Inpatient and postdischarge mortality among children with anaemia and malaria parasitaemia in Kenya: a cohort study

Abstract

Background: Anaemia and malaria are leading causes of paediatric hospitalisation and inpatient mortality in sub-Saharan Africa. However, there is limited empirical data on survival following hospital discharge. We aimed to estimate independent effects of anaemia and malaria parasitaemia on inpatient and 1 year postdischarge mortality among Kenyan children.

Methods: A retrospective cohort study among children admitted to Kilifi County Hospital (KCH) from 2010 to 2019 and followed-up for 1 year postdischarge in Kilifi Health and Demographic Surveillance System (KHDSS). The main exposures were anaemia and malaria parasitaemia at the time of hospital admission while inpatient and 1 year postdischarge mortality were the outcomes.

Results: We included 9431 admissions among 7578 children (43% girls), median (IQR) age 19 (9.9‒23) months. 2069 (22%), 3893 (41%) and 1140 (12%) admissions had mild, moderate and severe anaemia, whereas 366 (3.9%), 779 (8.3%) and 224 (2.4%) had low, medium and high malaria parasitaemia, respectively. Overall, there were 381 (4.0%) inpatient deaths: 317/381 (83%) and 47/381 (12%) among children with any level of anaemia and malaria parasitaemia, respectively. Moderate and severe, but not mild anaemia, were positively associated with inpatient death. Low and high level parasitaemia were positively associated with inpatient mortality, while medium level parasitaemia was negatively associated. There were 228 (3.1%) postdischarge deaths: 32.8 (95% CI 28.8‒37.3) deaths/1000 child-years. 180/228 (79%) deaths occurred within 6 months after index discharge and 99/228 (43%) occurred in the community. Overall, 180/228 (79%) and 10/228 (4.4%) postdischarge deaths occurred among children with any level of anaemia and malaria parasitaemia, respectively. Severe anaemia was positively associated with postdischarge mortality (adjusted HR 1.94 (95% CI 1.11‒3.40)), while medium level parasitaemia was negatively associated.

Conclusion: Interventions to create awareness of postdischarge risks, improve uptake of existing interventions and improved discharge processes targeting high-risk groups such as children admitted with severe anaemia, need to be prioritised.

Keywords: Anaemia; Child health; Global Health; Malaria; Paediatrics.

Figure 1. Study participants flow stratified by (a) anaemia levels and (b) malaria parasitaemia density. In panel (a), 3/1775 (0.17%), 1/3723 (0.03%), 1/1052 (0.10%) and 1/490 (0.20%) were missing follow-up data among admissions with none, moderate, severe and with no available anaemia status data, respectively. In panel (b), 4/7263 (0.06%), 1/351 (0.27%) and 1/465 (0.21%) were missing follow-up data among admissions with none, low and no available malaria parasitaemia, respectively. All the records missing follow-up data are shown in square brackets ([ ]) in the figures.

Figure 2. Cumulative hazards of postdischarge mortality stratified by (a) anaemia levels and (b) malaria parasitaemia density.