Missing information is a major drawback in analyzing data collected in many routine health care settings. Multiple imputation assuming a missing at random mechanism is a popular method to handle missing data. The missing at random assumption cannot be confirmed from the observed data alone, hence the need for sensitivity analysis to assess robustness of inference. However, sensitivity analysis is rarely conducted and reported in practice. We analyzed routine paediatric data collected during a cluster randomized trial conducted in Kenyan hospitals. We imputed missing patient and clinician-level variables assuming the missing at random mechanism. We also imputed missing clinician-level variables assuming a missing not at random mechanism. We incorporated opinions from 15 clinical experts in the form of prior distributions and shift parameters in the delta adjustment method. An interaction between trial intervention arm and follow-up time, hospital, clinician and patient-level factors were included in a proportional odds random-effects analysis model. We performed these analyses using R functions derived from the jomo package. Parameter estimates from multiple imputation under the missing at random mechanism were similar to multiple imputation estimates assuming the missing not at random mechanism. Our inferences were insensitive to departures from the missing at random assumption using either the prior distributions or shift parameters sensitivity analysis approach.

Background: Routine clinical data are widely used in many countries to monitor quality of care. A limitation of routine data is missing information which occurs due to lack of documentation of care processes by health care providers, poor record keeping, or limited health care technology at facility level. Our objective was to address missing covariates while properly accounting for hierarchical structure in routine pediatric pneumonia care. Methods: We analyzed routine data collected during a cluster randomized trial to investigating the effect of audit and feedback (A&F) over time on inpatient pneumonia care among children admitted in 12 Kenyan hospitals between March and November 2016. Six hospitals in the intervention arm received enhance A&F on classification and treatment of pneumonia cases in addition to a standard A&F report on general inpatient pediatric care. The remaining six in control arm received standard A&F alone. We derived and analyzed a composite outcome known as Pediatric Admission Quality of Care (PAQC) score. In our analysis, we adjusted for patients, clinician and hospital level factors. Missing data occurred in patient and clinician level variables. We did multiple imputation of missing covariates within the joint model imputation framework. We fitted proportion odds random effects model and generalized estimating equation (GEE) models to the data before and after multilevel multiple imputation. Results: Overall, 2,299 children aged 2 to 59 months were admitted with childhood pneumonia in 12 hospitals during the trial period. 2,127 (92%) of the children (level 1) were admitted by 378 clinicians across the 12 hospitals. Enhanced A&F led to improved inpatient pediatric pneumonia care over time compared to standard A&F. Female clinicians and hospitals with low admission workload were associated with higher uptake of the new pneumonia guidelines during the trial period. In both random effects and marginal model, parameter estimates were biased and inefficient under complete case analysis. Conclusions: Enhanced A&F improved the uptake of WHO recommended pediatric pneumonia guidelines over time compared to standard audit and feedback. When imputing missing data, it is important to account for the hierarchical structure to ensure compatibility with analysis models of interest to alleviate bias.

Background: The health seeking behavior in Kenya raises concerns in malaria case management at the private sector. Adherence to the national guidelines for the diagnosis, treatment and prevention of malaria is key in management of the disease. Presumptive treatment remains a major challenge in Kenya, especially in the private sector, with major gaps in literature identified on predictors of this treatment. Mixed-effects regression modelling considers county clustering, is more accurate in prediction and is more efficient and flexible. Methods: The study design was a cross-sectional, nationally representative, retail outlet survey secondary data analysis. The study populations included the health care providers in the retail outlets sampled randomly in both the rural and urban settings in Kenya. The primary outcome of interest was the proportion of health care providers who treated patients presumptively. Multivariable analysis was conducted for the significant variables, adjusting for clustering at the county level to determine the predictors of presumptive treatment. The best fitting model was examined using the Akaike Information Criterion (AIC). Results: Out of the 333 health care providers who treated patients, 190 (57%) treated patients presumptively. From the mixed effects logistic regression model, the predictors of presumptive treatment of uncomplicated malaria were case management training (AOR = 0.44; 95% CI = (0.18 – 1.09)), asked signs or symptoms (AOR = 0.19; 95% CI = (0.10 - 0.37)) and results presented (AOR = 0.08 95% CI = (0.03 - 0.19)). Conclusions: Presumptive treatment of uncomplicated malaria remains a challenge in the private retail sector. Malaria case management training and health care providers asking of signs and symptoms and results presented predicts presumptive treatment. To address the issue of presumptive treatment of Malaria, strengthening of malaria case management training is key for health care providers in the private sector.

According to the Kenya National School-Based Deworming program launched in 2012 and implemented for the first 5 years (2012–2017), the prevalence of soil-transmitted helminths (STH) and schistosomiasis substantially reduced over the mentioned period among the surveyed schools. However, this reduction is heterogeneous. In this study, we aimed to determine the factors associated with the 5-year school-level infection prevalence and relative reduction (RR) in prevalence in Kenya following the implementation of the program. Multiple variables related to treatment, water, sanitation, and hygiene (WASH) and environmental factors were assembled and included in mixed-effects linear regression models to identify key determinants of the school location STH and schistosomiasis prevalence and RR. Reduced prevalence of Ascaris lumbricoides was associated with low (< 1%) baseline prevalence, seven rounds of treatment, high (50–75%) self-reported coverage of household handwashing facility equipped with water and soap, high (20–25°C) land surface temperature, and community population density of 5–10 people per 100 m2. Reduced hookworm prevalence was associated with low (< 1%) baseline prevalence and the presence of a school feeding program. Reduced Trichuris trichiura prevalence was associated with low (< 1%) baseline prevalence. Reduced Schistosoma mansoni prevalence was associated with low (< 1%) baseline prevalence, three treatment rounds, and high (> 75%) reported coverage of a household improved water source. Reduced Schistosoma haematobium was associated with high aridity index. Analysis indicated that a combination of factors, including the number of treatment rounds, multiple related program interventions, community- and school-level WASH, and several environmental factors had a major influence on the school-level infection transmission and reduction.

Composite scores are useful in providing insights and trends about complex and multidimensional quality of care processes. However, missing data in subcomponents may hinder the overall reliability of a composite measure. In this study, strategies for handling missing data in Paediatric Admission Quality of Care (PAQC) score, an ordinal composite outcome, were explored through a simulation study. Specifically, the implications of the conventional method employed in addressing missing PAQC score subcomponents, consisting of scoring missing PAQC score components with a zero, and a multiple imputation (MI)-based strategy, were assessed. The latent normal joint modelling MI approach was used for the latter. Across simulation scenarios, MI of missing PAQC score elements at item level produced minimally biased estimates compared to the conventional method. Moreover, regression coefficients were more prone to bias compared to standards errors. Magnitude of bias was dependent on the proportion of missingness and the missing data generating mechanism. Therefore, incomplete composite outcome subcomponents should be handled carefully to alleviate potential for biased estimates and misleading inferences. Further research on other strategies of imputing at the component and composite outcome level and imputing compatibly with the substantive model in this setting, is needed.