Child Mortality Data Source and GPR Results

Published: 17 March 2017| Version 1 | DOI: 10.17632/9z3pzd6rmd.1
Farshad Farzadfar


There are three types of data that include the summary and complete birth history data. Census data, the Demographic and Health Survey (DHS) and the ARJ project (Iranian population growth rate measurement) contain SBH questions only; the DHS contains CBH questions as well. Subsequently, the census data of 1986, 1996, 2006, and 2011, the DHS data of 2000 and 2010, and those of the ARJ project in 1999 were determined as the sources of data. 1. Census The first and last ‘National Population and Housing Census’ in Iran took place in 1956 and 2011, respectively. As per the 4th Article of Iran’s Statistics Center, this census is to be performed every five years upon the President’s order. The next National Population and Housing Census in Iran takes place in 2016. The main goal of this census is to collect statistics and data from all over the country to identify demographic and housing characteristics. The latter is used to process data required for social development, economic and cultural planning, and also to prepare grounds for appropriate frameworks for statistical investigations. The census forms include questions on the status of morbidity and mortality. SBH –related questions are also included in these questionnaires. In Iran, these questions have been included in the questionnaires since 1981, in the following form: Total number of female/male children born alive; total number of female/male children currently alive (living with this or another household). We used the SBH data of the censuses of the years 1986, 1996, 2006, and 2011. 2. Demographic and Health Survey (DHS) This study has in fact been derived from global studies on fertility and prevalence of contraceptive use, in which, data related to fertility, pregnancy, maternal and child health and nutrition in developing countries are collected. However, data on malaria, AIDS, and domestic abuse are collected voluntarily. In order to estimate child mortality in this program, the SBH and CBH questions are included in the questionnaire as well. The SBH and CBH data of the 2000 DHS project were used to estimate child mortality. The quality of the 2010 data was poor, so they were put aside and not used for the study. 3. MOHME’s Death Registration System Currently, two organizations are responsible for registering deaths in Iran; the National Organization for Civil Registration and MOHME’s Death Registration System. The National Organization for Civil Registration data contains many errors and delays in reporting. In 1995, an effort was made in the Bushehr province to launch a registry using all possible sources and afterwards, this step was taken across the entire country. Initially (from 1995 to 2001), MOHME’s Deputy of Research was in charge of registering death data. Afterwards, MOHME’s Deputy of Health took charge and has continued to do so since then. Here, we have used the data from the years 1995 to 2010. The 2005 data were not available, hence they were not used.


Steps to reproduce

1. Identification of data sources 2. Data preparation 3. Analysis of data 1- Summary Birth History (SBH) data, using: a. Maternal Age Cohort (MAC) method b. Maternal Age Period (MAP) method c. Combining the estimates derived from the two abovementioned methods through the Loess method 2- Complete Birth History (CBH) data 4. Combining the results of different trends through Gaussian Process Regression (GPR)


Age, Mortality, Shock, Child, Census, Gaussian Process, National Surveys