1 INTRODUCTION

Dietary iron deficiency (ID) and vitamin A deficiency (VAD) are the two leading micronutrient deficiencies of public health nutrition importance that negatively impact women of reproductive age (WRA) residing in low- and middle-income countries (LMICs) of Africa and Asia (Abbafati et al., 2020; Gardner & Kassebaum, 2020; Mawani & Aziz Ali, 2016). However, the highest burden is among pregnant and lactating women because of the increased physiological demands for iron and vitamin A, which increase due to either fetal demands during pregnancy or breastfeeding during lactation (Gannon et al., 2020; Khayat et al., 2017). If vitamin A requirements during lactation are not met, it may be associated with clinical VAD characterized by night blindness and bitot's spots among lactating mothers (Baytekus et al., 2019). The increased iron demands among WRA coupled with the intake of low-iron foods are associated with ID anemia, the leading cause of maternal anemia in LMICs (Chaparro & Suchdev, 2019). Moreover, the progress on anemia in WRA (15–49 years) is insufficient to meet the World Health Assembly's global nutrition target to reduce anemia prevalence by 50% by the year 2030 in LMICs, including Uganda (Stevens et al., 2022).

The recommended dietary allowance (RDA) is the average daily dietary intake level that is sufficient to meet the nutrient requirements of nearly all healthy individuals in a particular life stage and sex group. It is worth noting that the RDA for vitamin A among lactating mothers is higher at 885–1300 μgRAE/day compared to pregnant women at 700 μgRAE/day (Gannon et al., 2020; Kominiarek & Rajan, 2016). Moreover, the RDA for iron needed by lactating mothers is high at 10–30 mg/day compared to 8 mg/day for adult males (Ross et al., 2014; World Health Organization, 2016). To contribute toward achieving the RDA for iron among lactating mothers, iron supplementation programs are recommended by the World Health Organization (WHO) during the postpartum period, beginning just after neonate delivery to the first 6 weeks after delivery (World Health Organization, 2016). In contrast, the WHO does not recommend vitamin A supplementation during postpartum because it has neither health benefits to the mother nor the infant (World Health Organization, 2011). However, it recommends that the lactating mother should consume a diversified diet including vitamin A-rich foods (World Health Organization, 2011). It is worth noting that vitamin A and iron supplementation during postpartum have the potential to improve the breast milk content of vitamin A and iron, respectively (Holm et al., 2016; Oliveira et al., 2016). This is important because it improves the vitamin A intake of the breastfeeding child, and therefore may improve the vitamin A and iron status of the breastfeeding infant. However, postpartum vitamin A and iron supplementation programs are not sustainable because they are done for only the first 6 weeks after delivery, yet lactation should go up to 2 years and beyond as recommended by the WHO (World Health Organization & United Nations Children's Fund, 2021). Therefore, a continued intake of either vitamin A and iron-fortified foods or vitamin A and iron animal source foods such as meat and liver would be necessary to improve the vitamin A and iron status of lactating mothers and their breast milk (Mwangi et al., 2021). However, vitamin A and iron-fortified foods are either expensive or not available on the market to be accessed by poor mothers who reside in LMICs and rural settings.

A high proportion of lactating mothers residing in rural Uganda are at an increased risk of developing VAD and ID anemia because they habitually consume staple plant-source foods (PSFs) that are low in PVA carotenoids and iron (Nankumbi et al., 2023). The staple PSFs are usually either tubers such as conventional non-PVA biofortified sweet potato (Ipomoea batatas), white-fleshed sweet potato (WFSP), or low-iron conventional non-iron biofortified common bean (NIBCB) such as Nambale. Plant biofortification is a nutrition-sensitive strategy that increases the concentration of target nutrients, particularly micronutrients, such as iron and vitamin A in the edible portions of staple food crops through conventional breeding, fertilizer application, or bioengineering (recombinant DNA technology) to improve the nutritional status of target consumers (Vasconcelos et al., 2017). Compared to other nutrition-specific strategies such as industrial fortification and supplementation traditionally used to combat VAD and ID, biofortification is relatively cost-effective because once biofortified seeds are released, the rural poor from LMICs can plant them season after season for sustained consumption (Bouis & Saltzman, 2017; Glahn & Noh, 2021).

To contribute toward improved vitamin A and iron intake among vulnerable groups such as lactating mothers, the government of Uganda through HarvestPlus introduced PVA biofortification and iron biofortification programs for its staple foods such as sweet potato (Ipomoea batatas) and common bean (Phaseolus vulgaris), respectively (Atero, 2019; HarvestPlus, 2012; Hotz et al., 2012). Therefore, the PVA-biofortified sweet potato, orange-fleshed sweet potato (OFSP), and iron-biofortified common bean (IBCB) were introduced in Uganda. Introducing iron-biofortified foods and provitamin A-biofortified foods is increasingly seen as a potential strategy for combating ID and VAD in LMICs of Africa (Finkelstein et al., 2017; Govender et al., 2019b; Gurmu et al., 2014). It is worth noting that understanding the acceptability of these biofortified foods is a prerequisite for future consumer utilization of such biofortified foods (Huey et al., 2023; Siwela et al., 2020).

In Uganda, a dish prepared from a combination of WFSP and NIBCB is commonly consumed among WRA including lactating mothers (Uganda Bureau of Statistics & Inner City Fund, 2018). However, a combination of WFSP and NIBCB is low in PVA carotenoids and iron, respectively (Drapal & Fraser, 2019). When such a combination is habitually eaten, it has the potential to increase the risk of VAD and ID among Ugandan consumers (Kyamuhangire et al., 2013; Tidemann-Andersen et al., 2011). To contribute toward improving PVA carotenoid and iron intake among lactating mothers in rural Uganda, a PVA carotenoid–iron-rich innovative dish was prepared from a combination of OFSP and IBCB released in Uganda. This is plausible because the former is rich in PVA carotenoids (Neela & Fanta, 2019), and the latter is rich in iron (Luna et al., 2020). Therefore, this study evaluated the perceptions and sensory acceptability of a PVA carotenoid–iron-rich OFSP+IBCB (test composite foods) against the low PVA carotenoid–iron WFSP+NIBCB (control composite foods) among lactating mothers in rural Uganda. Findings generated from this study inform whether a dish of OFSP+IBCB rich in PVA carotenoids and iron has the potential to replace the conventional dish WFSP+NIBCB, which is low in PVA carotenoids and iron among Ugandan lactating mothers.

2 MATERIALS AND METHODS

2.5 Description of the control and test composite dishes used in the study

2.5.1 Ingredients for the preparation of the study composite dishes

Composite foods are characterized by being multi-ingredient and include both ready-to-eat products and home-prepared dishes (Durazzo et al., 2020). This study used home-based cooking methods commonly used to prepare common beans and sweet potatoes in Uganda. The control composite dish was WFSP served with NIBCB (WFSP+NIBCB). This combination was selected because it is a non-biofortified dish habitually consumed in Uganda and it is characterized by low PVA carotenoids and iron. The innovative test composite dish was OFSP served with IBCB (OFSP+IBCB). The test composite dish was selected because the former is biofortified with PVA carotenoids, while the latter is biofortified with iron. Figure 1 shows the ingredients used to prepare the control and test composite dishes for the acceptability study.

2.5.2 Preparation of test and control composite dishes

The village health team members (community health workers) identified five expert peer mothers from the local community and invited them to Bwera Hospital kitchen to participate in the preparation of the study composite dishes. Expert peer mothers were encouraged to prepare the composite dishes using locally acceptable home-based methods used in the community to prepare common beans and sweet potatoes to be consumed by lactating mothers. The biofortified foods, OFSP (NKB 135) and IBCB (NARO bean 4c) were procured from the National Crops Resources Research Institute (NaCRRI), Namulonge, Uganda (National Crops Resources Research Institute, 2023). The commonly consumed WFSP (Ebiribwa) and NIBCB (Nambale) in the community were procured from the local market by expert peer mothers who prepared the composite dishes used in the study in the presence of the research assistants. The peer mothers soaked the NIBCB and IBCB separately overnight (approximately 8–10 h) before cooking them by boiling. The mothers chose to boil common beans with prior soaking because they argued that soaking softens common beans to reduce cooking time and saves fuel as reported in a previous Ugandan study (Buzigi et al., 2020c). The IBCB and NIBCB were cooked separately using a local stove (Esigiri) until they were soft and ready for human consumption. The cooking time for NIBCB and IBCB was approximately 1.5 and 3.5 h, respectively. Expert peer mothers prepared OFSP and WFSP separately by peeling, and washing, followed by steaming. The local method of steaming involved putting 1 liter of tap water into the cooking pot, followed by banana stalks, then banana leaves, in which sweet potato was wrapped before being put onto the hot charcoal stove for steaming. The role of the banana stalk was to create an elevation and separation between the water in the cooking pot from the banana leaves, where the sweet potatoes were wrapped. To this end, when water in the pot boiled, it released steam that vaporized through the banana stalk spaces to heat the banana leaves in which the sweet potatoes were wrapped. The cooking time for WFSP and OFSP was approximately 45 and 35 min, respectively. The duration of boiling and steaming was sufficient to soften the plant tissue as assessed by the penetration of the tip of a knife without resistance (Buzigi et al., 2020c).

2.5.3 Provitamin A carotenoids, vitamin A, and iron composition of cooked study foods

The PVA carotenoid composition of cooked sweet potato was analyzed by high-performance liquid chromatography (HPLC) as described in the HarvestPlus handbook for carotenoid analysis (Rodriguez-Amaya & Kimura, 2004). The vitamin A composition was calculated by using the Institute of Medicine (2001) bioconversion rates of PVA carotenoids to vitamin A (retinol activity equivalents), that is, 12 μg of β-carotene is equivalent to 1 μg of retinol, while 24 μg of α-carotene is equivalent to 1 μg retinol (Institute of Medicine, 2001). The iron composition of the common bean was analyzed by flame atomic absorption spectroscopy as described elsewhere (Buzigi et al., 2020c). Triplicate analysis for each variety of cooked sweet potato and common bean was conducted to get an average of each micronutrient in each food. Table 1 shows the PVA carotenoid and iron composition of the cooked sweet potato and common bean, respectively, used in the study.

TABLE 1. Provitamin A carotenoid (retinol activity equivalent) and iron composition of study cooked sweet potato and common bean, respectively, used in the study.

Micronutrient	OFSP/100 g	WFSP/100 g	IBCB/100 g	NIBCB/100 g
β-carotene (μg)	36,900	9200	Not tested	Not tested
α-carotene (μg)	300	110	Not tested	Not tested
Vitamin A (μgRAE)	3088	711	Not tested	Not tested
Iron (mg)	Not tested	Not tested	9.3	5.1

• Note: The values are presented on a cooked (fresh) basis.
• Abbreviations: IBCB, iron-biofortified common bean; NIBCB, non-iron-biofortified common bean; OFSP, orange-fleshed sweet potato; RAE, retinol activity equivalent (retinol); RAE, β-carotene (μg/100 g)/12+ α-carotene (μg/100 g)/24 (Institute of Medicine, 2001); WFSP, white-fleshed sweet potato.

2.6 Measurement of study variables

2.6.1 Measurement of sensory acceptability

Sensory acceptability was measured by the sensory evaluation method. The sensory evaluation sessions were held in a room with separate, isolated cubicles set up for each panelist. Each panelist received a spoon and a small polystyrene cup containing the control composite dish, approximately 50 g of NIBCB combined with 50 g WFSP (WFSP+NIBCB), and the test composite dish, approximately 50 g of IBCB combined with 50 g of OFSP (OFSP+IBCB). To prevent lactating mothers from making judgments based on labels, the samples were given random numbers as labels. The purpose of this was to ensure that panelists relied solely on their sensory experiences to assess the samples (Lawless & Heymann, 2010). To achieve this, the samples were randomly labeled with a unique three-digit code obtained from a table of random numbers and were served in a random sequence (Lawless & Heymann, 2010). The samples were warmed in a microwave oven for 10 seconds on medium heat, before serving. The lactating mothers were provided with a cup of water to rinse their palates between evaluating samples. Before each session, the sensory attributes of color, texture, aroma, taste, and overall acceptance were explained to the panelists (lactating mothers). An equally spaced five-point facial hedonic scale with ratings (1=“dislike very much”; 2 = “dislike”; 3=“neither like nor dislike”; 4 = “like”; 5= “like very much”) was used. It is worth noting that the WRA in the study district have lower literacy levels, with 67% of children dropping out of school before completing ordinary-level education (National Population Council, 2021). This coupled with low literacy levels established during the pilot study justifies the use of the five-point rating scale in the study participants as recommended by Stone and Sidel (2004). Each sensory attribute was described on the form with an accompanying facial hedonic scale. The ratings of the hedonic scale were verbally explained to the panelists in the local language during the sensory evaluation sessions. The participants were asked to rate the acceptance of each attribute by marking the appropriate response on the facial hedonic scale. A sensory attribute was considered acceptable if it was rated as “like to like very much.” Figure 2 shows the flow of sensory acceptability of the study composite dishes.

2.6.2 Measurement of perceptions

Qualitative data (perceptions) were collected using FGDs and a FGD guide. The FGDs were conducted to explore the lactating mothers' perceptions to understand the factors that may motivate them to consume the test composite dish (OFSP+IBCB). The FGDs were conducted about 30 min after the sensory evaluation study was completed. All the 94 lactating mothers who participated in the sensory evaluation study were eligible to participate in the FGDs. Through established community relationships, four facilitators, experienced in conducting FGDs in the local language (Lhukonzo), were recruited for 2 days of training in focus group discussion moderation concerning the specifics of this research study following guidelines explained elsewhere (Amico et al., 2011). A trained facilitator directed the discussions, using an unstructured FGD guide. The FGD guide consisted of a brief explanation of the samples that were tasted during the sensory evaluation as well as the question for initiating and facilitating the discussion. Guidelines for conducting FGDs with a structured set of open-ended questions were followed, as recommended elsewhere (Krueger & Casey, 2015). The open-ended question was “You have participated in the sensory evaluation of OFSP combined with IBCB. What would motivate you to consume the IBCB+OFSP composite dish?” Furthermore, the facilitators used probes such as “Would you explain further?” and “Would you give an example?” where it was deemed necessary. The FGDs were facilitated in the local language (Lhukonzo) by trained FGD facilitators. A digital voice recorder was used to record the FGDs after participants consented to the use of the voice recorder. A sample size for an FGD between 7 and 12 participants is appropriate for a qualitative nutrition-related study (Harris et al., 2009). Therefore, each focus group included 8–10 participants. The FGDs were conducted until data saturation was reached. Data saturation means that researchers are not finding any additional or new data from the FGDs (Hennink & Kaiser, 2022).

3 RESULTS

4 DISCUSSION

This study explored lactating mothers' perceptions and sensory acceptability of a PVA carotenoid–iron-rich composite dish prepared from a combination of OFSP and IBCB. A combination of WFSP and conventional NIBCB is commonly consumed in Uganda. However, this combination is low in PVA carotenoids and iron. The retinol activity equivalent (RAE) + iron composition in OFSP+ IBCB (test food) and WFSP+NIBCB (control food) used in this study was 3088 and 711 μgRAE+5.1 mg/100 g, respectively (Table 1). The RDA for iron and retinol for lactating women is 9.5 mg/day and 1300 μgRAE, respectively (Gannon et al., 2020; Kominiarek & Rajan, 2016; Ross et al., 2014). Therefore, approximately 100 g of IBCB and only 42 g from OFSP+IBCB would be needed to meet the RDA requirements for iron and retinol required for lactating mothers. In contrast, nearly 200 g of NIBCB and 200 g of WFSP from WFSP+NIBCB would be needed to meet the RDA requirements for iron and retinol required for lactating mothers. This suggests that compared to the WFSP+NIBCB, a smaller portion of OFSP+IBCB is needed to meet RDA requirements for iron and retinol for lactating mothers.

To our knowledge, this is the first attempt to understand the maternal acceptability of homemade composite dishes prepared from foods biofortified with PVA carotenoid and iron, versus the low iron-PVAC dishes prepared from conventional foods in Uganda. In the present study, the color of the IBCB+ OFSP dish was highly significantly accepted compared to that of the NIBCB+ OFSP dish, probably because of the deep orange color of the OFSP, which was attractive to the lactating mothers as they reported in the FGDs. In contrast, the lactating mothers seemed not to recognize any color difference between IBCB and NIBCB, as they emphasized during FGDs that every sensory attribute was similar between IBCB and NIBCB. Therefore, the color acceptability of OFSP+IBCB could have between attributed to the color of OFSP, but not IBCB. This finding is similar to the previous studies which reported that PVA-biofortified foods such as OFSP and either yellow maize or cassava were acceptable to the potential consumers because of the attractive deep orange color in the former and yellow color in the latter (Amod et al., 2016; Govender et al., 2014, 2019a; Talsma et al., 2013). It is worth noting that the orange and yellow colors observed in PVA-biofortified foods including OFSP are due to the increased PVA carotenoids added to these staples during biofortification (Yuan et al., 2015). Findings from this present study confirm that the yellow or orange color observed in PVA-biofortified foods improves the acceptability of other foods if they are combined with other foods with less attractive colors. In South Africa, Govender and colleagues demonstrated that because of the appealing color of OFSP and yellow PVA biofortified maize, consumers were accepting and had positive perceptions of non-biofortified foods, if they were combined with either OFSP or PVA-biofortified maize (Govender et al., 2019a). However, it is worth noting that the present study used a test composite dish prepared from a combination of biofortified foods including OFSP and IBCB.

During FGDs, mothers raised a suggestion of the likelihood of their children's acceptability of the complementary food prepared from the common bean and OFSP because of the attractive color they observed in OFSP. Such suggestions from lactating mothers confirm that lactating mothers are gatekeepers of complementary foods (CFs), they feed to their children (Buzigi et al., 2020a; Food and Agriculture of the United Nations, 2017). Moreover, previous studies have shown that, based on the appealing yellow or orange color observed in PVA biofortified foods, caregivers have shown interest in preparing CFs using PVA-biofortified foods (Amod et al., 2016; Govender et al., 2014). The lactating mothers' suggestions may also indicate the necessity of developing an innovative complementary food blend rich in PVA carotenoids and iron from OFSP+IBCB and testing its acceptability among their children. The lactating mothers' suggestion is plausible because 6- to 23-month-old children in low-income countries such as Uganda are at a higher risk of developing ID and VAD since they are predominantly fed CFs that are low in iron and vitamin A (Amaral et al., 2018; Ekesa et al., 2019; Uganda Bureau of Statistics & Inner City Fund, 2018). Moreover, developing and testing child acceptability of PVA carotenoid–iron-rich CFs and testing its acceptability among children in the age range of complementary feeding is feasible (Buzigi et al., 2020b).

The texture of IBCB+OFSP was not acceptable to the lactating mothers compared to that of NIBCB and WFSP (p = .02). The unacceptability of the IBCB+OFSP dish was linked to the soft texture of the OFSP but not IBCB, as study participants indicated during the FGDs that the sensory properties of IBCB and NIBCB were similar. This finding is not surprising because the majority of OFSP varieties released in low- and middle-income countries including Uganda have high moisture content, compared to the WFSP (Alam et al., 2016; Neela & Fanta, 2019), hence their softness. The softness reported by lactating mothers in the present study indicates the necessity of increasing the dry matter content of OFSP. Moreover, it is feasible to increase the dry matter content of OFSP (Ginting et al., 2017). In contrast, during the FGDs, lactating mothers used the softness of OFSP as a justification for the potential use of OFSP as a complementary food for their children. This justification is plausible because it is recommended that CFs should be soft to prevent choking in children (Food and Agriculture of the United Nations, 2017).

Aroma, taste, and general acceptability were scored highly by the lactating mothers and not significantly different between OFSP+IBCB and WFSP+NIBCB. Such findings indicate the potential of replacing conventional WFSP+NIBCB with OFSP+IBCB. Similar findings have been reported in South Africa (Govender et al., 2019a). However, the South African study did not have a composite dish having a combination of IBCB and OFSP as it only focused on the traditional composite dishes prepared with either OFSP or PVA-biofortified maize (Govender et al., 2019a). Moreover, the South African study recruited adult males and females (Govender et al., 2019a), compared to this present study, which recruited lactating mothers, a target group with high physiological demands for iron and vitamin A (Gannon et al., 2020; Khayat et al., 2017).

During FGDs, some mothers showed interest in knowing where they would access the vines and seeds for OFSP and IBCB, respectively, so that they could grow them for their home consumption, and even sell the excess to improve their incomes. These findings indicate the possibility of adopting the use of study biofortified foods in the study participants. Similar findings of the potential to adopt the growing OFSP have been reported elsewhere in Eastern Uganda (Ssebuliba et al., 2006). Moreover, the adoption of IBCB has the potential to improve household consumption of IBCB (Vaiknoras & Larochelle, 2021).

Considering the feasibility of preparing OFSP and IBCB, mothers were worried about the long cooking time for common beans and the high fuel costs associated with cooking common beans. This finding is consistent with previous studies conducted in Uganda, which showed that consumers were afraid to use common beans because of their long cooking time and high fuel consumption (Buzigi et al., 2020a), or they preferred common bean processing methods that reduce the cooking time of common bean and hence save fuel (Aseete et al., 2018). It is worth noting that the cooking time for NIBCB and IBCB used in this present study during the preparation of the study composite dishes was approximately 1.5 and 3.5 h, respectively. Such a high cooking time observed for IBCB compared to NIBCB would prevent the lactating mothers from cooking IBCB for household consumption. However, the peer mothers who prepared the study composite dishes first soaked common beans before cooking them. This was plausible because cooking IBCB with prior soaking may reduce the cooking time of common beans (Muroki et al., 2023).

During FGDs, lactating mothers indicated that they would consume OFSP+IBCB if they knew that they were nutritious or if eating them would provide any health benefits. For example, one study participant emphasized that she would consume a combination of OFSP and IBCB because she knew from a previous nutrition education program, that the OFSP provided health benefits to the eye. This finding informs that there is a need to promote the consumption of biofortified foods such as OFSP and IBCB through community nutrition education. The education messages should focus on the nutritional value of biofortified foods and the health benefits of their consumption. It is worth noting that previous studies have shown that the sensory acceptability and willingness to pay for biofortified foods may be improved through nutrition and health promotion program activities such as providing information, education, and communication about the nutritional and health benefits of eating biofortified food crops such as biofortified yellow cassava and IBCB (Bechoff et al., 2018; Oparinde, Banerji, et al., 2016; Oparinde, Birol, et al., 2016). For example, in Rwanda, the sensory acceptability of IBCB was higher among consumers who had prior sensitization and awareness about the nutritional benefits of consuming IBCB (Murekezi et al., 2017).

5 CONCLUSIONS

The PVA carotenoid-rich sweet potato, OFSP served with a high iron common bean, IBCB (OFSP+IBCB) was significantly accepted for color, but not texture, compared to the low PVA conventional sweet potato, WFSP served with the low iron conventional common bean, NIBCB (WFSP+NIBCB). However, for overall acceptability, OFSP+IBCB and WFSP+ NIBCB were equally acceptable, suggesting that the biofortified, OFSP+IBCB dish has the potential to replace the staple non-biofortified, WFSP+NIBCB dish among lactating mothers in the study area. Lactating mothers had positive perceptions of using OFSP and IBCB if they could cook fast, and provide health benefits upon consumption, with a stable availability and economic access in the food supply chain.

AUTHOR CONTRIBUTIONS

Edward Buzigi: Conceptualization (lead); data curation (lead); formal analysis (lead); investigation (lead); methodology (lead); project administration (equal); resources (equal); software (equal); supervision (equal); validation (equal); visualization (equal); writing – original draft (lead); writing – review and editing (lead). Kirthee Pillay: Conceptualization (equal); investigation (equal); methodology (equal); supervision (equal); validation (equal); visualization (equal); writing – review and editing (equal). Muthulisi Siwela: Conceptualization (equal); data curation (equal); methodology (equal); resources (equal); supervision (equal); writing – review and editing (equal). Blessing Mkhwanazi: Conceptualization (equal); funding acquisition (equal); methodology (equal); validation (equal); visualization (equal); writing – review and editing (equal). Mjabuliseni Ngidi: Conceptualization (equal); funding acquisition (equal); methodology (equal); supervision (equal); visualization (equal); writing – review and editing (equal). Babra Muhindo Mahinda: Conceptualization (equal); investigation (equal); methodology (equal); resources (equal); visualization (equal); writing – review and editing (equal). Joshua Ssozi: Conceptualization (equal); methodology (equal); writing – review and editing (equal). Evyline Barugahara Isingoma: Methodology (equal); writing – review and editing (equal).

FUNDING INFORMATION

This research was funded through the postdoctoral scholarship awarded to Edward Buzigi in the School of Agricultural, Earth, and Environmental Science within the College of Agriculture, Engineering, and Science at the University of Kwa-Zulu Natal, South Africa.

CONFLICT OF INTEREST STATEMENT

The authors declare that they have no competing interests.

ETHICS APPROVAL AND CONSENT TO PARTICIPATE

The study was performed following the ethical standards as laid down in the 1964 Declaration of Helsinki. Ethical approval was granted by the Research Ethical Committee at The AIDS Support Organisation Research Ethical Committee (Reference number TASO-2023-252). Informed and signed consent were obtained individually from the mothers in the postnatal clinicians who participated in the study. Informed consent was taken from legally authorized representatives and/or guardians of all participants who were below 18 years old and those without formal education.