Seasonal Malaria Chemoprevention in Guinea in 2022: coverage in four prefectures.

Seasonal Malaria Chemoprevention (SMC) was introduced in Guinea in 2015 in four cycles from July to October, for children aged 3 months to 5 years, in 6 prefectures, scaling up to 8 prefectures in 2016, 10 in 2017, 13 in 2018, and 17 prefectures from 2020. Household surveys have been conducted at the end of each year to monitor SMC uptake and use of LLINs. The average SMC coverage per cycle was 71.6% in 2018, 71.5% in 2019, and 78.2% in 2020, but fell to 51% in 2021. In 2021, SMC distribution teams were trained to document referral of children who were unwell, check vaccination records and refer under-vaccinated children, check receipt of a LLIN (long-lasting insecticide-treated bednet) in the last campaign, promote antenatal attendance for pregnant women including referral of women who had not received IPTi or an LLIN. The extra workload of these additional activities might have contributed to lower SMC coverage in 2021. In addition, in 2021, children who were missed at cycle 1 were unlikely to get SMC in later cycles, and children who did not receive SMC were also less likely to use a LLIN. In 2022, a survey was undertaken in four prefectures only, with the aims a) to compare SMC coverage in two areas (Koubia and Dalaba) where SMC distribution teams included an additional staff responsible for the additional activities listed above, and two areas (Kankan and Dabola) where the additional activities were undertaken without additional staff, and b) to determine if the association with LLIN use observed in 2021 was confirmed. Methods: The survey was designed to be able to estimate SMC coverage with a margin of error of about +/-7% overall and +/-15% in each prefecture. A total of 48 communities were selected, 12 in each prefecture, selected with probability proportional to population. Area sampling was used to select households. Each selected community was divided into segments, one segment chosen at random, and all eligible children within the chosen segment were included in the survey. Up to 2 call-back visits were undertaken to minimize non-response. Verbal consent of caregivers was sought after explaining the aims of the survey and the nature of the questions, and consent documented. Caregivers were interviewed about SMC treatments, dates of treatments were recorded from the SMC card, and SMC registers were checked to verify SMC treatments. Children under 7 yrs were included, to assess SMC coverage in children under 5 years and to determine if children above the age range received SMC. In addition, all persons who slept in the household the night before the survey were listed, all bednets owned by the household were also listed and inspected, and for each person, the net they slept under, if any, was noted. In 2022, SMC campaigns included documented referral of children who were unwell, checking vaccination records and referral of under-vaccinated children, and promotion of antenatal attendance for pregnant women. Questions related to implementation of these additional activities were included in the survey. Data were collected on tablet PCs. Results: Response rates: A total of 704 households were visited, of which 612 had eligible children, and 578/612 participated, a response rate of 94.4%. A total of 582 caregivers were included, and 1126 children under 7 years, of whom 753 children were eligible to receive four cycles of SMC. SMC documentation: SMC status was determined from SMC cards (available for 41.1% of children), from registers (records found for 32.5% of those without a card), or from caregiver recall. Documented SMC status, from card or register, was available for 60.3% of children. SMC coverage: With respect to the four main cycles in July, August, September and October, the percentage of children who received all four treatments was higher in Koubia and Dalaba where there were additional staff than in Kankan and Dabola, where all activities were managed by the same team. The percentage of children who received all four treatments was 78.5% (95%CI 69.5%,85.4%) in Koubia and Dalaba, and 57.6% (95%CI 45.5%,69.0%) in Kankan and Dabola. Although coverage was similar in all four prefectures in the first cycle, coverage fell in successive cycles in Kankan and Dabola but was maintained at a high level in all cycles in Koubia and Dalaba. Children who were missed at cycle 1 were unlikely to receive any SMC. Overall, 81% of children received SMC in July. Of the 19% who did not receive SMC in July, 98% of these received no SMC in later cycles. ITN use: Overall, 80.5% of children under 5 years of age slept under an insecticide-treated bednet (ITN) the night before the survey. Children who did not use an ITN were less likely to receive SMC (50% received four SMC cycles compared to 63% among children using an ITN). This difference was most marked in the poorest households (27% compared to 58%). The age 5-9 age group, who are vulnerable because they have stopped receiving SMC, had lower ITN use (63.3%). SMC adherence: According to caregivers, the first dose was generally administered by the drug distributor, rather than by the caregiver under observation. Caregiver-reported adherence to day 2 and 3 doses was high. For 98% of children who received SMC, the first dose was administered by the drug distributor, 1.4% by the caregiver under observation, and 0.6% were left with the caregiver but were not administered. Among children who received the first dose, 94.2% received both the day 2 and day 3 doses, according to caregivers. Referral of children who were unwell during SMC visits: According to caregivers, only 3.1% of children were unwell at the time of the last SMC visit. However when asked if the child was referred to the clinic by the SMC team, caregivers reported a high proportion of children referred, 40.2%. This could have included referrals for other reasons (e.g. vaccination). Vaccination referrals: In total, of children who received SMC in the final cycle, 69.3% had their vaccine record checked, and of those, 75.8% were referred for vaccinations, and of those, caregivers reported that 53.2% of children attended for vaccination. The percentage of children whose vaccine record was checked differed by prefecture (90.9% in Koubia, 77.2% in Dalaba, 74.0% in Kankan, and 36.5% in Dabola). Promotion of antenatal care during SMC visits: 32.5% of households reported that SMC teams had given advice about ANC attendance for pregnant women, during the last SMC visit. There were 18.2% of households where a pregnant woman was residing, and among these, 27% were reported to have been referred to ANC by the SMC team. Recommendations: 1.High coverage of SMC was achieved in the four prefectures surveyed, with over 80% of children reached in the July cycle, but coverage was lower in later cycles in Dabola and Kankan leading to fewer children receiving four monthly treatments, 58% compared to 78% in Koubia and Dalaba. This could have been associated with additional workload of checking vaccination status and other activities, but it is likely other factors could have also been involved, including less experience of the coordinating partner in Kankan (a new sub-recipient of GF funds), and drug distributor demotivation associated with payment delays to staff in an earlier LLIN campaign in Dabola. These factors should be reviewed with local teams to mitigate them in future. 2.Administrative estimates of coverage were higher than the survey estimates. The percentage of children who received four treatments was 60.8% (95% confidence interval 50.1% to 70.6%) according to the survey, but 94.2% according to administrative estimates. This may be partly explained by the fact that the administrative estimates include any treatments given to older children, whereas the survey is restricted to the eligible age range. It may also reflect population changes/migration. But where there is a larger discrepancy, the possibility of unreported wastage or errors in administrative data should be investigated. Survey estimates are more reliable. This highlights the importance of using nationally representative surveys to monitor coverage and not relying on administrative data. Caregiver recall in surveys could be unreliable, especially if the primary caregiver is not available during the survey and a proxy respondent is used, but we have found in previous surveys that recall with respect to whether the child has received any SMC, and with respect to the number of monthly treatments received, tends to agree well with what is recorded in registers and on cards. Recall bias is unlikely to explain the large discrepancy between survey and administrative estimates of coverage. However, card retention decreases with time, it is advisable for future surveys, to plan field work just after the last cycle, this will improve data quality as SMC cards are more likely to be available and caregiver recall will be more reliable. 3. Children who are missed at cycle 1, tend not to receive SMC in later cycles. This highlights the importance of reaching all children at cycle 1. There may also be reluctance to register new children at later cycles, this could be addressed during training and supervision. 4. ITN use is low in the 5-9 age group, these children are vulnerable as they have stopped receiving SMC. Strategies to improve access to ITNs in this age group should be pursued.