Assessing the knowledge–attitude–practice gap in prevention of mother-to-child transmission of HIV among pregnant and lactating women: a cross-sectional study in Musoma Municipality, Tanzania

Abstract

Background

The transmission of HIV from mother to child remains a significant public health challenge in sub-Saharan Africa. Although previous studies have shown high levels of awareness and knowledge among women, significant gaps persist between knowledge and the consistent adoption of preventive practices to reduce mother-to-child transmission (MTCT) of HIV. This study aimed to assess the knowledge–practice gap in PMTCT among pregnant and lactating women in Musoma Municipality, Tanzania, regardless of HIV status, and to explore sociodemographic factors influencing this gap. also considered the role of attitudes as a potential intermediary between knowledge and practice.

Methods

A hospital-based cross-sectional study enrolled 200 women (100 pregnant, 100 lactating) through systematic sampling at Musoma Municipal Hospital (Jan-Jul 2024). Data were collected using a structured questionnaire and analyzed using descriptive statistics, chi-square/Fisher’s tests, and binary logistic regression (unadjusted odds ratios with 95% CI) in R-4.3.0 to determine factors affecting HIV prevention practices and attitudes.

Results

While 98% demonstrated excellent HIV knowledge and 100% awareness, only 68% showed good preventive practices. Positive attitudes were held by 67% of participants. Women of older age (mean 30 vs. 27 years; p = 0.04) and Protestant participants (OR = 2.84, p = 0.04) were positively associated with better practices. Primiparous women were less likely to engage in preventive behaviors (OR = 0.26, p = 0.05). Although the Simbiti ethnic group showed a lower likelihood of engagement (OR = 0.24, p = 0.06), this result should be interpreted cautiously due to the very small sample size (n = 8) and wide confidence interval.

Conclusion

This study found that while older women and those affiliated with the Protestant faith were more likely to adopt PMTCT practices, a persistent knowledge–practice gap remained. Primiparous women and members of the Simbiti ethnic group were less likely to engage in preventive behaviors, underscoring the need for age-sensitive, culturally tailored, and community-based interventions to strengthen PMTCT efforts in Tanzania.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-025-25453-7.

Introduction

Throughout history, humanity has faced a number of pandemics, some of which have had enormous consequences on human populations [1–3]. One such ongoing global health crisis is acquired immunodeficiency syndrome (AIDS), caused by the Human Immunodeficiency Virus (HIV) [4]. The disease was first identified in the early 1980 s and has resulted in millions of deaths and has affected every region of the world [5]. The virus is primarily transmitted through sexual contact, blood, and mother-to-child transmission (MTCT), and it continues to excessively affect vulnerable populations, particularly women and children in developing countries [6, 7]. MTCT, which can occur during pregnancy, childbirth, or breastfeeding, remains the leading cause of HIV infection in children.

In 2024 World Health Organization (WHO) report estimated that in 2023 there were over 40 million people living with HIV, including 21 million women and 1.5 million children [8]. Despite significant advances in prevention of mother-to-child HIV transmission (PMTCT) initiatives, pediatric HIV is still a major public health issue, especially in sub-Saharan Africa, which accounts for 65% of HIV cases globally. Further, an estimated 1.3 million women living with HIV become pregnant each year, with 98% of these pregnancies occurring in sub-Saharan Africa [7, 9]. In the absence of interventions, it is estimated that between 15% and 45% of infants exposed to HIV will become infected themselves [7].

Since 2011, global initiatives have focused on eliminating pediatric HIV and improving the health outcomes of HIV-positive mothers and their children through PMTCT program [1]. PMTCT offers a comprehensive set of services, including HIV counseling and testing throughout pregnancy, antiretroviral therapy (ART) for mothers and infants, high-quality antenatal care, safe delivery practices, and guidance on infant feeding [10, 11]. Although developed countries have shown higher adherence and better outcomes from this program, PMTCT coverage remains limited in many developing countries, especially in sub-Saharan Africa [12, 13].

In Tanzania, despite significant governmental efforts to implement the PMTCT program and increase access to HIV testing and treatment for women [14], the rate of HIV transmission from mothers to children remains a challenge [15]. Factors such as limited awareness, stigma, inadequate healthcare access, and sociocultural beliefs contribute to the inconsistent adoption of PMTCT services [16]. Previous studies indicate that even when pregnant and lactating women have knowledge about HIV and its transmission, there is often a disconnect between their awareness and actual health-seeking behavior [17–19]. This knowledge–practice gap may result from barriers such as fear of disclosing HIV status, resistance from partners, financial constraints, and misconceptions about the safety and importance of antiretroviral therapy (ART) and related preventive measures [20, 21]. Therefore, it is important to assess both their level of knowledge and how well it is applied in real-life community settings.

PMTCT of HIV is a crucial strategy in reducing the global HIV/AIDS burden, having prevented an estimated 2.2 million infections and saved 1 million children’s lives since its adoption [1]. A comprehensive understanding of the knowledge, attitudes, and practices (KAP) related to HIV prevention among pregnant women is a critical and cost-effective component of PMTCT success [13, 22]. This study aims to to assess the knowledge–practice gap in the prevention of mother-to-child transmission (PMTCT) of HIV among pregnant and lactating women in Musoma Municipality, Tanzania, regardless of HIV status, and to identify the sociodemographic factors that may influence this gap. In addition, the study considers the role of attitudes as a potential intermediary between knowledge and practice.

Methods

Study population and design

This hospital-based cross-sectional study enrolled pregnant women presenting for their first ANC consultation and postpartum lactating mothers accessing services at Musoma Municipal Council Hospital. This study enrolled women regardless of their known HIV status, as the objective was to assess baseline PMTCT related Knowledge, Attitudes, and Practices (KAP) at the point of entry into the healthcare system. Musoma Municipality is an urban administrative unit situated in Tanzania’s Mara Region with an estimated population of 175,264. Musoma Municipal Council Hospital provides key maternal health services, with an average of 95 pregnant women attending first antenatal visits and 176 lactating mothers receiving care monthly. Exclusion criteria were: [1] individuals hospitalized for more than 48 h; and [2] individuals with cognitive or communication impairments precluding informed consents or reliable data collection.

Sample size determination and sampling technique

The intended sample size for our study was attained by using a standard technique commonly used in heath studies for estimating proportions in a population [23].

The minimum required sample size (n = 196) was determined using Cochran’s formula for proportion estimates at a 95% confidence level and 50% assumed prevalence and was increased to 200 to account for potential non-response. For the sampling procedure, the interval (k) was derived by dividing the average monthly first ANC attendance (95) by the desired monthly sample size (~ 14), yielding k = 7. To facilitate meaningful subgroup comparisons, the final sample was allocated equally between pregnant women (n = 100) and lactating women (n = 100), rather than proportionally to population size, thereby ensuring adequate statistical power for between-group analyses.

Sampling procedure

Eligible participants were recruited through systematic sampling at Musoma Municipal Hospital between January and July 2024. For each subgroup (pregnant and lactating women), the first participant (reference participant) was identified randomly on the first day of data collection by drawing lots among the first three eligible women presenting at the antenatal clinic (for pregnant women) or postnatal clinic (for lactating women). Subsequently, participants were selected at regular intervals (approximately every kth woman, where k was calculated based on patient flow) until the sample quota was met. This approach minimized selection bias and ensured that both subgroups were recruited independently.

Sampling continued daily during routine antenatal and postnatal clinic visits until the sample sizes of 100 pregnant and 100 lactating women were attained. Although proportional allocation to population size would have yielded more lactating mothers than pregnant women, equal allocation was chosen to allow balanced subgroup sizes and maximize power for between-group comparisons, which was central to the study objectives.

Data collection tool

The reviewed literature regarding HIV/AIDS and PMTCT included a subsection detailing biographical information on participants, including age, education, marital status, employment status, and number of children. The study also examines participants’ knowledge of HIV transmission and prevention, as well as participants’ awareness, attitudes, and practices related to HIV infection, specifically in pregnant and lactating women, using a nine-question Likert scale [24]. Additionally, the study investigated participants’ health-seeking behaviors related to HIV/AIDS and PMTCT, as determined by their history of HIV testing, participants’ willingness to disclose their HIV positive status, and attendance to recommended PMTCT interventions. To mitigate social desirability bias, participants were assured of anonymity, and questions were phrased neutrally. Future studies could triangulate self-reported practices with clinic records for validation. 

Key outcome variables were derived from the structured questionnaire and categorized as follows:

Knowledge: Knowledge was assessed through 12 questions on HIV transmission, prevention, and misconceptions (Q10, Q11, Q15–Q21). Each correct response was scored as 1 and each incorrect response as 0, yielding a total score ranging from 0 to 12. Participants with scores ≥ 10 (≥ 83% correct) were classified as having good knowledge, whereas those scoring < 10 were classified as having poor knowledge.

Attitudes: Attitudes were measured using a 5-point Likert scale (1 = Strongly disagree, 5 = Strongly agree) for five statements (Q22–Q26). The total score therefore ranged from 5 to 25. Participants with scores ≥ 20 were categorized as having positive attitudes, while those with scores < 20 were categorized as having negative attitudes.

Practice: Preventive practices were assessed using three key behaviors applicable to all women: [1] HIV testing during the current pregnancy or lactation (Q27) [2], consistent or occasional condom use with a partner (Q29), and [3] participation in HIV education or support groups (Q30). Each reported practice was scored as 1, and non-adherence as 0, producing a total score between 0 and 3. Participants with scores ≥ 2 were classified as having appropriate practice, and those with scores < 2 as having inappropriate practice.

Validity and Reliability

To uphold ethical standards in data collection, the data collectors received thorough training [25]. To guarantee linguistic precision, the study instruments were first created in English, subsequently translated into Swahili by professional translators, and then back translated into English. Completed questionnaires were examined every day to verify their completeness and any errors were rectified during this assessment to ensure the data reliability.

Data management and analysis

The dataset collected through structured questionnaires was analyzed using the R-4.3.0 Software [26]. Following data entry and cleaning procedures, rigorous quality control measures were applied to verify data completeness and internal consistency [27]. We provide descriptive statistics for all variables, with categorical variables presented as frequencies and percentages, and continuous variables represented as mean and standard deviations [28]. Fisher’s exact test and chi-square test were used to assess the connection between variables [29]. These results are presented in Supplementary Table S3 to avoid redundancy in the main text, as the primary analysis relied on unadjusted binary logistic regression to quantify associations. Binary logistic regression (unadjusted odds ratios with 95% CI) was used to examine the factors associated with individuals’ views and behaviors towards HIV. In this investigation, a p-value ≤ 0.05 was considered statistically significant. The statistical study presented data on factors that helped to understand PMTCT among pregnant women, and their following health behaviors [30].

In addition to descriptive and inferential analyses, the study incorporated a comparative framework to contextualize the findings with national policy targets. Specifically, study results were assessed in relation to Tanzania’s National HIV Prevention Road Map and the Elimination of Mother-to-Child Transmission (EMTCT) Strategic Plan II. These national frameworks outline high-risk populations, strategic interventions, and performance benchmarks relevant to pregnant and lactating women. To operationalize this comparison, indicators on knowledge, attitudes, and preventive practices from the present study were aligned with corresponding national targets and strategic pillars. This approach enabled us to evaluate the extent to which local realities in Musoma Municipality reflect progress toward national goals for HIV prevention and EMTCT. The detailed comparative tables are provided in the supplementary material (Table S1 and S2), while the main Discussion focuses on interpreting the implications of the findings.

Results

Socio-demographic characteristics

During the study period, 758 pregnant women registered for antenatal care, and 100 of them agreed to be interviewed. Similarly, from 1,408 eligible lactating mothers, 100 (7.1%) were systematically enrolled. The age range of the respondents was 17–42 years, with a mean age of 31 ± 4 years, most of the respondents are primary school 97(48.5%). The majority 83(41.5%) were Roman followers. One thirds 117(58.5%) were housewives, as summarized in Table 1.

Table 1.

Socio-demographic characteristics of the respondents in the survey (N = 200)

Variables	category	Frequency (n)	Percentage (%)
Age group	≤ 25	64	32.0
	25 − 35	91	45.5
	≥ 35	45	22.5
Marital status	Married	134	67.0
	Unmarried (single, divorced, separated, widowed)	66	33.0
Ethnic group	Jita	96	48.0
	Kwaya	39	19.5
	Kurya	31	15.5
	Luo	12	6.0
	Simbiti	8	4.0
	Other ethnic groups (Zanaki, Kiroba, Swata)	14	7.0
Religion	Muslim	37	18.5
	Roman Catholic	80	41.5
	Protestant	51	25.5
	Orthodox	32	14.5
Educational status	No formal education	21	10.5
	Primary	97	48.5
	Secondary	82	41.0
Occupation	Housewife	117	58.5
	Government employee	54	27.0
	Other/Informal (student, merchant, day worker)	29	14.5
Number of births	Primipara	64	32.0
	Multipara	117	58.5
	Grand multipara	19	9.5
Husband’s occupation	Farmer	34	17
	Government worker	94	47
	Private worker	30	15
	Merchant	38	19
	Others	4	2

Chi-square and fisher’s exact test

Comparative analysis revealed significant associations between MTCT prevention practices and key demographic and knowledge factors. As presented in Supplementary Table S3, women who adhered to prevention practices were older (mean age 30 years versus 27 years; t = − 2.02, p = 0.04) and had more knowledge of routes of HIV transmission (p = 0.01). Knowledge level was strongly associated with practices: 100% of adherents (136/136) had good knowledge compared with 90.6% (58/64) of non-adherents (χ² = 10.12, p = 0.001), and all participants with poor knowledge (6/6) fell into the poor practices group. Overall, 97.0% (194/200) of participants had good knowledge. A significant association was also observed between attitudes and practices (χ² = 157.10, p < 0.001), with 95.6% of those with appropriate practices showing positive attitudes compared to only 6.3% of those with inappropriate practices, while 67.0% of all participants held a positive attitude overall. Religion (χ² = 6.73, p = 0.08) and number of births (χ² = 5.81, p = 0.05) showed trends, but no significant differences were found for marital status, ethnicity, education, or occupation (all p > 0.05). These findings indicate that while good knowledge is necessary for adopting preventive practices, it is not sufficient on its own, and positive attitudes are strongly aligned with effective preventive measures.

Unadjusted binary logistic regression analysis of factors influencing practices and attitude on HIV infection among participants

In the unadjusted binary logistic regression analysis in Table 2, Protestant affiliation shows a positive association with the proactive actions compared to Muslim (OR = 2.84, P = 0.04). The magnitude of this positive association is notably stronger than other religious groups listed. The number of births also has a significant correlation with the outcome, and both multi-para and primiparity show a negative correlation trend. However, compared with multi-para, the negative trend of primipara is stronger (P = 0.05, OR = 0.26). Other significant factors consist of age (P = 0.05, OR = 1.04), and ethnic with individuals of Simbiti ethnicity potentially being less likely to act (P = 0.06, OR = 0.24). On the other hand, as shown in Table 3, the Protestant faith remained significantly linked to positive attitudes towards AIDS (P = 0.02).

Table 2.

Factors influencing practices on HIV infection among participants

Variables		β (SE)	Z	p-value	OR	95% C.I. for OR
						Lower	Upper
Age		0.04 (0.02)	1.99	0.05	1.04	1.01	1.08
Religion	Muslim (Ref)	-	-	-	1.00	(Reference)
	Orthodox	0.20 (0.50)	0.39	0.69	1.22	0.46	3.25
	Protestant	1.04 (0.50)	2.09	0.04	2.84	1.07	7.57
	Roman	−0.01 (0.41)	−0.01	0.99	0.99	0.44	2.22
Ethnic	Jita (Ref)	-	-	-	1.00	(Reference)
	Kiroba	0.19 (1.18)	0.16	0.87	1.21	0.12	12.14
	Kurya	−0.17 (0.45)	−0.37	0.71	0.85	0.35	2.03
	Kwaya	−0.55 (0.40)	−1.37	0.17	0.58	0.27	1.26
	Luo	0.30 (0.70)	0.42	0.67	1.34	0.34	5.26
	Simbiti	−1.42 (0.77)	−1.86	0.06	0.24	0.05	1.08
	Swata	0.70 (1.12)	0.63	0.53	2.01	0.22	18.06
	Zanaki	0.48 (1.14)	0.42	0.68	1.61	0.17	15.09
Duration of awareness	< 5 years (Ref)	-	-	-	1.00	(Reference)
	>5 years	0.44 (0.32)	1.36	0.17	1.55	0.82	2.90
Births Number	Grand multi-para (Ref)	-	-	-	1.00	(Reference)
	Multi-para	−0.78 (0.66)	−1.18	0.24	0.46	0.13	1.67
	Primipara	−1.36 (0.68)	−2.00	0.05	0.26	0.07	0.97
Attitude on HIV infection
	Negative					(Reference)
	Positive	5.78(0.66)	8.71	< 0.001	325.00	88.43	1194.44

OR Odds Ratio, CI Confidence Interval, β Logistic regression coefficient, SE Standard Error

Reference categories: Muslim (religion), Jita (ethnicity), < 5 years (awareness), grand multipara (birth number)

Table 3.

Factors influencing attitude on HIV infection among participants

Variables		β (SE)	Z	p-value	OR	95% C.I. for OR
						Lower	Upper
Age		0.03 (0.02)	1.58	0.11	1.03	0.99	1.07
Religion	Muslim (Ref)	-	-	-	1.00	(Reference)
	Orthodox	0.42 (0.50)	0.85	0.40	1.52	0.58	4.03
	Protestant	1.14 (0.48)	2.35	0.02	3.12	1.21	8.07
	Roman	0.27 (0.41)	0.67	0.50	1.31	0.59	2.91
Ethnic	Jita (Ref)	-	-	-	1.00	(Reference)
	Kiroba	14.76 (727.70)	0.02	0.98	-	0.00	∞
	Kurya	−0.07 (0.44)	−0.15	0.88	0.94	0.39	2.24
	Kwaya	−0.55 (0.39)	−1.40	0.16	0.58	0.27	1.25
	Luo	0.00 (0.64)	0.01	1.00	1.00	0.29	3.53
	Simbiti	−0.81 (0.74)	−1.09	0.28	0.45	0.10	1.91
	Swata	0.80 (1.12)	0.72	0.47	2.23	0.25	19.96
	Zanaki	0.58 (1.14)	0.51	0.61	1.78	0.19	16.67
Duration of awareness	< 5 years (Ref)	-	-	-	1.00	(Reference)
	> 5 years	0.47 (0.32)	1.47	0.14	1.60	0.86	2.99
Source of information	Doctor (Ref)	-	-	-	1.00	(Reference)
	Health worker	0.56 (0.49)	1.13	0.26	1.75	0.67	4.58
	Newspaper	1.19 (1.20)	0.98	0.33	3.27	0.31	34.72
	Others	0.02 (0.65)	0.03	0.97	1.02	0.28	3.68
	Television and radio	0.80 (0.57)	1.41	0.16	2.23	0.73	6.84
Births Number	Grand multi-para (Ref)	-	-	-	1.00	(Reference)
	Multi-para	−0.55 (0.60)	−0.92	0.36	0.58	0.18	1.86
	Primipara	−0.88 (0.62)	−1.42	0.16	0.42	0.12	1.40

OR Odds Ratio, CI Confidence Interval, β Logistic regression coefficient, SE Standard Error 

Reference categories: Muslim (religion), Jita (ethnicity), < 5 years (awareness), grand multipara (birth number)

Respondents’ general Practices, attitudes, knowledge, and awareness regarding HIV infection

Regarding preventive practices Figure 1 A shows that 68% (n = 136) of the participants demonstrated appropriate preventive practices, while 32% (n = 64) had suboptimal practices. Attitudes toward HIV infection Figure 1B indicates that 67% of the participants had positive attitudes toward HIV infection, whereas 33% (n = 66) had negative attitudes. This suggests that although the majority of participants had positive views, one-third still need to improve their attitudes. Knowledge about HIV infection was excellent overall Figure 1C reveals that nearly all participants (98%) had good knowledge of HIV infection, and Awareness of HIV infection was universal Figure 1D shows that 100% of the participants were aware of HIV infection. While knowledge and awareness levels are satisfactory, there remains substantial room for improvement in both preventive practices and attitudes. Specifically, the one-third of participants with suboptimal practices or negative attitudes represent important targets for future intervention programs, highlighting the need for enhanced interventions in HIV prevention and education to guide practices and change attitudes.

Fig. 1.

Figure A. Respondents practices regarding HIV infection Figure B. Respondents attitude regarding HIV infection Figure C. Respondents knowledge regarding HIV infection Figure D. Respondents awareness regarding HIV infection

Discussion

This study assessed the knowledge-practice gap in the prevention of mother-to-child transmission of HIV among pregnant and lactating women in Musoma Municipality, Tanzania and identified sociodemographic factors associated with this gap, yielding three key findings. First, a clear knowledge–practice gap was observed: while awareness of HIV and MTCT was nearly universal (100%) and knowledge levels were very high (98%), only 68% of participants reported good preventive practices. This significant knowledge–practice gap highlights that awareness alone does not necessarily translate into behavior and suggests the presence of additional barriers. In our data, primiparous women were less likely to adopt preventive practices, whereas older women and those identifying as Protestant were more likely to do so. Although the Simbiti ethnic group appeared less likely to adopt preventive practices, this finding should be interpreted with caution given the very small sample size, which limits its reliability. These findings highlight that social (religious affiliation), cultural (ethnic background), and experiential factors (parity and age) may act as facilitators or constraints to behavior change. Similar gaps between knowledge and practice have been documented across sub-Saharan Africa, where social, cultural, and structural influences continue to limit optimal uptake of HIV prevention recommendations despite widespread awareness [31, 32].

Maternal age emerged as an important predictor of HIV prevention among pregnant and lactating women engagement, with older women more likely to adopt preventive behaviors than younger or adolescent mothers. This may reflect accumulated life experience, greater confidence, and repeated exposure to the health system. Comparable age-related disparities in PMTC uptake have been reported in Nigeria, Uganda, and South Africa, where adolescent mothers often face stigma, limited support, and reduced service access [33–36]. These findings underscore the importance of age-sensitive interventions that address the unique challenges faced by adolescents and young women, including fear of disclosure, lack of partner support, and inadequate knowledge, which collectively heighten their vulnerability.

Parity was also associated with preventive behaviors. Primiparous women were less likely to engage in PMTCT practices compared with multiparous women, which is consistent with findings from Zambia and Malawi [37, 38]. Limited antenatal exposure, absence of supportive peer networks, and reduced confidence may all contribute to this gap. Interventions such as targeted counseling during the first ANC visit, structured peer support, and mentorship programs linking experienced mothers with first-time mothers could help overcome these barriers and narrow the divide between knowledge and practice.

Another major contributor to attitudes related to prevention of PMTCT of HIV was religious affiliation, with Protestant women more likely than Muslim women to report positive attitudes and adherence to preventive practices. This finding should be interpreted with caution. Religious affiliation in this context is best understood as a proxy variable, reflecting broader socio-cultural and contextual influences rather than a direct causal determinant. Denominational teachings, faith-based counseling, or the presence of church-linked health programs may help explain the observed association, as reported in Nigeria [39] and Uganda [40, 41]. Moreover, underlying contextual influences, such as ethnicity (including the Simbiti group in our study area), socio-economic status, and access to health services, were not controlled for in this analysis. As such, the findings should be considered exploratory, pointing to areas where further research using adjusted multivariable analysis is needed to disentangle the relative contributions of religion, ethnicity, and other contextual determinants of PMTCT engagement.

Collectively, these findings confirm that high knowledge alone is insufficient to achieve consistent PMTCT engagement. Age, parity, ethnicity, and religious affiliation interact with broader socio-cultural and structural dynamics to shape preventive practices. Addressing these factors requires interventions that integrate biomedical knowledge dissemination with targeted, context-sensitive strategies designed to overcome psychosocial and cultural barriers to behavior change.

Comparison of study findings with the national HIV prevention road map and tanzania’s elimination of MTCT of HIV strategic plan II

The findings of this study provide important insights when considered in the context of Tanzania’s National HIV Prevention Road Map and the Elimination of MTCT of HIV Strategic Plan II (Table S1 and S2). Both national frameworks emphasize improving knowledge, attitudes, and practices among women of reproductive age as essential to eliminating pediatric HIV. While our study found near-universal HIV awareness, consistent with national survey data, gaps remain in translating this awareness into preventive practices, particularly condom use and engagement in HIV support programs. These results underscore the challenge of bridging knowledge and practice, a gap explicitly identified in the Road Map as a barrier to progress.

Our analysis also revealed that socio-demographic factors, including maternal age, parity, and religious affiliation, influenced preventive practices. This finding aligns with the national strategies’ recognition that socio-cultural contexts shape service uptake and highlights the importance of tailored, differentiated service delivery models. For example, the association of Protestant faith with better preventive practices supports the strategic emphasis on faith-based and community engagement, while the vulnerability of primiparous women suggests the need for targeted support for first-time mothers.

In terms of national targets, the proportion of women in this study with good knowledge (98%) is close to the target set in national campaigns. However, positive attitudes (67%) and preventive practices (68%) fall short of the ≥ 80% and ≥ 85% benchmarks set for 2025. These gaps illustrate the urgent need for intensified community education, strengthened support services, and culturally sensitive interventions to meet national elimination goals.

Study recommendations

Based on the study’s findings and evidence from the regional and national policy frameworks, it is recommended to enhance working partnerships with Protestant and other faith-based groups to strengthen their positive impact on reducing stigma and promoting HIV prevention behaviors in line with Tanzania’s National HIV Prevention Road Map priorities [42]. Interventions should target first-time mothers while recognizing their unique informational and psychosocial needs through peer support and mentorship programs. In line with the reviewer’s observation, our study specifically highlights primiparous women as a vulnerable group, and peer-based mentorship models such as the PEPFAR-supported “Mentor Mothers” approach represent a practical and evidence-based strategy to improve their understanding and consistent use of PMTCT practices. Scaling up this model would directly respond to the gaps identified in our findings.

Moreover, to build trust with first-time mothers, HIV prevention messages and services must be relevant within a cultural context and account for the beliefs and practices of ethnic groups, such as the Simbiti. This can be achieved in part through collaborations with local leaders and engagement through culturally acceptable communication strategies. Older women, who could also be favorable advocates and have a greater willingness to receive new ideas, can be widely trained to serve as community advocates and role models, although services that are friendly to youth will be developed to reach young mothers more effectively. Increasing community health worker reach will address disconnection between local communities and their health facilities, which is critical in helping to ensure people utilize what they learn to prevent complications. Finally, in respect to mother-caring roles, antenatal and postpartum visits should take a combined approach supporting ongoing, context-specific counseling to consolidate knowledge while addressing barriers to the adoption of PMTCT in the face of stigma and lack of understanding. The primary objective will address gaps in knowledge and practice while ultimately increasing the momentum towards the end of mother-to-child transmission of HIV in Tanzania.

Limitation

This study has several limitations that should be considered when interpreting the findings. First, the cross-sectional design limits the ability to establish causal relationships, as data were collected at a single point in time and can only demonstrate associations. Second, the reliance on self-reported information may have introduced social desirability bias, with some participants potentially overstating their knowledge or preventive practices. Third, HIV serostatus was not collected, as the study included all pregnant and lactating women regardless of HIV status; while this reflects the universal entry point of PMTCT programs through routine testing and counseling, it means that conclusions were drawn primarily from social, cultural, and behavioral perspectives rather than stratified outcomes among women living with HIV/AIDS. Fourth, some subgroup analyses, particularly for the Simbiti ethnic group (n = 8), were limited by very small sample sizes, which restricts statistical power and the generalizability of those findings. Finally, the absence of adjusted multivariable models means that some observed associations may be confounded by unmeasured factors. Future studies should incorporate HIV status data, use adequately powered subgroup samples, and apply multivariable analyses to better disentangle the relative contributions of biological, social, and cultural influences. Despite these limitations, the study provides valuable insights into current patterns and associations, offering a foundation for further research and for informing targeted, context-specific public health interventions to strengthen PMTCT efforts in Tanzania.

Conclusion

This study demonstrates a persistent knowledge–practice gap in the prevention of mother-to-child transmission of HIV among pregnant and lactating women in Musoma Municipality, with nearly universal awareness and high knowledge levels not consistently translating into preventive behaviors. Older women and those affiliated with the Protestant faith were more likely to engage in PMTCT practices, while primiparous women and members of the Simbiti ethnic group were less likely, underscoring vulnerabilities linked to parity and cultural context. These findings emphasize the need for age-sensitive, culturally appropriate, and community-driven interventions, including faith-based partnerships, tailored strategies for ethnic groups, and peer mentorship models such as Mentor Mothers. Although this study did not restrict participation to women living with HIV/AIDS, the results remain highly relevant to PMTCT programming, as universal HIV testing and counseling are the entry points for all women. Despite its limitations, the study provides valuable insights that can inform targeted public health interventions and future research, with a focus on first-time mothers, culturally distinct groups, and community engagement to bridge the divide between knowledge and practice and accelerate progress toward eliminating MTCT of HIV in Tanzania.

Abbreviations

ART

Antiretroviral Therapy

ANC

Antenatal Care

HIV/AIDS

Human immunodeficiency Virus/Acquired immunodeficiency Syndrome

HCT

HIV counseling and testing

KAP

knowledge, attitudes, and practices

MTCT

Mother-To-Child Transmission

PMTCT

Prevention Mother-To-Child Transmission

Authors’ contributions

**S.H.S.** Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. **K.T.** Writing – review & editing, Writing – original draft, Methodology, Investigation, Data curation, Conceptualization. **S.M.S.** Formal analysis, Writing – original draft, Writing – review & editing. **Z.Z.** Writing – review & editing, Methodology, Investigation. **S.C.** Writing – review & editing, Conceptualization. **J.S.** Writing – review & editing, Methodology. **J.K.** Writing – review & editing, Investigation. **S.S.** Writing – review & editing, Project administration. **J.Z.** Project administration, Supervision, Funding acquisition, Resources, Conceptualization. All authors contributed to validation and visualization, reviewed the final manuscript, and agreed to be accountable for the work’s integrity.

Funding

There was no funding for this study.

Data availability

The anonymized dataset supporting the conclusions of this article is available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The study was reviewed and approved by the Musoma Municipal Council Hospital (Re: HMM/D.30/4/08). All research was performed in accordance with the relevant guidelines and regulations as outlined in the Declaration of Helsinki. Written informed consent was obtained from all prospective participants prior to their enrollment in the study. The consent process involved a detailed verbal and written explanation of the studies purpose, procedures, potential benefit to public health, and the assurance of anonymity, confidentiality, and that no personal identifiers would be collected or stored. Participants were explicitly informed that their participation was entirely voluntary, that they had the right to withdraw at any time without any consequence to their medical care.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.