Background: Insecticide-treated nets are a key intervention for malaria prevention. While mass distribution can rapidly scale up ITN coverage, multiple channels may be needed to sustain high levels of ITN access and ownership. In Ghana’s Eastern Region, a continuous ITN distribution pilot, started in October 2012, 18-24 months after a mass campaign. The pilot distributed ITNs through antenatal care services (ANC), child welfare clinic services (CWC) through the Expanded Programme on Immunization, and to students in two classes of primary schools. Methods: ITN ownership and access were evaluated through two cross-sectional surveys, conducted at baseline in April 2012, 11-15 months after the mass campaign, and at endline in December 2013, after 1 year of continuous distribution. A representative sample was obtained using a multi-stage cluster sampling design. Household heads were interviewed using a structured questionnaire. Results: Household ownership of at least one ITN was 91.3% (95% CI 88.8-93.9) at baseline and was not statistically significant at endline 18 months later at 88.3% (95% CI 84.9-91.0) (p = 0.10). Ownership of at least 1 ITN per two people significantly decreased from 51.3% (95% CI 47.1-55.4) to 40.2% (95% CI 36.4-44.6) (p < 0.01). Population access to an ITN within the household also significantly decreased from 74.5% (95% CI 71.2-77.7) at baseline to 66.4% (95% CI 62.9-69.9) at endline (p < 0.01). The concentration index score for any CD channel was slightly positive (0.10; 95% CI 0.04-0.15). Conclusion: Thirty-one months after the mass campaign, the 15 months of continuous distribution activities had maintained levels of household ownership at least one ITN, but household ownership of one ITN for every two people and population access to ITN had declined. Ownership and access were higher with the CD programme than without. However, the number of ITNs delivered via ANC, CWC and two primary school classes were insufficient to sustain coverage targets. Future programmes should implement continuous distribution strategies fully within 1 year after a campaign or widen eligibility criteria (such as increase the number of classes) during the first year of implementation to make up for programme delays.
Although six channels had been envisaged, only three channels were implemented during the pilot period. Students in primary class 2 and 6 received ITNs in October 2012. In November 2013, primary class 4 and 5 received ITNs, shown in Fig. 1. This change in classes was made because nation-wide school distribution was expected in May 2014, toward the end of the school year, serving classes 2 and 6; planners wanted to avoid distributing ITNs twice in a single school year to the same classes. The health facility and school distributions used existing structures within ANC and CWC clinics and schools, namely the storage, records, beneficiaries, and staff. Implementation required coordination between three main government agencies, the Ghana Health Service’s NMCP, the division of Reproductive and Child Health, the division of EPI, and the Ghana Education Service’s GES School Health Education Programme (SHEP). Representatives from each agency were involved in planning, execution and supervision at all levels: national, regional, districts, circuits/sub-districts and school or health facility level. Timeline of study and intervention activities in Eastern Region, 2010–2013 The number of ITNs distributed through schools was quantified using student enrolment data for the target classes. Twenty-six district SHEP coordinators and 180 circuit supervisors were trained on how to collect and compile enrollment data from schools and how to distribute ITNs and complete reports. ITNs were transported to district education stores and then directly to schools by circuit supervisors. Radio messaging and dramas accompanied school distributions to promote awareness, answer questions about eligibility, and encourage ITN use. For the final 10 months of the pilot period, there was an increased focus on improving ITN distribution through health facilities through the provision of training, supervision visits and ITNs. All public and private health facilities that offered ANC and CWC services in Eastern Region participated in the programme. Every pregnant woman who visited ANC for the first time was entitled to receive an ITN, and given information about the effects of malaria in pregnancy and the need for proper use of nets. At CWC clinics, every child aged 18–24 months receiving measles II booster dose was entitled to an ITN. The choice to provide the ITN at the measles booster was made partly to boost rates for this new vaccination. To facilitate record-keeping, the Maternal Record Book and Child Health Card was modified to include “ITN given” and ITNs issued were recorded in ANC and CWC registers. During the study period, 135,070 ITNs were distributed in the first round of school distribution in October 2012, and 136,000 were distributed in the second round in November 2013. Students in 2682 public and private primary schools received ITNs. During the same period, 114,000 ITNs were allocated across all health facilities in the 26 districts. All told, 385,070 ITNs were distributed during the pilot. NetCALC modelling indicated that this would have been sufficient to maintain levels of 90% household ownership of at least 1 ITN, and 77% population ITN access for 2012 and 2013, on the assumption that the Eastern Region campaign had achieved those coverage targets in 2011 based on distribution data. A baseline survey was conducted in April 2012, 12–16 months after the mass campaign, and about 4 months before the start of CD activities (Fig. 1). The endline survey was carried out in December 2013, after 1 year of CD implementation, including two rounds of school distribution and 10 months of health facility distribution. Ghana has two rainy seasons each year, from April to July and from September to November; baseline data collection took place at the beginning of the first rainy season, while the endline data collection fielded near the end of the second rainy season. The primary outcomes of interest were ownership of at least one ITN and population access to an ITN, as defined by RBM’s Malaria Evaluation Reference Group [21]. Secondary outcomes included levels of over and under-supply within households, the relative contribution of each channel to ITN ownership, and ITN use. A multi-stage cluster sampling design was used in both surveys to ensure comparison across time. A cluster was defined as a community and 60 clusters were selected using systematic sampling with probability proportionate to size (PPS) based on population data from the campaign’s household registration lists. Within each cluster, a list of households was prepared by the survey team and households were then randomly selected for interview. If a cluster had more than 200 households, an equal-size section approach was used and one section was randomly chosen from the household list. Households were defined as “people eating from the same pot” which was the definition used in the mass campaign. Seventeen households per cluster at baseline and fifteen households per cluster at endline were targeted. To demonstrate that household ownership of 1 ITN for every 2 people was maintained between baseline and endline surveys, the standard formula for an equivalence study was used [22]. Sample size was calculated using an alpha error of 95%, a beta error of 80%, a design effect of 1.75, an anticipated non-response rate of 5%, and the expectation that there would be 5.0 persons per household, 15% of the population under 5, 4% of the proportion was pregnant, and that the percentage of households with 1 ITN for two people would be 49.8% at baseline and 48% at endline. The population estimates were based on the 2008 Ghana Demographic and Health Survey [23]. For data collection, the same pre-tested questionnaire was used for baseline and endline data collection. The primary respondent was the head of household or his/her spouse and the person who was present during the visit of mass campaign team. The questionnaire was based on the Malaria Indicator Survey and focused on household ownership and use of ITNs. Questions were added to capture several processes specific to continuous distribution such as the number of ITNs received through school, ANC, or CWC as well as the number of eligible students within the household. ANC/EPI (CWC) nets from the pilot were nets reported from those sources that were also obtained during the pilot period, as determined the question on “how many months ago did you obtain this net”. Double entry of all records was done using EpiData software version 3.1. Both data sets were then compared and any discrepant record was verified from the original questionnaires. Data were then transferred to Stata 14.0 statistical software package for further consistency checks and preparation for analysis. All analysis was done adjusting for the cluster sampling by using the “svy” command family in Stata. Concentration index and concentration curves were used to analyse outcome differences by wealth. Standard errors and confidence intervals for the concentration indices were calculated using the formula suggested by Kakwani et al. [24].
N/A