Working together to improve research methods

13 Sep 2017 03:20 PM

Blog posted by: Sam Rowley, programme manager, MRC and Rosie Lovett, senior scientific adviser, NICE, 13 September 2017.

Sam and Rosie discuss how the MRC and NICE have teamed up to improve research methods in health and social care.

Sam:  The Medical Research Council (MRC) and NICE have the same goal -- to improve the quality of evidence used in health and social care.

To generate good quality evidence we need to practice research with high standards.

Our Methodology Research Programme (co-funded by the National Institute for Health Research) seeks to improve how research is done.

NICE are key partners of the programme. They work with us to identify the research that will bring the most benefit to patients.

So far the programme has funded projects ranging from the assessment of using computer software to predict how an infectious disease might spread, to innovative ways of making clinical trials more efficient.

By working together we can make sure the evidence underpinning health and social care policy is of the highest quality.

We can also help NICE to solve the challenges their committees may face. For example we are currently asking experts for their help in the development of methods to better assess data.

Rosie: NICE committees often have to decide whether new treatments are better than existing ones. These decisions are not made lightly, and require the best quality evidence to inform them.

The “gold standard” of evidence usually comes from randomised controlled trials (RCTs) where people are allocated to the treatment or control group at random. Neither the patient nor their doctor will know which group they are in. This means we can be confident that any difference between the groups is caused by the treatment, the results are unbiased.

However, RCTs are not always available, which means our committees must then use other kinds of evidence, such as historical studies or patient registries. This kind of evidence is called ‘observational data’ and it is known to have a higher risk of bias.

Observational data is still useful, but we need better methods of analysing its worth and reliability.

The MRC has joined us in calling for academics and researchers to help us address this challenge.

And we also welcome direct partnership applications from researchers.