Study Design: Stratified, matched, cluster-randomized, controlled trial

Unit of Randomization: Healthcare facility

Study Duration: 3 years; prevalence of latent Tuberculosis infection (LTBI) in healthcare workers (HCWs) will be at measured at baseline, and LTBI incidence will be measured among susceptible HCWs at 12 and 24 months. Secondary outcomes will be measures at 0 (pre-intervention) 6, 12, 18, and 24 months. In year three, results will be analyzed and disseminated.

Study Components: Assessment of institutional safety culture; observations/audits of Tuberculosis (TB) patient flow (wait times) and HCW TB infection control (IC) practices; documentation of time intervals for processing sputum smears and initiation of TB treatment; facility assessments; random allocation and implementation of enhanced Tuberculosis infection control (TB IC) package; testing of HCWs to determine LTBI at 0, 12, 24 months; cost evaluation of intervention.

Sample Size: For the cluster randomized design, we estimate that 11 clusters per group will allow for 77 percent (%) power to identify a 30% reduction in LTBI incidence in the intervention vs. control clusters. This assumes LTBI incidence 5% per year in the control group, design effect for clustering of 2.0, and cluster size of 300 (average 600 HCW per cluster with 50% LTBI prevalence at baseline).

TB remains a cause of substantial morbidity and mortality, affecting an estimated 13.7 million persons and resulting in 1.8 million deaths worldwide. TB transmission has been well-documented in a wide variety of healthcare settings. Moreover, the global expansion of HIV care programs may inadvertently increase TB transmission in healthcare settings by congregating highly susceptible individuals with those likely to have TB disease. The urgency of reducing TB transmission in healthcare facilities has been intensified by the emergence of drug-resistant TB strains, including extensively resistant TB strains, and the high mortality of these strains in people living with human immunodeficiency virus (HIV).

Healthcare workers are at higher risk of both TB infection and disease compared to the general population, with estimates that 63-94% of TB infection and up to 89% of TB disease in this population is due to occupational exposure.

The World Health Organization (WHO) has identified institutional TB IC as one of the core "3 I's" interventions required to reduce the burden of TB among people living with HIV. Although TB IC guidelines exist and a "package" of interventions has been shown to successfully interrupt TB outbreaks in U.S. hospitals, there is limited information on feasibility, impact or cost of TB IC programs in middle- and low-income countries where TB burdens are high and nosocomial TB transmission has been well-documented.

Currently recommended TB IC strategies are complex and multi-faceted and include: administrative controls (e.g., early identification, treatment, and isolation or cohorting of infectious TB patients); effective engineering/environmental controls (such as, general ventilation or ultraviolet germicidal irradiation); and appropriate use of respiratory protection (N-95 particulate respirators) to protect HCWs. Implementation of many of these recommended measures require administrative/managerial support and sustained behavior change of frontline staff; some require substantial healthcare expenditures. There is an urgent need for simple, evidence-based and cost-effective strategies to help guide implementation of TB IC programs and reduce institutional TB transmission in resource-limited settings where TB and HIV are endemic. A recent call to address gaps in the TB IC evidence base identified key priorities including operational research to investigate the efficacy and cost-effectiveness of TB IC measures, and behavioral research to develop effective strategies to inform, motivate and provide skills to HCWs to implement and sustain effective airborne IC procedures and practices. This study directly addresses these identified priorities.

At root, ensuring good implementation of all TB IC procedures is a challenge of HCW behavior change. Even appropriate use of simple environmental control measures, requires a substantial element of behavior change to ensure effectiveness; for example, keeping needed windows open, ensuring needed fans are on and directed appropriately, and ensuring performance of routine maintenance checks of equipment. In this evaluation, the proposed intervention package focuses on tools and techniques that support the development of an institutional culture of safety and HCW behavior change regarding TB IC practices.

The theoretical framework for this intervention package is based on evidence showing that certain interventions favorably impact HCWs' IC practices and related patient outcomes, specifically 1) audits and feedback of IC performance and outcome data, 2) participation in IC collaborative (including mentoring), and 3) use of standardized IC checklists. Audit and feedback of performance have been used for decades as a strategy to improve implementation and adherence to clinical practice guidelines. Performance feedback has similarly been shown to be an effective intervention for improving IC practices. Also, there is a growing body of evidence to support the use of simple, evidence-based checklists as an effective IC strategy. When studied, use of checklists has fostered adoption of best practices, resulting in significant and sustained reductions in the targeted healthcare-associated infections (such as, surgical site infections and catheter-related bloodstream infections). Checklists are intended to be practical, easy-to-use tools that are designed to improve recall, prompt providers to perform recommended infection prevention steps, and make clear minimum expectations for IC. While the checklist approach has been used widely in other aspects of hospital IC, it has not yet been used widely for airborne IC. Lastly, collaboratives have been used to address a variety of health care issues and when studied in randomized trials, their efficacy has ranged from -16% to 70%. In Thailand, IC collaboratives have been associated with lower rates of healthcare-associated infections and better IC practices. In this study, we propose to use a robust study design to implement a multi-faceted TB IC package and to assess the impact of its implementation on TB transmission in hospitals and clinics where care is provided to patients with TB or other potential airborne respiratory infections.

Inclusion Criteria:

• General provincial or regional hospital of at least 300 beds OR
• TB or Respiratory hospitals of at least 100 beds
• Geographic location that allows for ground transport of blood specimens to a designated reference laboratory within 16 hours of blood draw
• Hospital director is willing to commit staff time to study participation, including designating personnel to oversee TB IC and EnTIC study activities, as evidenced by a letter of support for the study

Exclusion Criteria:

• Specialty hospitals (such as, pediatric, infectious diseases, maternity)
• Recent (within the past 3 years) or current participation in a TB IC initiative

This study will be conducted at 22 hospitals, 10 in Thailand and 12 in Vietnam. The study facilities will be hospitals that typically provide care to patients with confirmed or suspected TB, namely TB or general hospitals.