The social gradient in health inequalities
The social gradient in health inequalities

It’s become evident in recent years that, even within wealthy countries like the UK, there are wide disparities between different regions, cities and even within cities in levels of serious illness and premature deaths. The so-called ‘Glasgow effect’ has, for example, been much reported on. Within Cardiff, there is a documented difference in life expectancy between Lisvane and Fairwater of around 12 years. In this Cafe, Dr Jeremy Segrott asked us to consider why such health inequalities matter, and presented some options that have been suggested for tackling them, each with different ethical and political implications.

While measures such as life expectancy often hit the headlines, these inequalities are more complex than just these numbers alone, and reflect the influence of other inequalities – such as in educational opportunities, access to healthy food, the quality of housing, the extent of pollution in a neghbourhood, and so on. These other inequalities in turn are the product of broader social forces, which mean that tackling health effects at the ‘end of the pipe’, as it were, may not be the most effective or efficient way of addressing the problem.  While the focus of media reporting of health inequalities often focuses on how to change individual behaviour, social science research indicates that the deeply entrenched nature of these inequalities will make inteventions that focus on getting individuals to do things differently will be largely ineffective. The ‘social gradient’ in health outcomes remains.

So what measures can work? Jeremy took us through a discussion of three options. The first is targeting interventions (in Jeremy’s research, he focused on measures aimed at helping parents with family issues) at those who suffer worst from a given inequality. While this may make sense on paper, in reality it tends to have little impact and can even have negative effects.  It can lead to stigma for those identified as most in need, while not doing more than just shifting the bottom end of the social gradient in health inequalities up a little. Those slightly further up the slope, but still suffering bad outcomes, are not affected at all.

An alternative is universal provision, of which the NHS and child benefit stand as good examples. Here, services are available on the basis of need. Stigma is no longer a problem, but at the same time those most in need may not be aware of the existence of services or may have other reasons for not accessing them. While universal approaches offer a vision of equal access, they do not necessarily guarantee equitable access. Once again, they fail to shift the social gradient upward.

Another option, discussed by Sir Michael Marmot in his 2010 review (the Marmot Review) of health inequalities in the UK, is what has been called proportionate universalism. Here, services are provided universally, but the ‘intensity’ of intervention can vary depending on need. This is intended to avoid stigma (like universal approaches) but to increase equity and efficiency. Jeremy gave two examples. The first was the the Triple P programme in Australia, which couples universal provision with a range of levels of intervention, ranging from the provision of information via various media to ongoing intensive work between practitioners and parents. The second was the Strengthening Families Programme in Wales, which aims to tackle substance misuse among teenagers in Wales pre-emptively. Here, once again, different levels of intervention from information provision to close working with groups of families experiencing similar issues were carried out.

In each case, Jeremy suggested, recruitment for these programmes aimed at universal reach, while also assessing those who chose to participate for their level of need. Barriers to taking part were consciously lowered – in the Welsh case, through funding families’ travel costs and providing food.

In discussion, attendees raised the question of how we determine what kinds of processes we need to target to effectively shift the social gradient. What, in other words, are the causal factors which actually contribute most to them. There are political investments in particular interventions (such as the much-criticised Troubled Families programme), which ensures that they continue even when they do not work. It is also important to note that there are some targeted inteventions (such as providing healthcare to asylum seekers) which have significant effects on populations higher up the social gradient (by preventing the spread of infectious diseases, for example). Finally, the provision of resources to tackle issues can run into problems when mutiple agencies with different responsibilities are involved, as where fostered children move (as they grow older) between local authorities, and patterns of intervention are interrupted. Finally, Marmot’s own difficulty in specifying what he means by proportionate universalism was reflected in the difficulty some found in telling it apart from targeted interventions – how exactly what the identification of subgroups who were more in need differ from ‘means testing’ and other measures that might introduce stigma?

Leave a Reply

Your email address will not be published. Required fields are marked *