Severe anemia in Ugandan children under 5 years : Implementation research on its prevalence, management and outcomes

Abstract

Background: Severe anemia (SA) is a common clinical condition that is a public health problem in children less than 5 years of age in resource limited countries. Although management of SA is known, inpatient mortality is reportedly much higher than expected, and survivors suffer significant outpatient morbidity in the immediate post-discharge period. This thesis examined the burden, effect of level of care on inpatient mortality and factors associated with severe anemia and poor outpatient outcomes in SA children in Uganda. Methodology: We carried out three sub-studies. Sub-study 1 was a situational analysis of inpatient care practices in the management of SA children. Randomly selected inpatient files over a one-year period were reviewed to determine the burden of SA and the level of care provided to SA patients. Qualitative data was also collected from health workers to provide context and explanation for the level of care provided. Sub-study 2 was an implementation study in which we assessed adherence to clinical guidelines in the management of SA children before and after implementation of a quality improvement (QI) intervention. We then compared inpatient mortality in SA children managed versus those not managed according to clinical guidelines. Sub-studies 1 and 2 were carried out in children aged 0-5 years admitted in Lira and Jinja Regional Referral Hospitals. Sub-study 3 was a prospective cohort study in which children, aged 0-5 years, admitted to Jinja Regional Referral Hospital with SA (hemoglobin (Hb) level ≤ 5.0 g/dl) were evaluated for a range of demographic, clinical and laboratory factors at admission and followed up for 6 months after hospitalization. In this sub-study logistic analysis was used to determine factors associated with SA and readmissions or death in the immediate post-discharge period. Results: Sub-study 1: Overall, 574/2,275 (25.2%) selected files of children admitted in Lira and Jinja hospitals were assigned a diagnosis of SA. However only 204 (8.9%) were diagnosed appropriately with pre-transfusion Hb ≤ 5.0 g/dL. Blood transfusion in the management of SA children was given appropriately as per guidelines in only 245 (44.5%) of the SA cases. SA children transfused appropriately had lower inpatient mortality compared to those transfused inappropriately, 2.9% vs. 7.2%, p = 0.03. Major issues identified by health workers as affecting management of SA children included late presentation to hospital and unreliable availability of equipment for measurement of Hb. Sub-study 2: There was an improvement in the level of care after QI intervention with more children being managed according to clinical guidelines compared to the period before, 218/510 (42.7%) vs 158/621 (25.4%) (p < 0.001). Overall, children managed according to clinical guidelines had reduced risk of inpatient mortality compared to those not managed according to clinical guidelines, [OR 0.28, (95%, CI 0.14, 0.55)]. Sub-study 3: In a cohort of 284 children with SA admitted at Jinja hospital, common etiological factors were malaria parasitemia 103 (36.5%), blackwater fever (BWF) 93 (32.7%) and vitamin B12 deficiency 44 (15.5%). Factors associated with SA included malaria parasitemia, OR, 4.3; (95% CI, 1.4 to 13.8), unit increases in total white blood count (OR, 1.3; 95% CI, 1.1 to 1.4); C-reactive protein, OR, 1.8; (95% CI, 1.3 to 2.4); and ferritin, OR, 2.7; (95% CI, 1.9 to 4.0). Amongst the 279 SA children discharged from hospital 128 (45.9%) were readmitted at least once while 22 (7.9%) died during the 6 months follow up period. Malaria and recurrence of SA were the most common reasons for readmissions. After adjusting for age, sex, nutritional status and parasitemia SA children with BWF had higher risk of readmissions, HR 1.68 (95% CI, 1.1, 2.5), and a greater risk of death, HR 3.37 (95% CI, 1.3 to 8.5), compared to those without blackwater fever (BWF). Conclusion: Severe anemia is a common diagnosis with one in four children admitted to Lira and Jinja hospitals being assigned this diagnosis. The diagnosis of SA is not always supported by laboratory confirmation of Hb level nor management done according to clinical guidelines. Consequently, over half of suspected SA patients were given a blood transfusion presumptively. However, management of children with suspected severe anemia according to clinical guidelines reduced the risk of inpatient death by about 70%. Malaria and inflammation were factors associated with severe anemia in children in Jinja Hospital. Post-discharge, nearly half of SA children required readmission in the immediate 6 months while one in 12 died during this period. Blackwater fever was associated with increased risk of readmission and death during the post-discharge period. Efforts to reduce inpatient mortality in SA children in resource-limited settings should focus on training and supporting health workers to adhere to clinical guidelines. Given that malaria was the most common reason for readmission, post-discharge malaria chemoprophylaxis is urgently needed for SA children living in malaria endemic areas. Blackwater fever (BWF) remains poorly understood, and pathogenesis of BWF should be studied further to determine why it is associated with increased readmissions.