Background: Implementation fidelity which is defined as the degree to which programmes are implemented as intended is one of the factors that affect programme outcome, thus requiring careful examination. This study aims to acquire insight into the degree to which nutritional counselling and Iron and Folic Acid supplementation (IFAs) policy guidelines during pregnancy have been implemented as intended and the challenges to implementation fidelity. Methods: Data were collected in rural Uasin Gishu County in the western part of Kenya through document analysis, questionnaires among intervention recipients (n = 188) and semi-structured interviews with programme implementers (n = 6). Data collection and analysis were guided by an implementation fidelity framework. We specifically evaluated adherence to intervention design (content, frequency, duration and coverage), exposure or dosage, quality of delivery and participant responsiveness. Results: Coverage of nutritional counselling and IFAs policy is widespread. However, partial provision was reported in all the intervention components. Only 10% accessed intervention within the first trimester as recommended by policy guidelines, only 28% reported receiving nutritional counselling, only 18 and 15% of the respondents received 90 or more iron and folic acid pills respectively during their entire pregnancy period, and 66% completed taking the IFAs pills that were issued to them. Late initial bookings to antenatal care, drug stock shortage, staff shortage and long queues, confusing dosage instructions, side effects of the pills and issuing of many pills at one go, were established to be the main challenges to effective implementation fidelity. Anticipated health consequences and emphasis by the health officer to comply with instructions were established to be motivations for adherence to nutritional counselling and IFAs guidelines. Conclusions: Implementation fidelity of nutritional counselling and IFAs policy in Kenya is generally weak. There is need for approaches to enhance early access to interventions, enhance stock availability, provide mitigation measures for the side effects, as well as intensify nutritional counselling to promote the consumption of micronutrient-rich food sources available in the local environment to substitute for the shortage of nutritional supplements and low compliance to IFAs.
We adopted ‘programme theory’ as the conceptual frame work to meet the main objective of this study. Programme theory as defined by Bickman is the construction of a plausible model of how a programme is supposed to work [26]. It involves the construction of a causal model linking programme inputs and activities to a chain of intended or observed outcomes and then using this model to guide the evaluation. There is no uniform way of developing such models because each is developed for a particular programme and does not represent the “off-the-shelf” use of a single established social science theory [26]. We used the program theory of MIYCN and next designed a programme implementation fidelity framework as presented in Fig. 1 to guide the evaluation. ‘Implementation fidelity’ (also termed programme integrity) is defined as the degree to which programmes are implemented as intended [25, 27]. Programme theory of process pathways to maternal nutrition intervention in Kenya To assess implementation fidelity, various main dimensions can be identified in literature, which Caroll et al. [28] grouped into two components: For this study, the conceptual interpretation of the programme theory and two core components of implementation fidelity – adherence and moderating variables – was guided by MIYCN implementation guidelines and messages as indicated in Fig. Fig.1.1. This framework became a guiding tool in the design of research instruments, data analysis and interpretation of the results as further described in “data collection” section. This institution-based descriptive cross-sectional survey used qualitative and quantitative methods of data collection. Researcher administered questionnaire was used to collect data from intervention recipients and a semi-structured interview guide was used to collect data from programme implementers. The study was conducted between March and June 2017 in rural Uasin Gishu County in the western part of Kenya (one of the 47 counties of Kenya). Uasin Gishu is predominantly inhabited by the Kalenjin who are the third largest ethnic community out of the 44 ethnic communities in the country [29]. The predominant settlement pattern in Uasin Gishu County is rural (64.1%) [29] and malnutrition in Kenya is highly prevalent in the rural areas [1, 2], hence the reason why this study was rural based. Data for this manuscript is part of a larger research project whose main objective was to investigate the social cultural context of nutrition in pregnancy and the utilization of nutrition intervention services in rural Uasin Gishu County [30]. All major health facilities (sub-County hospitals) found in the rural areas of Uasin Gishu County – six in total – were selected for the study [31]. These health facilities normally serve as rural referral hospitals. Sub-County hospitals as a result receive the highest rural patient/client population from various locations in a given sub-County hence the reason why they were selected for the study. Respondents were recruited amongst the mothers seeking maternal care at the six sub-County hospitals of Uasin Gishu County. Documenting women’s own experiences and perceptions on nutrition interventions and ANC services provided during their latest pregnancy was part of the main objectives of this study. As a result pregnant mothers who had at least one previous ANC services at a health facility and mothers who had delivered a baby within 1 month preceding the study were included in the study. To enhance cultural homogeneity, only Kalenjin mothers (the predominant cultural group in Uasin Gishu County) were included in the study. These selection criteria excluded: non-Kalenjin women, Kalenjin women seeking ANC for the first time or postnatal care past 1 month of delivery, as well as those unable or unwilling to participate. Systematic sampling technique was used to select study participants whereby every second woman who met the inclusion criteria was recruited until the minimum desired sample size of 188 out of on average total of 795 women who were seen per month in these hospitals, was attained. Maternal care attendance registration database of 6 months preceding this study were used to determine the average number of women seeking maternal care per month in these health facilities. Researcher administered questionnaire, containing closed and open-ended questions, was used to collect quantitative and qualitative data, respectively, from the sampled respondents (intervention recipients). The conceptual framework guided the design of research instruments and data collection approaches. Prior to the study, a detailed literature search was conducted to understand programme theory guidelines and intervention strategies that are used in the implementation process. The framework was operationalized as follows. The questionnaires were administered by the first author and trained research assistants to all mothers who met the inclusion criteria and consented to the study. The four research assistants were recruited with a background in social sciences, specifically degree in sociology. Before the actual data collection, 2 days theory and practical training regarding the objectives of the study and ways of administering the questionnaire were given to the data collectors by the first author. The prepared questionnaire was pre-tested prior to the actual data collection amongst ten pregnant women and eight post-natal mothers (n = 18) in one of the health facilities that was not included in the main survey. The questionnaire was edited and checked for completeness daily and before the data entry. All the research team was engaged in the pre-test and review of research instruments. The first part of the questionnaire largely contained close-ended questions. This part assessed the socioeconomic and demographic status of the respondents. It also assessed adherence, i.e. coverage, content, frequency and duration of the interventions, as well as anaemia status of the respondents. This information was retrieved from the respondent’s appointment card and the results were validated by the respondents. The second section of the questionnaire largely contained open-ended questions and it mainly assessed the quality of delivery and participant responsiveness. Interviews were conducted in a room within the health facility, ensuring privacy. Detailed note taking and tape recording was done during the interviews which lasted for about 30–45 min. Respondents were encouraged to talk freely and clarifying questions were only asked when the women did not understand the questions or when the interviewers wanted to get insight story of the responses. The women were interviewed in Swahili or in their local language. The interviews were later translated and transcribed into English by one of the research assistants who understands both languages and validated by a professional translator. Semi-structured interviews (SSIs) with programme implementers (nursing officers in charge of Maternal and Child Health (MCH) or whoever was on duty in the MCH section) were conducted to triangulate the research findings obtained from intervention recipients and document analysis. We specifically acquired information on programme theory and implementation as well as the challenges they face in the implementation of the MYICN guidelines. In total there were six programme implementers selected from each of the six sub-County hospitals where the study was conducted. The health workers were interviewed by the first author in the language of their preference (English or Swahili). The interviews were conducted in a quiet private room at their place of work to avoid distractions, ensure privacy and anonymity of the responses [32]. Information gathered from the SSIs was recorded, transcribed and further manually analysed to explore meanings and enrich the responses obtained from the interviews with mothers. Parts of these narratives have been presented in the results section. Collected statistical data were edited, coded, entered in Excel and exported to SPSS version 23. These were analysed to establish descriptive statistics such as frequency, percentage and mean which were used to describe studied variables. The first author performed the data coding and entry. Qualitative data were transcribed verbatim into Microsoft Word files and translated into English with each participant being identified with a pseudonym. Thereafter the transcripts together with field notes were transferred into MAXQDA 12.3.2 software for coding, analysis, and identification of major themes and sub-themes. Codes were based on main themes derived from the conceptual framework presented in Fig. Fig.11 as the initial coding guide. The data were further coded based on recurring themes as identified in the transcripts of SSIs and the recurring issues raised by participants derived from the open-ended questions in the questionnaire. Thematic analysis was used to identify the most common recurring themes and issues [32, 33]. Several researchers with different backgrounds provided input in the analysis to increase its validity [34].
N/A