key: cord-0704803-kiv5lsyv
authors: Bergen, Nicole; Cata-Preta, Bianca O.; Schlotheuber, Anne; Santos, Thiago M.; Danovaro-Holliday, M. Carolina; Mengistu, Tewodaj; Sodha, Samir V.; Hogan, Daniel R.; Barros, Aluisio J. D.; Hosseinpoor, Ahmad Reza
title: Economic-Related Inequalities in Zero-Dose Children: A Study of Non-Receipt of Diphtheria–Tetanus–Pertussis Immunization Using Household Health Survey Data from 89 Low- and Middle-Income Countries
date: 2022-04-18
journal: Vaccines (Basel)
DOI: 10.3390/vaccines10040633
sha: 299706380c4f70913c43b5037ea95dede0332f8a
doc_id: 704803
cord_uid: kiv5lsyv

Despite advances in scaling up new vaccines in low- and middle-income countries, the global number of unvaccinated children has remained high over the past decade. We used 2000–2019 household survey data from 154 surveys representing 89 low- and middle-income countries to assess within-country, economic-related inequality in the prevalence of one-year-old children with zero doses of diphtheria–tetanus–pertussis (DTP) vaccine. Zero-dose DTP prevalence data were disaggregated by household wealth quintile. Difference, ratio, slope index of inequality, concentration index, and excess change measures were calculated to assess the latest situation and change over time, by country income grouping for 17 countries with high zero-dose DTP numbers and prevalence. Across 89 countries, the median prevalence of zero-dose DTP was 7.6%. Within-country inequalities mostly favored the richest quintile, with 19 of 89 countries reporting a rich–poor gap of ≥20.0 percentage points. Low-income countries had higher inequality than lower–middle-income countries and upper–middle-income countries (difference between the median prevalence in the poorest and richest quintiles: 14.4, 8.9, and 2.7 percentage points, respectively). Zero-dose DTP prevalence among the poorest households of low-income countries declined between 2000 and 2009 and between 2010 and 2019, yet economic-related inequality remained high in many countries. Widespread economic-related inequalities in zero-dose DTP prevalence are particularly pronounced in low-income countries and have remained high over the previous decade.

The Immunization Agenda 2030 (IA2030) envisions a world where everyone, everywhere, at every age, fully benefits from vaccines to improve health and well-being [1] . There have been notable gains in reaching children who have been partially vaccinated by making better use of existing vaccination delivery channels and addressing missed opportunities for vaccination among children who have some access to health services [2] . Yet, there are large numbers of children who have not received any routine vaccines (zero-dose children).

Disaggregated data were sourced from 88 Demographic Health Surveys (DHSs) and 66 Multiple Indicator Cluster Surveys (MICSs) in 89 low-and middle-income countries. DHS and MICS are large-scale, nationally representative household surveys that collect data through standardized face-to-face interviews with women aged 15-49 years, repeated approximately every three to five years. The data used in this analysis are the product of reanalysis of DHS and MICS microdata by the International Center for Equity in Health at the Federal University of Pelotas.

Zero-dose DTP prevalence was defined as the percentage of children aged 12-23 months who did not receive any of the three DTP-containing vaccine doses recommended in infancy at the time of the DHS or MICS. In certain countries, the sample specified different age ranges to align with national vaccination schedules and ensure that our indicator is consistent with results from national survey reports and recent publications on this topic. Surveys with an age range of [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] Household economic status was based on a household wealth index, which considered ownership of household assets and access to certain services. Principal component analysis was used to construct country-specific indices and generate wealth quintiles ranging from quintile 1 (poorest) to quintile 5 (richest). Information about wealth quintiles is available in the DHS or MICS datasets.

Countries were considered for inclusion based on the availability of data for children with zero-dose DTP from a DHS or MICS conducted between 2010 and 2019. The analysis included countries categorized by the World Bank Group as low-, lower-middle-, or uppermiddle-income countries in July 2021. Countries included in the change over time analysis had additional data available from a survey conducted between 2000 and 2009 (allowing for a range of 5-15 years between the two surveys). Ten country surveys with an insufficient sample size (less than 25 for any subgroup) were excluded from the analysis.

A subset of 17 countries was identified for further analysis, based on (a) reporting more than 25% national zero-dose DTP prevalence according to the latest DHS or MICS between 2010 and 2019 or (b) being in the top 10 countries with the highest numbers of children with zero-dose DTP in 2019 [17] .

To assess the latest situation of inequality (using data from 2010-2019), we calculated summary measures of inequality using zero-dose DTP prevalence data disaggregated by wealth quintiles. Difference, a measure of absolute inequality, was calculated as prevalence in the poorest quintile minus prevalence in the richest quintile. Ratio, a measure of relative inequality, was calculated as prevalence in the poorest quintile divided by prevalence in the richest quintile.

Absolute and relative complex summary measures of inequality (slope index of inequality and concentration index, respectively), which take into account the situation across all wealth quintiles, were also calculated [18] . The slope index of inequality represents the difference in estimated zero-dose DTP prevalence between the poorest and richest quintiles using a regression model that accounts for the prevalence and weighting across all quintiles. The concentration index depicts the gradient across quintiles on a relative scale, indicating the extent to which zero-dose DTP is concentrated among the poorest subgroups. Concentration index accounts for the weighting of subgroups by population share. Negative values of the slope index of inequality and the concentration index indicate that zero-dose DTP is more prevalent among the poor than the rich.

To assess how inequalities have changed over time, annual absolute change was calculated using two data points, from 2010 to 2019 and from 2000 to 2009 (spanning about ten years). Annual absolute change was calculated as the difference in zero-dose DTP prevalence in the recent versus older surveys, divided by the number of years between the two surveys. It was calculated to quantify change in national average, and in the most and least advantaged subgroups.

Annual absolute excess change compares the annual absolute change in the least and most advantaged subgroups. For example, to arrive at absolute excess change for economic status, the annual absolute change in quintile 5 is subtracted from the annual absolute change in quintile 1, yielding annual absolute excess change in percentage points. A positive excess change indicates change favoring the richest subgroup (e.g., faster declines or slower increases in zero-dose DTP prevalence among quintile 5). A negative annual excess change indicates change favoring the poorest subgroup (e.g., faster declines or slower increases in zero-dose DTP prevalence among quintile 1).

A survey sampling design was taken into account when calculating point estimates of disaggregated data and corresponding 95% confidence intervals (95% CI). We ascertained statistical significance with 95% CIs.

Analyses were conducted using Stata 16 and Stata 17 software and data visuals were developed using Tableau software (version 2021.4).

The analysis of the latest situation of economic-related inequality included 89 countries, where the median national zero-dose DTP prevalence was 7.6%, ranging from 0% in Republic of Moldova to 72.7% in South Sudan (note: the South Sudan MICS 2010 occurred in the period leading up to South Sudan gaining independence in 2011) and 45.0% in Central African Republic (Table 1) . While over a fifth of countries (19 of 89) showed a difference of 2.0 percentage points or less between zero-dose DTP prevalence in the poorest and richest quintiles, an equal number of countries (19 of 89) had a difference of at least 20 percentage points. Five countries showed higher zero-dose DTP prevalence among the richest than the poorest by a gap of over 5 percentage points, reaching a maximum difference of 10.7 and 10.1 percentage points in Tunisia and Namibia, respectively. The majority of countries (54 of 89) had prevalence that was at least twice as high in the poorest quintile than the richest. 3.6 (1.9-6.7) Across 89 countries, the median zero-dose DTP prevalence among the poorest quintile (9.9%) was higher than the median prevalence among the richest quintile (5.0%) (Figure 1 ). Across all countries, the range of prevalence values in the richest quintile was smaller than the range of values in the poorer quintiles. In low-income countries (n = 22), the difference between the median prevalence in the poorest quintile (20.7%) and richest quintile (6.3%) was 14.4 percentage points. Lower-middle-income countries (n = 41) and upper-middleincome countries (n = 26) had comparatively smaller gaps, at 8.9 percentage points and 2.7 percentage points, respectively. Across 89 countries, the median zero-dose DTP prevalence among the poorest quintile (9.9%) was higher than the median prevalence among the richest quintile (5.0%) (Figure 1) . Across all countries, the range of prevalence values in the richest quintile was smaller than the range of values in the poorer quintiles. In low-income countries (n = 22), the difference between the median prevalence in the poorest quintile (20.7%) and richest quintile (6.3%) was 14.4 percentage points. Lower-middle-income countries (n = 41) and upper-middle-income countries (n = 26) had comparatively smaller gaps, at 8.9 percentage points and 2.7 percentage points, respectively. Slope index of inequality demonstrated similar patterns of inequality as difference across country income groupings, with the absolute extent of inequality tending to be slightly higher when taking all subgroups into account. Concentration index demonstrated similar patterns as ratio.

The change over time analysis included 65 countries with available data ( Table 2 ). The median absolute annual change in national average across countries was −0.20 per year, indicating that, over a period of ten years, the national zero-dose DTP prevalence declined by 2.0 percentage points (median across 65 countries). Across countries, inequality hardly changed over time, with a low median change in both the poorest quintile (−0.16, or a decline of 1.6 percentage points over ten years) and the richest quintile (−0.04, or a decline of 0.4 percentage points over ten years). Slope index of inequality demonstrated similar patterns of inequality as difference across country income groupings, with the absolute extent of inequality tending to be slightly higher when taking all subgroups into account. Concentration index demonstrated similar patterns as ratio.

The change over time analysis included 65 countries with available data ( Table 2 ). The median absolute annual change in national average across countries was −0.20 per year, indicating that, over a period of ten years, the national zero-dose DTP prevalence declined by 2.0 percentage points (median across 65 countries). Across countries, inequality hardly changed over time, with a low median change in both the poorest quintile (−0.16, or a decline of 1.6 percentage points over ten years) and the richest quintile (−0.04, or a decline of 0.4 percentage points over ten years). Overall, changes in inequality and national average over time demonstrated different patterns across countries (Figure 2) . Several countries showed the most desirable trend: decreased national zero-dose DTP prevalence, with larger improvements among the poorest households. Fourteen countries had a statistically significant decrease in absolute annual excess change. Most markedly, in Niger, the national zero-dose DTP prevalence decreased by 4.6 percentage points per year between 2006 and 2012, with much more reduction among the poorest quintile than the richest (absolute annual excess decline of 4.1 percentage points). Similarly, Burkina Faso, India, and Turkey decreased their national zero-dose DTP prevalence with narrowing inequality, while Namibia had no change in national prevalence alongside narrowing inequality (driven by increased zero-dose DTP prevalence in the richest quintile). Guinea and Madagascar both showed increased national zero-dose DTP prevalence driven by faster increases in the poorest quintile than the richest.

Many countries showed little or no absolute change in national average or economicrelated inequality over the past ten years. Egypt and Mongolia sustained low zero-dose DTP prevalence (around 1% and 3%, respectively), with absolute economic-related inequality remaining around 2 percentage points or less. In other countries, such as Benin, high zero-dose DTP prevalence remained unchanged (above 15%), with a sustained gap of at least 25 percentage points between the poorest and richest quintiles.

In Congo, Côte d'Ivoire, and Haiti, the national zero-dose DTP prevalence remained largely unchanged between the two surveys (in the range of 14-20%), though the situation across wealth quintiles changed in divergent ways. In Haiti, the gap between the rich and poor widened due to increased prevalence among the poorest quintile (by more than 6 percentage points across ten years) and prevalence decreasing among the richest (by more than 7 percentage points across ten years). In Congo and Côte d'Ivoire, the zero-dose DTP prevalence increased among the richest quintile (by more than 9 percentage points across ten years) and decreased among the poorest quintile (by around 4-6 percentage points across ten years).

A subset of 17 countries had zero-dose DTP prevalence of more than 25% according to the latest available survey since 2010 (n = 13) and/or highest numbers of children with zero-dose DTP (n = 9). This subset consists of seven low-income countries, nine lowermiddle-income countries, and one upper-middle-income country. All 17 countries reported higher prevalence in the poorest quintile than the richest, with difference ranging from 2.7 percentage points in Mexico to 53.2 percentage points in South Sudan (median difference of 32.1 percentage points across 17 countries).

Eleven of these countries had data from two or more surveys (Figure 3 ). Of these, four countries-Central African Republic, Guinea, Madagascar, and Philippines-had a statistically significant increase in national zero-dose DTP prevalence, with prevalence tending to increase in all wealth quintiles (except Central African Republic, where the prevalence among the richest quintile was unchanged). The gap between the richest and poorest increased by up to 12.2 percentage points over ten years in Guinea. Six countries had a statistically significant decrease in national prevalence. In India and Indonesia, there were large reductions in prevalence in the poorer subgroups which narrowed absolute inequality by 24.7 percentage points and 17.7 percentage points, respectively, over ten years. Nigeria and Pakistan reported improvements in all quintiles, with faster declines among the poorest than the richest, while Ethiopia reported faster improvements among the richest. In Lao People's Democratic Republic, the zero-dose DTP prevalence declined among all except the poorest quintile. centage points). Similarly, Burkina Faso, India, and Turkey decreased their national zerodose DTP prevalence with narrowing inequality, while Namibia had no change in national prevalence alongside narrowing inequality (driven by increased zero-dose DTP prevalence in the richest quintile). Guinea and Madagascar both showed increased national zero-dose DTP prevalence driven by faster increases in the poorest quintile than the richest. 

In this study, we described economic-related inequalities in zero-dose children across 89 low-and middle-income countries. Our results indicate that widespread economic-related inequalities to the detriment of the poorest countries, similar to previous multi-country analyses of inequalities in DTP3 coverage [19] [20] [21] . Our focus on zero-dose DTP prevalence, the flagship indicator of IA2030 and Gavi 5.0, yielded new insights into changes in inequality over time. However, a previous analysis of trends in economic-related inequality in DTP3 coverage showed narrowing inequality over the previous decade [20] , while the present analysis of zero-dose DTP shows that economic-related inequalities have been largely sustained over the previous ten years. These divergent findings underscore the importance of dedicated monitoring and tracking the unvaccinated, using zero-dose DTP as the tracer indicator. The stark increase in the number of zero-dose children in 2020 further underscores the importance of the equity orientation of Gavi 5 .0 and IA2030 and the need to restore, strengthen, and expand the reach of immunization systems.

Our study was the first, to our knowledge, to report socioeconomic inequality in zero-dose DTP across country income groupings. We reported more pronounced economic-related inequality in low-income countries than lower-middle-and upper-middleincome countries; however, there was variation across countries in the extent of inequality. The group of low-income countries demonstrated greater variation than lower-middle-or upper-middle-income countries. We found no apparent differences between country income groupings for changes in economic-related inequality over time. Previous research addressing trends in national DTP3 coverage in 46 African countries compared coverage in 2000 and 2015 by country income grouping and reported substantial improvements in the low-income countries; however, unlike our study, their analysis allowed for different country classification at the two time periods, and it did not report on socioeconomic inequality [22] .

Multi-country studies such as ours are useful to describe global trends and inform where further studies and/or targeted resource distribution are needed. Our initial investigation of inequalities in a subset of 17 countries with a high prevalence or a high number 

In this study, we described economic-related inequalities in zero-dose children across 89 low-and middle-income countries. Our results indicate that widespread economicrelated inequalities to the detriment of the poorest countries, similar to previous multicountry analyses of inequalities in DTP3 coverage [19] [20] [21] . Our focus on zero-dose DTP prevalence, the flagship indicator of IA2030 and Gavi 5.0, yielded new insights into changes in inequality over time. However, a previous analysis of trends in economic-related inequality in DTP3 coverage showed narrowing inequality over the previous decade [20] , while the present analysis of zero-dose DTP shows that economic-related inequalities have been largely sustained over the previous ten years. These divergent findings underscore the importance of dedicated monitoring and tracking the unvaccinated, using zero-dose DTP as the tracer indicator. The stark increase in the number of zero-dose children in 2020 further underscores the importance of the equity orientation of Gavi 5 .0 and IA2030 and the need to restore, strengthen, and expand the reach of immunization systems.

Our study was the first, to our knowledge, to report socioeconomic inequality in zero-dose DTP across country income groupings. We reported more pronounced economicrelated inequality in low-income countries than lower-middle-and upper-middle-income countries; however, there was variation across countries in the extent of inequality. The group of low-income countries demonstrated greater variation than lower-middle-or upper-middle-income countries. We found no apparent differences between country income groupings for changes in economic-related inequality over time. Previous research addressing trends in national DTP3 coverage in 46 African countries compared coverage in 2000 and 2015 by country income grouping and reported substantial improvements in the low-income countries; however, unlike our study, their analysis allowed for different country classification at the two time periods, and it did not report on socioeconomic inequality [22] .

Multi-country studies such as ours are useful to describe global trends and inform where further studies and/or targeted resource distribution are needed. Our initial investigation of inequalities in a subset of 17 countries with a high prevalence or a high number of zero-dose children (including countries with low performance, but also better performing countries with large birth cohorts) revealed divergent patterns, particularly with regards to how inequality changed over time. Notably, India and Indonesia achieved substantial reductions in economic-related inequality, while the situation in Guinea and Madagascar became worse, in terms of both inequality and national average. Country-specific studies are required to explore contextual factors that underlie existing socioeconomic inequalities and identify lessons learned in areas where the situation is improving. While cross-cutting strategies to improve equity in vaccination coverage have been identified, including improving the relationships between the health system and mothers who are socioeconomically disadvantaged, setting-specific approaches should be tailored to account for local considerations [11, 23] .

Since the inception of the Expanded Program on Immunization in the 1970s, all countries have included DTP3 in the national vaccination schedules, as per the recommendation of the WHO. There are complex reasons why children from socioeconomically disadvantaged subgroups lack access to vaccinations. Women's low social status is a barrier that cuts across family, community, and health facility environments, and manifests through educational attainment, income, decision-making, and resource allocation [14, 24] . Deliberate measures are needed to mitigate women's lack of access to immunization services and encourage greater involvement of fathers and communities in immunization initiatives. In sub-Saharan Africa, attributes related to parents or caregivers that are most frequently cited as barriers to vaccination include lack of knowledge about immunization, distance to access point, financial deprivation, lack of support from partner and distrust in vaccines, and immunization programs, whereas pertinent barriers related to health systems include human resource limitations, inadequate infrastructure, and supply chain inadequacies [25] .

Others have explored the reasons underlying socioeconomic inequalities in vaccine coverage. For instance, across 46 low-and middle-income countries, pro-rich inequality in the receipt of four basic vaccines in childhood was associated with higher antenatal care visit attendance among wealthier women [26] . A study of country-level determinants reported national political stability, gender equality, and smaller land surface as factors that were predictive of more equitable DTP3 coverage in terms of wealth, education, and multidimensional poverty [27] . With a novel approach to measure multi-dimensional equity in immunization coverage, Patenaude and Odihi (2021) reported maternal education, socioeconomic status, and health insurance coverage as the primary drivers of unfair disadvantage in zero-dose status in India (accounting for 31%, 16%, and 7% of inequality, respectively) [28] . Further study of factors specific to "hard-to-reach" zero-dose children is warranted [29] .

Deriving estimates and assessing inequalities of zero-dose children, a population defined by its lack of contact with health systems, is subject to measurement limitations [30] . Our use of household survey data helped to overcome inaccurate denominator values and absence of information about socioeconomic variables, i.e., major limitations of administrative data sources. Surveys, however, are subject to other limitations and biases, including exclusion of settings that are conflict-affected or hard to reach, respondent recall/selfreporting inaccuracies, and inaccurate or outdated sampling frames. Furthermore, children who were delayed in receiving vaccines may not have been captured. For these reasons, the reported estimates may underestimate zero-dose DTP prevalence.

Our analysis used a lack of DTP vaccination as a measure of zero-dose children, following the operational definition used by IA2030, with reports by World Health Organization and UNICEF [17] which can be measured with different data sources, and emphasized the reach of routine immunization services. Our study yielded similar results as a previous analysis that defined zero-dose children by the lack of vaccination of four basic vaccines universally recommended in infancy (BCG, DTP, MCV, and polio) [9] . 

Alongside indicators of vaccination coverage and drop-out, monitoring and reporting on zero-dose children is a stated priority of IA2030 and Gavi 5.0. The indicator provides valuable insights on children who lack access to routine immunizations and can help to inform actions to equitably expand the reach of immunization programs and basic healthcare services. In other words, actions should be equity-oriented to achieve national improvements alongside narrowing socioeconomic inequality (faster improvement among the poorest subgroups). To this end, our analysis of economic-related inequalities in zerodose DTP prevalence serves as an important baseline assessment to benchmark future efforts to reduce the number of zero-dose children. Socioeconomic inequalities in zero-dose children, alongside other dimensions of inequality, should be monitored regularly as new data become available. Investments should expand and strengthen data collection and analysis, with a focus on disaggregated data. Further in-depth analysis of country-level inequalities is warranted, including exploration of the drivers of inequality and contextspecific strategies to address them.

Immunization Agenda 2030: A Global Strategy to Leave No One Behind

Essential Programme on Immunization: Reducing Missed Opportunities for Vaccination (MOV)

Measuring routine childhood vaccination coverage in 204 countries and territories, 1980-2019: A systematic analysis for the Global Burden of Disease Study 2020, Release 1

Impact of the SARS-CoV-2 pandemic on routine immunisation services: Evidence of disruption and recovery from 170 countries and territories

Estimating global and regional disruptions to routine childhood vaccine coverage during the COVID-19 pandemic in 2020: A modelling study

Routine vaccination coverage-Worldwide

Finding children with high risk of non-vaccination in 92 low-and middle-income countries: A decision tree approach

Factors associated with vaccination status of children aged 12-48 months in India

Zero-dose children and the immunisation cascade: Understanding immunisation pathways in low and middle-income countries

Reaching Zero-Dose Children

Children: Key Issues for Consideration. Equity Reference Group for Immunization

Child vaccination in sub-Saharan Africa: Increasing coverage addresses inequalities

Unvaccinated children in years of increasing coverage: How many and who are they? Evidence from 96 low-and middle-income countries: Unvaccinated children in years of increasing coverage

The influence of gender on immunisation: Using an ecological framework to examine intersecting inequities and pathways to change

Progress in reaching unvaccinated (zero-dose) children in India, 1992-2016: A multilevel, geospatial analysis of repeated cross-sectional surveys

Demand-and supply-side determinants of diphtheria-pertussis-tetanus nonvaccination and dropout in rural India

World Health Organization and UNICEF; UNICEF Immunization Unit. Immunization Coverage: Are We Losing Ground?

Summary measures of health inequality: A review of existing measures and their application

Access to skilled attendant at birth and the coverage of the third dose of diphtheria-tetanus-pertussis vaccine across 14 West African countries-An equity analysis

State of inequality in diphtheria-tetanus-pertussis immunisation coverage in low-income and middle-income countries: A multicountry study of household health surveys

Monitoring equity in vaccination coverage: A systematic analysis of demographic and health surveys from 45 Gavi-supported countries

State of equity: Childhood immunization in the World Health Organization African region

Addressing the persistent inequities in immunization coverage

Gender determinants of vaccination status in children: Evidence from a meta-ethnographic systematic review

Barriers to childhood immunization in sub-Saharan Africa: A systematic review

Socioeconomic inequalities in child vaccination in low/middle-income countries: What accounts for the differences?

Country-level predictors of vaccination coverage and inequalities in Gavi-supported countries

A Standardized Approach for Measuring multi-Dimensional Equity in Vaccination Coverage, Cost-of-Illness, and Health Outcomes: Evidence from the Vaccine Economics Research for Sustainability & Equity (VERSE) Project. SSRN Electron

Defining hard-to-reach populations for vaccination

Challenges in measuring supplemental immunization activity coverage among measles zero-dose children

Acknowledgments: Katherine Kirkby and Cecilia Vidal Fuertes provided inputs into the early conceptualization of the paper and contributed to data preparation.

The authors declare no conflict of interest. The authors alone are responsible for the views expressed in this publication and they do not necessarily represent the views, decisions or policies of their institutions.