Welcome to the Health Equity Evidence Centre                

Despite decades of evidence about health and care inequalities, there has been little progress in addressing them. Stand-out successes in the UK include the cross-government strategy on health inequalities in the 2000s which reduced the gap in life expectancy and infant mortality and the roll out of the COVID-19 vaccinations across minority ethnic groups. 

A fundamental reason why progress is slow is the lack of a comprehensive evidence base. For example, whilst we know that the cross-government strategy on health inequalities was successful, we don’t know why – what were the key policy changes that made the difference and what can we learn from them? Part of the problem is that evidence on effective strategies to address inequalities is difficult to navigate because it is substantial and crosses multiple academic disciplines. 

The Health Equity Evidence Centre aims to build the evidence base of what works to address health inequalities. Our work currently focuses on primary care because 90% of contacts with the NHS occur in this domain. 

We use machine-learning software to build Living Evidence Maps of what works to address inequalities in primary care. Rather than using traditional academic methods of literature review, which take time, effort and go out of date quicky, we use machine learning to identify relevant articles quickly and efficiently. We tag these articles and present them in evidence maps which are updated every month and, in this way, they are ‘Living’. The evidence maps are open access; anyone around the world can view them. 

Based on these maps, we produce evidence briefings on what works to address inequalities in primary care. Topics include how to address inequalities in cervical screening and mitigating inequalities in digital and telephone triage. You can see our latest briefings here. 

We handle evidence differently, using the principle of transferrable evidence. Where traditional academic reviewing methods only report evidence that meets a strict eligibility criteria, we believe that it is better to logically transfer evidence from other population groups to develop evidence-informed recommendations. For example, we know that services requiring more patient effort (or agency) to benefit are more likely to increase inequalities – the more hoops a patient must jump through, the greater the inequality. We can apply this principle to many different services – from diabetes care to cancer screening. 

We have a team of researchers who maintain the evidence maps and develop evidence briefings. We also have a citizen’s panel who ensure that we have a strong community voice in what we do. 

We also have lots more on the website, including How-To guides for practitioners and data insights.

This is a start of our journey using machine-learning to build the evidence-base of what works to build a fairer society. We’d love you to join us. Please contact us to discuss opportunities for collaboration, follow us or sign-up to our newsletter.