Statistics from Altmetric.com
Mackenbach has thoughtfully and objectively critiqued the effect of the English 1997–2000 strategy to reduce health inequalities (see page 568).1 His analysis (consistent with others) is sobering; held up against endpoint targets of reducing inequalities in life expectancy and infant mortality, the strategy has not succeeded, despite a large government investment and effort that surpasses any other (European) country. However, it might be that inequalities would have deteriorated even further in the absence of the English strategy.
Mackenbach proposes three ingredients for the way forward: more and better advocacy to ensure governments have the mandate for more fundamental policy shifts (eg, income redistribution); more research on the differential effectiveness of interventions; better and more focused policy planning and implementation, matching activity with modifiable and influential determinants of health inequalities. In this commentary, we provide some reflections on Mackenbach's analysis and recommendations from the antipodes of New Zealand.
New Zealand had a centre-left government from 1990 to 1999 that also attempted to address health inequalities—albeit not to the same extent of budget, planning and monitoring as England. Activities ranged from fresh fruits and vegetables in deprived schools to increased primary healthcare funding (preferentially targeted at deprived populations in initial roll-out at least), and from the enshrining of the need to address health inequalities (ethnic largely) in legislation to some income redistribution through welfare to work and family support programmes. Indications are that socioeconomic inequalities in mortality have remained constant in relative terms (perhaps reducing in absolute terms),2 and, thankfully, that gaps in life expectancy between Māori and non-Māori have reduced by 1–2 years since the mid-1990s (following widening inequalities in the 1980s to 1990s).3 4 While changes in social determinants and health policy may be responsible for some of this improvement, it must be noted that mortality rates are also driven by many more longer-run determinants than just contemporaneous or recent policy. Mackenbach argues that one of the major limitations of the English strategy was that it was essentially set up to fail as it could not pick up the long time lags between changes in policy (if effective) and changes in mortality or life expectancy.
We completely concur with Mackenbach's analysis that evidence on what actually changes inequalities in health (as opposed to average health per se) is desperately needed. Without wanting to dampen enthusiasm, we do, however, want to point to the lessons from a recent New Zealand experience published in this journal.5 In a randomised trial of price discounts on healthy foods and nutritional education among 1100 supermarket shoppers (price discounts increased healthy food purchasing, but not tailored nutritional education), we attempted to recruit equal numbers of Māori, Pacific and European/other shoppers. By the time of final analysis, numerous factors (eg, the necessity to use ethnic-specific recruitment strategies, attrition) had conspired to make the planned comparisons of intervention effectiveness across social groups problematic to interpret. That is, there is an extra layer of considerations, effort and resource required to achieve the subpopulation comparisons that Mackenbach is highlighting as necessary, and we as a research community (and also research funders) need to grapple with and solve these challenges. Additionally, wise use of indirect evidence (eg, if the price elasticity for tobacco is higher among lower socioeconomic groups, then the default starting position is that the price elasticity on healthy food is probably also higher among lower socioeconomic groups until proven otherwise) and modelling is needed in addition to a sole reliance on direct empirical evidence.
Mackenbach hints that more serious income redistribution is perhaps the best example of ‘large-scale policy change’ necessary to reduce health inequalities. As two of the ‘many social epidemiologists’ identified by Mackenbach as proponents of this recommendation, we also caution against placing exclusive faith in this mechanism. The evidence on the association of income with health is mostly cross-sectional, and the little evidence there is on actual changes in income with changes in health suggests much weaker associations—at least in the short run.6 The evidence on the association of income inequality with health at the population level (adjusting for personal income) is mixed,7 8 and cross-national comparisons (led by Mackenbach) fail to find a close correlation of health inequalities with income inequalities or egalitarianism.9 Modelling of the likely health impacts of sizeable income redistribution of mortality inequalities suggests ‘useful’ but far from large reductions in health inequalities.10
What other ‘large-scale policy change’ might be warranted to reduce health inequalities? In terms of proximal determinants, Mackenbach points to alcohol as a missed entry point. We would also point to the need to tackle tobacco in a far more serious way than just scaling up cessation programmes. For example, following strong political leadership from the Māori Party, the New Zealand parliament has now committed to a goal of a tobacco-free New Zealand by 2025.11 Such a goal is feasible12 and consistent with growing international momentum for a tobacco endgame.13 If the endgame goal is achieved in New Zealand, we estimate an improvement in Māori and non-Māori life expectancy of about 5 and 3 years, respectively (compared with current smoking prevalence continuing indefinitely), a closing of the gap by an impressive 2 years.14 That is, focusing seriously on old-school risk factors (or ‘ill-health prevention’ in Marmot Review terms15) has substantial potential to reduce health inequalities. We would add to this list population-wide action on salt and saturated fat as a mechanism to both improve overall health and reduce health inequalities, consistent with international recommendations to the United Nations High-Level Meeting on Non-Communicable Diseases in September 2011.13
Funding Health Research Council of New Zealand.
Competing interests None.
Provenance and peer review Not commissioned; not externally peer reviewed.
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.