Martin MBONYE

Thesis summary

Infectious diseases continue to be a major public health problem causing nearly 35% of all deaths globally. Moreover, there is a critical shortage of health workers to address this disease burden with over 30% of the needed critical workforce vacant in most health facilities and most of physicians’ positions unfilled. In absence of physicians, the mid-level health providers are performing roles usually reserved for them such as antiretroviral therapy (ART) initiation. These providers however report insufficiently trained to handle these tasks. The literature on the effect of training programs for mid-level providers is sparse and so is the improved patient-level Health Management Information System (HMIS) data to evaluate these training programs and use in estimating disease burden faced by these health providers.

An analysis of the pre-intervention disease burden in the Integrated Infectious Diseases Capacity Building Evaluation project (IDCAP) health facilities showed over 500 distinct diagnoses were made at primary care facilities. Infectious diseases overwhelmed the disease burden and were diagnosed 76.3% of the time with malaria (48.3%) being the most frequently diagnosed illness. A high proportion of patients were diagnosed with multiple illnesses while use of laboratory diagnostics was limited. Further analysis indicated that malaria parasitemia positivity rate (MPPR) prior to implementation of IDCAP training interventions was high among outpatient adults (56%-64%) and children aged under five years (≅ 40%). An analysis of data collected during the last year of IDCAP implementation also showed a high MPPR among infants aged less than six months (36.1%).  

We evaluated pre-intervention health worker practices in malaria case management. We observed gross underutilization of laboratory support during disease diagnosis. Even when the laboratory tests were performed, with exception of malaria, a lot of positive or abnormal results did not have a record of diagnosis for a related illness. In many instances, clinicians ignored laboratory results and prescribed antimalarials for patients with a confirmed negative malaria test result. Clinicians were rarely taking patient’s history and conducting basic physical examinations. This is despite the fact that making an accurate diagnosis needs in-depth patient’s history and conducting rigorous physical examination, and for illnesses that require laboratory support, conducting a laboratory exam.

An analysis of practices and predictors of malaria diagnosis and treatment in a vulnerable population of infants aged under six months was conducted. We noticed high rates of malaria suspicion in this population in which malaria is perceived to be rare. However, the testing rates were quite low while a lot of patients with confirmed negative malaria test results and those without any laboratory confirmation were often prescribed antimalarials. Off-label antimalarial prescriptions were common among infants aged less than four months and weighing <5kgs. Our results showed that, malaria in this population may not be rare as previously thought, while the common practice of off-label Artemisinin-based Combination Therapy (ACT) use, low testing of malaria suspects, and prescription of antimalarials to patients with negative malaria test results shows the challenge health workers face in managing malaria in this group in absence of evidence-based guidelines.

To improve facility performance in infectious diseases care, IDCAP developed and tested two training interventions for mid-level health workers. One was a classroom-based course for two mid-level practitioners (MLP) called Integrated Management of Infectious Diseases (IMID) and the other was an educational outreach program with an on-site support (OSS) and continuous quality improvement (CQI) activities for all health workers. OSS was implemented in a phased approach beginning with 18 of 36 health facilities randomly assigned to the first phase and the rest in the second phase. We tested whether this training had an effect on case management of fever and malaria. A combination of IMID and OSS significantly increased the estimated proportion of patients who received an appropriate antimalarial among those who were prescribed any antimalarial and reduced the proportion of patients who were prescribed an antimalarial despite a negative diagnostic test result for malaria. IMID alone did not have a significant effect on any facility performance indicators for case management of malaria, nor was the incremental effect of OSS statistically significant.

We did a secondary analysis to test for the effect of OSS on workload and facility performance in infectious diseases care separately for two subgroups of health providers 1) with IMID training (IMID) and 2) without (No-IMID). We analyzed only records with data of which cadre of provider that managed the patient. We noticed that effect of OSS were heterogeneous across subgroups. The effects of OSS among the IMID subgroup tended to be statistically significant for indicators that reflected clinical decision-making and among the No-IMID subgroup for indicators that required a broader team effort to identify and manage patients appropriately.

Amidst a severe shortage of health workers and logistics in Uganda, health workers face a daunting challenge to effectively tackle a huge burden of infectious diseases. This study showed that a training for mid-level health workers combining both classroom and educational outreach efforts would be effective in improving facility performance in infectious diseases care. However, our results also showed a significant effect of educational outreach on some aspects of facility performance.