Damned if you do, damned if you don't: subgroup analysis and equity

Abstract

The final report from the WHO Commission on the social determinants of health recently noted: 'For policy, however important an ethical imperative, values alone are insufficient. There needs to be evidence on what can be done and what is likely to work in practice to improve health and reduce health inequities.' This is challenging, because understanding how to reduce health inequities between the poorest and better-off members of society may require a greater use of subgroup analysis to explore the differential effects of public health interventions. However, while this may produce evidence that is more policy relevant, the requisite subgroup analyses are often seen as tantamount to statistical malpractice. This paper considers some of the methodological problems with subgroup analysis, and its applicability to considerations of equity, using both clinical and public health examples. Finally, it suggests how policy needs for information on subgroups can be met while maintaining rigour.

Could scale-up of parenting programmes improve child disruptive behaviour and reduce social inequalities? Using individual participant data meta-analysis to establish for whom programmes are effective and cost-effective

“…In addition, trial outcome data are rarely broken down by equity factors; when trials do test socioeconomic or other predictors of treatment effects, statistical models are specified in varying ways, 80 for example, some calculating interaction effects but others only within-group predictors, which renders synthesis meaningless. 23,82 These problems apply no less to parenting intervention trials, 46,50 and can be overcome by use of pooled data. Pooling individual-level data is an exciting new approach to data synthesis, 80,83 increasingly common in medicine in recent years, 84 but rarely used in public health or psychosocial fields.…”

“…Our design represents a unique opportunity to overcome the problems of low power and reporting bias that beset subgroup analyses in most individual trials, and hence to enhance our understanding of how parenting interventions may reduce or widen health and social inequalities. 82,109,110 Moreover, our study will illustrate the extent of this 'ecological fallacy', which can arise when aggregate-level data are used to infer individual-level effects, as we will be able to directly compare trials and individual-level moderator effects.…”

Human well-being impacts of terrestrial protected areas

“…Assessing the impact of PAs on health or social inequalities would require either individual sound studies with justifiable subgroup analyses, or a set of comparable studies which describe in detail the socioeconomic position of the populations studied [103,104]. Neither was available from the extant literature.…”

Benefits of environmental inequality assessments for action

“…Dichotomous approaches only allow the identification of an exposure gap between two groups but may, in the case of income, fail to elucidate potential inequalities associated with the middle categories (see figure 1). Data availability thus often restricts adequate assessment of inequalities between subgroups19 and thereby frequently limits targeted action (see discussion by Benach et al 12). …”

“…While there are many studies describing targeted and well-intentioned interventions and policies, they are often weak in assessing their equity effects17 19 24 and there is a specific gap of evidence for the equity impact of interventions on the wider social determinants of health including the environment 23. Some of the limitations and challenges are illustrated by the examples below (covering both environmental and health equity interventions).…”