Effect of a Community Health Worker-Delivered Parental Education and Counseling Intervention on Anemia Cure Rates in Rural Indian Children: A Pragmatic Cluster Randomized Clinical Trial

Abstract

Importance: Iron deficiency anemia, the largest cause of anemia worldwide, adversely affects cognitive development in children. Moreover, the imperceptible childhood anemia prevalence reduction in response to anemia control measures is associated with tremendous social and economic cost.

Objective: To evaluate the effects of community-based parental education/counseling when combined with usual treatment on children's anemia cure rate.

Design, setting, and participants: A pragmatic cluster randomized clinical trial in children aged 12 to 59 months from 55 villages from the rural Chamrajnagar district in southern India was conducted between November 2014 and July 2015; 6-month follow-up ended in January 2016. Villages were randomly assigned to either usual treatment (n = 27) or to the intervention (n = 28). Among 1144 participating children, 534 were diagnosed as having anemia (hemoglobin levels <11 g/dL and >7.9 g/dL; to convert to grams per liter, multiply by 10) and constituted the study sample in this analysis. Data were analyzed between July 2016 and September 2017.

Interventions: Iron and folic acid (IFA), 20 mg/d, 5 times daily per week, for 5 months (usual treatment) or health worker-delivered education/counseling combined with usual treatment (intervention).

Main outcomes and measures: The primary outcome was anemia cure rate defined as hemoglobin level at or greater than 11 g/dL during follow-up.

Results: Of the children included in the study, the mean age was 30 months, with a slightly higher ratio of boys to girls. Of 534 children with anemia (intervention n = 303; usual treatment n = 231), 517 were reassessed after 6 months (intervention n = 298; usual treatment n = 219) while 17 were lost to follow-up (intervention n = 5 and usual treatment n = 12). Anemia cure rate was higher in children in the intervention group compared with children receiving usual treatment (55.7% [n = 166 of 298] vs 41.4% [n = 90 of 219]). The risk ratio derived through multilevel logistic regression was 1.37 (95% CI, 1.04-1.70); the model-estimated risk difference was 15.1% (95% CI, 3.9-26.3). Intervention-group children demonstrated larger mean hemoglobin increments (difference, intervention vs control: 0.25 g/dL; 95% CI, 0.07-0.44 g/dL) and improved IFA adherence (61.7%; 95% CI, 56.2-67.3 vs 48.4%; 95% CI, 41.7-55.1 consumed >75% of tablets provided). Adverse events were mild (intervention: 26.8%; 95% CI, 21.8-31.9 vs usual treatment: 21%; 95% CI, 15.6-26.4). To cure 1 child with anemia, 7 mothers needed to be counseled (number needed to treat: 7; 95% CI, 4-26).

Conclusions and relevance: Parental education and counseling by a community health worker achieved perceivable gains in curing childhood anemia. Policy makers should consider this approach to enhance population level anemia control.

Trial registration: ISRCTN identifier: ISRCTN68413407.

Figure.. Flow Diagram of the Karnataka Anemia Project 2 Study From Recruitment to Analysis

Sixty clusters were randomly assigned to the intervention and to usual treatment. Lay health workers in the intervention arm were trained to deliver the intervention. After obtaining informed consent from a parent, eligible participants were recruited and enrolled. Parents/guardians were surveyed on socioeconomic and demographic characteristics and anemia risk factors at baseline, and children were followed up at 2 times (at baseline and after 6 months at the end of the trial) for measurements of hemoglobin, ferritin, prior 24-hour dietary recall, and anthropometry. Adherence to iron tablets and adverse events in the last 30 days were recorded at 6 months when the trial concluded. Anemia cure rate, postintervention hemoglobin change, and adherence among children with anemia in both intervention and usual treatment groups were compared. To convert hemoglobin to grams per liter, multiply by 10.