Isoniazid preventive therapy (IPT) is a well studied clinical intervention for primary and secondary prevention of active tuberculosis (TB) after infection with Mycobacterium tuberculosis. It is widely used in industrialized countries in TB outbreak management, focusing on high risk groups such as close contacts in the family, in congregate settings, and in the workplace amongst others. Individuals infected with Human Immunodeficiency Virus (HIV) have a markedly higher risk of acquiring a TB-infection and developing consequently active TB, making HIV-infected individuals a target population for IPT. Studies of IPT in HIV infected persons in the nineties demonstrated the efficacy of IPT in the prevention of active TB in Sub -Saharan Africa and more recent studies suggest that the protective effect remains present in individuals on antiretroviral therapy.

Despite the proven efficacy of IPT this intervention has not been taken up by most HIV and TB control programmes in Africa where the burden of TB/HIV is highest. The reasons for the low uptake of IPT are many and varied but include fears of expansion of isoniazid resistance and subsequently the development of multi -drug resistant TB with widespread use of IPT. Additionally screening protocols for excluding active TB and selecting persons for IPT have not been uniformly agreed upon. There have also been concerns that programmes designed to provide IPT may shift TB control programmes from their primary responsibility of finding and treating active TB. Finally it has been unclear as to which programme, between the HIV and the TB control programme, has the primary responsibility of managing the provision of the IPT intervention.

The World Health Organization and other technical agencies engaged in global TB control have recently re-emphasized the need to scale up IPT. In this proposal we outline an operational research study to evaluate the introduction of IPT at community level and to measure its effectiveness at preventing TB. The study is based on the context of expansion of Community-Based Direct Observed Therapy Short Course (CB-DOTS), home-based care and the concept of HIV prevention with positives (PwPs), where there is a real opportunity to focus on the household as a source of HIV-associated tuberculosis.

The study is designed as a cluster randomized trial. It compares the incidence of TB in household contacts including children under 5 of identified TB/HIV co-infected patients, who received IPT through proactive community intervention and those in a control group where the community was handled in the "usual way". In the intervention group household contacts of index cases of HIV positive, smear positive PTB will be visited at home and consenting contacts will be screened for active TB using a simple questionnaire. Those found to be fit will receive isoniazid 300mg (5 mg per Kg for children) once daily for 6 months, regardless of the HIV-status. Those found not to be fit will be referred for further evaluation at the nearest TB diagnostic centre. In the control group, routine care following national guidelines will be offered. This consists of contact invitation and assessment of eligibility for IPT, especially, in children less than 5 years. Both groups will be followed up monthly through household visits. Follow up will be for a total of 24 months including the six months when IPT is provided.

A confidential HIV screening test will be provided to all consenting contacts in both intervention and control group after appropriate counseling.

The primary outcome is the incidence of TB in the intervention and control household contacts. The difference in incidence between the two groups is a measure of efficacy of the intervention. In addition the efficacy of the intervention will be estimated stratified by HIV status of household contacts if data allows. Secondary outcomes are the incidence of adverse events, the incidence of TB-related symptoms, measures on the uptake of IPT (proportion of contacts starting and discontinuing IPT, treatment adherence) and programmatic indicators, i.e. percentage of persons eligible for IPT and resources needed.

Index cases

• New sputum smear positive pulmonary tuberculosis (PTB) patients initiating treatment for TB
• HIV infected
• Resident in the catchment area of the TB diagnostic facility
• Written informed consent
• Age equal or greater than 15 years.

Contacts

• Household contacts of a HIV infected smear positive PTB
• No plans to relocate to another area for at least the next two years
• Asymptomatic (Fit) with no cough in last 2 weeks, fever, weight loss (or failure to gain weight in children), haemoptysis or chronic diarrhea at least for the last one month.
• Written informed consent