7 Point Plan
Data used in this report
The data used in this report were derived from a
range of sources, which are summarized below.
Childhood diarrhoea burden
Estimates of the global and regional number of diarrhoea cases and deaths are based on the Global
Burden Disease project, and are for the year 2004 (the latest year estimates are available). The project
provides a comprehensive assessment of mortality and loss of health due to diseases, injuries and risks
for all regions of the world. When reviewing these estimates, it is important to note that the distribution
of under-five deaths by cause refers to the primary cause of death. The estimated percentage
distribution of cause-specific mortality for the year 2004 was applied in this report to the 2008 envelope
of total under-five deaths worldwide (8.8 million in 2008) to arrive at the number of under-five deaths
due to diarrhoea globally.
Prevention and treatment coverage
Data on prevention and treatment interventions were derived largely from national-level household
surveys, notably the Multiple Indicator Cluster Surveys (MICS), supported by UNICEF, and the
Demographic and Health Surveys (DHS), supported by the United States Agency for International
Development (USAID). Information from the surveys are compiled by UNICEF Headquarters
and made available in a series of public-access databases found at
are also published annually in The State of the World’s Children
Multiple Indicator Cluster Surveys
are nationally representative, standardized sample surveys to
which UNICEF provides financial and technical support. Since their inception in 1995, nearly 200
MICS have been carried out globally. The latest round of surveys (MICS3) was conducted in more
than 50 countries between 2005 and 2006. The next round is scheduled for 2009-2010. More
information is available at
Demographic and Health Surveys
are also nationally representative, standardized surveys that are
usually implemented every five years with funding from USAID. The DHS is designed to collect a variety
of data on a broad range of demographic and health issues and to be comparable over time and across
countries. More information is available at
The indicators to monitor prevention coverage
(such as immunization, nutrition, and water and sanitation) presented in this report are based
on well known and long-standing child survival indicators that are regularly used to monitor
progress towards global goals and commitments. For example, data on water supply and sanitation
are based on the work of the WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation
). Further information on
these prevention indicators, such as definitions and data sources, are available at
Oral rehydration therapy with continued feeding
– Proportion of children aged 0-59 months
with diarrhoea receiving oral rehydration salts or recommended homemade fluids or increased fluids,
and continued feeding during the diarrhoea episode.
This indicator reflects the UNICEF and WHO programme
recommendations for diarrhoea treatment (with the exception of zinc). The recommendations
were developed on the basis of broad consensus by leading experts in the field during a UNICEF and
WHO advisory meeting in 2004,45
in a 2007 follow-up technical conference.46
– Proportion of children aged 0-59 months with diarrhoea receiving oral rehydration
salts during the diarrhoea episode.
This indicator is based on information provided by mothers or
caregivers. Caregivers are asked whether their children suffered from diarrhoea in the two weeks
prior to the survey. If so, they are then asked if the child received a fluid made from a special packet of
ORS or a pre-packaged ORS fluid (where applicable) during the diarrhoea episode.
Recommended homemade fluids
– Proportion of children aged 0-59 months with diarrhoea receiving
a government-recommended homemade fluid (to be customized based on national guidelines)
during the diarrhoea episode.
This indicator is based on information provided by mothers or caregivers.
Caregivers are asked whether their children suffered from diarrhoea in the two weeks prior to the survey.
If so, they are then asked if the child received a government-recommended homemade fluid during
the diarrhoea episode. This question should be customized prior to starting survey work to include
the specific fluids recommended by national guidelines to treat diarrhoea.
– Proportion of children aged 0-59 months with diarrhoea receiving more to drink
during the diarrhoea episode.
This indicator is based on information provided by mothers or caregivers.
Caregivers are asked whether their children suffered from diarrhoea in the two weeks prior to
the survey. If so, they are then asked if (during this illness) the child received none, much less,
somewhat less, about the same, or more to drink than usual. Children reported to have received
more to drink than usual during the illness are considered to have received this intervention.
– Proportion of children aged 0-59 months with diarrhoea receiving more, about
the same or somewhat less food during the diarrhoea episode.
This indicator is based on information
provided by mothers or caregivers. Caregivers are asked whether their children suffered from diarrhoea
in the two weeks prior to the survey. If so, they are then asked if (during the illness) the child
received none, much less, somewhat less, about the same, or more food than usual. Children reported
to have received either somewhat less, about the same, or more food than usual during the illness
are considered to have received this intervention.
Regional and global estimates
– These are based on population-weighted averages, weighted by the
total number of children under five years of age. These estimates are presented only if available data
cover at least 50 per cent of total children under five years of age in regional or global groupings. The list
of countries included in these groupings is available
Trends over time
– Changes in treatment indicator
definitions over the years have resulted in a relative lack of comparable data from the 1990s in order to
assess trends over time. To the extent possible, data collected through previous surveys have been reanalysed
for the purposes of this report to conform to the current indicator definition (ORT with continued
feeding) to monitor treatment coverage.
For each of the treatment indicators, regional assessments of trends over time were conducted on
the basis of a subset of countries with two or more comparable data points around the time periods
2000 and 2007. A linear regression line was then fitted through all available data points for each
country included in the assessment to derive an estimate for the earlier (2000) and later time periods
(2007). A regional estimate was then presented in this report if the subset of countries included in
the trend analyses represented at least 50 per cent of the total children under five in the regional or global grouping.
Interpreting treatment coverage data from household surveys
The interpretation of treatment coverage must take
into account a number of important issues:
First, the indicator to monitor current treatment
guidelines (ORT with continued feeding) reflects the multiple components of this recommendation (with
the exception of zinc). As mentioned previously, this indicator definition was developed and agreed upon
by leading experts at a UNICEF and WHO meeting in 2004, and was recently re-confirmed during a
Countdown to 2015 technical meeting in 2007.47
is important to evaluate the contribution of each individual component of the indicator to the overall
coverage value, and this assessment was presented in this report.
Second, for some countries, comparisons of treatment coverage based on the current indicator
definition with previously used indicators may result in markedly different values. There may also be a different assessment of trends over time
depending on the indicator used. It is therefore important for countries to adopt and promote the
latest recommendations for treating diarrhoeal diseases, and to monitor these programmes using
the appropriate indicators.
Third, prevalence estimates derived from national-level household surveys may vary markedly by
season and by timing of outbreaks (such as cholera). Prevalence estimates are also affected
by the survey respondents’ understanding of what constitutes a diarrhoea episode. The survey does
not measure the type of diarrhoea experienced by the child (including its length and severity),
nor the extent of dehydration resulting from the diarrhoea episode. These prevalence estimates
are used to derive the denominator for diarrhoea treatment coverage values.
Fourth, information is not collected on the number and timing of interventions used during the diarrhoea
episode, including whether children received early ORS administration, the number of ORS
packets received during the course of the illness, or whether homemade fluids were correctly prepared.
In addition, different countries have different guidelines on what constitutes a recommended
homemade fluid. These policies are not always clearly defined, and survey questions may therefore
not be customized for countries according to their specific national guidelines prior to starting
survey work. In these cases, survey respondents must decide for themselves if the fluid the child
received was a government-recommended one, leading to major data quality issues.
Fifth, while questions on diarrhoea treatment have been incorporated into major national-level household
surveys, such as the DHS, since the 1980s, there have been a number of slight changes to the
construction of these questions over time, as well as their response categories. Here again, it is important
to note that survey questionnaires are translated into different languages, which may also result in
slight differences in the wording of questions across countries and over time – affecting data collected
not only for the diarrhoea treatment indictors, but for other information as well. Further research is
needed to determine the extent to which these slight wording changes may have affected overall
The change in response categories for the continued feeding indicator around 2000 has particularly
affected the availability of data to report on this indicator as well as the ORT with continued feeding
indicator. For example, prior to 2000, caregivers were asked if the child received more, the same
or less to eat during the diarrhoea episode. After that time, the response categories were revised to
include ‘somewhat less’ and ‘much less’ in addition to the other categories. Children are considered to
have received the continued feeding intervention if they received the same, more or somewhat less
to eat during their illness. This leads to a lack of available data to measure this indicator definition
prior to 2000, given that the caregiver responses did not include a ‘somewhat less’ category.