Evidence and science in health policy: Nigel Edwards Q&A

The past year has thrust the relationship between evidence and policy-making into the spotlight. With the subject under debate at this year’s Summit series, before the event we asked Nigel Edwards for his views on it. When are health policy-makers most and least likely to listen to the evidence and science?

Blog post

Published: 09/03/2021

How well do you think the UK uses evidence and science within its policy-making?

The strength of the British system is that we do have a big depth in public health, health services research and other sciences to call upon. The weakness of the system is that the civil service over the years has lost some of its own expertise in how to use and interpret that. It has perhaps not had enough subject matter experts, and instead become overly reliant on a generalist model and external consultants.

The other problem is that politicians have occasionally been reluctant to listen to what the science says. The government in the 1980s, for instance, refused to pay for a large public survey on sexual behaviour, and initially shied away from publicity on dealing with HIV and AIDS. We've had periods of time where the politicians haven’t necessarily got the balance right in the trade-offs between scientific and policy judgements. We've seen that a bit in the current pandemic, haven't we?

How would you consider the UK’s regard for science and evidence during the Covid-19 pandemic?

The scientific advice has been mainly followed, but often with a delay, which has brought costs with it. Churchill said about America that it could always be relied upon to do the right thing, but only after it has exhausted every other possibility. While that’s perhaps unfair here, there has been a repeating pattern of the advice not being fully heeded at first, and then the politicians having to play catch up.

They have found the tension between what the science says and what they politically want to do difficult to manage. For example, SAGE were keen on a circuit-breaker lockdown in September, and then there was that prevarication over Christmas. We then opened the schools and then shut them the next day – which all suggests that the politicians have found the challenge of the advice they got rather difficult.

That’s despite the opinion polls repeatedly showing that the public are much more cautious on easing restrictions than the politicians think they are. So not only have we not used the basic science as well as we might, we've also not necessarily been listening to the people who know about behavioural science.

More generally, are there any areas within the NHS and social care where the evidence doesn't always seem to be listened to?

With the exception of sexual health services, the rule of thumb would appear to be that the more the policy is based on clinical matters, the less the politicians and the policy-makers feel the ability to interfere with it. The more it is about how the NHS is run, structured and organised, the more it is that pre-existing prejudices can be brought into play and the evidence ignored.

I remember Stuart Rose and Don Berwick both being commissioned to give advice on better management in the NHS, and their subsequent reports said some quite uncomfortable things about performance management. But having got the evidence, those reports were not then used, because they contradicted people’s pre-existing ideas.

None of the reform initiatives have been subject to rigorous evaluation, apart from the 2012 Health Act, which they started and then cancelled, mainly because it wasn't really being implemented. Policies that are based on belief don't seem to be very well evaluated.

How might we better deal with issues within the NHS where clear evidence is difficult to produce?

Evaluations can generally tell you whether drug A works better than drug B, but asking whether this model of integrated care works better than that model of integrated care turns out to be a complex and tricky question. One way to deal with that is to be more honest about it and have more quasi-experimental approaches, with evaluation built in.

One question might well be “how does an ICS work?”, and the answer to it could be that “we don't have evidence on that, but we have some clues on what might work, so let's start and evaluate it as we go” – and not be afraid to say six months later that it’s not working terribly well and that we need to change it.

The same might well be true of a complex hospital at home or virtual ward model. You want to continually build in a learning approach rather than just say that you’ve designed the perfect model.

Which health policies in the UK do you think have successfully taken on board the right evidence and science?

The various cancer policies over the last decade certainly fit into that category, as do the national service frameworks from the mid-1990s. The London Stroke Project is another particularly good example.

What can the UK health system learn from how other countries use evidence and science in their decision-making?

There is a cultural difference in other countries. For instance, academic experts such as David Blumenthal or Don Berwick might well leave their work to serve in an administration, before then going back to academia. There can be more traffic backwards and forwards between think tanks, universities and the policy world – the boundaries are more fluid. In the Netherlands, for instance, the chief scientific adviser to the Dutch health care system is a professor in Utrecht, and Israel also has more of that cross-pollination.

But I don’t think we should beat ourselves up too much. The extent to which NHS England and the Department of Health generate and use evidence is good. It's more how it gets from production to implementation, and the number of opportunities to misstep along the way.

The Nuffield Trust Summit series runs from March 9 until March 18. To register and for more programme details, visit the Summit series page.

To watch the full recordings of our Summit sessions so far, including both sessions on evidence and science in health policy, please click here.

Suggested citation

Edwards N (2021) “Evidence and science in health policy: a Q&A”, Nuffield Trust comment.

Comments