Trends in immunisation inequity: evidence, rights, and planning

In The Lancet Global Health, Ahmad Hosseinpoor and colleagues provide a status report on global disparities in rates of immunisation with three doses of diphtheria– tetanus–pertussis (DTP3) vaccine in children of diff erent socio economic status (measured by quintiles of estimated physical assets). As in previous similar studies, there were large and signifi cant disparities between children from the richest and poorest households. As suffi cient comparable household surveys became widely available for cross-country trend analysis at the turn of the century, data for immunisation and other health indicators have consistently pointed in the same, disheartening direction. These inequities are not declining through time. The paper advances knowledge about inequity trends in immunisation in two complimentary ways. First, it contributes novel and hitherto unassessed empirical evidence from more than 50 countries about the magnitude of disparities among children of diff erent socioeconomic groups. Additionally, it off ers updated confi rmation of trends that have been outlined and documented for more than 10 years: progress in immunisation rates is all too often slower (at best) for poorer children than richer ones—often the observed changes among the poorest children are not statistically signifi cant at all. This lack of progress for the most marginalised groups leads to two seemingly contradictory messages. On the one hand, increasing immunisation does not naturally, automatically, or often occur in an equitable way. For various reasons (political, economic, social, cultural), the opposite is the case. On the other hand, however, there are suffi cient cases of countries that have succeeded in reducing these inequities. Even if they are a minority, they show it is indeed possible to accelerate national average immunisation rates while reducing disparities. At least two issues require further elucidation in the analysis of inequities in immunisation in particular, and health in general. There is a need to explore and carry out sensitivity analyses using diff erent methods to group and rank households according to their wealth. For instance, the use of principal component analysis (PCA), as Hosseinpoor and colleagues did, has been criticised, given the kind of variables and indicators usually found in typical household surveys—ie, they are dichoto mous variables that do not follow a normal distribution. Although simple alternatives exist, they are not applied as often as PCA-based indices. Clearly, using a diff erent method to determine quintiles would result in diff erent ranking of households. How stable are they to changes in method? Are the estimated absolute and relative gaps the same when diff erent groups are formed by using alternative methods? Moreover, are the substantial conclusions in terms of the trends maintained under diff erent household rankings? Second, additional work is required to analyse the correlation and reinforcing mechanisms across dimensions of disparities other than income or wealth— ie, gender, ethnicity, location (including urban), parents’ education, parents’ employment status, and type of economic activity. The ways in which the various forms of exclusion interact with each other is related to the concepts of multiple burden and intersectionality in the antidiscrimination legal literature. These points are not purely of academic interest. They are important for practice and to guide “bottom up” interventions, formulating health policy goals in distributional terms, and progressive realisation of the right to health. People do not live in quintiles (whichever way they are measured or analysed). They live in communities with criss-crossing axes of exclusion and disparities. Moreover, while no government agency, public servant, or politician is responsible for quintiles, they are mandated to serve (and they can be held accountable to) citizens who live in particular districts. Thus, it is important to identify them (through the joint analysis of quintiles and geographic location) to be able to properly design focused interventions within a universal approach. Another practical implication refers to the way that work-plans, targets, and milestones in programmes and inter ventions by non-governmental organisations, international agencies, and governments are usually set up on the understanding that universality cannot be achieved in the short run. Consequently, they aim to achieve coverage of X% of the population. In a Published Online August 3, 2016 http://dx.doi.org/10.1016/ S2214-109X(16)30189-9


Trends in immunisation inequity: evidence, rights, and planning
In The Lancet Global Health, Ahmad Hosseinpoor and colleagues 1 provide a status report on global disparities in rates of immunisation with three doses of diphtheriatetanus-pertussis (DTP3) vaccine in children of diff erent socio economic status (measured by quintiles of estimated physical assets).As in previous similar studies, 2 there were large and signifi cant disparities between children from the richest and poorest households.As suffi cient comparable household surveys became widely available for cross-country trend analysis at the turn of the century, data for immunisation and other health indicators have consistently pointed in the same, disheartening direction. 2These inequities are not declining through time.
The paper advances knowledge about inequity trends in immunisation in two complimentary ways.First, it contributes novel and hitherto unassessed empirical evidence from more than 50 countries about the magnitude of disparities among children of diff erent socioeconomic groups.Additionally, it off ers updated confi rmation of trends that have been outlined and documented for more than 10 years: progress in immunisation rates is all too often slower (at best) for poorer children than richer ones-often the observed changes among the poorest children are not statistically signifi cant at all.This lack of progress for the most marginalised groups leads to two seemingly contradictory messages.On the one hand, increasing immunisation does not naturally, automatically, or often occur in an equitable way.For various reasons (political, economic, social, cultural), the opposite is the case.On the other hand, however, there are suffi cient cases of countries that have succeeded in reducing these inequities.Even if they are a minority, they show it is indeed possible to accelerate national average immunisation rates while reducing disparities.
At least two issues require further elucidation in the analysis of inequities in immunisation in particular, and health in general.There is a need to explore and carry out sensitivity analyses using diff erent methods to group and rank households according to their wealth.For instance, the use of principal component analysis (PCA), as Hosseinpoor and colleagues did, has been criticised, given the kind of variables and indicators usually found in typical household surveys-ie, they are dichoto mous variables that do not follow a normal distribution.Although simple alternatives exist, 3-6 they are not applied as often as PCA-based indices.Clearly, using a diff erent method to determine quintiles would result in diff erent ranking of households.How stable are they to changes in method?Are the estimated absolute and relative gaps the same when diff erent groups are formed by using alternative methods?Moreover, are the substantial conclusions in terms of the trends maintained under diff erent household rankings?
Second, additional work is required to analyse the correlation and reinforcing mechanisms across dimensions of disparities other than income or wealthie, gender, ethnicity, location (including urban), parents' education, parents' employment status, and type of economic activity.The ways in which the various forms of exclusion interact with each other is related to the concepts of multiple burden and intersectionality in the antidiscrimination legal literature. 7hese points are not purely of academic interest.They are important for practice and to guide "bottom up" interventions, formulating health policy goals in distributional terms, and progressive realisation of the right to health. 8People do not live in quintiles (whichever way they are measured or analysed).They live in communities with criss-crossing axes of exclusion and disparities.Moreover, while no government agency, public servant, or politician is responsible for quintiles, they are mandated to serve (and they can be held accountable to) citizens who live in particular districts.Thus, it is important to identify them (through the joint analysis of quintiles and geographic location) to be able to properly design focused interventions within a universal approach. 9nother practical implication refers to the way that work-plans, targets, and milestones in programmes and inter ventions by non-governmental organisations, international agencies, and governments are usually set up on the understanding that universality cannot be achieved in the short run.Consequently, they aim to achieve coverage of X% of the population.In a misapplication of progressive realisation, they seldom realise that attempting a goal of X% is equivalent to having an explicit goal of excluding 100-X%.Moreover, this is done with complete disregard to mentioning or analysing which 100-X% of the population is doomed to wait for coverage or even which selection criteria will be applied to decide it.It is no wonder, then, that very few countries have managed to reduce disparities , as planners and policy-makers constantly fail to apply an equity approach. 10Studies like the one by Hosseinpoor and colleagues 1 should help to change these dismal trends in policy design and results for children.
Copyright © The Author(s).Published by Elsevier Ltd.This is an Open Access article under the CC BY license.

1
Hosseinpoor AR, Bergen N, Schlotheuber Measuring household wealth with latent trait modelling: an application to Malawian DHS data.Soc Indic Res 2014; 118: 877-91.7 Crenshaw K. Mapping the margins: intersectionality, identity politics, and violence against women of color.Stanford Law Rev 1993; 43: 1241-99.8 Gwatkin D. Health inequalities and the health of the poor: what do we know?What can we do?Bull World Health Organ 2000; 78: 1. 9 Evans TM, Whitehead M, Diderichsen F, Bhuiya A, Wirth M, eds.Challenging inequities in health: from ethics to action.New York: Oxford University Press, 2001. 10 Vandemoortele J. Equity begins with children.In: Minujin A, Nandy S, eds.Global child poverty and well-being: measurement, concepts, policy and action.Bristol: Policy Press, 2012.
A, et al.State of inequality in diphtheria-tetanus-pertussis immunisation coverage in low-income and middle-income countries: a multicountry study of household health surveys.Lancet Glob Health 2016; published online Aug 3. http://dx.