WHO Statistical Information System (WHOSIS)

Distribution of causes of death among children aged < 5 years (percentage of total)

Rationale for use

The target of Millennium Development Goal 4 is to “Reduce by two thirds, from 1990 to 2015, the under-five mortality rate”. Efforts to improve child survival can be effective only if they are based on reasonably accurate information about the causes of childhood deaths. Cause-of-death information is needed to prioritize interventions and plan for their delivery, to determine the effectiveness of disease-specific interventions, and to assess trends in disease burden in relation to national and international goals.

Definition

The causes of death are those that are entered on the medical certificate of cause of death in countries and recorded by the civil (vital) registration systems. For the analyses, we have used the concept of the 'underlying cause of death' as defined by ICD (WHO, 1992). In countries with incomplete or no civil registration, causes of death are those reported as such in epidemiological studies that use verbal autopsy algorithms to establish cause of death.

Associated terms

The under-5 mortality rate is strictly speaking not a rate (i.e. the number of deaths divided by the number of population at risk during a certain period of time), but a probability of death derived from a life table and expressed as rate per 1000 live births.

Underlying cause of death has been defined by ICD (WHO, 1992) as “(a) the disease or injury which initiated the train of morbid events leading directly to death, or (b) the circumstances of the accident or violence which produced the fatal injury”.

Data sources

WHO regularly receives mortality-by-cause data from Member States, as recorded in national civil (vital) registration systems. These statistics were therefore analysed to obtain the distribution of child deaths by cause in 72 countries where those systems are judged to be sound (on the basis of reliable diagnostic procedures and standard application of cause coding that follows ICD rules as applied to death certificates) and have coverage rates of 85% or above, all of which are high- or middle-income countries. For countries with an incomplete or no vital registration system, epidemiological studies and statistical modelling were used extensively.

Methods of estimation

Cause-of-death data from civil registration systems were evaluated for their completeness. Complete and nationally-representative data were then grouped by ICD codes into the cause categories, and the proportions of these causes with regard to the total number of deaths of children aged less than 5 years were then computed. For countries with incomplete data or no data, the distribution of deaths by cause was estimated in two steps. In the first step, a statistical model was used to assign deaths to one of three broad categories of causes: communicable diseases; noncommunicable diseases; or injuries and external causes. In a second step, cause-specific under-five mortality estimates from the Child Health Epidemiology Reference Group (CHERG), WHO technical programmes, and UNAIDS were taken into account in assigning the distribution of deaths to specific causes. A variety of methods, including proportional mortality and natural history models, were used by CHERG and WHO to develop country-level cause-specific mortality estimates. All CHERG working groups developed comparable and standardized procedures to generate estimates from the databases.

Disaggregation

Country level, age groups—neonatal (0–27 days) and 28 days to 59 months.

References

Database

Comments

Percentage of deaths among children aged less than 5 years was estimated for eight cause categories only. There are still estimates of some of the major causes of child deaths (e.g. injuries) that have not yet been developed using the CHERG methods. Also, a better understanding of the indirect contributions of diseases to child deaths is needed in order to assess disease control priorities and evaluate interventions.

Share