Temporal Patterns in Vaccination Disparities in 23 Low- to Middle-Income Countries: Tracking Progress and Potential Impact on Health Gains

Developed by:
Richard Rheingans, PhD
John Anderson, MS
Camila Pazos, MDP
Erica Odera, MDP

Click here to view the interactive model.

Background and Objectives

This research and examines temporal patterns of vaccination disparities in 23 middle- to low-income countries using data from Demographic and Health Surveys (DHS) national household surveys [1] in sub-Saharan Africa, South Asia and Latin America. We include coverage for three doses of DPT (diphtheria, pertussis, and tetanus), four doses of the oral polio vaccine (OPV), one dose of Bacillus Calmet-Guerin (BCG) against tuberculosis, and one dose of measles vaccine.

The objectives are to use the interactive model identify countries that have significantly improved vaccine equity and those that are lagging behind by assessing trends by economic status within individual countries. More specifically, this research addresses the following questions:

1.  Which countries have made improvements in coverage and disparities?

2.  Does equity improve with coverage improvements?

3.  Does equity improve with higher national income?

4.  Which countries have high disparity in both vaccine coverage and high mortality?

5.  Are these patterns the same for all vaccines?

Data Description

In order to examine temporal changes in this relationship, countries were selected based on availability of data after 2005 for countries that also had previously been surveyed within the last 5 to 6 years. While the dates and period between them differ among countries, they provide an indication of change over time, with the limitation that additional changes may have occurred subsequent to the most recent survey years.

For all countries we extracted aggregated data from the DHS data portal that provides coverage estimates by region or wealth status.  In order to compare vaccine coverage, BCG, DPT3, OPV3, and measles vaccines for children 12-23 months of age were examined. We also extracted aggregate data on child mortality as a proxy for risk of vaccine preventable diseases.

Equity Ratio and Concentration Index

Several measures are available to characterize the level of disparity in outcomes or levels of service. Equity ratios and concentration indices were constructed as overall measures of disparities in coverage. The equity ratio is the ratio of coverage in the poorest quintile compared to that of the richest. A concentration index, analogous to the GINI Index, was created for each country using the wealth index and vaccination coverage data [2].

The concentration index is derived concentration curve graphs that display the cumulative fraction of outcomes (such as vaccination) on the vertical axis and the cumulative fraction of subjects by economic status on the horizontal axis. For an evenly distributed outcome, the bottom 40% of children ranked by wealth would account for 40% of vaccinated children. This equal line is referred to as the ‘line of equity’. The concentration index is defined as 2 times the area between the line of equity and the actual concentration index.

While the concentration index captures more information about the level of disparity, the equity ratio is somewhat simpler to interpret. As a result the equity ratio is the default measure in most figures, but the concentration index results are reported in our supplementary materials.

Results

The concentration index, equity ratios, and temporal coverage data were examined to determine whether improvements in coverage were associated with reductions in disparities. For each vaccine, we compared national level equity ratios and concentration index values to examine whether improvements in coverage occurred between the two time periods and if these were associated with reductions in disparities.  Between the early and late 2000s, the 23 countries experienced different trajectories in coverage and disparities in BCG, DPT3, OPV3 and measles vaccinations.

Overall, nationwide vaccination coverage has increased over time for almost all countries for each of the four vaccines considered. For all four vaccines considered, the majority of countries made substantial improvements in coverage or were already at relatively high levels. However some countries experienced limited increases and remained relatively low in overall coverage for specific vaccines. For many countries, improvements in overall coverage levels resulted in reductions in disparities (higher equity ratios). Other countries made limited progress in reducing disparities (i.e., increasing equity ratios) for the various vaccines.

Demographic and Health Surveys Used in the Analysis

Country

Years

Abbreviation

Bangladesh

2004, 2007

BD

Bolivia

2003, 2008

BO

Cambodia

2005, 2010

KH

Colombia

2005, 2010

CO

Dominican Republic

2002, 2007

DR

Ethiopia

2000, 2011

ET

Ghana

2003, 2008

GH

Haiti

2000, 2005-6

HT

India

1998-9, 2005-6

IA

Indonesia

2002-3, 2007

ID

Kenya

2003, 2008-9

KE

Madagascar

2003-4, 2008-9

MD

Malawi

2004, 2010

MW

Mali

2001, 2006

ML

Namibia

2000, 2006-7

NM

Nepal

2001, 2011

NP

Niger

1998, 2006

NI

Nigeria

2003, 2008

NG

Philippines

2003, 2008

PH

Rwanda

2005, 2010

RW

Tanzania

2004-5, 2010

TZ

Uganda

2000-1, 2006

UG

Zimbabwe

1999, 2011

ZW

References

1.  Demographic and Health Surveys: STATcompiler.  [cited 2012 May 21]; Available from: http://www.statcompiler.com/.

2.  O’Donnell et al. 2008.  Analyzing Health Equity Using Household Survey Data: A guide to techniques and their implementation. Washington, DC: The World Bank.