Do we have more data than insight?

Calls for new ways to use information to support better health services are fairly common. However, much less attention is paid to the skilled analysts who can make sense of the data being produced. Investing in those people is essential if we are going to use information to create a more efficient and effective health care system.

My recent thought paper was based on interviews from across the UK to assess the state of analytical capability in health care – and identify some potential solutions. There was consensus that the past few years have seen enormous changes in the way we use data and information, helping transform care delivery not only in terms of individual care, but also the management of health care services. For example, in England, many critical items on the policy agenda make important demands on new ways to use information – whether in looking for efficiencies following Lord Carter's review of efficiency in hospitals, the creation and testing of new care models, tackling problems of flow, or creating learning health care systems.

To deliver these benefits, a skilled workforce that is able to manipulate, analyse and interpret data is essential. However, health services in the UK often cannot access the analytical skills that they require to support decision making. In some parts of the NHS, there are too few analysts, with the available people being too heavily focussed on routine data manipulation (‘lifting and shifting’), rather than on the other analytical tasks required to improve service delivery. There are excellent analysts, but often they work in small units with little chance to learn from peers and develop professionally; as a result, skills can be limited. In addition, the term ‘analyst’ itself can be a little misleading, since very often we are talking about a range of skills that exists in a team rather than an individual.

To improve analytical capability, approaches must be multi-faceted and long term, with action at a number of levels in the health service. Some of it is about supporting professional development and training that is focused on the application of analysis to improve the quality and efficiency of health care. Some analytical networks have already been formed, and are important to help overcome the fragmentation and isolation that analysts in health services can face. But coverage across the country is patchy – we’re working with the Association of Professional Health Care Analysts to help it grow and develop.

Since analytical communities are fragmented, there are strong arguments in favour of working across current organisational boundaries. This could be in regional groups or through ad-hoc consortia of providers, commissioners or regulators. Linked with this is the ability to create environments that help germinate new analytical approaches and support their adoption in health care. So for example, in the Improvement Analytics Unit, we are working with NHS England to support the adoption of the analytical methods needed to evaluate new models of care. We are doing this by bringing together people with the analytical skills needed in one unit, spanning two organisations.

One of the most effective ways to improve the skills of analytical teams is to establish links with topic experts, who typically work in academic settings. There are examples where senior academics have worked with the NHS on problem solving outside of the traditional commissioned research models, including projects jointly developed by the NHS and academic groups. ‘Researcher-in-residence’ models are increasingly being used. Such initiatives are to be encouraged, since they represent one way to bridge the sometimes conflicting demands of an academic career and the needs of the service.

But while the supply of analytical skills must be addressed, it needs to be appropriately matched with demand for analysis. Many of the people I interviewed suggested that senior managers are either unaware of quantitative methods, or lack confidence in the potential of using data to help to improve care for patients. There are a number of ways that this can be addressed. At a local level, for instance, this can be done through training or raising awareness, with compelling examples of how good analysis can enable better decision making. Other approaches include assessing or auditing analytical development, as well as making more explicit central requirements about the use of particular analytical methods.

A recurring theme from the interviews was the importance of analytical leadership – and, in particular, that tribe of people who understand the possibilities of good analysis yet can also engage with managers at the highest levels to influence and shape demand. Those organisations that have a better developed analytical workforce tend to have strong analysts who are themselves influential leaders within their organisations. To develop the profile of good quality analytics in health care, we need to recognise the current generation of leaders in analytics as well as investing in the next generation.

To create more efficient and effective health care systems, we need to build a better infrastructure for health intelligence. This means not just investing in the new technology of health information, but also in the people who know how to make the most of that information to support a better service and deliver better care to patients.

Martin Bardsley is a Senior Fellow at the Health Foundation and the Nuffield Trust