We conducted a cross-sectional comparative mixed-methods study with quasi-experimental features following TREND reporting guidelines, in Ziniaré district, Burkina Faso, in August 2024. The study sites were selected based on the operational implementation strategy of the Living Goods DESC initiative in collaboration with the Ministry of Health. The design compared non-randomly selected intervention communes where task-shifting of family planning (FP) services to community health workers, Agents de Santé à Base Communautaire (ASBCs), was actively implemented and supported, with control communes where FP services remained primarily health facility-based (Centres de Santé et de Promotion Sociale, CSPS). While ASBCs in both arms provided FP counseling and received supervision for routine community health activities, in intervention communes they were additionally trained and authorized to dispense short-acting contraceptives (oral pills, subcutaneous DMPA, condoms) within the community, receive supportive supervision explicitly extended to FP, and participate in systematic performance monitoring. In this study, task-shifting refers specifically to the formal delegation of FP service delivery to ASBCs, including counseling and provision of short-term contraceptive methods, whereas supportive supervision and coaching are considered complementary mechanisms to ensure quality and sustainability. Quantitative and qualitative data were collected in parallel during the same period across both intervention and control communes. The design allows for the assessment of factors associated with family planning uptake, as well as the implementation of task-shifting family planning services to Agents de Santé de Base Communautaire (ASBCs) in Ziniaré District, while recognizing that causal attribution is limited due to its quasi-experimental cross-sectional nature. This study was implemented within the framework of Burkina Faso's national FP policy, which supports task-shifting of contraceptive services to ASBCs.

In the intervention group (Nagréongo, Loumbila, Dapélogo, and Ourgou Manéga), ASBCs were supported under the Living Goods DESC model (Digitally enabled, Equipped, Supervised, and Compensated) and received additional FP-specific inputs over an eight-month period (January–August 2024). These inputs included the provision of family planning commodities, ensuring a regular supply of oral contraceptive pills, condoms, and subcutaneous depot medroxyprogesterone acetate (DMPA-SC). ASBCs also benefited from supportive supervision and coaching, with structured mentoring on FP counseling, community engagement, and client follow-up. Supervisors conducted monthly field visits that combined direct observation, feedback, and problem-solving sessions. In addition, ASBCs were engaged in structured performance monitoring specifically for FP, where monthly targets were set, progress was regularly reviewed, and feedback was systematically provided through supervision sessions. Monthly performance targets for FP service delivery were supported by data review sessions (feedback loops) between ASBCs and supervisors to enhance accountability and facilitate problem-solving. Implementation fidelity of the family planning intervention across communes, including ASBC density, supervision frequency, stockout occurrence, intervention start dates, and attainment of performance targets, is summarized in Table 1.

In the control communes (Absouya, Zitenga, and Ziniaré town), ASBCs were part of DESC for general health services but did not receive FP-specific commodities, supervision, or structured FP performance monitoring. Family planning services in these areas remained primarily facility-based, without targeted community-level performance management for FP. The geographic distribution of intervention and control communes within Ziniaré District is presented in Figure 1.

The quantitative survey targeted women of reproductive age (15–49 years) residing in intervention and control communes. A village-based sampling strategy was employed, beginning with a household listing to identify eligible women in each village. From this list, ten households per village were randomly selected, covering a total of 27 villages, 13 in the intervention group and 14 in the control group. In addition, purposive considerations were applied to ensure socio-demographic diversity, particularly with respect to age, marital status, and education.

Based on a priori power calculations (α = 0.05, 80% power), a minimum of 252 women was required to detect a meaningful difference in FP use between groups, assuming a baseline prevalence of 32%, a detectable difference of 18%, and an intra-cluster correlation coefficient (ICC) of 0.02. These assumptions and corresponding detectable effect parameters are summarized in Appendix Table A1.

A total of 282 women were enrolled and interviewed (136 intervention, 146 control), exceeding the required minimum. Cluster sizes ranged from 1 to 20 participants (mean = 9.10, SD = 3.53, CV = 0.39). post-hoc detectable effect calculations were performed using these achieved sample sizes and observed cluster sizes, with ICC = 0.02, applying the standard design-effect–adjusted formula for two-sample comparisons of proportions. Appendix Table A1 presents the full post-hoc calculations, and Appendix Table A2 provides a sensitivity analysis for ICC values ranging from 0.01 to 0.03. The resulting minimum detectable differences substantially exceed the observed 5-percentage-point difference (48% vs. 53%), confirming that the study was underpowered to detect this small effect.

The qualitative component purposively sampled men, ASBCs, and health facility managers to explore FP barriers, male involvement, and implementation experiences. Participants were selected to maximize variation in age, sex, and roles, and data collection continued until thematic saturation was reached, that is, when no substantially new insights emerged. Although women were not directly included in the qualitative interviews, a limitation acknowledged in this study, the survey data served as an important supplementary source, providing valuable insights into women's perspectives.

As commune-level pre-intervention data on modern contraceptive prevalence and women's education were not available (EDS 2021 provides only district- or regional-level estimates), ASBC system capacity was the only comparable indicator accessible across all communes. Appendix Table A3 presents pre-intervention data on the total number of ASBCs and the ratio of Women of Reproductive Age (WRA) per ASBC for each commune. These figures serve as proxies for baseline ASBC coverage. Baseline ASBC coverage was broadly comparable between intervention and control communes (mean WRA/ASBC: 171 vs. 171, respectively), suggesting no major imbalance in community health system capacity prior to implementation.

To guide our quantitative analyses and clarify assumptions regarding potential confounding, we constructed a Directed Acyclic Graph (DAG) illustrating the hypothesized relationships between intervention exposure, socio-demographic factors, and family planning outcomes (Figure 2). The DAG depicts the pathways through which the task-shifting intervention, individual characteristics (e.g., age, education, marital status), and contextual factors (e.g., commune-level implementation differences) may influence both FP knowledge and modern FP use.

Importantly, partner communication about FP was modeled as a mediator rather than a confounder. This decision reflects our theoretical assumption that the intervention may enhance partner communication, through increased knowledge, empowerment, or counseling effects, which in turn can influence contraceptive uptake. In the DAG, partner communication lies on the causal pathway between exposure to the intervention and FP outcomes, rather than preceding the intervention.

Modeling it as a confounder would imply that partner communication independently affects both exposure and outcome (i.e., that it predates exposure), which is inconsistent with the program's logic model and implementation timeline. Aligning with the DAG, we therefore treated partner communication as a mechanistic pathway linking the intervention to behavioral outcomes, rather than a source of confounding bias.

The quantitative component consisted of a structured household questionnaire administered to women of reproductive age (15–49 years). The tool captured socio-demographic characteristics, knowledge and use of family planning (FP) methods, partner communication about FP, exposure to FP messages, and access to FP services. Questionnaires were pre-tested, locally adapted, and administered in August 2024 to ensure temporal alignment between intervention and control areas.

The qualitative component included key informant interviews (KIIs) with facility managers (n = 7) and focus group discussions (FGDs) with community health workers (ASBCs, n = 18) and male community members (n = 7). Women were not directly included in the qualitative interviews. While survey data provided insights into women's perspectives, this exclusion may bias qualitative interpretations toward provider and male perspectives.

The primary outcomes of interest were the current use of modern family planning (FP) methods and knowledge of FP methods. Modern FP use was defined as the current use of at least one modern contraceptive method, including sterilization, implants, intrauterine devices (IUDs), injectables, pills, or condoms. Women reporting the use of multiple methods were classified as users for the binary outcome variable (yes/no). Knowledge of FP methods was assessed spontaneously by asking women to name all modern contraceptive methods they knew without prompts, and the outcome was operationalized as the total number of modern methods correctly identified by each participant.

Ethical approval was obtained from the National Ethics Committee for Health Research (CERS), Ministry of Health of Burkina Faso (Ref. No. 2024-07-211). Written informed consent was obtained from all participants; illiterate participants were assisted by a witness of their own choosing. For adolescents aged 15–17 years, parental or guardian consent was obtained alongside the adolescent's written assent using age-appropriate language. To safeguard privacy, particularly for sensitive topics such as reproductive health and family planning, interviews were conducted in private settings without parental presence. All responses were anonymized, confidentiality was strictly maintained, and data were securely stored with access limited to the research team.

The study participants exhibit several key socio-demographic characteristics that influence their FP utilization. As shown in Table 2, slightly over half (53.2%), of the women surveyed are illiterate, with a slightly higher rate (60.3%) in the control group compared to the intervention group (45.6%). Overall, 94.7% of the women are married or cohabiting with no statistically significant difference in marital status between the groups (p-value = 0.9). The average age of the participants is 27.5 years, indicating a relatively young demographic with specific reproductive health needs. These demographic factors collectively shape the context within which FP services are accessed and utilized.

Analysis of contraceptive methods known reveals that women in the control group are familiar with a slightly higher average of 5.8 methods compared to 5.2 methods in the intervention group. This difference was statistically significant (p-value = 0.0046). This suggests that awareness of available contraceptive methods was broadly high but somewhat more diversified in the control sites (Table 3).

Interestingly, men's knowledge of contraceptive methods is reported as quite limited, as illustrated by participant quotes:

Regression analysis identified several factors influencing the number of contraceptive methods known. Participation in gatherings for educational talks and events was significantly associated with higher knowledge: women who attended such gatherings reported knowing 17% more methods than those who did not (IRR = 1.17, 95% CI:1.08–1.27, p < 0.001). Educational attainment also appeared to play an important role: compared to women with secondary or higher education (reference category), those who were illiterate or had only primary schooling reported knowing 12% and 10% fewer methods, respectively (IRR = 0.88, 95% CI: 0.79, 0.97; and IRR = 0.90, 95% CI: 0.82, 0.99) (Table 4).

Cluster-level variation was low (ICC_commune = 0.02; ICC_village = 0.025), indicating that only 2%–2.5% of the variance in contraceptive knowledge was attributable to differences between communes or villages. However, the design effects were non-negligible (DEFF_commune = 1.77; DEFF_village = 1.20), reflecting inflation of the standard errors due to clustering. The full variance decomposition, including ICC and design effect estimates, is provided in Appendix Table A4. These results justify the use of multilevel modeling to obtain valid standard errors and confidence intervals.

These findings indicate that participation in community gatherings and higher educational attainment are independently associated with greater contraceptive knowledge, after accounting for clustering and individual-level confounding.

Analysis of the primary sources of information on family planning (FP) revealed that community-based communication channels were the dominant sources of information among women. As shown in Table 5, social gatherings were the most frequently cited source overall, reported by 59% of respondents (67% in the control group vs. 49% in the treatment group, p = 0.008). This highlights the central role of interpersonal and community networks in the dissemination of FP information within the study population.

Radio ranked as the second most common source, mentioned by 49% of women (45% in the control group vs. 53% in the treatment group, p = 0.3), confirming its importance as a widely accessible traditional medium.

In contrast, modern digital and print media were significantly less impactful: Television was reported by 13% of respondents, while information received via mobile phone/SMS stood at 10%. Print media, such as newspapers and magazines, were the least common source, cited by only 8.5% of the women.

Overall, these findings underscore the importance of leveraging social gatherings and radio as primary communication channels for FP messages. Strengthening these interpersonal and mass media platforms remains essential for expanding the reach and impact of FP information in rural communities.

These findings, coupled with the limited knowledge expressed by male participants, suggest that while overall knowledge exists, targeted efforts are needed to address knowledge gaps related to specific methods, particularly among less educated women and in areas where task delegation is implemented. Leveraging gatherings as a key information channel appears crucial for enhancing awareness of diverse contraceptive options for both women and men.

The study revealed that 48% of women in the intervention group, where family planning tasks were delegated to ASBCs, reported using a contraceptive method. This was slightly lower than the control group, where 53% of women reported contraceptive use. Implants and injectables emerged as the most used methods across both the intervention and control groups, indicating their relative popularity and acceptance within the community (Figure 3).

Analysis of the sources of family planning (FP) methods revealed a strong reliance on formal health facilities across both study groups. The majority of women obtained their contraceptive methods from Centres de Santé et de Promotion Sociale (CSPS), accounting for 82% in the intervention group and 90% in the control group. By contrast, only 17% of women in the intervention group reported receiving their FP methods directly from Agents de Santé à Base Communautaire (ASBCs), compared to 1% in the control group, where ASBCs primarily provided counselling or referrals.

These findings indicate that, although the intervention successfully expanded ASBCs’ roles to include the provision of certain FP methods (oral pills, DMPA-SC, and condoms), health facilities remained the dominant source of contraceptives for most women. The introduction of community-based distribution through ASBCs thus represented an additional access point rather than a substitution for facility-based services.

Figure 4 presents the distribution of contraceptive sources among users, illustrating the predominant role of CSPS in method provision and the comparatively limited, though emerging, contribution of ASBCs within the intervention arm.

To gain a deeper understanding of the determinants of family planning (FP) adoption, multilevel mixed-effects logistic regression analysis was employed (Table 6). The analysis identified several key factors associated with modern FP use, including the number of contraceptive methods known by a woman, marital status, and ability to discuss FP with one's partner.

Greater awareness of a wider range of contraceptive options was associated with higher odds of FP use (p < 0.001). Being married or cohabiting with a partner was also strongly associated with contraceptive use; women in union had approximately 21 times higher odds of using a contraceptive method compared to single women (OR = 21.3, 95% CI: 2.1–215.9, p = 0.009). The extremely wide confidence interval reflects the small number of single women in the sample, which leads to instability in the estimate. To improve interpretability beyond the unstable odds ratio, we additionally derived marginal predicted probabilities from the adjusted model. The adjusted predicted probability of modern contraceptive use was 41% (95% CI: 30%–54%) among married/cohabiting women compared with 3% (95% CI: 0%–26%) among single women, corresponding to an absolute difference of 38 percentage points. These predicted probabilities provide a clearer and more interpretable representation of the association than the OR alone. The detailed results are presented in Appendix Table A5.

The ability to discuss FP with one's partner was strongly associated with contraceptive use (OR = 7.57, p < 0.001). Although initially considered a potential mediator, its inclusion in the model did not alter the estimated effect of the intervention (Model with partner communication: OR = 0.68; Model without: OR = 0.96). The detailed estimates for the model excluding partner communication are provided in Appendix Table A6, confirming that partner communication was not on the causal pathway. Sensitivity analyses confirmed the robustness of the associations for FP knowledge and marital status.

The intervention itself did not have a statistically significant effect on modern contraceptive use. In the fully adjusted model, the intervention was associated with OR = 0.68 (95% CI: 0.27–1.68, p = 0.4), indicating no measurable improvement in FP uptake.

The mixed-effects logistic model showed substantial clustering at the village level, with an intraclass correlation coefficient (ICC) of 0.19, corresponding to a design effect of 2.59. This indicates that nearly one-fifth of the total variance in contraceptive use was attributable to differences between villages, justifying the use of multilevel modeling. No commune-level covariates were included, as the Directed Acyclic Graph (DAG) identified only individual-level confounders as minimally sufficient for adjustment. Random slopes were tested but did not improve model fit and were therefore not retained.

Overall, while individual-level knowledge, marital status, and partner communication were key determinants of contraceptive use, the task-shifting intervention implemented through ASBCs did not yield a detectable effect after accounting for clustering and confounding.

The mixed-methods findings indicate that sociocultural norms and gender relations strongly shape FP behaviors. The strong association between partner discussion and FP use (OR = 7.57, p < 0.001) suggests that spousal communication is a key enabling factor. The large odds ratio for marital status (OR = 21.3) highlights the particular barriers faced by unmarried women, for whom FP use may be socially discouraged or stigmatized.

These findings collectively point to the importance of social and relational dynamics in shaping FP outcomes and suggest that behavioral and cultural factors may influence the effectiveness of community-based delivery strategies.

There were several factors that influenced men's attitudes towards contraception, while some acknowledge its potential benefits, such as facilitating improved childcare and better birth spacing, they also harbor concerns about possible adverse health effects for women, including illnesses and sterility, and the potential financial burden associated with managing complications. Furthermore, religious and traditional beliefs play a crucial role in shaping their views. Some men express that family planning practices conflicted with religious principles.

Also, one male participant conveyed his apprehension about contraceptive methods, stating that “For me contraceptive methods are good but not totally, it is not good because the implants used by women move in the body. You can place it and then come to look for it and not find it and that can make you sterile.”

Women's perceptions of contraceptive methods demonstrate variability, particularly when comparing the intervention and control groups. As shown in Table 7, in the intervention group, 8.1% of women strongly agree that using contraceptives will make it difficult for them to conceive in the future, whereas this figure rises to 29.5% in the control group. Additionally, while 4.4% of women in the intervention group express the belief that contraception could lead to abnormal births, this concern is held by 10% of women in the control group. Overall, these findings suggest that women residing in areas where task delegation for family planning services is implemented tend to exhibit a more favorable perception of contraceptive use.

ASBCs are strong proponents of family planning, emphasizing its numerous advantages, including enabling optimal pregnancy spacing, preventing unintended pregnancies, and promoting both maternal and child health, as well as women's empowerment. ASBCs generally dismiss sociocultural and religious barriers, asserting that acceptance exists even within the Muslim community, and they highlight the historical presence of traditional contraceptive methods. An ASBC articulated this view, saying that “family planning allows the couple to have the child when they want, [and that] also it does not make women sterile”. However, it is important to note that some ASBCs do acknowledge potential side effects associated with contraceptive use, such as disturbances in the menstrual cycle.

In addition to their support for family planning, ASBCs emphasize the importance of ongoing training and supervision. One ASBC noted that despite having received initial training, periodic refreshers are crucial for maintaining high-quality service, allowing them to confidently explain contraceptive options even without reference materials. Furthermore, they appreciate the consistent supply of contraceptive products, stating that they rarely face stock shortages thanks to reliable restocking from health facilities and support from Living Goods supervisors, ensuring they can meet community needs effectively.

Health facility managers report encountering various challenges in the implementation of FP services, primarily stemming from community reluctance and prevailing sociocultural barriers. A key obstacle identified is the pronounced preference among women for consulting female ASBCs. This preference is supported by quantitative data presented in Table 8, which demonstrates that 61% of women prefer consulting female ASBCs, compared to only 33.3% who express a preference for male ASBCs. The influence of social dynamics on service delivery is further illustrated by a key informant who explained that “it is difficult for a woman to confide in an ASBC man for a desire for FP given that her husband and the ASBC are in the same community and sometimes they are even friends [adding that] that is a difficulty”.

The study reveals that there is a large majority (89%). of women using FP had informed their partners, indicating a general awareness among men. Furthermore, a considerable proportion of women (87%) reported that they consulted their partners before initiating contraceptive use.

The data analysis, presented in the study and illustrated in Figure 5, shows varying levels of partner engagement. As detailed in Table 9, in the intervention group, where task delegation in FP was implemented, 72.1% of women reported they could negotiate with their husbands about the timing of future pregnancies, compared to 57.5% of women in the control group. Additionally, 78.7% of women in the intervention group stated they could discuss FP with their husbands, whereas only 63% of women in the control group reported the same. While these figures collectively suggest a higher degree of male involvement in FP decisions within the intervention group, causation cannot be inferred due to the cross-sectional nature of the data.

Contrary to the perception on spousal communication on family planning, focus group discussions highlighted that most men do not accompany their wives to health facilities when choosing a contraceptive method. This observation is supported by accounts from some women who reported that they do not inform their partners about their decision to adopt FP methods; instead, they directly visit the facility to obtain the method. For example, some men expressed that “I have never accompanied my wife and I will never accompany her”. In discussions about improving communication, some men suggested that “The woman should normally inform you before using contraceptive methods, but if she does not inform you, how will you know to support her?”. It was also noted by several men that “Some inform the husband but on the other hand some women do not inform, however logic would dictate that you discuss to make a decision”. This reflects the varied approaches to communication within couples regarding family planning, highlighting the influence of individual and cultural differences in decision-making.

To provide a comprehensive understanding of the factors shaping family planning knowledge and uptake, we developed a joint display linking quantitative predictors with illustrative qualitative quotes. This integration highlights how statistical associations observed in the survey are reinforced, nuanced, or challenged by perspectives from men, ASBCs, and health facility managers. As shown in Table 10, the joint display synthesizes these complementary insights to offer a richer interpretation of family planning behavior, knowledge, and use.

Low literacy levels remain a key barrier to FP adoption in Burkina Faso, limiting women's ability to access, interpret, and act upon reproductive health information. Illiteracy was lower among women in the intervention group (45.6%) compared to the control group (60.3%), which may reflect slightly higher baseline health literacy or improved understanding through community-based counseling in local languages. This aligns with previous evidence showing that community-delivered information can enhance FP knowledge and uptake in low-literacy settings (11). Nearly all participants were married or cohabiting, reflecting entrenched cultural norms of early marriage and underscoring the critical role of male partners in FP decision-making. Given that Burkina Faso's fertility rate remains among the highest globally (5.4 children per woman) (12), community-based programs that engage couples and normalize FP discussions are essential.

Women in both study groups primarily reported receiving FP information through radio and social gatherings, reflecting the dominant channels for health communication in rural Burkina Faso. Radio's wide accessibility and low cost make it a key source, while social gatherings leverage community norms and peer influence to disseminate information, consistent with evidence from sub-Saharan Africa (11, 13). The study found minimal reliance on other sources such as print media, television, or mobile/SMS messaging across both groups, highlighting that digital and text-based channels currently play a limited role in this setting. This underscores the continued importance of broad-reach and interpersonal communication strategies for FP programming, particularly in low-literacy populations.

The analysis of contraceptive sources revealed a strong and persistent preference for established health facilities, underscoring their central role in contraceptive provision despite the implementation of a community-based intervention. The vast majority of women across both groups relied on Centres de Santé et de Promotion Sociale (CSPS) as their primary source, 82% among users in the intervention arm and 90% in the control arm, as expected given the absence of community-based supply options in the latter. Only 17% of users in the intervention group reported obtaining their contraceptive methods directly from ASBCs.

This pattern suggests that, while ASBCs play an important role in counselling, education, and demand generation, they are not yet the main service providers for most women. The intervention effectively activated ASBCs as a new access point within the community (17% vs. 1% in control), but the overall low utilization of this channel highlights that enhanced service availability did not necessarily translate into increased community uptake.

Several factors help explain this limited reach. One of the most salient is the perception of health facilities as more confidential and trustworthy environments for obtaining FP services. In the sociocultural context of Burkina Faso, where male opposition to contraception remains common, many women prefer to seek FP services at facilities to avoid being seen or questioned by their husbands or community members. By contrast, obtaining methods through ASBCs, who live within the same communities, increases the risk of social scrutiny or gossip, which discourages use. This dynamic reflects broader gender and social norms that shape women's comfort with different service delivery channels. It is particularly pronounced among younger and less educated women, who report greater constraints on reproductive decision-making and stronger concerns about discretion (14, 15). Women in this study emphasized proximity, reliability, and above all, privacy and confidentiality as key factors influencing their choice of service source.

These findings underscore that while task-shifting successfully decentralized contraceptive provision and improved local access, its impact remains limited by perceptions of confidentiality and persistent gender norms. To enhance the reach of community-based providers, future interventions should integrate components that strengthen confidentiality safeguards, support gender-sensitive provider training, and address community norms that discourage women from openly engaging with ASBCs for family planning services.

Differences in contraceptive method mix illustrate how the task-shifting intervention influenced service delivery. Women in the intervention group showed greater reliance on injectables, a short-acting method easily provided at the community level by ASBCs, whereas women in the control group favored implants, which require facility-based insertion. This pattern aligns with prior evidence indicating that short-acting methods are more readily integrated into community-based models (16).

A key observation was that 61% of women expressed a preference for female ASBCs, highlighting the importance of gender concordance between provider and client for sensitive services such as family planning (17). In contexts where male ASBCs predominate, this preference may structurally limit the reach of task-shifting, particularly among women less comfortable discussing reproductive intentions with male providers.

These findings suggest several considerations for program design within the context of community-based family planning service delivery. First, the gender composition of ASBCs is critical for acceptability, indicating that communities with a sufficient proportion of female providers may achieve higher engagement. Second, method suitability influences utilization patterns, with short-acting methods being more amenable to community provision and thus more likely to be adopted when delivered by ASBCs. Finally, male ASBC deployment may require additional supportive strategies, such as community sensitization or targeted engagement initiatives, to optimize acceptability in contexts where prevailing gender norms constrain interactions between women and male providers.

Overall, these findings emphasize that both the type of contraceptive method and the gender of the ASBC affect uptake patterns, while practical recommendations for addressing these factors are presented in the “Programmatic Implications” section.

The absence of a significant difference in contraceptive prevalence between intervention and control groups (48% vs. 53%, OR = 0.68, 95% CI: 0.27–1.68, p = 0.4) contrasts with global systematic reviews reporting strong effects of task-shifting on family planning uptake (18, 19). This apparent discrepancy is likely explained by several important contextual factors specific to Ziniaré:

Short intervention duration: The eight-month implementation period may have been insufficient for measurable behavioral changes in contraceptive use to manifest, whereas most studies demonstrating strong task-shifting effects report follow-up periods of 12–24 months or longer.

Our multilevel regression analyses identified several correlates of modern contraceptive notably marital status, FP knowledge, and partner communication. Married or cohabiting women were significantly more likely to use modern methods (OR = 21.3, p = 0.009), and higher knowledge of FP methods was positively associated with use (OR = 1.45, p < 0.001), consistent with findings from Orjingene and Morgan (20), who emphasized education and knowledge as fundamental drivers of contraceptive adoption in community-based interventions.

Notably, women reporting discussion of FP with their partners exhibited markedly higher odds of contraceptive use (OR = 7.57, p < 0.001), aligning with multiple studies cited in the International Journal of Public Health (21) and underscoring spousal communication as one of the strongest predictors of FP uptake. This strong association also highlights the critical importance of male engagement strategies, as emphasized by the Ouagadougou Partnership framework for francophone West Africa, particularly in contexts where social norms influence reproductive decision-making.

Considering partner communication as a mediator clarifies the mechanism through which task-shifting interventions are expected to influence contraceptive behavior. By increasing access to FP services through ASBCs, the intervention may have enhanced opportunities for partner dialogue and engagement. Such communication is known to facilitate contraceptive adoption, although this pathway did not translate into a measurable difference in overall contraceptive use in this study. These findings emphasize that task-shifting strategies may be most effective when coupled with interventions promoting spousal communication and male engagement, rather than focusing solely on service delivery, to maximize behavioral impact.

The evaluation of community-based family planning (FP) services in the Ziniaré district, Burkina Faso, highlights both encouraging progress and critical limitations that inform future program adaptation. Moving from high awareness to sustained uptake requires targeted modifications to the national task-shifting model. Based on study findings, the following programmatic recommendations are proposed: (1)

Extend the Implementation and Evaluation Horizon

The strength of this study lies in its mixed-methods approach, which provided the qualitative data necessary to interpret the null quantitative finding, revealing the critical, non-logistical barriers of confidentiality and male opposition, thereby generating contextually relevant, data-driven programmatic recommendations.

This study has several limitations. Its cross-sectional and quasi-experimental design limits causal attribution, and the study was underpowered to detect the observed prevalence difference. Methodologically, the exclusion of women from the qualitative interviews means that women's experiential voices are not directly represented, and the depth of narrative understanding regarding barriers, such as preferences for facility privacy, confidentiality, or social constraints, is primarily derived from provider and male perspectives. This may bias the interpretation of qualitative findings and highlights the need for future studies to include women's qualitative perspectives.

Future research should investigate the differences between women who discuss family planning (FP) with their partners and those who do not, with particular attention to the potential influence of intimate partner violence (IPV) dynamics and reproductive autonomy. Longitudinal studies are also needed to assess the sustained impact of task-shifting on FP uptake, including its scalability to other districts. Additionally, future work should explore effective strategies to enhance men's knowledge, engagement, and shared decision-making in family planning.