Recommendations for research

The guideline committee has made the following recommendations for research. The guideline committee's full set of research recommendations is detailed in the full guideline.

1 Universal compared with risk‑based approach to using rapid diagnostic tests

In people with suspected TB, what is the relative clinical and cost effectiveness of universal and risk‑based use of rapid nucleic acid amplification tests?

Why this is important

The guideline committee noted that there were 2 possible approaches to using rapid nucleic acid amplification tests for suspected TB. The current approach is to use them only if TB is strongly suspected and rapid information about mycobacterial species would alter the person's care. Another approach is to use them in anyone with a possible diagnosis of TB. There is a trade‑off between ensuring that all people with active TB are diagnosed and avoiding a large number of false positives, which leads to unnecessary treatment. This trade‑off may lead to differences in the cost effectiveness of each approach. NICE's systematic review of the diagnosis of active TB did not identify any robust evidence on this, nor did the health technology assessment on using nucleic acid amplification tests to detect drug resistance. Cost‑effectiveness studies are needed to improve understanding in this area.

2 Diagnosis in children

Apart from culture, what other diagnostic tests or combinations of tests are effective in establishing an accurate diagnosis of active respiratory TB in children and young people with suspected active TB?

Why this is important

The guideline committee noted the lack of evidence on the diagnosis of active TB in children. The disease manifests differently in children than in adults, and more evidence would have been useful to the committee. Cross‑sectional studies are needed to examine the relative accuracy of different tests, and the most appropriate specimen type for these tests, compared with tests currently in use. In particular, the poor accuracy of many tests in children means that diagnostic strategies that is, combinations of tests, should be investigated, including both tests with high sensitivity and tests based on host response.

3 Treating isoniazid‑resistant TB

For isoniazid‑resistant TB, what is the most effective regimen for reducing mortality and morbidity?

Why this is important

There is little evidence for the treatment of isoniazid resistant TB. This is the most common form of drug resistance in the UK, occurring in 7.5% of TB cases. Currently, treatment is not always successful, even when the recommended drugs are given for the recommended time and there are no adherence issues. It is particularly difficult to treat if there are treatment interruptions or if the central nervous system is involved. Randomised controlled trials are needed to compare different anti‑TB regimens for isoniazid‑resistant TB, assessing mortality, treatment success or treatment failure, rates of relapse and adverse events.

4 Impact of infection control measures on quality of life

What effects does isolation have on the quality of life of people being treated for TB?

Why this is important

Isolation is known to significantly affect a person's quality of life. Despite this, the guideline committee identified no reliable data on the impact of isolation on quality of life. This information is essential in producing economic models that reflect the real costs of isolation. Data on the impact of isolation on quality of life need to be collected and reported.

5 Treatment interruptions caused by adverse events (specifically hepatotoxicity)

For people with active, drug susceptible TB who experience treatment interruptions because of adverse events, particularly hepatotoxicity, what approach to re‑establishing treatment is most effective in reducing mortality and morbidity?

Why this is important

There is little evidence on re‑establishing treatment after interruptions because of adverse events. This is key to ensuring treatment success without relapse or the emergence of drug resistance, but avoiding further adverse events is also important. Randomised controlled trials are needed to compare approaches to re‑establishing treatment for active, drug susceptible TB after it is interrupted because of adverse events, particularly hepatotoxicity. These trials should assess mortality, treatment success or failure, rates of relapse, the recurrence of adverse events and the emergence of drug resistance. Approaches evaluated could compare, for example, restarting regimens with lengthening their duration, as well as sequential reintroduction. Approaches should vary depending on the proportion of doses missed and the stage of treatment (initial or continuation phase) in which the interruption occurred. Prospective observational cohort studies with multivariable analyses may also be useful.