Friday, 22 November 2013

Dancing cheek to cheek: Public health and personalisation

Dancing cheek to cheek: A public health perspective on personalisation

I don’t know if this is a common experience, but I’ve got involved in most of the best things in my life without really understanding what I was getting myself into. You’ll be pleased to know that I won’t be writing about most of them here, but to my surprise one of them has been getting involved in public health. I’ve been a researcher for many years working mainly with people with learning disabilities, which got me into working with In Control and many others on trying to track people’s experiences of personalisation in England. But the work on personalisation isn’t my day job, so to speak. Most of my working life is now spent working in a public health observatory concerned with people with learning disabilities (this is where I shamelessly plug Improving Health and Lives

Recently, it seems to me that personalisation and public health have been like two teenagers at a school disco – they’ve noticed each other, they like what they see, but they don’t really know each other and it’s hard to know what they’re going to say to each other, let alone how they’re going to get to the point of dancing together. There is an increasing recognition that personalisation has to be a community enterprise if it is to become embedded and affordable at scale, and also that personalisation contains useful components way beyond the population of people currently eligible for personal budgets (whether education, social care, health, or any combination of the three). Alex Fox’s blog on local area co-ordination ( is a particularly good recent example of broadening thinking in this respect. On the public health side, as much of the public health capacity in England moves to local authorities in the shape of Health and Wellbeing Boards, there seems to be a renewed appetite for moving public health into new and more collaborative terrain.
Although my dubious metaphor casts me in the role of the go-between between the two parties at the school disco (“No, no, they really do like you, just go over there and talk to them”), this blog outlines some of the things I’ve been learning about a public health perspective, and what utility this perspective might have for the personalisation project.

What is health?

Although this is probably obvious to everyone else, it took me a while to appreciate just how expansively health is defined in the public health world. The original World Health Organization statement Health is a state of complete physical, mental, and social well-being and not merely the absence of disease or infirmity’ (From the Preamble to the Constitution of the World Health Organization which entered into force on 7 April 1948) is both radical and, it has to be said, problematic in a whole host of ways (not least that dangling word infirmity at the end there). What it does do, though, is put into play the idea that public health should be concerned about (and responsible for?) issues such as self-determination as an important aspect of complete physical, mental, and social well-being. It also sets ambitious goals about what states of health we should be aspiring to. Taken together, this way of thinking about health should have great resonance for an ambitious personalisation agenda.

What are health inequalities?

It also took me a while to understand how major strands of public health are focused on health inequalities, and what the word ‘inequalities’ means in a public health context. A widely used definition of health inequalities has been provided by Margaret Whitehead, in referring to differences in health status between social groups ‘which are unnecessary and avoidable but, in addition, are also considered unfair and unjust’ (Whitehead, 1992). Allied to the expansive view of health, my guess is that many people involved in personalisation could sign up to the view that most people using social care (and eligible for personal health budgets too) are experiencing states some way short of complete physical, mental, and social well-being compared to much of the rest of the population, that these differences in health states are mostly or entirely unnecessary and avoidable, unfair and unjust, and that personalisation could be a vehicle to help these groups reduce (or eliminate?) these inequalities.

For me, this definition of health inequalities poses some challenges to the personalisation world, which in its operational details has often not been clear about its view on inequalities (how would you know if equality had been achieved, on what dimensions, between which groups, for example?). I think it also helps to clarify some of the terrain on which debate about personalisation occurs, for example: What states of health are legitimate areas of intervention (in the broadest sense) by public services? Where there are differences in health status (particularly when broadly defined) between groups, to what extent are these differences seen as unnecessary and avoidable or not? Also, are these differences seen as unfair and unjust or not? And finally, is reducing inequality a legitimate function of public services anyway?

However, personalisation also poses a challenge to mainstream public health. Given that I work with people with learning disabilities, I have been continually struck by how little attention mainstream public health pays to disability. Inequalities are often framed in terms of socio-economic position, gender and/or ethnicity, for example, but the health inequalities experienced by disabled people are typically absent from public health discourse. Evaluations of public health interventions (from smoking bans through Sure Start through place-based programmes such as the New Deal for Communities) rarely consider their success or otherwise in including disabled people and in ensuring that they reduce inequalities between disabled people and other groups. Working with Health and Wellbeing Boards through local authorities, personalisation can bring these issues front and centre to public health thinking and action.

What are the causes of health inequalities?

If we accept that health inequalities are unnecessary and avoidable, unfair and unjust, then it is our responsibility to understand how these inequalities happen and to do something about them. A major strand of public health work has been involved in trying to understand what Sir Michael Marmot calls ‘The causes of the causes’. We know that smoking causes lung cancer, but what causes people in some groups to be more likely to take up smoking and less likely to stop? Here we’re in the realm of the social determinants of health and health inequalities, which are neatly summarised in the rainbow of health inequalities (Dahgren and Whitehead, 1991, see below). The rainbow has been developed as a very general representation of the social determinants of health inequalities both within societies and across societies on a global scale.
In my view there is much in this representation of health inequalities for those working in the personalisation world to chew on, particularly those engaging with community and place-based approaches. Most obviously from a UK perspective, social care is not represented in this diagram, presumably because it is not a global phenomenon and the functions taken up by social care in the UK are either carried out by other elements of societies in other places or are maybe not carried out at all. Although this can seem like a threatening absence, the relatively minor place also allocated to health services carries the broader message that achieving health equity goes far beyond fixing service systems.

The rainbow of social determinants of health inequalities (Dahlgren & Whitehead, 1991)

Allied to this, the individual factors that people bring with them are at the centre of the rainbow, but the impact of these factors on people’s health is mediated through a whole host of nested social factors, all of which we can do something about. For example, what do personalised support plans say about promoting healthy lifestyles? Are factors such as nutrition, physical activity, smoking, alcohol and drug intake, healthy sleeping patterns and so on an integral part of the discussion when it comes to support planning? And if they are, how can they be supported by the social and community networks that surround people – are people’s PAs smokers, for example, and do they smoke when they’re out with a person; do people have friends and family who want to join in with physical activities? And then moving outwards through the rainbow to people’s living and working conditions, do people’s environments support good health? People may have the motivation and the networks encouraging them to eat a healthy diet, but if healthy, fresh food is not available at the local shop and fresh food is only obtainable by a car trip or internet shopping, or if cooking facilities in the person’s home are not accessible, a healthy diet as part of a healthy life is going to be difficult to achieve. Poor housing, unemployment and insecure and meaningless work environments are all likely to contribute to poor health. And finally, at the outer layer of the rainbow, are broader social and economic conditions. Poverty determines so much about a person’s life (where people live, what they eat, what they are able to do with their time, social relationships), and hostile cultural conditions (for example the harassment and crime that many disabled people experience) are also powerful forces determining and reinforcing health inequalities.

My thinking is that something like the rainbow of health inequalities could be very useful in thinking through what a personalisation agenda means both in broad policy terms and potentially for individuals or communities. This could bring some real clarity for the personalisation project in identifying what the priorities are for dealing with health inequalities, and also what the levers are for addressing these priorities. Much of the exciting work around community building and local area co-ordination could easily form part of such a project, and potentially provide an ambitious yet boundaried set of priorities that Health and Wellbeing Boards would recognise as priorities for them too.

 What can we do about health inequalities?

Understanding the causes of health inequalities is one thing, but doing something about them is quite another. When thinking about where to put our energies in this respect, one seemingly simple diagram has really helped me. This is the Health Impact Pyramid (Frieden, 2010). This proposes that there are five levels of intervention designed to improve people’s health; 1) Counselling and education; 2) Clinical interventions; 3) Long-lasting protective interventions; 4) Changing the context to make individuals’ default decisions healthy; 5) Socio-economic factors.

The health impact pyramid (Frieden, 2010)

According to Frieden, there are two reasons why this is a pyramid. First, as you go further up the pyramid, the intervention requires more active participation, additional effort and informed consent from the person on the receiving end. Second, as you go further up the pyramid the interventions become more expensive, intensive and focused on fewer and fewer people.

So in public health terms, there are questions about the extent to which counselling and education interventions (such as intensive diet, exercise and weight loss programmes) are effective in improving population health, as only a relatively small proportion of people take them up and the additional effort involved makes them not sustainable for most people. Similarly, relatively expensive clinical interventions typically only occur once people already have health problems. Long-lasting protective interventions (such as vaccination and screening programmes) are lower cost, require less effort from people on the receiving end, and may be relatively effective at preventing the onset of some health problems for almost entire populations.
The bottom two levels to me are the most interesting, as they move beyond any sort of health service to thinking about how communities, environments and societies can be arranged to promote the health of populations, without people having to actively sign up to them. Some of the biggest advances in public health (for example, clean water, or toilets, as I was reminded by Toilet Day recently) have been at this level, where the default position in the UK (having clean water through the tap) is the healthy one and the unhealthy option (getting your water from a polluted river) would require a lot more effort. There is considerable work in public health around trying to arrange environments to make people’s default decisions healthier (by, for example, discouraging car use, encouraging physical activity, making street environments safer at night etc), although the evidence base for their effectiveness is small and there has been little effort to consider disabled people in the design and evaluation of these environments. The final, socio-economic level, has great potential but as Frieden acknowledges can be the most controversial as interventions at this level are highly politicised, often with little consensus, and can be open to charges that they are not within the legitimate realm of public health activity.

Can the personalisation project learn anything from the health impact pyramid? I think it can, and recent directions in moving towards community approaches suggest that the lessons are already being learned. For example, the full-on personalisation approach, confined to people already eligible for social care (and for personal health budgets?), might be seen as analogous to a public health intervention somewhere near the top of the pyramid – it’s intensive, applied to a relatively small proportion of people in population terms, and requires active additional engagement and effort from the people involved.

Are there problems if this becomes the exclusive focus of the personalisation project? Well, there are some that spring to mind. First, it results in a focus on ‘downstream’ rather than ‘upstream’ interventions (in other words, it concentrates on trying to fish people out of the river rather than trying to stop people falling in the river in the first place). Second, it’s operating on a terrain (who is ‘eligible’ for services) set by others and that sets up divisions between people (it’s not fair that they get that personalisation thing when I have to pay for support myself without any help in planning what I want). Third, it can start with a focus on services and tries to encourage people to think beyond services, rather than starting from the viewpoint of what people want from their lives and then thinking if and how services are part of the answer. Fourth, the additional effort required from people and their circles of support is felt by some people to be burdensome, which may reduce its sustainability for some people. Finally, in this full-on form it’s never going to be scalable to entire populations and communities.

For me, this is where the community focus of the personalisation project becomes really exciting. The ‘full-on’ version is absolutely necessary for many people, but are there aspects of a personalisation approach that are the equivalent of a population vaccination or screening programme, or that work to help communities become more inclusive and mutually supportive such that people don’t end up so isolated and in environments that are more likely to result in crises for them. And, can personalisation speak to socioeconomic factors – most obviously in terms of housing, neighbourhood quality and employment, for example?

All of the above community-based initiatives are exactly what Health and Wellbeing Boards should be interested in, particularly with the potential within local authority settings to knit together communities in ways far beyond ‘services’. 

I hope this blog has shown some of the ways in which personalisation and public health can both challenge each other and help each other, to the potential benefit of whole communities. These two diagrams might be useful tools for the personalisation project to think through what it wants to do and how it wants to do it, while explicitly acknowledging a social justice agenda and making its activity intelligible to public health. Just dance together, you know you want to…


Dahlgren, G., & Whitehead, M. (1991/2007). Policies and strategies to promote social equity in health. Background document to WHO - Strategy paper for Europe ( Stockhome: Institutet för Framtidsstudier
Frieden, T. R. (2010). A framework for public health action: The health impact pyramid. American Journal of Public Health, 100(4), 6.
Whitehead, M. (1992). The concepts and principles of equity and health. International Journal of Health Services, 22, 429–445.