February 2014
Research into reducing tuberculosis deaths: small projects can make a big differenceTB randomisation clinics

We are very proud of the fact that our funds contribute towards medical research. In 2008, £7m of charity funds, mainly from legacies, was made available to the UCLH/UCL National Institute for Health Research Biomedical Research Centre, to be used in conjunction with NHS and other grant money to support ground-breaking clinical and translational research.

But that is by no means the whole story. Some of our discretionary funds also undertake their own work, raising money for areas of interest to our clinicians and to the wider medical world. We are delighted to be able to report back on one such project, where a relatively small sum – just £12,500 – helped the team to undertake a pilot study, which in turn has brought in additional funding for a bigger project.

The team’s project summary and interim report are attached. We will be following the project with interest, and will aim to publish further updates in due course.

TB Fast Track: Summary

Background and study aims

The increased use of antiretroviral therapy (ART) for HIV-positive people in low-income countries has saved many thousands of lives. However, the death rate among people starting ART remains much higher than in wealthy countries, and many of these deaths are due to tuberculosis (TB). TB is hard to diagnose, especially among people with HIV and in low-income countries. The traditional tests for TB are either not very good or are slow to give results, so TB is often identified late, or missed altogether. Newer tests for TB are becoming available. One of these, the LAM test, can be done in clinics using a urine sample. The LAM test works best to pick up TB in people with advanced HIV disease, but even in this group it will miss up to a quarter of TB cases. Other simple tests which predict both TB and risk of death in HIV-positive people include haemoglobin (the red blood cell count) and weight for height (body mass index, BMI).

The aim of this study is to find out whether we can reduce early deaths among HIV-positive people by rapidly identifying and treating those at high risk of having TB, using tests that can be done on site, with immediate results, by nurses in primary health clinics.

What does the study involve?

The study will run in primary health clinics in South Africa. Ten "intervention" clinics will offer the intervention to HIV-positive men and women, aged 18 or above, who are not already taking ART or TB treatment and have CD4 cell counts below 150x106/l.

In the intervention clinics, HIV positive people who are eligible and agree to take part will be assessed by a study nurse. We will do a urine LAM test; take a finger prick blood sample to measure haemoglobin; and measure BMI.  We will assess the risk of TB based on symptoms and these test results. People who are at high risk of TB will start TB treatment as soon as possible, and then ART about two weeks later. People with a medium risk of TB will have further tests sent, according to South African guidelines, with an early review to decide if they should start TB treatment, or start ART. People who are at low risk of TB will start ART as soon as possible.

We will follow up all participants for six months; the main outcome of interest is whether people are still alive at six months. We will compare survival at six months among people taking part in the study in the intervention clinics with similar HIV-positive people in ten "control" clinics, who will be looked after in the standard way according to South African national guidelines.

We will also measure the cost of the intervention, and if the intervention works, we will measure how much it costs per life saved.

What are the possible benefits and risks?

We think that by treating people early for TB, followed by ART, we will reduce the number of people who die around the time of starting ART, and that this will benefit people in the intervention clinics. However, it is likely that some people in the intervention clinics will start TB treatment even though they do not have active TB, and these people may get side effects of TB treatment that they would not have had if they were looked after in the standard way.  We think it is likely that, overall, the intervention will do more good than harm, but we need to do this trial to find out for sure.

Where is the study run from?

The study is run from the Aurum Institute and the Foundation for Professional Development, South Africa. The study is led by the London School of Hygiene and Tropical Medicine.

When is the study starting and how long is it expected to run for?
The study started in December 2012, and we expect to recruit patients until March 2014. Participants will be followed up for 6 months. We hope to complete the study in January 2015.

Who is funding the study?

Funds for the pilot study came in part from our discretionary fund within UCLH Charity. The main study is funded by Global Health Trials, which is a collaboration between the Wellcome Trust, the UK Department for International Development and the UK Medical Research Council.

Who is the main contact?

Professor Alison Grant: This email address is being protected from spambots. You need JavaScript enabled to view it.

Progress Report

Between June and December 2012 we completed the phase I pilot and made significant progress with the phase II pilot.  

Pilot phase I involved visits to all 20 study sites (10 intervention and 10 control) and detailed review of clinic registers and records, looking for individuals with similar profiles to those we would like to recruit for the main study.  We extracted a great deal of (anonymous) information about these selected individuals, and this helped us to trace their journey through the clinic and guided the way we structured the phase II pilot.  The information gathered here will also enable us to establish a baseline and see how much of a difference our intervention will make.

In between phase I and phase II we started training our staff (research nurses and research assistants).  This involved an intensive programme covering the basics of HIV & TB pathology, treatment and side-effects, general research protocols, data management and research ethics.  The specifics of our study were discussed in much greater detail and there was rigorous training in the study protocol and procedures.  Many of the staff had no prior experience of research and, while their training took a little longer, this meant we were able to establish a very solid foundation for their future development.

Pilot phase II began in September 2012.  We started in 4 clinics: two intervention sites, each staffed by a research nurse; and two control sites, each staffed by a (non-clinical) research assistant.  Phase II involved the recruitment of real patients, very similar to those suitable for recruitment to the main study (our criteria were slightly less strict, with an upper CD4 count of 200 cellsx106/l rather than the 150 cells x106/l specified for the main study), but with no changes made to their clinical management.  This enabled us to further explore the organisation of each clinic (they are all different!) and allowed a trial run of the documents and procedures used to collect information from potential participants.  The information gathered was itself extremely useful in guiding the way we structured our interactions with the participants and allowed us to further adapt the questions we were asking.

By the end of December we had completed recruitment of pilot participants in five of the twenty clinics and we had seven more trained staff, ready to recruit.  Two clinics had begun recruiting participants to the main study.

The work proposed to complete this project involves rolling out the phase II pilot to the remaining sites and to complete the training of the remaining research staff.  A major part of the work will be organising and analysing the data collected during the pilot phases and assimilating them into a report for feedback to local health authorities and eventual publication.

We have also applied for further funding in order to continue work on this project and to support a sub-study focusing on participants who die after enrolment.