Republic of Mozambique RAcE Multi-Country Results Dissemination Meeting Endline Survey Results Republic of Mozambique Ministry of Health Abuja, Nigeria 24-27 October, 2017

Survey Design Survey was conducted by the National Health Institute (INS) with technical assistance from ICF and financial & logistical support from SC; protocol & instruments were approved by the MoH’s IRB Design: 30x30 multi-stage cross-sectional cluster based household survey with children 2-59 months who had recenty been sick with diarrhea, fever or fast breathing in the past 2 weeks prior to the survey; target was 300 surveys for each disease

Survey Instruments The instruments consisisted of 7 modules collecting data on 22 key indicators: Caregiver & hhld background information Caregiver knowledge of iCCM activities A module for each major childhood illness Standard demographic & health survey data Data were collected via mobile devices; ICF programmed the questionnaire using CommCare An APE survey questionnaire was implemented, which was developed using tools from the previous CIDA-funded iCCM programme and the Quality of Care survey

Survey Preparations & Implementation Data collection took place between 8-30 October, 2016, in the 4 provinces The survey was translated and uploaded into the mobile devices in Portuguese; enumerators translated it into the appropriate local languages during data collection A total of 12 survey teams implemented the study, each team composed of 1 supervisor and 2 enumerators There were various delays throughout, primarily due to the distance between clusters and poor road & connectivity issues

Data Limitations Despite quality control measures, the dataset had a number of errors, requiring additional cleaning and some reconstruction Inhambane (MC) province had been excluded at baseline so had 2 additional clusters sampled to help provide programme-wide information - therefore endline results are based on 3 out of 4 RAcE provinces Both Inhambane and Nampula had deployed APEs at the time of the baseline that had been supported by Malaria Consortium and Save the Children, respectively Many enumerators did not record household identifying data in the data sheets and ICF & INS had to manually link information Households were interviewed that did not have a child that was sick in the past 2 weeks and were given the caregivers questions only At the time of the baseline, there were 124 APEs in Nampula in April 2013 and 656 by the end of Dec 2016

Profile of Caregivers: Distance to Nearest HF The average distance of households to health facilities in the APE communities is between 13.7 and 14.5 km. The slides on distance and time walked to a HF is a good reminder of why remote communities need basic life-saving services in their communities

Profile of Caregivers: Walking Time to Nearest HF Families walk on average about 2 hours and 27 minutes each way to reach their closest health facility

Caregiver Knowledge & Perceptions of iCCM trained APE in their Community Perceptions have not changed significantly between baseline and endline, although it is evident that caregivers have consistently viewed APEs positively; communities seek their care where available

Caregiver Knowledge of Childhood Illnesses There is a significant difference in caregiver knowledge over the course of the programme in all areas except the cause of malaria, which was not statistically significant; caregivers are more educated about malaria symptoms and treatment from baseline to endline.

Care Seeking using APE as First Source of Care (of those who sought care) As highlighted in the earlier slides, the data indicate that where an APE is present, 67.4% of community members use them as the first source of care; the percentage nearly doubled from baseline to endline. At baseline, 56% of caregivers sought care from health facilities and 23% sought care from APEs. At endline, this percentage flipped to 57% caregivers seeking care from APEs and 23% seeking care from a health facility, suggesting an increase in access to care in the RAcE project areas.

Fever cases in which care was sought from APE While more people saught fever care from an APE and had their blood drawn by an APE from baseline to endline, of those cases who were RDT positive, less cases were treated by the APE between baseline and endline. At the baseline, Nampula had a bumper stock of RTDs and ACT purchased from the previous CIDA-funded program. By the endline, the significantly lower treatment rate was likely due to insufficient AL stock. The APE survey noted that stockouts of ACTs and RDTs were widespread, with only 28% of APEs reporting continuous stock of ACTs and 44% reporting continuous stockouts of RDTs during the month prior to the survey.

Cases that Received Treatment from an APE vs Other Provider The treatment of diarrhea and ARI increased significantly by APEs between baseline and endline

Cases that Received Treatment from an APE vs Other Provider The treatment of diarrhea and ARI increased significantly by APEs between baseline and endline

Caregivers who did not seek any care Overall, 15% of caregivers did not seek any care for the 3 iCCM illnesses, both at baseline and endline, meaning that the health system and the APE programme needs to do more to further influence care-seeking behavior. The reduction in careseeking for fever is concerning, although there are overlapping confidence intervals between baseline and endline so it is possible that the difference is not significant.

APE Survey: Medicines and Diagnostics Availability Indicator Result % of APEs with all CCM medicines and diagnostics in stock on day of assessment (artemether-lumefantrine, amoxicillin, ORS, zinc, RDTs, timer) 38% Artemether-lumefantrine (1x6 or 2x6) 44% RDTs 75% Amoxicillin 81% ORS 91% Zinc Timer 78% % of APEs reporting no stockouts of essential iCCM supplies lasting seven days or more in the month before the survey (artemether-lumefantrine 1x6 and 2x6, RDTs, amoxicillin, ORS, zinc) 16% Only 38% of the APEs surveyed had all the iCCM diagnostics and medicines in stock on the day of the survey. Only 16% reported that they had no stockouts of essential iCCM supplies lasting 7 days or more in the month prior to the survey. Stockouts are still a concern.

Main Conclusions (1) Caregivers live far from HF; the best way to ensure the early classification and treatment for malaria, diarrhea and pneumonia is through APEs APEs are well recognized and appreciated in their communities; those who sought care for their sick children were more likely to approach their APEs as first point of care There has been a significant increased treatment by an APE of diarrhea, pneumonia and testing for malaria; treatment of malaria has decreased, likely due to stockouts (confirmed by the APE survey)

Main Conclusions (2) It can be argued that APEs that are well-stocked and supervised for service quality can provide even greater treatment rates, and the government and partners must continue to invest in the APE programme in order to contribute towards the reduction in <5 morbidity and mortality rates in Mozambique

OBRIGADO KANIMAMBO THANK YOU