Statistics from Altmetric.com
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.
Some contextual effects can be discovered only through international comparisons
In contrast with the plethora of studies concerning the social inequalities in health that have been conducted since the 1980s, there are relatively few related to injuries.1,2 There are even fewer studies comparing the influence of these inequalities in different countries. So, the publication of Borrell et al’s study on social inequalities in transportation injury mortality across European countries3 is most welcome and timely [see page 138]. Their study elucidated the effect of differing national contexts on the influence of social disparities on road traffic injury (RTI) mortality.
In this commentary, we will first explain what a so-called “contextual effect” is and then illustrate that many such effects can be discovered only through international comparisons. We contend that “dissimilarities” may provide more useful information than “similarities”. Because the contextual determinants of social inequality in RTI mortality between and within countries are different, they have different implications for injury prevention.
WHAT IS A CONTEXTUAL EFFECT?
In linguistics, context refers to the text surrounding a word, giving a better understanding of what the word means. In art, contextualism refers to the way a work of art may only be understood by knowing the historical, political, or cultural circumstances when it was produced. In health research, context similarly refers to the wider situation surrounding the association between an exposure and outcome and how this wider situation may confer meaning on that association.4
The study of the effects of collective or group characteristics on individual level outcomes is termed contextual or multilevel analysis. The effects of group level properties on individual level outcomes are contextual effects.5 Incorporating group level variables in contextual analysis is important because this approach provides information not captured by individual level data.6 For example, national income level gives information different from individual level income. The national income level may be a marker for countrywide factors potentially related to RTI, such as road conditions, user patterns, or medical care. Such factors affect everyone in the country regardless of their income. Similarly, the national unemployment level may be a marker of the effects of globalization, which affects everyone in a country, employed or unemployed.
Macro level variables, such as political regime, urban planning, road design, regulations, population density, traffic volume, and so on, affect individuals directly and also constrain the choices they make.6 Only through international comparisons can these contextual RTI risk factors in any one country be revealed. This kind of information could not be derived from individual level studies conducted within a country. If everyone in a country was exposed to the same contextual risk factors, there would be no interpersonal variation among study subjects.7–9
“SIMILARITIES” OR “DISSIMILARITIES”?
In their international comparisons, Borrell et al concentrated mostly on “similarities”—for example, in all countries less educated men had higher death rates compared with better educated men. More might have been learned had they examined why the magnitude of social inequalities differed in different countries. However, a better understanding of “dissimilarities” between countries could provide more information for injury prevention. Each country has its own historical and cultural context that shapes social inequalities in health and this is why Kunitz argued for “particularism” to understand the social determinants of population health.10
For example, a recent study revealed that in China, higher socioeconomic position (SEP) was related to a less healthy lifestyle, whereas in the United States, higher SEP was related to a healthier lifestyle.11 The contrast was partially due to cultural factors and partially to level of economic development.
Using Borrel et al’s data, we found the magnitude of relative inequality in RTI mortality between countries to be larger than the relative social inequality within a country.3 The age adjusted death rate ratios (RR) among men with low, middle, and high educational achievement ranged from 1.09 in Spain (Madrid) to 1.66 in Switzerland. However, the RR between Madrid and Finland was about 3.0, regardless of educational level. Thus, in explaining inequalities in RTI mortality rates, social contextual differences between countries are more important than social group differences in any one country.
We have simplified and modified Borrel et al’s figure 1 to illustrate these differences. The figure shows that the slope of social inequality between countries was steeper than the slope within countries. Though country C has the lowest RTI mortality compared with country A or B, country C has the largest social inequality compared with A or B. Why?
Determinants of social inequality in RTI mortality between countries
To explain these findings, we used Macintyre’s conceptualisation.12 She suggested three explanations for geographical variations in health: (1) compositional explanations, which draw attention to the characteristics of individuals concentrated in particular places; (2) contextual explanations, which focus on structures in the local physical and social environment; (3) collective explanations, which concentrate on sociocultural and historical features of communities.
Macintyre gave an example relevant to injury prevention. Children in deprived areas may not play outdoors because their families do not have gardens or the resources to take them to parks (a compositional resource based explanation). Or they may not do so because too few parks are provided or there are no easy public transport links to parks that do exist (a contextual resource based explanation). She also suggests that within the prevailing culture either play is not seen as important to children or it is not considered safe for children to play with other (strange) children in public places (a collective explanation).
Analogously, high RTI mortality in a country might be because of a higher percentage of dangerous drivers (a compositional explanation); poorer road infrastructure or post-crash care (a contextual explanation); or lower standards of safety—that is, less thorough policy implementation and enforcement (a collective explanation).
Determinants of social inequality within countries
To explain social inequality in RTI within countries, Laflamme proposed a model that illustrates upstream (social context) and downstream (individual level) mechanisms.1,13 At an individual level, the model depicts a pathway from social position, through exposure, to health hazards/risk factors, and on to health outcome. Two downstream mechanisms—differential exposure and differential susceptibility—may come into play in the relation between social position and health outcome. Contextual influences are highlighted by four entry points, each of which refers to different mechanisms on social stratification, on differential exposure, on different susceptibility, and directly on health.
Finally, Lynch and Kaplan propose a Marxian-Weberian view of how socioeconomic position affects health.14 Some examples pertaining to RTI are seen in table 1 (see http://www.injuryprevention.com/supplemental for table 1). This framework reminds us not to overlook the unequal power relationships that affect the implementation or effectiveness of traffic safety policies. We combined both the above frameworks in table 2 to contrast the determinants of social inequality between and within countries.
International comparisons provide a prism through which the wider variations in inequalities between countries come into focus. These comparisons highlight the importance of contextual effects on social inequalities in RTI. Individual risky behaviors are embodied in the social context and the unequal power relationships that affect the adoption of countermeasures, as well as acting to shape different exposures and resources experienced across social groups. Clarifying the determinants of social inequality in RTI mortality between and within countries will help us develop better injury prevention strategies.
Some contextual effects can be discovered only through international comparisons
The table is available as a downloadable PDF (printer friendly file).
If you do not have Adobe Reader installed on your computer,
you can download this free-of-charge, please Click here
Files in this Data Supplement:
- [view PDF] - Table 1. Exposure resources framework to explain the relations between socioeconomic position (SEP) and road traffic injury.