Male partner-assisted contact tracing for HIV and tuberculosis in Malawi: an adaptive multi-arm multi-stage randomised trial (mPATCH-TB)

In Africa, tuberculosis and HIV (the virus that causes AIDS) are major health problems, and both affect millions of adults every year. HIV slowly weakens the immune system, and makes individuals very susceptible to developing severe tuberculosis. Although effective treatments are available for both tuberculosis and HIV, many individuals who need treatment don't receive it because they are unaware that they have the infection, or experience difficulties in accessing testing and treatment at clinics. Whilst they are not receiving treatment, they can transmit infection to others in their household or communities, and may die of the consequences of infection.

Men in Africa experience particular difficulties in accessing testing and treatment for tuberculosis and HIV. Our previous research in Africa has shown that two-times as many men than women have untreated tuberculosis in the community. Men experience substantially longer delays than women in accessing diagnosis for HIV and TB (on average up to one year longer). This is for a number of reasons including masculine care-seeking behaviour, competing demands with employment, and perceptions that health centres are unwelcoming for men. We have additionally shown that men are responsible for up to 2/3rds of new tuberculosis infections in Africa. This means that improving timely diagnosis of TB and HIV among men could have important health benefits for men, women and children in the community by reducing infection risk.

Building upon our substantial experience in undertaking HIV and tuberculosis research in sub-Saharan Africa, in this research we aim to rigorously identify strategies that could improve mens' rates of testing for tuberculosis and HIV. Women are substantially more likely than men to attend health centres with symptoms of tuberculosis, and are therefore a key group through which male partners with undiagnosed infection can be accessed. We have previously successfully demonstrated this approach in a recent study in which pregnant women attending clinics were asked to distribute HIV self-test kits to their partners.

In the first stage of this research, we will conduct detailed interviews with men, women and health providers to identify and refine interventions that are likely to be feasible and acceptable when delivered by women to their male partners. We will base interventions upon those that we have experience of testing through previous studies. Types of interventions include: availability of a "male-friendly fast-track clinic"; written information and reminder phone-calls/text messages; home collection of sputum by partners; home HIV test delivery by partners; and varying levels of financial incentives.

We will then conduct an innovative trial to identify which of these interventions has greatest potential to be effective, affordable and safe at increasing male partners' rates of TB and HIV testing. We will do this by recruiting a total of 445 women attending a health centre in Blantyre with symptoms of tuberculosis. Women participants will be tested for TB and HIV, and then randomly allocated to one of the five intervention groups according to the day they attend the clinic. After 30 days, we will assess how many male partners in each group have completed screening for tuberculosis and HIV. (To ensure equity of care, all household members will be able to access screening at the clinic, however the primary focus will be on male partners because of their high risk of disease). We will compare rates of testing between groups, and interventions that do not achieve rates of competing testing higher than the standard of care will be discarded, leaving only interventions with high potential for public health effectiveness.

By the completion of the study, we will have identified 1-2 interventions that can be evaluated in a larger definitive trial that will investigate public health effectiveness in reducing the amount of undiagnosed TB and HIV.

Men are currently a neglected group in responses to the HIV and TB co-epidemics in sub-Saharan Africa, despite having high prevalence of undiagnosed disease, and facing considerable barriers to testing, treatment and prevention services. Through epidemiological and meta-analysis research, we have demonstrated that men in Africa have 2-fold greater prevalence of tuberculosis than women, and can spend up to a year longer than women infectious in the community before achieving a diagnosis. Efforts to engage men in TB and HIV care earlier are likely to have substantial health benefits for men, women and children.

In this development grant, we will use qualitative methods and an innovative multi-arm adaptive study design to identify the most promising candidate interventions to increase TB/HIV screening among male partners of women attending primary health care in Malawi with symptoms of TB. Eligible women will be randomly allocated by day of clinic attendance to either standard of care (including access to a Fast Track TB screening clinic), or one of 4 interventions. Interventions will be based on previous research we have done to increase partners' HIV self-testing study, and will refined by qualitative interviews. Interviews will likely include: phone reminders, partner delivery of sputum cups; HIV self-testing; and financial incentives. The primary outcome will compare the proportion of male partner household contacts that complete TB screening (defined as receipt of result of a sputum Xpert or smear result) within four weeks of recruitment in standard of care with each intervention group.

By the completion of this development grant we will have strong evidence to indicate which interventions have potential to achieve public health effectiveness in Africa. Selected interventions will then subsequently be evaluated in a Phase 3 clinic cluster-randomised trial.

Impact will be generate through the following benefits (see Pathways to Impact for delivery approach):

- Improved knowledge base: The evidence generated by the proposed research will raise the profile of men's access to TB/HIV care and prevention services in sub-Saharan Africa. By understanding the intervention preferences of women and their male partners, and by rigorously evaluating their potential cost-effectiveness, we will document how a comprehensive and responsive public health research programme can be implemented to address a critical evidence gap. The evidence generated will contribute to local, regional and global effort to understand how best to engage men in HIV and TB care. Our expertise in qualitative method, health economics and MAMS trial methods will be augmented, and will contribute to more robust evaluations of public health interventions from researchers in the field in low-resource settings.

- Improved policy-making: We will provide national and international policy-makers with clear evidence of the public health impact of interventions to improve TB and HIV screening rates among men. Our previous research has shown that there is a substantial policy gap in this area. By the conclusion of this research, policymakers will have a greater understanding of the need for cost-effective, acceptable, and scalable interventions that could be rapidly implemented in Malawi and in other high HIV/TB burden countries. Our previous research into HIV self-testing - which has now been included in national HIV guidelines in a number of countries in the region - shows that we have a strong track track record at influencing policy.

- Improved health systems: In Malawi, as in many other countries in sub-Saharan Africa, there is considerable ongoing debate about how men can be reached by already-overstretched health systems. Through this research, we will inform the Malawi National Tuberculosis and HIV Programmes about novel and cost-effective strategies that will support the design of male-friendly health TB and HIV services, and will provide evidence as to how interventions can be implemented through the routine HIV and TB care programmes at scale.