Study design and location
The South Sudan’s healthcare system operates three tiers of healthcare service: community, primary, secondary and tertiary levels (29,30) (30). The community and primary healthcare shares a significant portion of the country’s health pooled fund is spent on the community and primary healthcare. The community healthcare system (also called Integrated Community Case Management - iCCM or Boma Health Initiative) operates a delivery mechanism at both community and household level, which is focused on the integrated essential child health care (IECHC) and community based child survival. While the primary healthcare services provide Basic Package of Health Services (BPHS). The BPHS covers preventive, curative, health promotion and managerial activities, which care services are delivered at both the primary healthcare unit and the primary healthcare centre. When compared to the primary healthcare centre, the primary healthcare unit operate only five days per week, eight hour per day and is less staffed with a community healthcare worker, community midwife, a vaccinator and 2 other non-healthcare staff and provides less service which is focused on diagnosis and treatment of common community illnesses, immunisation and therapeutic supplementary feeding programmes with a target to cover 15,000 population. The primary healthcare centre (PHCC) is the first level referral health facility having qualified health professionals including clinical officers, nurses/midwives, pharmacy technician, vaccinators, nutritionist and nutrition assistant offering a wider range of immunisation, basic emergency obstetric and neonatal care (BEmONC), diagnostic, curative, minor surgical services and 24-hour-in-patient care for severe common community illness like severe acute malnutrition , severe malaria,etc. A PHCC is designed and located to cover a population of 25,000 and receives referred cases from 4 PHCUs. It receives cases referred by the health workers of the community healthcare system (also called the Boma health workers) and the primary healthcare unit (30). The immunisation unit of the PHCC is usually a room that is not located in the same building as the nutrition unit, and far (>30m) from each other . Similarly, the immunisation unit is usually not close to the out-patient curative consultation department; but sometimes are located in the same building with each other (26).
South Sudan consists of 10 states and 79 counties with a population of 11,062,000 people in 2019. The Lakes State includes Rumbek East and Rumbek Centre counties with a population of 153,550 and 232,752 in 2017 respectively. This was a retrospective intervention study conducted in selected primary health care centers (PHCCs) in Rumbek East county (Aduel PHCC, Pacong PHCC, and Paloc PHCC) and Rumbek Centre county (Amongpiny PHCC, Malual Bab PHCC, and Matangai PHCC) of South Sudan.
Immunisation service integration
Between July to December 2019, a strategy was implemented to integrate the Expanded Programme on Immunization (EPI) into nutrition programmes in all the primary health care centers in Rumbek Centre and EPI was integrated to under 5-year-old out-patient departments in selected primary health centres in Rumbek East.
The intervention adopted used both the functional and infrastructural service integration approaches were implemented in a mutualistic relationship between the programmes’ service providers with a fully decentralized local planning context which had no much external forces on it (31,32). In this study, we refer to immunisation service integration as the adoption and assimilation of immunisation services into the nutrition service unit and out-patient department of the primary healthcare centre wherein, the static immunisation operational activities were co-located, co-delivered and co-reported in the nutrition service unit and the out-patient department of the primary healthcare centre. Weekly re-orientation/on-job training on quick adoption of the functional approaches and practice of service integration into their daily practice during this integration period were conducted to staff of these two programmes. The practices adopted by these staff included: staff retention practice during the period of intervention, trust between staff, staff-to-staff support, user-centered culture, shared belief and co-ownership between the two programmes’ staff, joint-decision-making between the two programmes’ staff, regular communication and meetings were regularly monitored. As there are two implementing partners managing the nutrition service programme; and the curative and preventive health services. Programme managers of these two programmes at county level were oriented and accepted to lose territories’ ownership of their concerned programs in the immunisation service integration intervention period (32). Lastly, immunisation screening system was also adopted in the nutrition service unit and the out-patient department of the primary healthcare centre, where children between the age of 0 -23 months were screened for zero or under-immunisation before accessing the nutrition services and the curative services; and where applicable, these children were directed to the vaccinators for immunisation. The primary health care centers were regularly monitored for the uptake of 1st, 2nd and 3rd dose of the pentavalent vaccine for 6 months during July-December 2019 (26).
The pentavalent vaccine is recommended for children at 6, 10, and 14 weeks for the 1st, 2nd and 3rd doses respectively as per the national immunisation schedule in South Sudan. Children who miss out on the recommended vaccination schedule are still offered the pentavalent vaccine upto the age of two years (less than 24 months of age).
District health information system
The study population includes children between 6 weeks -23 months of age who received one, two or all of the doses of the pentavalent vaccine. All the monthly data of the pentavalent vaccine was solely extracted from the South-Sudan’s District Health Information System 2 (DHIS2) website (33),and was used in the analysis. No sampling method was adopted. We strongly believe that the readily available DHIS2 data represents the general population to which the results of the study is applied; in addition to that, it has given our study the power to detect a valid estimate through its large sample size irrespective of socio-economic status of the population. Also, our monitoring and supportive supervision to the vaccinators for proper recording helped in improving the quality of data by ensuring completeness, accuracy and timeliness of data submission.
The records were retrieved on a monthly basis for a 12-month period in 2019 (6 months before and 6 months after intervention) across all the three monitored primary health care centers in each county and was jointly checked with the county health department for data quality and validation. The data was then classified into two groups using the WHO immunisation age groups classification – infant for under-1 year old age group (≥ 6 weeks and < 12 months) and toddler for 1 year and above age group (≥ 12 months and < 24 months), as in this way it is possible to compare the “timely coverage” versus “delayed coverage” respectively. Considering other confounding factors was not possible as data from the dhis2 is strongly restricted and limited to certain variables like age, sex, geographical location - counties, communties.
Immunisation (or vaccine uptake) rate was calculated using the number of children in each targeted age group that received the last recommended pentavalent vaccine dose as the numerator while the estimated number of target population were assumed to be the total number of surviving children in the target group for each vaccine, and was taken as denominator (34,35). This is because the Southern Sudan Centre for Census, Statistics and Evaluation is yet to publish the most accurate and reliable data for the number of surviving children in the local communities of the country. The number of surviving children was obtained by subtracting estimated deaths (using the country’s infant mortality rate) from the total live births per month in each county (36). Mathematically, the total number of surviving children at time t as:
Surviving children = Total number of birth delivery - total number of not surviving infants in the considered period
Number of surviving children at time t: P(t) = P(0) - (P(0) *r), where:
P(t) is the dead infants after t years
P(0) is the total number of birth delivery at the base month
r is infant mortality rate (63.7%)
t is time considered
Comparative effectiveness analysis
We compared the uptake for the three doses of the pentavalent vaccine before immunisation service integration intervention during January-June 2019 and after immunisation service integration into nutrition sites and children’s outpatient departments during July-December 2019, using standard normal distribution test. To address seasonally induced effectiveness bias, we also compared the intervention period (July - December) in 2019 with the same period (July - December) in 2018. Data were stratified by age and immunisation rates were obtained in each age group to adjust for the confounding effect of age. Rate ratios (before and after intervention) were computed to ascertain percentage contribution of intervention programmes in both counties. We calculated rate ratios to estimate the effectiveness of the integration of the Expanded Programme of Immunisation in nutrition settings of Rumbek Centre county in comparison to pediatric outpatient department settings of Rumbek East county. Immunisation drop-out metric was calculated by subtracting the number of pentavalent vaccine 3rd dose uptake from the number of the pentavalent vaccine 1st dose uptake in each county. The immunisation drop-out metric was divided by the uptake of the pentavalent vaccine 1st dose to estimate the immunisation drop-out rate (34,35). Student t-test was used to compare uptake of immunisation before and after intervention. Standard normal distribution test (Z-test) was used to generate confidence intervals (with 95% confidence level) for rate ratios. Estimates with p-value less than significance level (5%) were reported as strongly associated. We conducted our statistical analysis using SPSS version 25.
Ethical approval for the study to be conducted and published was granted by the Health Research Ethics Committee of the State Ministry of Health (SMoH) of Lakes State (formerly Western Lakes State), South Sudan (Reference Number: MOH/WLS/14/09/2019). The research was carried out in accordance with the principles of the Helsinki declaration.
Patient and Public Involvement statement:
The immunisation coverage data used in this study is based on children eligible for pentavalent vaccination in the Rumbek East and the Rumbek Centre counties of South Sudan. Children and parents/adult carers were not involved in setting the research question or the outcome measures.