternal mortality reduction matching global SDG 3.1 goals. Other Sub-Saharan African studies show similar findings government support is essential for achieving Sustainable Development Goals (SDGs) related to maternal health ([1]).