Article Text

Download PDFPDF

We must move on: taking stock (yet again) of the evidence for socio-economic differentials in health
  1. G David Batty, Deputy Editor1,2
  1. 1Department of Epidemiology & Public Health, University College London, London, UK
  2. 2Centre for Cognitive Ageing and Cognitive Epidemiology, Department of Psychology, University of Edinburgh, Edinburgh, UK
  1. Correspondence to David Batty, Wellcome Trust Fellow, Department of Epidemiology & Public Health, University College London, 1-19 Torrington Place, London WC1E 6BT, UK; david.batty{at}

Statistics from

While working in Scandinavia for a year during my first spell of economic migrancy, I was told by locals that the Danish calendar had 14 months owing to the recurrent “November, November, November”—an indication of the inevitably damp, cold and prolonged Copenhagen winter characterised by the desertion of all but the most hardened of drinkers from the hostelries and coffee shops of this picturesque city. No such bleak offerings in this, the penultimate issue for 2011 of the Journal of Epidemiology & Community Health however, where we present the same eclectic collection of reports that have come to define the journal since its inception as the British Journal of Social Medicine in 1947. Divided into three invariably overlapping themes—ageing,1–5 lifecourse,6–12 cancer13–16—this trichotomy of content is rather arbitrary on occasion but serves to make sense of our particular branch of science which, like all others, is increasingly voluminous to the point of overwhelming. As always, our contributing authors have wide geographical reach, so maintaining the international flavour of the journal. In each of their papers you will find exposures and outcomes that cover the full gamete of psychological, social, behavioural, physiological indices as we seek to make sense of the world around us.

A section dedicated to ageing in any self-respecting epidemiology journal would not be worth its salt if we did not give a vigorous nod towards normal cognitive decline and its pathological extreme, dementia. The long-held suggestion that cognitive impairment and dementia may be the product of vascular processes has found support in correlations with established cardiovascular disease risk factors (raised blood pressure, smoking, obesity, elevated blood cholesterol), so raising the possibility that a normalisation of their levels through pharmacological intervention may delay failing cognition. The apparent differential protective effects of statin use—widely used to treat hyperlipidaemia—on dementia but not mild cognitive impairment1 is surprising. Given the heavily confounded nature of the association in this observational study–users of, and adheres to, medication are fundamentally different in their characteristics from those who are less compliant–further investigation using that gold cadillac of research designs, the randomised controlled trial by further piggy-backing of statin trials that may have been designed only with cardiovascular disease outcomes in mind is warranted. Mendelian Randomisation, where a gene variant for hyperlipidaemia is used as a (hopefully) unconfounded proxy, would also have utility in this scenario. Elsewhere,2 but along the same theme, there was a suggestion that remaining in employment rather than retiring may be more beneficial for cognitive performance. This partial support for the ‘disuse’ hypothesis will provide a soupcon of solace for those academics whose universities have declared that there is now no official age for retirement.

Unsurprisingly given the journal's roots, socio-economic status is a spine running through other papers under the ageing rubric, and those in the themes of cancer and lifecourse. It certainly reflects the weight of current submissions to the Journal. Favourable social circumstances—often assessed across several decades in the featured studies—would also appear to confer protection against4 low quality of life,3 hip fracture,5 allostatic load (a proxy for accumulated damage to system integrity),6 low self-rated health (twice8 11), partner violence,9 reduced walking speed,10 advanced breast cancer,13 cancer survival (if one considers the English North-South divide to be a proxy for poverty),14 suicide,7 and an array of other health indicators.12

As Sander Greenland reminded us at the 2011 International Epidemiology Association Congress in Edinburgh, and as featured in the proceedings sponsored by this journal,17 any investigator has bias, and it is through my tainted spectacles and my work in cognitive epidemiology that I gravitate to the role that education—and by association, IQ–might have in the aetiology of suicide. That people with basic education had an elevated risk of completed suicide relative to better performing students in a study comprising close to one million members,7 places in stark relief the findings of early studies showing that students at esteemed UK universities had higher suicide rates than the general population.18 Also brought into question is the observation from group-level studies that IQ assessed across nations using a variety of tests is somehow of use in understanding suicide aetiology at the level of the individual.19 The latter is a useful teaching illustration of the ecological fallacy.

The hefty collection of papers herein in the realm of socio-economic status certainly help to further define the nature of ‘the gradient’ as it has come to be known to social epidemiologists, and crystallises how poverty ‘gets under the skin’ to generate health inequalities. However, as has been noted in other fields of epidemiology, the study of socioeconomic variations in health has largely become occupational therapy for social scientists and epidemiologists, me doubtless included. Surely enough is known now, and in truth has been for some years, to hope for a more substantial evidence base from trials or natural experiments which aim to reduce this differential. When I relocated to Scotland in 2004 after sampling those Novembers in Copenhagen there was simultaneously a call for funding from the Scottish Chief Scientist's Office and a call to arms in this very journal for more experimental evidence—quasi or otherwise—of what actually worked in ameliorating poverty differentials. The former, I was told some time later, yielded effectively zero response from the research community, and the latter has little altered the landscape of this journal and that of others. We will shortly announce a call for papers describing the methods used in intervention studies that aim to narrow the gap, and we hope that you will contribute.



  • Competing interests None.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.