Maternal and health care workers’ perspectives on exclusive breastfeeding in the context of maternal HIV infection, in Busia county, western Kenya: a mixed methods cross-sectional survey

Background: World Health Organization recommends exclusive breastfeeding (EBF) for 6 months with maternal active antiretroviral therapy (ART) to prevent mother-to-child transmission (PMTCT) of HIV. However, EBF in low resource settings remains low. We explored perspectives of EBF by HIV-infected mothers and health care workers in Busia County with a high prevalence of HIV to understand factors influencing the practice. Methods: A mixed methods cross-sectional survey using concurrent quantitative and qualitative data collection methods was conducted at PMTCT clinics. Data on socio-demography, young infant feeding practices, maternal and infant health was collected between February 2013 and August 2015 from 371 purposively sampled HIV-infected mother-infant dyads using a semi-structured questionnaire. Focus group discussions with mothers, in-depth interviews and passive observation of health care workers during interaction with mothers were conducted. Significance of difference between mothers practicing EBF or not was tested by Chi-square and Fisher’s exact tests setting significance level at 5%. Qualitative data was coded and content analyzed to generate themes. Results: Three hundred and forty-nine (94%) mothers practiced EBF. Maternal comprehension of EBF to PMTCT of HIV influenced choice and practice of EBF (P value = 0.019 and < 0.001 respectively). Health care workers emphasized adherence to ART and offered nutritional supplementation during EBF. Health care workers’ nutritional counseling in the context of maternal HIV was poor. Mentor mothers shared their experiences with mothers and offered live case demonstrations of their successfully EBF, healthy and HIV-uninfected children. The main threats to EBF were teenage motherhood, low maternal education and working during EBF. Conclusions: EBF among HIV-infected mothers in Busia County, Kenya was high. Health education and counselling by health care workers, maternal comprehension of ART adherence to PMTCT of HIV, nutritional supplementation and mentor mothers’ peer counseling using live case demonstrations of HIV-uninfected EBF children promoted and sustained practice of EBF for 6 months. Teenage motherhood, low maternal education and having to work threatened EBF.

The study was conducted in Busia County, western Kenya where infants are at high risk of MTCT of HIV due to higher prevalence of HIV estimated at approximately 10%, compared with the national average of about 7% [25]. The setting was PMTCT of HIV clinics in 3 hospitals and 7 health centers supported by United States Urgency for International Development and Academic Model Providing Access to Health Care Project. An average of 8–10 and 5–7 mother infant-dyads among other patients are seen at each of the hospitals and health centers respectively. All pregnant and breastfeeding mothers receive active ART regardless of the CD4 count. Breastfeeding infants are given prophylactic ARVs to prevent MTCT of HIV. Monitoring of infant growth and adherence to active ART is done monthly. Nutritional support to the pregnant and breastfeeding HIV-infected mothers is in the form of vitamins, hematinic and 9 kg monthly rations of corn soy blend flour for porridge which is supplied by World Food Program. Psychosocial support is offered by health care workers and peer social support groups where mothers disclose their HIV status and share experiences. A mixed methods cross-sectional survey applying concurrent qualitative and quantitative data collection methods was conducted from February 2013 to August 2015. To explore maternal and health care workers’ perspectives on EBF during the first 6 months in the context of maternal HIV infection in order to understand factors positively and negatively influencing the practice. Mothers with infants aged 6 weeks to 6 months and health care workers at PMTCT of HIV clinics in Busia, Kenya. Inclusion Criteria: Exclusion Criteria: Fischer’s formula was used to estimate the minimum sample size of mothers required for the questionnaire based on the prevalence of EBF among HIV-infected mothers of 35% in Kitale district hospital, western Kenya [26]. This is because there was no data on exclusive breastfeeding among HIV-infected mothers in Busia County and that Kitale District hospital was the nearest hospital where a study on infant feeding practices among HIV infected women receiving PMTC services had been conducted. A minimum sample of 349 mother-infant dyads would allow estimation of the proportion of mothers who EBF with a precision of 5%. USAID-AMPATH-Plus Project facilities which included 1 county hospital, 2 sub-county hospitals and 7 health centers were the sampling frame from which HIV-infected mothers were sampled. Three hundred and seventy-one 15–45 year old mothers with 6 month old infants attending PMTCT clinics were sampled purposively by the principal investigator and clinical officers at PMTCT clinics and consecutively recruited in the study to enable complete data set for analysis. Thirty two health care workers at PMTCT of HIV clinics for at least 3 months during the study period were purposively sampled for in-depth interview. Eight focus group discussions (FGDs) were conducted till the point of saturation when no new information emanated from the discussions. A questionnaire was used in order to maintain consistency of the questions posed to the mothers and hence validity of the responses obtained. Focus group discussions (FGDs) were used to identify the experiences and perspectives of the mothers on factors influencing EBF. They aided to uncover unique perspectives on EBF due to the group environment in which data are collected [27]. FGDs prompted members to give more detailed account of issues influencing EBF among them and other mothers in the community. In-depth interviews were used to get the rare experiences of mothers practicing EBF during interaction with the health workers [27]. A questionnaire developed by the researcher in English was translated into Kiswahili – a language which subjects in the setting understood. This was back translated into English by a second person experienced in translation of Kiswahili to English in order to confirm consistency of the meaning. This was validated by pilot testing on a sample of 30 HIV-infected mothers with infants not more than 6 months of age. This facilitated necessary corrections to ensure that the subjects understood the questions in order to ensure consistency of the responses and hence reliability of the data generated. Interviewer-administered structured questionnaires with closed and open ended questions were used to collect quantitative and qualitative data from 371 mothers on demographic characteristics and advice on EBF during pregnancy and postnatal periods till infants attained the age of 6 months. The infant’s exact age was derived from the birthday indicated in the well-baby clinic booklet. Using maternal recall, information on morbidity from common infant illnesses such as upper respiratory tract infections, pneumonia, diarrhea and malaria since birth was collected to determine the health status of the infants during the first 6 months of life based on the feeding option. FGDs with the mothers, in-depth interviews and passive observations of health care workers were conducted to triangulate this data. Mothers with 6 weeks-6 months old infants not involved in the survey using the questionnaire were eligible for FGDs. They were booked to come to the clinic on a convenient date when at least 8–12 mothers would be available for FGD. Eight FGDs using a topic guide were conducted at PMTCT of HIV clinics to understand the perspectives of mothers on EBF. The focus of the FGD was to establish whether HIV-infected mothers understood the meaning of EBF; which mothers were likely or unlikely to practice EBF and why; factors that influence EBF, support received to ensure that their infants were EBF and the influence of health care workers on choice of infant feeding. This was conducted by the researcher and an assistant moderator for 1 -1·5 h using a pretested FGD topic guide. The FGD were conducted and audio recorded at a private spot away from other mothers and health care workers in order to understand one another and for assurance of confidentiality. Thirty two health care workers at PMTCT of HIV clinics were interviewed and audio recorded for 1–1·5 h by the principal investigator and an assistant moderator using an interview guide on knowledge and perspectives of EBF in the context of maternal HIV infection. In-depth interview guide focused on the experience of health care workers during their interaction with HIV-infected mothers with infants from birth till the age of 6 months; which mothers are most or least likely to practice exclusive breastfeeding; which factors have positive and negative influence on EBF; what support was offered to ensure that most infants were fed according to the latest WHO recommendation; how health workers influence the choice of feeding babies; what they would recommend to promote, support and sustain exclusive breastfeeding. The interviews were continued until no new themes emerged from the data. Thirty two health care workers were observed during their interaction with the mothers by passive participation of the researcher. The aim was to assess their practice and accuracy of counselling information on infant feeding given to HIV-infected mothers and find out whether they perform nutritional counseling or not. Their knowledge and skills on young infant feeding counseling were assessed and scored against a check list by the researcher using a counseling flow chart derived from HIV and infant feeding counseling tools based on United Nations policies and guidelines [28]. These state that all HIV infected mothers should receive counseling which includes provision of general information about the risks and benefits of various infant feeding options and specific guidance in selecting the option most likely to be suitable for their situation. Whatever the mother decides she should be supported in her choice. The aspects that were observed were 15 in total. The score was carried out after the health care workers had interacted with several mothers and forgot that they were being watched. Each aspect of knowledge and skill was assigned a score of 1 and a percentage out of a total score of 15 computed. A score less than 50% was rated poor, 51 – 64% pass, 65 – 74% credit and 75 – 100% distinction. The National Council of Science and Technology through Moi University and Moi Teaching and Referral Hospital Research and Ethics Committee approved the study – Formal Approval number: FAN: IREC 000,916. Permission was sought from the Director of USAID-AMPATH-Plus Project and the Medical Officer of Health of Busia County. Informed written consent was sought from health care workers and mothers. Principles of confidentiality, respect for persons, beneficence and justice were adhered to during the study period. All mother-infant dyads received standard health care according to WHO and USAID-AMPATH-Plus Project guidelines. Quantitative data was cleaned, coded and entered in IBM SPSS version 20 for storage and analysis by the principal investigator with the assistance of a biostatistician. Descriptive statistics were used to summarize the data. Chi-square test and Fisher’s Exact tests were used to determine the differences between infants and mothers subjected to EBF or not. A 5% level of statistical significance was used. Qualitative data was audio recorded, transcribed, labeled and content analysis done to generate themes.

N/A

• Dima Health