Assessment and evaluation of in-patient diagnosis and discharge protocols of Ghanaian infants and children (0 - 59 months) diagnosed with severe acute malnutrition - the SAMAC study

Abstract

Introduction: Complicated severe acute malnutrition (SAM) poses an enormous threat to the survival of children 0 - 59 months old. Even though the World Health Organisation (WHO) has provided diagnostic guidelines for in-patient management of SAM, these are not based on strong evidence. Furthermore, knowledge of the protocols used in various hospitals and the actual clinical practices regarding SAM diagnosis, admission and discharge remains limited. In addition, information on the relationship between the admission characteristics of children with complicated SAM and length of hospital stay (LOS), mortality and weight gain is scant. This study, therefore, aimed to assess the protocols of the hospitals compared with actual clinical practices and the WHO guidelines, with regard to the diagnosis, in-patient admission and discharge of children with SAM in relation to patient treatment outcomes. Methods: Data on the protocols of two Ghanaian hospitals for SAM diagnosis, admission and discharge were collected and compared with the WHO guidelines. The study also involved a retrospective collection of data from the medical records of children aged 0 - 59 months, diagnosed and admitted with SAM, treated and subsequently discharged between January 2013 and June 2017 at the Princess Marie Louis Children’s hospital and the Komfo Anokye teaching hospital in Ghana. Anthropometric and clinical characteristics on admission were assessed in relation to LOS, mortality, daily weight gain, weight-for-age z-scores (WHZ) and improvement in mid-upper arm circumference (MUAC). Data was analysed using SAS version 9.4. Spearman’s correlation and linear regression were used to determine the association between admission characteristics and the treatment outcomes. Results: There were some similarities but a few disparities between the hospitals’ protocols and the WHO guidelines. In the hospitals’ protocols, MUAC was the only anthropometric tool recommended for SAM diagnosis as compared with the use of MUAC or WHZ as anthropometric tools recommended by the WHO. Additionally, children 6 - 59 months weighing < 4 kg are considered severely malnourished and require in-patient treatment according the protocols of the hospitals. There were also some differences between the protocols of the hospitals and the practices as observed in the medical records. Although, MUAC is the only anthropometric tool recommended by the hospitals for SAM diagnosis, only 34.3% of the children had MUAC assessed at admission. Among the discharged children, only 58.5% had their infections resolved and 74.1% had good appetite while only 44.2% had both good appetite and all infections treated at the time of discharge. In total, 289 medical records were included in the study. Discharge, death and abscond rates were 77.7%, 17.7% and 3.8% respectively. Average LOS was 11.4 days with 5.8 g/kg/day weight gain. Oedema and WHZ < -5 SD were associated with longer LOS (14.1 days; 95% CI: 11.5; 17.2; P = 0.02 and 14.0; 95% CI: 11.1; 17.7; P = 0.02 respectively). Average time to death was 4.8 days, with infants < 6 months dying earlier (1.5 days; 95% CI: 0.7; 3.2; P = 0.001) than the 6 - 59 month group (5.9 days). No significant changes in MUAC (P = 0.31) and WHZ (P = 0.69) were observed from admission to discharge. MUAC < 11.5 cm and WHZ < -3 SD were associated with a 3.9 and 1.7 times increased risk of death, although this increase was not statistically significant (P = 0.38 and P = 0.53 respectively). Shock, oedema and HIV-positive status were associated with an increased risk of death that was 7.1 (95% CI: 2.7; 20.5; P < 0.001), 2.5 (95% CI: 1.2; 5.5; P = 0.02) and 3.1 (95% CI: 1.3; 7.2; P = 0.03) higher. Conclusion and recommendations: There were differences between the hospitals’ protocols and the WHO guideline. The clinical practices observed regarding SAM diagnosis, admission and discharge in the hospitals are not adequately commensurate with the hospitals’ protocols. MUAC was not assessed in a large proportion of the children, although it was the only anthropometric diagnostic indicator for SAM, recommended in the protocols of the hospitals. The average weight gain was 5.8 g/kg/day over a LOS of 11.4 days. Furthermore, the high mortality rate was beyond the acceptable SPHERE minimum. These results emphasise the need for detailed health systems research and an in-depth investigation into additional factors contributing to the results found in this study in order to curb the high SAM mortality rate. They also highlight the urgent need for the Ghana Health Service, the Ministry of Health, the administration of the hospitals and the healthcare professionals to implement strategies to optimise the healthcare system in order to reduce SAM deaths in the hospitals by early identification, diagnosis and optimum treatment.