Factors associated with HIV infection among infants born to mothers with HIV in Blantyre, Malawi: a mixed methods study

Abstract

Background

HIV infection is a major cause of mortality and morbidity among infants in Malawi. Despite the scale-up of elimination of mother-to-child transmission (e-MTCT) services in Malawi, the mother-to-child transmission rate was 6.5% in 2021. Therefore, the study aimed to assess factors that are associated with HIV infection among infants born to mothers with HIV in Blantyre, Malawi.

Methods

A cross-sectional study was conducted using a mixed-methods convergent (triangulation) approach. The study was conducted at Bangwe, Limbe, and Ndirande Health Centres which are primary care health facilities in Blantyre City. The study participants were mothers with HIV and their infants exposed to HIV, aged 0 to 24 months, receiving e-MTCT care and e-MTCT service providers at the health facilities. The study population were nine purposively-selected e-MTCT service providers for the qualitative component of the study and 271 mothers with HIV and their infants exposed to HIV for the quantitative component sampled using probability proportional to size across the three facilities and simple random sampling within each facility. In-depth qualitative interviews were conducted with the service providers and tape-recorded while quantitative data were retrospectively extracted from e-MTCT logbooks and hospital records for the mothers and infants receiving e-MTCT care from January 2020 to December 2021. Quantitative data was entered and analyzed in Stata Version 17. Descriptive statistics (frequency, mean and percentage) were used to summarise data for each variable. Binary logistic regression was employed to determine association variables of HIV in infants. A p-value of less than 0.05 was considered significant in deciding on the association of variables. Interviews were transcribed verbatim, and the cleaned data were imported into NVivo software for thematic coding and analysis.

Results

Of the 271 infants exposed to HIV reviewed, 12.2% (95% CI = 8.5% − 16.7%) were HIV-infected. Factors that were significantly associated with infant HIV infection included: mixed feeding before six months (aOR = 5.8, 95% CI = 1.38–24.29), high maternal viral load (aOR = 78.6, 95% CI = 22.74–271.7) and receiving ARV prophylaxis at birth (aOR 0.2, 95% CI = 0.05–0.56). A qualitative study found that maternal and service provider individual-level and general health system factors were associated with suboptimal uptake of e-MTCT intervention. These included negative attitudes among mothers, tardiness by healthcare workers and mothers, the provision of incorrect contact details, defaulting on treatment, frequent relocation, religious beliefs, poor adherence to treatment protocols and late initiation of antiretroviral therapy, multiple service delivery locations, lack of proper training or knowledge gaps, high workload and lack of resources.

Conclusion

HIV infection among infants exposed to HIV remains unacceptably high, mostly due to sub-optimal e-MTCT program implementation. This requires intensified monitoring and evaluation (M&E) of the e-MTCT programme implementation and use of M&E data for targeted quality improvement.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-026-27798-z.

Introduction

HIV infection is a significant health problem causing morbidity and mortality among children worldwide. Approximately 1.7 million children are living with HIV/AIDS worldwide, with more than 80% of them living in Sub-Saharan African countries [1]. Eastern and southern Africa remain the highest regions with a high prevalence of HIV, accounting for approximately 55% of all people and two-thirds of all children living with HIV [1]. Malawi is one of the countries in the Sub-Saharan Africa region that has been severely affected by the HIV/AIDS epidemic, with a total of 569,000 children that have been infected since the beginning of the HIV epidemic [2].

HIV infection in children may occur during pregnancy, labour and delivery or breastfeeding, which is known as mother-to-child transmission (MTCT) and accounts for 14% of all new HIV infections worldwide [3]. Over 90% of new infections in infants and young children occur through MTCT. Without any intervention, transmission rates range from 15% to 45% (WHO) [4], with 10% to 25% of transmission occurring during pregnancy, 35% to 40% during labour, and 35% to 40% during breastfeeding [5]. In Malawi, MTCT accounted for approximately 6.5% of new HIV infections in 2021 [6].

Some of the major risk factors for MTCT include the absence of antiretroviral intervention for the pregnant and breastfeeding mothers living with HIV, failure to initiate nevirapine prophylaxis for the infant, the mother’s advanced HIV clinical disease and mixed infant feeding practices [4, 7, 8]. To address the risk factors, Malawi adopted the option B+ strategy in 2011 which involves the introduction of lifelong ART for all HIV-positive pregnant and breastfeeding women regardless of clinical staging or CD4 count [9]. By 2019, the transmission rates decreased to 2% at six weeks after birth and 7.6% at the end of the breastfeeding period, resulting in 3,336 new infections among children in Malawi [2]. In addition, infants exposed to HIV receive antiretroviral prophylaxis at birth and the WHO further recommends co-trimoxazole prophylaxis from four to six weeks of age, continued until HIV infection is definitively excluded following cessation of breastfeeding [10]. Malawi achieved 95% coverage for infant Nevirapine prophylaxis among known infants exposed to HIV born at health facilities and 94% start cotrimoxazole prophylaxis within two months of birth [9]. A study conducted in 2021 found that late antenatal clinic attendance, late initiation of maternal ART and failure to initiate infant prophylaxis were among the risk factors associated with MTCT [2].

In 2020, Malawi set ambitious targets of reducing new childhood HIV infections by 90% and HIV-related child deaths by more than 50% by 2030 [11, 12], reducing MTCT rate at the end of breastfeeding to less than 5% and new pediatric HIV infections to ≤ 50 per 100 000 live births [11]. Achievement of these targets requires high coverage of quality e-MTCT interventions. Thus, the objectives of this study were to assess factors that are associated with HIV infection among infants born to mothers with HIV in Blantyre, Malawi.

Methods

Study design and setting

We used a mixed methods convergent (triangulation) approach [13] which combined a cross-sectional quantitative study with a grounded-theory qualitative study. The study was conducted from January 2022 to June 2023 at three primary care health facilities (health centres) in Blantyre City, Southern Malawi, namely Bangwe, Limbe, and Ndirande. Blantyre City has one of the highest HIV prevalence in Malawi, estimated at 14.2% in 2021 [14]. The catchment populations for the health centres are high; Limbe health centre: 105,347, Ndirande: 145,187 [15] and Bangwe: 143,187 [16]. The health centres provide e-MTCT services to pregnant and breastfeeding mothers living with HIV of all ages. In 2021, it is estimated that 336, 267 and 309 mothers living with HIV received e-MTCT services at Bangwe, Limbe, and Ndirande health centres, respectively. During e-MTCT clinics, the health centres routinely collect various data which are recorded in patient logbooks as well as follow-up cards at the antenatal and under-five clinics.

Study participants and recruitment

The study population for the cross-sectional quantitative study was infants exposed to HIV and their mothers who had enrolled in e-MTCT at the three health centres from January 2020 to December 2021. Infants exposed to HIV who were not confirmed to be HIV positive and infant-mother pairs who had no complete data were excluded. The study populations for the qualitative study were nine purposively selected e-MTCT service providers who were selected using a non-probability sampling technique to provide in-depth insights, with data collection continuing until no new themes emerged. To enhance representativeness, at least three providers were chosen from each facility, ensuring variation in cadre and experience and capturing diverse perspectives across the study sites. In addition, e-MTCT service providers with at least two years of work experience were included. The sample size for the quantitative study was 271 which was calculated using the Cochrane formula [17]: n₀ = z²pq/e² where n₀ is the minimum sample size required, z is the standard variable corresponding to 95% confidence interval (1.96), p is for the estimated prevalence (assumed to be 0.5) among infants exposed to HIV based on the primary objective, e is the marginal of error (0.05) and q is calculated as (1-p).

Substituting these values into the formula:

Since the study population was finite (N = 912), the sample size was further adjusted using the finite population correction formula:

Thus,

n = 384.16 / [1 + (384.16 − 1) / 912 = 270.5, which was rounded up to a final sample size of 271 participants.

Probability proportional to size was applied to allocate the sample proportionally across the three health centres based on the total number of eligible patients in each facility, ensuring that larger centres contributed more participants. Subsequently, simple random sampling was employed to select individual records from the register from each facility, giving each eligible patient an equal chance of being included. This approach helped ensure representativeness across facilities while minimizing selection bias at the individual level.

Data collection procedures

Study preparation

Prior to data collection, the research assistant (RA) recruited and the researcher were both trained on the study protocol, good clinical practices and data collection protocols.

Quantitative data collection

Data were collected between 1st and 19th May 2023 by the RA (JI) and the researcher (AT) by retrospectively reviewing mothers with HIV e-MTCT and infants exposed to HIV care and follow-up records at Bangwe, Limbe and Ndirande health centres which are routinely collected from under-five, maternity and ART clinics. The following data were extracted from the records: breastfeeding practice, socio-demographic characteristics, information on Antiretroviral (ARV) prophylaxis for the mother and infant, place of delivery, the clinical characteristics of the mothers with HIV and the infants exposed to HIV as well as infant’s final serostatus. The final HIV result and medical records for infants exposed to HIV were obtained from logbooks and maternal medical records.

Qualitative data collection

Data for qualitative study were collected through individual in-depth interviews (IDIs) conducted by the RA and researcher. The interview guides were developed in English and translated into Chichewa by the researcher (AT) and were reviewed by the second author (VM). Semi-structured guides were used to conduct the IDIs which lasted approximately 30 to 45 min and were audio-recorded. These IDIs were conducted in a quiet, private room provided at the study facility to safeguard the participants’ confidentiality and to ensure good quality of the audio-recordings.

Data management and analysis

Quantitative data

Extracted participants’ data were entered and analysed using Stata version 17. Descriptive statistics were used to summarize data for each variable. Binary logistic regression was employed to determine the association of variables with HIV in infants. A p-value of less than 0.05 was considered significant in deciding on the association of variables.

Qualitative data

The tape-recorded data was transcribed verbatim in Chichewa. The researcher ensured accuracy by comparing each transcription to its original audio recording through listening to the tapes and reviewing the transcriptions multiple times. Afterwards, the Chichewa transcriptions were translated into English for analysis. Data were stored securely, and audio files were saved using a password in a computer folder.

Data were analyzed using thematic content analysis. Initially, we familiarized ourselves with the data by reading interviews, observations, and field notes, and generated initial codes to identify potential connections. The initial coding scheme was validated through discussions with qualitative data analysis experts. Using NVivo 14 software, we created a codebook to organize the data, employing both deductive coding based on the study’s objectives and theoretical framework [18] and inductive coding from the data itself. Themes were identified from coded text segments that exhibited a consistent, coherent, and recognizable pattern. These segments were then organized to develop overarching themes. The researchers subsequently reviewed and refined the potential themes, ensuring they aligned with the entire dataset. The overarching themes, represented by various categories, were presented as findings, culminating in the final analysis and description of the results.

Ethical consideration

This study was conducted following the Declaration of Helsinki [19]. Ethical approvals were obtained from the College of Medicine Research and Ethics Committee (COMREC; reference number P.11/22/3889). Permission was sought from and granted by the Blantyre District Health Office (DHO) to access the participants’ medical records. A waiver of informed consent was granted for the quantitative data because this study involved secondary analysis of existing data routinely collected as part of standard medical care. Obtaining fresh consent was not required since there was no direct interaction with the participants. For qualitative data, participation was voluntary, and informed consent was obtained from the study participants (e-MTCT service providers). To ensure their confidentiality, anonymity, privacy, and numbers were assigned to maintain anonymity.

Results

Quantitative data

Sociodemographic and clinical characteristics of study participants

Two hundred and seventy-one records of mothers with HIV and their infants exposed to HIV were reviewed. Table 1 summarises the characteristics of mothers with HIV and infants exposed to HIV. At least one-half of the mothers were in the 25–35 age group. The proportions of male and female infants were almost equal. Most infants were reported to have received ARV prophylaxis (80.4%) and to be on exclusive breastfeeding (88.6%) before six months of age.

Table 1.

Sociodemographic and clinical characteristics of mothers with HIV and infants exposed to HIV (N = 271)

Variable	Frequency	Percentage
Maternal Age
< 18	7	2.6
18–24	21	26.2
25–35	140	51.7
Above 35	53	19.6
Health Facility
Bangwe	100	36.9
Limbe	79	29.2
Ndirande	92	34
Sex of the Infant
Female	135	49.8
Male	136	50.2
Age of infants at enrolment to e-MTCT care
≤ 6 weeks	232	85.6
> 6 weeks	39	14.4
Age of the infant at HIV diagnosis
< 18 months	268	98.9
>=18 months	3	1.1
Infant Cotrimoxazole preventive therapy initiation
> 6 weeks	77	28.4
≤ 6 weeks	194	71.6
Infant feeding option before 6 months
Exclusive breastfeeding	240	88.6
Mixed feeding	31	11.4
HIV stage for mothers
WHO stages I and II	267	98.5
WHO stages III and IV	4	1.5
Viral load during pregnancy or lactation for mothers
< 1000 copies	236	87.1
> 1000 copies	35	12.9
Infant ARV prophylaxis
No	53	19.6
Yes	218	80.4

Prevalence of HIV infection in infants

Of the 271 Infants exposed to HIV who were tested for HIV, 33 were infected, representing an HIV prevalence of 12.20% (95% = CI 8.5% − 16.7%). HIV prevalence among infants < 18 months was 9.9% (95% = CI 6.4% − 14.5%) and those older > 18 months was significantly higher at 25.6% (95% = CI 13% − 42.1%). The distribution of HIV prevalence by the health facility was as follows: Bangwe; 17.0% (95% CI = 10.2% − 25.8%), Limbe; 6.3% (95% CI = 2.1% − 14.2%) and Ndirande; 12.0% (95% CI = 6.1% − 20.4%).

Factors associated with HIV infection

An additional file [See additional file 1] shows factors that were assessed for their association with infant HIV infection. The bivariate analysis showed that infant cotrimoxazole prophylaxis, infant feeding practices (mixed feeding) and high maternal HIV viral load were positively associated with infant HIV infection. In contrast, maternal age and WHO clinical stage were not significantly associated with infant HIV infection.

In multivariable logistic regression, infants who were mixed-fed had 5.8 times higher odds of HIV infection than those who were exclusively breastfed. Furthermore, infants whose mothers had a viral load of ≥ 1000 copies/mL had 78.6 times higher odds of being infected with HIV than infants whose mothers had a viral load of less than 1000 copies/mL. Notably, Infants who took ARV prophylaxis were 80% less likely to be infected with HIV than infants who did not take ARV prophylaxis.

Qualitative data

Of the nine health workers who participated in the IDIs, seven were above 35 years old. Six were females and five were married. Four of the service providers went to college, and only 1 had less than five years of work experience in the e-MTCT program.

e-MTCT providers’ perspective regarding e-MTCT service delivery

Several factors were identified among healthcare workers as barriers that negatively impact e-MTCT service delivery. These barriers have been categorized into individual-level factors and health system factors.

Individual-level factors

Healthcare workers identified several individual-level barriers affecting the effective delivery of e-MTCT services. These include negative attitudes towards e-MTCT among some mothers, Healthcare workers’ tardiness in reporting for duties and mothers being late in attending clinic sessions, the provision of incorrect contact information, the negative influence of certain religious beliefs in decision making and frequent relocation. Additional challenges reported include poor adherence to treatment protocols and late initiation of antiretroviral therapy. These factors hinder consistent engagement in care and compromise the effectiveness of e-MTCT interventions.

Healthcare workers observed that some mothers seek to be attended to more quickly than those who arrived earlier while others intentionally keep a low profile, trying to avoid being seen by neighbors or colleagues. Mothers who are in a hurry or who try to avoid being seen may skip essential steps in antenatal care, such as HIV testing, counseling, or initiation of antiretroviral therapy (ART) which increases the risk of transmitting HIV to the infant during pregnancy, childbirth, or breastfeeding.

“These people, like I have said, have issues like wanting to be in a hurry and refusing to stay in one place. They might come here but end up hiding. On my own, I see this as a special issue. Like others may say maybe my neighbour will see me, they hide so that they should not be seen. So, when they come here, they walk in hiding. We are supposed to accept that if someone is hiding, we ask her what her problems are so that we can especially help her. Maybe she is afraid to appear in public with her neighbour, maybe there are her workmates.”………. (Healthcare worker 8, Limbe)

Most healthcare workers pointed out that women come to the hospital late for e-MTCT services. In addition, healthcare workers’ tardiness in reporting for duties has been observed to hinder e-MTCT service delivery leading to women leaving and not receiving the care needed such as medication.

“I can say that when women come in late, they end up finding that the doctors had already left or have knocked off.” (Healthcare worker 5, Bangwe).

“It can happen that here at [Health Centre X], the nurse comes around 9 or 10 and people are waiting to be given medication. That is the challenge, and most of the time, it is like that, but if the healthcare workers were to report for duties on time, Women on e-MTCT would not have difficulties staying in long queues or staying longer in a place to receive services.” (Healthcare worker 7, Limbe).

Healthcare workers complained that some mothers enrolled in the e-MTCT program provided false or inaccurate contact information. During enrollment, they are asked whether they prefer to be contacted by phone or followed up at home if they discontinue medication. However, when the details provided are incorrect, it becomes challenging for healthcare workers to trace and follow up with these women, particularly when appointments have been missed.

“As we initiate these women on medication, they give us the wrong physical addresses and phone numbers. When the individual fails to show up on their scheduled visit day, we try to search for her. We often discover that the location we are searching for this person, is not where she lives. She gave us the wrong address and sometimes the wrong name…” (Healthcare worker 1, Ndirande).

Healthcare workers reported that some mothers decline to receive care for themselves and their babies due to religious beliefs, leading to delayed initiation of ART. This delay increases the risk of HIV transmission to their infants, as necessary treatments are not administered in time to prevent mother-to-child transmission of HIV.

“I remember a case in which a woman came in. So, when I told her that she was HIV infected, she responded that “she has faith, she cannot have HIV”. The woman denied the services that we were offering her including giving her Nevirapine for the baby and later came back when the baby had already been infected…” (Healthcare worker 1, Ndirande).

Poor adherence to treatment and delayed initiation of antiretroviral (ARV) therapy were identified as key contributors to HIV infection among infants exposed to HIV. Healthcare workers noted that inconsistent medication use, delayed treatment initiation due to spousal approval, and late presentation to antenatal care, often in the third trimester, increase the risk of transmission. Defaulting from care further limits retention of mothers and infants exposed to HIV, while frequent relocation and displacement disrupt follow-up and continuity of care. One key issue that emerged during the interviews was the impact of Cyclone Freddy, which was a major cause of displacement among women receiving care.

“What happens is that when the parent is not taking the medication properly can transmit the disease to the baby. Since the mother’s immunity is reduced. When the immunity is reduced other things will make the baby get the virus from the mother like when breastfeeding, thus the reason that I can say can make a baby get the virus from the mother.”……. (Healthcare worker 7, Limbe)

“Like here in town recently, due to Cyclone Freddy, there are some mothers whose houses collapsed, and eventually they had nowhere to stay, so they relocated to an unknown place. These were known after follow-up visits were done. So, despite giving information to mothers that their child is supposed to be tested in these selected months: six weeks, 12 months, and some will not come. They will not come to collect drugs for themselves and the child. Then the child also missed testing at 12 months.” (Healthcare worker 9, Limbe).

Health system factors

The health system barriers included multiple service access points, lack of proper training or knowledge gaps, high workload, and lack of resources. Notably, knowledge gaps and high workload were common concerns reported by all participants. Most healthcare workers reported that most clients receive e-MTCT at multiple locations which affects the follow-up of these clients.

“Most mothers on e-MTCT in Blantyre have several hospitals where they receive services. They can come here for antenatal, but if you were to ask them, they receive ART medication somewhere. In addition, they would want the baby to receive services at a different hospital. So, it is a challenge because the follow-up is difficult.” (Healthcare worker 6, Bangwe).

Most of the e-MTCT providers noted a knowledge gap among healthcare workers regarding e-MTCT. They further added that only a few people were trained in ART and cited that absenteeism among those people who were trained disrupted quality e-MTCT services provision to both mothers with HIV and Infants exposed to HIV.

“I am of the view that the knowledge gap is still there. Most nurses don’t know the pink card properly. Some training for this program happened a long time ago. ART training is combined with e-MTCT. But not everyone goes for ART training. Maybe they will take three people from a big facility like here. A good example, for mothers who are at high risk, as soon as their baby is born, we are supposed to give 2P. But if you go to the maternity ward where they deliver, maybe nurses don’t have that knowledge.” (Healthcare worker 6, Bangwe).

Healthcare workers delivering e-MTCT services face a heavy workload as they juggle multiple responsibilities for a growing number of mother-infant pairs. The increasing cohort size makes it challenging to offer focused, individualized care. As a result, critical interventions may be delayed or overlooked, increasing the risk of HIV transmission to exposed infants.

“… There is a lot of work because our clinic offers services to many people. We also meet a lot of people on this day. There is a huge amount of work because when this woman comes to the clinic, she brings a card for her husband, her card, a card for the firstborn child, maybe who is also on medication, and she has brought another card with her for the baby on HCC. Meaning if there are 30 people, 30 times 4, it is 120. Sometimes, people may feel like we are going slow since it takes longer if one person goes into the room, but they don’t understand why the individual is taking so long. Because for one individual it means we are attending to several individuals.” (Healthcare worker 1, Ndirande).

Healthcare providers highlighted that a major challenge in delivering e-MTCT services is the lack of resources, particularly stockouts and drug shortages such as cotrimoxazole preventive therapy. This disrupts treatment and increases the risk of HIV transmission to infants. Furthermore, stockouts of dry blood spot bundles for retesting infants, delays diagnosis and treatment especially for infant who are HIV-positive, and a lack of resources such as airtime and transportation fees to do tracing for those women who have missed appointments.

“It might happen that some medication that we are supposed to give women we don’t have, we can say it is another burden because previous it could happen that there is no Bactrim”…………(Healthcare worker 7, Limbe).

“The constant availability of resources, because there are other instances, of course, they are not common. The baby has grown and the time for testing has come but no available materials for testing such as DBS bundles not available”………(Healthcare worker 3, Ndirande).

Discussion

This mixed-methods study assessed factors that are associated with HIV infection among infants born to mothers with HIV in Blantyre, Malawi. In this study, the overall prevalence of HIV infection amongst infants was 12.2%. Factors that were associated with HIV infection among infants exposed to HIV include taking cotrimoxazole, feeding option, viral load and infant ARV prophylaxis. The qualitative enquiry showed that individual-level and health system factors; negative attitudes among mothers, tardiness by HCW and mothers, the provision of incorrect contact details, defaulting on treatment, frequent relocation, religious beliefs, poor adherence to treatment protocols and late initiation of antiretroviral therapy, multiple service delivery locations, lack of proper training or knowledge gaps, high workload and lack of resources contributes to suboptimal uptake of e-MTCT associated with MTCT of HIV.

The infant HIV prevalence in the study area exceeded the national and WHO target of < 5% [2] and was higher than the prevalence reported in Oromia Regional State, Ethiopia (7.7%) [3], Uganda (5.1%) [20], and Bahir Dar administration, Ethiopia (5.8%) [21]. This could be due to the different geographic distribution of HIV prevalence and cultural practices, such as mixed feeding, which could increase the risk of HIV infection as well as differences in the implementation of the e-MTCT program.

In Malawi, the recommended e-MTCT interventions include viral load testing for mothers with HIV, maternal antiretroviral therapy (ARVs), early infant diagnosis (EID), provision of infant ARV prophylaxis for 6 weeks, exclusive breastfeeding for the first 6 months with continued breastfeeding up to 12 months or more while the mother is on ARVs, and support services like counselling [2]. This study found that poor or delayed uptake of these interventions was linked to a higher risk of HIV infection in infants. Infants who started CXT after 6 weeks, mixed feeding, infants whose mothers had a viral load of ≥ 1000 copies/mL had high odds of being infected with HIV, reinforcing the role of high maternal viral load in increasing transmission risk. This result aligns with evidence from Ethiopia [3, 4, 21] and Zimbabwe [22], although it contrasts with findings from Mozambique [5], possibly due to the smaller sample size in that study. Additionally, Infants who received ARV prophylaxis at birth had lower rates of infection, which is consistent with findings of another study [7]. This protective effect is likely due to the viral suppression activity of nevirapine (NVP), which inhibits HIV replication in the early stages of life.

HCWs reported challenges in e-MTCT intervention delivery which could explain the uptake of interventions among some mother-infant pair. Firstly, most mothers provide false contact information, making it difficult for HCWs to follow up on missed appointments or deliver critical information like infant HIV test results. Secondly, negative attitudes towards e-MTCT among some mothers such as impatience and demanding preferential treatment disrupt clinic workflows and create additional stress for already overburdened staff. Similar attitudes were documented in a previous study [23]. Thirdly, tardiness in attending clinic sessions by mothers and HCWs reporting late for duties has been observed which leads to missed services, with women sometimes arriving after staff have left, and others leaving due to long waiting times, which compromises continuity of care. A major issue identified is the high default rate among HIV-positive mothers, with an overall rate of 15.5% reported in a study conducted in Malawi [26]. This is often driven by relocation and displacement caused by natural disasters, such as Cyclone Freddy [27]. Displacement disrupts access to healthcare services, leading some women to miss appointments for drug refills and testing, which is associated with poor adherence and increased risk of poor immunological outcomes [28]. Interruptions in treatment also raise the risk of viral resistance, resulting in higher viral loads and ultimately increasing the risk of MTCT of HIV.

The study findings also revealed that HCWs face health system barriers that hinder the effective implementation of e-MTCT of HIV. The significant health system barriers mentioned were a shortage of essential drugs and supplies, staff shortages, poor training in e-MTCT protocol and high workload. The study findings are similar to the study done in Ghana [24] and India [23] which highlighted that frequent drug shortages and stock-outs, especially of NVP and infant HIV testing kits, were among the challenges encountered. In addition, the HCWs are not equipped with enough knowledge, plus few staff are trained on e-MTCT protocols. The study findings are similar to the findings of another study, which highlighted that a shortage of health workers and a lack of training are significant challenges experienced by most health facilities in the district in Ghana [24]. The knowledge gap leads to the provision of wrong information to those receiving the services, which may result in a higher number of children being infected with HIV from their mothers. This resonates with the study conducted in South Africa which demonstrated that increased workload leads to burnout among healthcare providers [25]. Increased workload may have been exacerbated due to staff shortages, increased catchment area and population being served as well as lack of training on e-MTCT protocol. Excessive work causes errors like misdiagnosis, leads to poor patient education, reduced interaction time, and incomplete documentation.

Study strengths and limitations

The study’s main strength was its use of a mixed-methods approach, which provided comprehensive and valid findings. Since this was not a cohort study, it could not assess the MTCT rate (incidence), infants of different ages were tested at different times. However, it faced several limitations, including missing variables in routine medical records that could have been associated with MTCT, participant refusal, recall bias, limited generalizability due to the focus on three specific health centres, and the inability to determine causality from the cross-sectional design. Furthermore, the qualitative component did not include mothers with HIV, which limits the breadth of perspectives captured.

To mitigate potential study design bias, validation and triangulation of collected data were employed. Randomization helped reduce sample selection bias, and recall bias was minimized by relying primarily on medical records and structured data collection tools. While participant refusal may have introduced selection bias, efforts to ensure confidentiality and obtain informed consent were undertaken, and its potential impact on representativeness was acknowledged.

Conclusion

This study reveals a concerningly high prevalence of HIV infection among infants exposed to HIV in the study area, underscoring the persistent challenges in the e-MTCT of HIV. Key factors associated with MTCT of HIV included mixed feeding before six months of age, maternal viral load of greater than or equal to 1000 copies/mL, and not receiving infant ARV prophylaxis at birth.

In addition, e-MTCT service providers face several challenges in delivering care to HIV-infected mothers and HIV-exposed infants. These include individual-level issues such as mothers’ negative attitudes, tardiness, incorrect contact information, treatment defaulting, frequent relocation, religious beliefs, poor adherence, and late initiation of antiretroviral therapy. Health system-related challenges include multiple care locations, inadequate training, high workloads, and resource shortages. These challenges hinder the effective implementation of e-MTCT interventions. Addressing these factors through improved maternal care, timely infant interventions, strengthening healthcare systems, and enhanced community engagement is vital for reducing HIV transmission and improving outcomes for both mothers and their infants.

Abbreviations

AIDS

Acquired Immunodeficiency Syndrome

ART

Antiretroviral Therapy

COMREC

College of Medicine Research Ethics Committee

CPD

Continuous Professional Development

DHO

District Health Office

e-MTCT

elimination of Mother-to-Child Transmission

HIV

Human Immuno-deficiency Virus

MTCT

Mother-to-Child Transmission

NVP

Nevirapine

WHO

World Health Organization

Authors’ contributions

AWT planned and designed the study, developed the study methods, interview guides, conducted data collection, analyzed the data and drafted the manuscript. VM supervised the planning of the study, development of methods, data collection, data analysis, reviewed and edited the manuscript. VM supervised data analysis, reviewed and edited the manuscript. All authors read and approved the final manuscript.

Funding

The study was funded by the first author.

Data availability

The datasets generated and/or analyzed during the current study are not publicly available due to participant privacy but are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approvals were obtained from the College of Medicine Research and Ethics Committee (COMREC) reference number P.11/22/3889 before the commencement of data collection. Permission was granted to access data and participants from the Blantyre District Health Office (DHO). A waiver of informed consent was granted for the quantitative data because this study involved secondary analysis of existing data routinely collected as part of standard medical care. Obtaining fresh consent was not required since there was no direct interaction with the participants. For qualitative data, participation was voluntary, and informed consent was obtained from the study participants. To ensure their confidentiality, anonymity, privacy, and numbers were assigned to maintain anonymity. To avoid coercion, e-MTCT service providers did not participate in the recruitment process of study participants. All methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.