Background: Severe acute malnutrition (SAM) among children under five years of age remains a huge public health and economic burden in Sub-Saharan Africa. We investigated time to recovery and its predictors among children aged 6 to 59 months admitted into Community-based Management of Acute Malnutrition (CMAM) stabilisation centres for complicated severe acute malnutrition and whether the outcomes met the minimum Sphere standards. Methods: The study was a retrospective cross sectional quantitative review of data recorded in six CMAM stabilization centres registers in four Local Government Areas, Katsina state, Nigeria from September 2010 to November 2016. Records of 6925 children, aged 6–59 months with complicated SAM were reviewed. Descriptive analysis was used to compare performance indicators with Sphere project reference standards. Cox proportional hazard regression analysis was used to estimate the predictors of recovery rate at p < 0.05 and Kaplan–Meier curve to predict the probability of surviving different forms of SAM. Results: Marasmus was the most common form of severe acute malnutrition (86%). Overall, the outcomes met the minimum sphere standards for inpatient management of SAM. Children with oedematous SAM (13.9%) had the lowest survival rate on Kaplan–Meier graph. The mortality rate was significantly higher during the ‘lean season’—May to August (Adjusted Hazard Ratio (AHR) = 0.491, 95% CI = 0.288–0.838). MUAC at Exit (AHR = 0.521, 95% CI = 0.306–0.890), marasmus (AHR = 2.144, 95% CI = 1.079–4.260), transfers from OTP (AHR = 1.105, 95% CI = 0.558–2.190) and average weight gain (AHR = 0.239, 95% CI = 0.169–0.340) were found to be significant predictors of time-to-recovery with p values < 0.05. Conclusion: The study showed that, despite a high turnover of complicated SAM cases in the stabilization centres, the community approach to inpatient management of acute malnutrition enabled early detection and reduced delays in access to care of complicated SAM cases. In the face of health workforce shortage in rural communities to provide pediatric specialist care for SAM children, we recommend task shifting to community health care workers through in service training could bridge the gap and save more lives of children dying from the complication of SAM in rural communities in Nigeria.
The study was a retrospective cross sectional quantitative review of data recorded in CMAM inpatient centres registers between September 2010 and November 2016. This study was undertaken in two secondary healthcare facilities (General Hospital Daura and General Hospital Baure) and four primary healthcare facilities (Comprehensive Health Centre Daura, Comprehensive Health Centre Dutsi, Comprehensive Health Centre Zango, Maternal and Child Health Centre Zango) designated as SCs across four Local Government Area (LGA), namely Baure, Daura, Dutsi and Zango in Katsina state, Northwest region of Nigeria. The stabilisation centres are situated in rural communities and admit only under five children with complicated SAM. The communities served by the health facilities are predominantly Hausa/Fulani-Muslims living in rural areas. The majority are agrarian and nomad. The total bed capacity in the six SCs was 60. In the four primary healthcare facilities, the wards were managed by Community Health Workers (CHWs) trained in the inpatient management of SAM, while the two secondary health facilities were managed by nurses. These health workers received yearly refresher training in the inpatient management of SAM. All the children were managed using the WHO 10 step to inpatient management of SAM. For children 6–59 months, once the child regain appetite and were clinically stable, they were transferred to the OTP closest to the child’s home, to continue nutritional rehabilitation with peanut based ready-to-use food until full recovery. HIV serology for children older than 19 months of age was measured using a rapid diagnostic test. However, the majority of the participants were not screened because of the inconsistency in the supplied of the kits. All children found to be HIV infected were referred to a secondary health facility for antiretroviral therapy (ART). All children aged 6–59 months admitted to the SCs between 2010 and 2016 were eligible for the study. Admission to SCs is based on the presence of bilateral pitting oedema and /or MUAC < 11.5 cm for children 6–59 months with medical complications. The initial sample size was 7789 children age zero to 60 months on the CMAM registers. However, the de facto eligible sample was 6925 due to excluding records with incomplete information like weight gain, length of stay and HIV status. Children were either admitted directly from the community or transferred from OTP centres. The sampling procedure is shown in Fig. 1. Flow chart showing sampling procedure All Children in the register were eligible for the study while excluding infants less than 6 months and children age greater than 59 months. The data were extracted from the CMAM inpatient records directly into Microsoft Excel for cleaning and then exported to the Statistical Package for Social Science (SPSS) version 23 for analysis. Severe Acute Malnutrition: Mid upper Arm circumference < 11.5 cm and/or oedema (excluding non-nutritional cause of oedema). Outpatient therapeutic programme: A component of CMAM that care for uncomplicated SAM. Inpatient therapeutic programme: A component of CMAM that care for complicated SAM in stabilization centres. Length of stay: The number of days a patient stays in the health facility before achieving an outcome. Outcome: Stabilized/recovered, death or defaulted. Recovery: SAM patient that has regained a good appetite and evidence of medical complications being resolved. Defaulter: SAM patient that left treatment before recovery. Death: SAM patient who died during the course of inpatient therapy. Dependent variables: outcome variables were stabilised/recovered; death and defaulter, length of stay (days) and average weight gain (g/kg/day). Independent variables: the independent variables were socio-demographic data (age, gender, and place of admission), anthropometry at admission and discharge (MUAC, weight), forms of malnutrition (marasmus, marasmus-kwashiorkor and kwashiorkor), facility of admission, and month of admission. The Kolmogrov Smirnov test of normality was used to check the normality of the distributions for the continuous variables which indicated that the variables were normally distributed The Multicollinearity test between independent variables indicated that there was no correlation between variables based on a variance inflation factor of less than 10. Data were described using frequency distribution and measures of central tendency and dispersion. Kaplan–Maier Curve and Long rank tests were used to estimate cumulative survival probability and to compare survival status probability across different groups. Bivariable and multivariable Cox regression were used to identify predictors of time-to-recovery and outcome. Variables with a p value less than 0.05 during bivariate analysis were included in the multivariate analysis. A sensitivity analysis was done to ascertain whether the findings of this study would have been different if the missing data for weight at exit and type of SAM (Table (Table1)1) had been included in the analysis. A dichotomous variable was then created using the total number of missing and present data in the dependent variable mortality. The analysis shows that generally the known data was more than the missing. Consequently, it can be inferred that the results would not have changed significantly if all the missing data were added to the analysis. Bivariable Cox proportional hazard regression model for predictors of death from complicated SAM in Northwest Nigeria Owing to missing data, values may not add up *P value < 0.05
N/A
DIMA AI Care
