Malaria indicator survey 2009, South Sudan: Baseline results at household level

Background: South Sudan has borne the brunt of years of chronic warfare and probably has the highest malaria burden in sub-Saharan Africa. Malaria is the leading cause of morbidity and mortality in the country. This nationally representative survey aimed to provide data on malaria indicators at household level across the country. Methods. In 2009, data were collected using a two-stage random cluster sample of 2,797 households in 150 census enumeration areas during a Malaria Indicator Survey (MIS) in South Sudan. The survey determined parasite and anaemia prevalence in vulnerable population groups and evaluated coverage, use and access to malaria control services. Standardized Roll Back Malaria Monitoring and Evaluation Reference Group (RBM-MERG) MIS household and women’s questionnaires were adapted to the local situation and used for collection of data that were analysed and summarized using descriptive statistics. Results: The results of this survey showed that 59.3% (95% CI: 57.5-61.1) of households owned at least one mosquito net. The proportion of the population with access to an ITN in their household was 49.7% (95% CI: 48.2-51.2). The utilization of insecticide-treated nets was low; 25.3% (95% CI: 23.9-26.7) for children under five (U5) and 35.9% (95% CI: 31.9-40.2) of pregnant women (OR: 1.66 (1.36-2.01); P =0.175). Prevalence of infection was 24.5% (95% CI: 23.0-26.1) in children U5 and 9.9% (95% CI: 7.4-13.1) in pregnant women. About two thirds (64%) of children U5 and 46% of pregnant women were anaemic. Only 2% of households were covered by indoor residual spraying (IRS) the previous year. Data shows that 58% reported that malaria is transmitted by mosquitoes, 34% mentioned that the use of mosquito nets could prevent malaria, 41% knew the correct treatment for malaria, and 52% of the children received treatment at a health facility. Conclusion: The observed high malaria prevalence could be due to low levels of coverage and utilization of interventions coupled with low knowledge levels. Therefore, access and utilization of malaria control tools should be increased through scaling up coverage and improving behaviour change communication. © 2014 Eyobo et al.; licensee BioMed Central Ltd. South Sudan covers 650,000 sq km of land between 8° and 18° degrees south latitude and between 20° and 35°degrees east longitude (Figure 1) with a population of 8.3 million and almost 900,000 refugees, returnees and internally displaced persons [7]. While the population density is 13 p/sq km, only 17% of the total population resides in urban areas. South Sudan comprises ten states in three regions: 1) Greater Equatoria: Eastern Equatoria, Western Equatoria and Central Equatoria, 2) Greater Bahr el Ghazal: Western Bahr el Ghazal, Northern Bahr el Ghazal, Warrap and Lakes, and, 3) Greater Upper Nile: Unity, Upper Nile and Jonglei. The MIS was conducted between November and December 2009 in accordance with the RBM MERG protocol [6] adapted to local settings. Map of South Sudan showing state boundaries (Source: SSCCSE, 2010). World Health Organization (WHO) recommended case management and vector control tools have been implemented expansively in South Sudan [4]. The WHO-led integrated vector management (IVM) has been adopted as the main approach for vector control. The NMCP developed a draft strategic plan for IVM for the period 2007–2012 [8]. The approach is to consolidate the use of LLINs while introducing additional interventions, i.e. IRS and larval source management (LSM), where applicable. Presently, the distribution of LLINs remains the only key operational vector control intervention with limited use of IRS and larviciding by Mentor Initiative, an NGO in Malakal County [9]. To date over 9.0 million LLINs have been distributed through mass distribution campaigns and health facility based routine distribution. The NMCP is putting in place implementation arrangements for operational deployment of targeted IRS and larviciding. The survey was designed to provide nationally representative estimates of key malaria indicators and utilized a two-stage cluster sample design. The sample was stratified into three survey regions: Greater Equatoria, Greater Bahr el Ghazal and Greater Upper Nile. First, a total of 150 census enumeration areas (EAs) as primary sampling units, stratified by three domains and degree of urbanization, were selected with probability proportional to size and a complete listing of all households in each cluster was carried out. 50 clusters (EAs) were allocated to each of the three domains and stratification by urban/rural was done within the domains. The sample was proportionally allocated hence self weighting in the three strata. Second, 20 households per EA were selected for interviewing using equal probability systematic random sampling making a total sample of 3,000 households. To minimize potential bias, up to three visits were made to ascertain compliance in case of absence of all eligible respondents or subjects. Standard MIS questionnaires based on the RBM MERG guidelines with modification to reflect relevant issues of malaria in South Sudan were used [6]. The household questionnaire was used to list all the usual members and visitors in the selected households and to identify eligible women for the individual interview and children aged 0–59 months for anaemia and malaria testing. It also collected basic information on the characteristics of each person listed, including age, sex, household’s residence and assets, and ownership, type and use of mosquito nets. The woman’s questionnaire was used to collect information from all women aged 15–49 years on background characteristics, full reproductive history, prenatal care and preventive malaria treatment for most recent birth, prevalence and treatment of fever among children under five years, including knowledge about malaria causes, prevention and treatment. The survey designing, planning and implementation was a collaborative effort by multiple individuals from local and international malaria stakeholders. Prior to data collection, two levels of training, combining both course work and practicals, were conducted, a five-day central training of trainers (ToT) for the principal trainers, and ten-day state level cascade training for interviewers. Each interviewer had been given a detailed MIS manual, which was designed in accordance with WHO recommendations [6,10]. A total of 182 field staff participated in the survey. Surveyors were organized in 26 teams, each team consisted of a supervisor, three interviewers and three laboratory technicians and a driver. In addition, ten field operations managers, one per state, were recruited. All teams in a single state were coordinated through their supervisors and the state field operations manager, who collaborated with the central team in Juba. The biomarkers in the survey included rapid diagnostic test (RDTs) and blood slides for microscopic examination for malaria and haemoglobin level testing for anaemia. Blood samples were collected from a finger-prick using a single-use, spring-loaded, sterile lancet. All the three tests were performed simultaneously from a single finger prick. Children aged 0–59 months and pregnant women were tested for anaemia because of the strong correlation with malaria infection. Haemoglobin concentration analysis was carried out on site using a battery-operated portable spectrophotometer (Hemo-Cue 201, Anglom, Sweden). Plasmodium falciparum malaria testing was done using the Paracheck Pf™ RDT, which has shown good sensitivity and specificity in operational settings [11]. Test results for both RDT and anaemia were provided to the child’s parent/guardian verbally and were recorded on the household questionnaire. Two blood slides, thick and thin films, were taken for each participant by a laboratory technician as per standard WHO-approved protocol [12]. The blood slides were air-dried, fixed (thin films), stained with Giemsa and transported to Juba Teaching Hospital reference laboratory for reading. Based on standard laboratory malaria microscopy procedures, the microscopists determined the presence, density (thick blood film) and species of the malaria parasites (thin blood film). If no parasites were found after examination of 200 high power fields, the thick blood smear was considered negative and the corresponding thin blood film was not read. For external quality control, all positive blood slides plus 10% of the negative slides were sent for cross-checking at a WHO prequalified laboratory in the Republic of Oman. The parasite prevalence rates were harnessed for the production of Map of Epidemiological Stratification of P. falciparum malaria among U5s using geo-statistical modeling. Children who tested positive for malaria using the RDT were offered a full course of treatment according to the standard protocol for treating malaria in South Sudan, i e, artesunate-amodiaquine-combination therapy (ACT) [13]. For children diagnosed with moderate-severe anaemia (ie, haemoglobin <8 g/dl), results were shared with the parent/guardian and the children were given artesunate-amodiaquine, albendazole (if >24 months of age as per national protocol for integrated maternal and child illnesses [13]) and supplemental iron. All severe cases with a positive RDT result and children with a haemoglobin level of under 8 g/dl were given a referral card and taken to a health facility for follow-up evaluation and treatment. Further to the successful collection of data during fieldwork, data processing staff were recruited and trained; these consisted of a supervisor from Southern Sudan Centre for Census Statistics and Evaluation (SSCCSE) and data entry operators. Data were entered twice using the CS Pro computer package and cleaned by checking missing cases and inconsistent entries. Descriptive statistics (ie, frequencies, percentages) were used to describe the characteristics of the sample and calculate coverage, use and access estimates. Data analysis was carried out in SPSS 16.0 (SPSS Inc, Chicago, IL, USA). To ensure high quality data collection, the teams were visited daily by central and state supervisors and monitored by principal investigators during the survey period. The teams randomly inspected completed households to confirm correctness of records obtained from the survey and completion of supervisory checklist and observing a team’s overall performance as well as providing feedback and sharing the experiences of other teams. Additionally, the quality of data entering and analysis was performed by highly qualified statisticians. The survey protocol received ethical clearance from the South Sudan Ministry of Health Ethics Committee. Written or verbal, informed consent was obtained from the heads of households and each eligible individual before conducting the household questionnaires. Additional informed consent from a child’s parent or guardian and the pregnant women for blood films and anti-malarial treatment with ACT was provided by a nurse or physician when participants had a positive rapid test result.