Safety and practicability of using mid-upper arm circumference as a discharge criterion in community based management of severe acute malnutrition in children aged 6 to 59 months programmes
The study design was a standards-based trial. There are no internationally agreed standards for post-discharge relapse and mortality in CMAM programs. The standard adopted for the study was informed by a review by Ashworth [6, 7], which used a standard of??10 % relapse as a ‘criterion for success’ for community-based nutritional rehabilitation programmes prior to the development of the CMAM treatment approach. A standard of less than 10 % for combined post discharge relapse and non-accidental, non-violent death was adopted for the study through consultation with experts in CMAM programming at FANTA/AED. This standard was more stringent than the standard used by Ashworth. In this article we present estimates for post-discharge relapse and mortality separately and combined so as to allow comparison against other standards.
The study was conducted in five outpatient health facilities run by the Government of Malawi Ministry of Health (MoH) and located in Lilongwe District, Malawi, with care delivered by MoH staff. A research team from Valid International (comprised of a research coordinator, a registered nurse, and a community mobilisation officer) trained MoH staff in study protocols and assisted directly in the measurement and treatment of all study subjects. The Malawi National Guidelines for Community-based Therapeutic Care [8] formed the standard of care. At the start of the study the admission criterion for the treatment of SAM was MUAC??110 mm according to the Malawi National guidelines. These guidelines were in the process of being revised to meet the recommendation of the UN Joint Statement of 2009 [2] to raise the admission criteria to MUAC??115 mm. The clinics in which the study subjects were recruited changed the admission criteria to MUAC???115 mm ahead of the publication of the revised guidelines with the consent of the Ministry of Health.
The study enrolled children aged between 6 and 59 months with uncomplicated severe acute malnutrition (SAM) as defined by MUAC???115 mm and without medical complications. Children with a MUAC???115 mm with concurrent bilateral pitting oedema or medical complications were referred to inpatient care according to Malawi National Guidelines and enrolled into the study on their return from inpatient care.
Subjects were recruited between February 2011 and March 2012. Age was verified by reference to the child’s ‘health passport’. Children whose caregivers were resident outside of Lilongwe District were excluded from the study but were admitted to the CMAM program and provided with standard care according to the Malawi National Guidelines. MUAC, weight and height were measured on admission. An appetite test was conducted to confirm the appropriate consumption of Ready to Use Therapeutic Food (RUTF) which is required in order to receive treatment as an outpatient.
Routine medications given on admission followed the Malawi National Guidelines and included a broad spectrum antibiotic (amoxicillin or co-trimoxazole) and vitamin A. Each child was assessed for malaria using a rapid diagnostic test, (Paracheck Pf®), and if positive, was treated with antimalarials (lumefantrine and artemether). If the Paracheck Pf®, rapid test was unavailable, the patient was treated empirically for malaria if they presented with a fever of unknown origin. Each caregiver and child was routinely offered serial testing for Human Immunodeficiency Virus (HIV) using the Determine HIV-1/2 Assay™ and, if tested positive, Uni-Gold Recombigen HIV assay™. HIV positive cases were included in the study and referred for appropriate treatment. It was not monitored whether the child went on to receive treatment for HIV. At each visit the child was given a ration of RUTF according to their weight (average 180–200 kcal/kg/day) and at the second visit anthelminthic treatment (albendazole) was given. The dosage for all medicines was age or weight related and given according to the Malawi National Guidelines.
Study subjects were requested to attend outpatient treatment for SAM on a weekly basis. At each clinic visit the child was measured for weight and MUAC and their appetite for RUTF was assessed. Due to the rotation of MoH clinical staff in attendance at the health centres, a single observer from the research team conducted each of the measurements of weight, height, and MUAC in order to minimise inter-observer variation. The standardisation of the measurements was assessed using the method of Habicht [9], using the principle investigator as the standard, and was reviewed every 3 months.
If the status of the child had not improved or had deteriorated since the previous visit then the child was referred to the MoH clinician for further assessment according to Malawi National Guidelines. Any child absent from treatment was followed up at home by a Health Surveillance Assistant (HSA) or community-based volunteer and encouraged to return for treatment.
Each study subject was discharged ‘cured’ from outpatient treatment when they achieved a MUAC of 125 mm or greater on two consecutive clinic visits. On discharge the caregiver was given a fourteen-day supply of RUTF calculated according to the weight of the child. The child was referred to a Supplementary Feeding Program (SFP) if such a program were operating in the health area of residence. A study subject was also discharged from the study if the child died, ‘defaulted’ (i.e. was absent for three consecutive visits) or was ‘non-cured’ (i.e. had not reached the discharge criteria) after sixteen weeks (112 days). All admissions and discharges from the study were approved and verified by a member of the Valid International research team. Any study subject discharged from the study as ‘non-cured’ was also assessed according to Malawi National Guidelines and referred back to the standard Ministry of Health programme for further treatment if the child had not reached the discharge criteria as ‘cured’ according to national guidelines.
Previous studies had indicated that most mortality following treatment for SAM occurs within three months of discharge [10]. Each child that was discharged as ‘cured’ was followed up every two weeks for a three-month period by an HSA trained in the study protocols. At each follow-up visit the child was assessed for illness (cough, fever, diarrhoea or vomiting) and their MUAC was measured by the HSA in accordance with their training in the Integrated Management of Childhood Illnesses (IMCI) program. Any child who had relapsed (i.e. met the program admission criteria of MUAC???115 mm or had bilateral pitting oedema) or had deteriorated clinically was referred back to the health centre for assessment and treatment. Relapsed children were eligible for re-admission to the study if their MUAC was equal to or below 115 mm. Any child absent from their home for two consecutive home visits was verified as being lost to follow-up (LTFU) by consulting with members of the local community regarding the whereabouts of the family. The accuracy of the MUAC measurements and clinical status of the child during follow-up was assessed by a member of the research team. Verification visits were made randomly to children in the follow-up phase of the study and visits were conducted in all cases where relapse was detected or LTFU had occurred.
No preferential treatment was given to study subjects compared with non-study subjects.
Data collection and analysis
On admission and at each subsequent visit, the child’s weight was measured (naked or wearing light undergarments and without shoes) to the nearest 100 g using Salter™ scales (Salter-Brecknell 235-6S series), the model used by the health centre for the clinical management of SAM cases. MUAC was measured to the nearest millimetre using a standard non-elastic MUAC tape (UNICEF supply code S0145620 “MUAC, Child 11.5, Red/PAC-50”). On admission and discharge the child’s height was measured to the nearest millimetre using a standard paediatric height board (supine length was measured for children less than 24 months of age).
After discharge as ‘cured’, the child was followed up at home every two weeks for a three-month period by a HSA trained in the study protocols. At each follow-up visit the child was assessed for illness (cough, fever, diarrhoea or vomiting) and their MUAC was measured by the HSA.
Data were collected on paper forms designed for the purpose and were entered into an EpiData Version 3.1 (EpiData Association, Odense, Denmark) database using both interactive checking for range and legal values and double-entry validation. Statistical analysis was performed using the R Language for Data Analysis and Graphics [11]. Height-for-age, weight-for-age, and weight-for-height z-scores were calculated according to the WHO Child Growth Standard [12]. The Kruskal-Wallis non-parametric test was used for comparisons of lengths of stay and proportional weight gain by categories of MUAC at admission and categories of height at admission and for comparison of height at admission for ‘cured’ and ‘non-cured’ cases.. The null hypothesis for the Kruskal-Wallis test is that the data in each group have the same distribution of the variable of interest. The strength of association between cure and MUAC at admission and attendance rate was investigated using logistic regression.