Effectiveness of Positive Deviance Approach in Promoting Appropriate Breastfeeding Practice in Jimma, Southwest Ethiopia: A Cluster Randomized Controlled Trial

Abstract

II Abstract Background: Breast feeding (BF) is one of the most important postpartum care to the newborn to survive which should be initiated within an hour of birth and continue exclusively in the first six months. However, suboptimal BF contributes to a significant number of infant deaths in developing countries. Positive Deviance Approach (PDA) was not tested and studied so far, whether it improves the poor practices of Exclusive Breast feeding (EBF) and Timely Initiation of Breastfeeding (TIBF) in Jimma town. Along with, understanding the end-user satisfaction and its drivers was essential to determine whether this approach will be viable in the long run. In a certain intervention, measuring implementation outcomes (the effects of deliberate actions) to explain the process and desired outcomes is paramount. In this regard, to what extent the participants’ perceive/rate these outcomes were not known. Moreover, these psychometric outcome measures lack conceptual clarity, reliability and validity to the context.Very importantly; this behavioral intervention was not evaluated for its usefulness. As a result, we designed a study with the following objectives; (1) determine the effectiveness of PDA in increasing the proportion of EBF practice, (2) determine the effectiveness of PDA in increasing the proportion of TIBF, survival probability of timely initiation and its pooled predictors, (3) determine the level of end-user satisfaction with PDA intervention and identify multi-level predictors to explain variability, (4) measure implementation outcomes, determine valid and reliable scales and analyzing their correlation and variation, and (5) evaluate the cost- effectiveness of PDA in promoting EBF. Methods: A cluster-randomized controlled trial was employed in Jimma town from February 01 to September 30, 2018. Six randomly selected clusters(Kebeles) were randomized into two arms. Then 260 pregnant women who met the eligibility criteria were enrolled either into the intervention or control arm depending on where they lived. Women in the intervention group received informational counseling and social support in addition to the usual service to promote EBF, from women-identified and trained as positive deviants in their community, while those in the control groups received the usual community-based services from urban health extension professionals. Data on primary and secondary outcome variables were collected at three points, and statistical difference was estimated using Chi-X2 or Fisher exact test. The net effect of the intervention was determined. The magnitude of the intervention effect was estimated using the relative risk (RR). General Linear Model (GLM) for repeated measures, while log-binomial regression-Generalized Estimating Equations (GEE) for a point measure considering binary outcomes were used. The baseline, midline and end-line data were analyzed to see the effect on EBF, while only the midline data were used for TIBF. The survival data were summarized using estimates of survival functions with the Log-Rank test and compared by Kaplan-Meier (KM) curves. The predictors of time to initiate BF were estimated by Cox-regression model and interpreted using adjusting hazard ratio. The data for the end-users’ satisfaction with and Implementation Outcomes (IOs) of PDA were collected between September 01-25, 2020, and from August 28, 2020 to September 30, 2020 respectively. The study participants for the satisfaction study were mothers and their relevant others (n=260 end-users) who are living in the intervention clusters, while intervention implementation participants/stakeholders (n=384) were for the IOs. End-users were invited to assess individual and community-level factors and their satisfaction. The validity and reliability of the satisfaction measurement scales was checked. The satisfaction mean scores were standardized using Percentages Scale Mean Score (PSMS) formulae. Two-level mixed-effects linear regression models were performed to fit individual, community, and mixed-level variables. III As a follow-up study, implementation participants were assessed for their perception toward implementation outcomes of PDA. The developed tool was validated. The determined mean scores indicates to what extent each scales were rated. Pearson’s correlation, independent sample t-test or One Way ANOVA was used to see the relationship among these outcomes and mean variabilities against selected background characteristics. The explained variation of scales was checked using coefficient of determination (r2). A cost-effectiveness analysis was done considering the two trial arms such as intervention (n=130) and control (n=130). The design for this study was a mixed of alongside a cluster randomized controlled trial (cRCT) with decision models built on common causes of childhood morbidity and mortality, and maternal Health Related Quality of Life (HRQoL) for the first 59 and 6 months respectively. Checklist was used to collect input/cost data for both control (routine care cost) and intervention costs. Infant outcomes were analyzed from the literature and our trial report, while maternal HRQoL measured using an adapted EuroQol (EQ-5D) tool. Those costs and outcomes were compared between the arms. The main outcome measures were the intervention cost, Disability Adjusted Life Years (DALYs) for infants and maternal Quality Adjusted Life Years (QALYs). Data were analyzed using Excel-spread sheet 2013, and SPSS 21.0. Independent samples t-test and One Way ANOVA were used to compare the mean scores, and mixed-effects linear regression model was fitted to determine the effect size. In analyzing data for all of these studies, assumptions were checked as appropriate. For every statistical tests, 95% CI with a P-value of < 0.05 was used. Results: While analyzing the data to evaluate the effectiveness of PDA on EBF and TIBF practices, EBF was significantly different between the groups at mid and end points, while no difference at baseline. A higher proportion of mothers in the intervention group reported EBF compared to the control group at mid and end-line. The rate of EBF increased by 18.5% (P=0.01) in the intervention group while 0.2% in the counterparts; with a net effect of +18.2%. The probability of practicing EBF was significantly higher for the intervention group compared to the control group. At baseline, the relative risk of avoiding EBF was similar among the two groups. However, at follow-up, mothers in the intervention group were 2 times more likely to practice EBF compared to those in the control group. Similarly, TIBF was significantly different between the groups at midline but not at baseline. A higher proportion (60.47%) of mothers in the intervention group initiated timely BF compared to the control group (27.1%) at midline, with a net effect of +27.3%. Unlike the baseline, mothers in the intervention group were almost 2 times more likely to initiate BF timely (RR:1.64, 95% CI:1.268-2.121; P:0.000) compared to the counterparts. The life table showed that half of the BF initiation occurred in the first hour of birth in the intervention group while takes longer time in the control group. The timely initiation survival probability was 65% in the intervention group whereas 40% in the control group. The median time to initiate BF was 1 and 2 hours in the intervention and control group respectively. Cox-Regression revealed that parity, sex preference, mode of delivery, the health status of mother, support of relevant others, knowledge, and attitude of EBF were pooled predictors of TIBF (p<0.05). Findings from the analysis of end-users’ satisfaction showed that the overall level of end-users’ satisfaction (PSMS) with PDA as an intervention to improve EBF was 50.9% with a maximum score of 99% and a minimum of 8%. Of the emerged satisfaction measuring scales, the standardized mean score for the user empowerment scale was the highest (53.7%). Five scales were emerged with 84.2% of the total variability explained in users’ satisfaction. The mixed- effect model revealed that age, occupation, experience of breastfeeding(BF), knowledge, IV attitude, self-efficacy, main source of BF information, previous home visit/support received from HEPs, participation in any social activities, and perceived community support for BF were independent two-level predictors of satisfaction. Whereas, the findings from the assessment of IOs of PDA revealed that the total variability explained by the emerged scales was 72.1%. Valid and reliable 52 items were developed to measure these outcomes. The maximum mean score was for ‘appropriateness scale’ 27.81 (6.5SD), while the minimum was for ‘implementation cost scale’ 11.37 (5.2SD). The overall mean score was 164.18(26.8SD). Majority (66.7%) of outcomes of PDA were highly rated. Implementation fidelity, penetration, organizational readiness, and sustainability scores were positively and significantly correlated with acceptability of the approach. Furthermore, significant mean differences were observed between sexes, educational status, roles of the participants and level of engagement. Of all the explained variations among the measurement scales, the maximum variation explained was observed between acceptability and implementation fidelity (36%). Finally, the cost-effectiveness analysis revealed that the mean PDA intervention cost per person (infant-mother pair) was $5.9 more than the usual care estimated for 6 to 59 months. The mean incremental DALYs and QALY of the intervention were 0.40 and 0.18 respectively. The Incremental Cost-effectiveness Ratio (ICER) for infants and mothers were $14.75/DALY averted and $32.8/QALY gained respectively. Each extra DALY averted for infants and QALY gained for mothers by the PDA as an intervention to promote EBF relative to the control group costed (or will cost) $14.75 and $32.8 USD respectively. Conclusion: PDA is an effective intervention strategy to promote EBF and TIBF. It was also effective in shorting the time to initiate BF, median time, and the survival probability among the intervention group. More than half of the end-users were satisfied with PDA, and the variabilities were predicted by multi-level factors. This study also identified nine valid, reliable and well explained IO measures that help to understand the intervention reality. Majority of outcome measures were highly rated by intervention participants which suggest as it is a promising approach to promote EBF by PDs. However, addressing the perception of participants about feasibility, adaptability and cost issues need great attention before and during implementation. Objectively, this approach is feasible and cost-effective. Therefore, we recommend the use of PDA to promote EBF as an added strategy to the existing maternal and child urban health program. Keywords: Effectiveness, positive deviance approach, exclusive and timely initiation of breastfeeding, survival probabilities, user satisfaction, multi-level predictors, implementation outcome measurement scales, cost-effectiveness, cluster randomized trial, follow-up study, Ethiopia