How do we decide what works in wellbeing?

Posted in: Evidence and policymaking, Health

Emily Rempel is an interdisciplinary PhD student in the Department of Psychology and the Institute for Policy Research at the University of Bath.

The region around the University of Bath is a relatively ‘well’ local authority. By this I mean that Bath and North East Somerset (B&NES) consistently ranks at or above average on the ONS measures for wellbeing, which include population scales of life satisfaction, anxiety, worthwhileness and happiness. B&NES, and other surrounding authorities, are committed to providing services that address these kinds of measures and seek to increase the wellbeing of the people that use their services. Examples include the Wellbeing College, Developing Health & Independence and Second Step Housing Association. While one can assume there is significant worth in increasing the wellbeing of a community for that community’s own cohesion and happiness, there are also more economically minded ideals behind the push for wellbeing. The main driving concept is that happier, more stable people are healthier people and that healthier people use public services less and ultimately save money for institutions like the NHS. This sounds all well and good: people are happier, governments are spending less money and we all end up better off. But the reality is far more complex. That basic assumption that happiness leads to cost savings requires, for lack of a better word, evidence.

What constitutes evidence in wellbeing is both politically and practically challenging. An essential first question is: what do we mean by wellbeing? Professor Sarah White, Professor of International Development and Wellbeing at the University of Bath, has written extensively on this issue. She argues that wellbeing includes a variety of factors from subjective assessments of happiness to objective ‘quality of life’ measures1. Professor White also describes key differences between individual concepts of wellbeing and collective wellbeing – personal happiness versus national economic health, for example. While there is no easy answer to the question of what wellbeing consists of, local and national governments have provided several frameworks in order to measure wellbeing (and provide services that address it). The aforementioned ONS scales are one example of measuring and defining wellbeing at the national level. Another commonly used framework is the New Economics Foundation’s commissioned work on the Five Ways to Wellbeing. This distils subjective wellbeing into five key areas: be active, take notice, connect, giving and keep learning. Although they consist of a frustrating mix of participles and infinitives, the ‘Five Ways’ are often used by both third sector organisations and local authorities to conceptualise wellbeing. However, none of these definitions offer a holistic and comprehensive concept of wellbeing. To add a bit more complexity, when it is difficult to define it is also difficult to measure – how, for example, do we measure the Five Ways? And more specifically, how do we measure whether programmes like those listed above address these areas? What constitutes success?

From my own experience, measures of wellbeing are at least as complicated as definitions. Assessments of local wellbeing services often contain a battery of different kinds of measures. Much of this diversity is due to diversely vested interests in what these wellbeing services do. Is the purpose of the service to address personal wellbeing, save money or improve health? Or all three?  From a subjective wellbeing perspective, measures like the Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS) are appropriate. However, this does not capture economic wellbeing or objective measures of personal environment. For example, knowing that someone has a score of 56 out of 70 on the WEMWBS tells us nothing about whether they live in a stable and safe home environment. From the health and economic perspective, administrative data is often used to see if individuals who participated in wellbeing courses or activities used health services less. But can we be sure that a decrease in health services use is not due to other factors? There are seasonal, regional and personal variations in health services use that are difficult to capture with administrative data. Furthermore, how can we be sure that using health services less indicates people are healthier? From an individualistic perspective, the Five Ways to Wellbeing could be measured simply by asking people how they ‘connect’ or how much they ‘connect’ in their day-to-day lives. But does every individual interpret ‘connecting’ in the same way? As always, it is difficult to create consistency in subjective, individualistic measures. More broadly, how can we expect a wellbeing service to show success on all of these different kinds of measures? When there is a lack of conceptual clarity, muddles in measurement will follow. The solution is not to throw everything at the wall and hope that something will stick in the name of evidence. There needs to be a better understanding of what these services and activities offer if ‘wellbeing’ is to be addressed in the community.

It would be logical to assume that the solution to this lack of evidence in wellbeing services is simply more evidence. However, as outlined above, the answer is not so simple. Instead of more evidence, there must be a shift towards the right kind of evidence on whether services improve individual, and therefore community, health and wellbeing. First, there must be a critical review around the causative assumptions of wellbeing activities. This means picking apart the impact a typing course has on wellbeing versus a yoga class. Each has value, but measuring them the same way would be a misstep. This means taking a step back from the assumption that improved subjective wellbeing leads to, and necessitates, cost savings. For example, administrative health records are unlikely to tell us if a basket weaving course is effective. Furthermore, basket weaving is unlikely to put the NHS in the black. It may be more effective to evaluate the impact of basket weaving via measures of social isolation. We need to unpick and critique the aspects of wellbeing that individual activities and courses address, and then apply the appropriate metrics to assess whether wellbeing improves. After those steps are achieved, there may be an opportunity to truly test if that personal wellbeing leads to improved health leads to cost savings causative narrative persists. In the end, measuring and commissioning wellbeing services is a national initiative that will continue whether we agree on what wellbeing means or not. And commissioners will likely project onto wellbeing services the kinds of changes they want to see in health services in general – specifically, less money, fewer patients and better population health. It is impossible to predict where this push for wellbeing will take us, but – as with any change in community health and public services – we need to be critical of how ‘well’ we really are doing.

1 White, Sarah (2014). Wellbeing and Quality of Life Assessment: A Practical Guide, Rugby, UK: Practical Action Publishing.

Posted in: Evidence and policymaking, Health


  • (we won't publish this)

Write a response