A mid-level health manager intervention to promote uptake of isoniazid preventive therapy among people with HIV in Uganda: a cluster randomised trial

HIV epidemiology, prevention, treatment, and implementation strategies for public health

The global HIV response has made tremendous progress but is entering a new phase with additional challenges. Scientific innovations have led to multiple safe, effective, and durable options for treatment and prevention, and long-acting formulations for 2-monthly and 6-monthly dosing are becoming available with even longer dosing intervals possible on the horizon. The scientific agenda for HIV cure and remission strategies is moving forward but faces uncertain thresholds for success and acceptability. Nonetheless, innovations in prevention and treatment have often failed to reach large segments of the global population (eg, key and marginalised populations), and these major disparities in access and uptake at multiple levels have caused progress to fall short of their potential to affect public health. Moving forward, sharper epidemiologic tools based on longitudinal, person-centred data are needed to more accurately characterise remaining gaps and guide continued progress against the HIV epidemic. We should also increase prioritisation of strategies that address socio-behavioural challenges and can lead to effective and equitable implementation of existing interventions with high levels of quality that better match individual needs. We review HIV epidemiologic trends; advances in HIV prevention, treatment, and care delivery; and discuss emerging challenges for ending the HIV epidemic over the next decade that are relevant for general practitioners and others involved in HIV care.

The importance of using disease causal models in studies of preventive interventions: Learning from preeclampsia research

OBJECTIVE

Interventions aimed at preventing disease are commonly studied as strategies of primary or secondary prevention. Unfortunately, this dichotomy can be misleading, and studies might unknowingly exclude people at high risk of the disease that could benefit from the intervention. Here I use the example of aspirin for prevention of preeclampsia to illustrate this problem.

METHODS

I use directed acyclic graphs to represent several causal models of aspirin and preeclampsia, each making different assumptions regarding the causal relation between previous preeclampsia, aspirin, and subsequent preeclampsia. Afterwards, I discuss the implications of each model.

RESULTS

Aspirin started being recommended to pregnant women that had presented preeclampsia in previous pregnancies, but not to women at high risk due to other factors. Studies started evaluating aspirin in women at high risk due to these other causes and found it also reduced the risk of preeclampsia in them. Thanks to a shift towards risk-based interventions, guidelines started recommending aspirin to all women considered at high risk of preeclampsia. Furthermore, recent studies have begun using blood markers in women without classic risk factors to identify additional women that might benefit from aspirin. With such advances, performing "secondary prevention" once the first event occurred will increasingly represent a failure to intervene on time.

CONCLUSIONS

Explicitly illustrating disease causal models helps to identify those individuals that are most likely to benefit from risk reduction, regardless of whether they were previously afflicted by the disease. This is beneficial when designing studies and when implementing preventive interventions.

Geographical, social, and political contexts of tuberculosis control and intervention, as reported by mid-level health managers in Uganda: 'The activity around town'

Training district-level health officers and other mid-level health system managers revealed multiple contextual factors across political, administrative, and social axes affecting tuberculosis (TB) and TB control in Uganda. Individual relationships between local health, political, and media leaders affect efforts to inform the public and provide services, yet greater administrative coordination between national-level logistics, implementing partner funding, and local needs is required. Social challenges to TB control include high population mobility, local industries, poverty with high-density living and social venues, and misinformation about TB. Capitalizing on implementation knowledge and sharing data can overcome social geographic challenges to TB-prevention planning through strategic healthcare capacity-building at the district level.

Defining and estimating effects in cluster randomized trials: A methods comparison

Across research disciplines, cluster randomized trials (CRTs) are commonly implemented to evaluate interventions delivered to groups of participants, such as communities and clinics. Despite advances in the design and analysis of CRTs, several challenges remain. First, there are many possible ways to specify the causal effect of interest (eg, at the individual-level or at the cluster-level). Second, the theoretical and practical performance of common methods for CRT analysis remain poorly understood. Here, we present a general framework to formally define an array of causal effects in terms of summary measures of counterfactual outcomes. Next, we provide a comprehensive overview of CRT estimators, including the t-test, generalized estimating equations (GEE), augmented-GEE, and targeted maximum likelihood estimation (TMLE). Using finite sample simulations, we illustrate the practical performance of these estimators for different causal effects and when, as commonly occurs, there are limited numbers of clusters of different sizes. Finally, our application to data from the Preterm Birth Initiative (PTBi) study demonstrates the real-world impact of varying cluster sizes and targeting effects at the cluster-level or at the individual-level. Specifically, the relative effect of the PTBi intervention was 0.81 at the cluster-level, corresponding to a 19% reduction in outcome incidence, and was 0.66 at the individual-level, corresponding to a 34% reduction in outcome risk. Given its flexibility to estimate a variety of user-specified effects and ability to adaptively adjust for covariates for precision gains while maintaining Type-I error control, we conclude TMLE is a promising tool for CRT analysis.