Estimation of Child Mortality Rate Using Brass-Trussell Technique and Hypothetical Cohort Approach between Two Censuses in Iran

This paper aims to evaluate and adjust the error in estimation of infant mortality rate (IMR) and under five mortality rate (U5MR) using Brass Model when fertility rate has a declining trend. This paper applies a hypothetical cohort of changes of average children ever born and children died between two censuses/surveys (1986-2011) in Iran. The results indicate an overestimation of 14 percent in the midpoint of the period 1986 to 1996 and 1996 to 2006 and about 37 percent over-estimation for the period 2006 to 2011 both in IMR and U5MR. With omitting non random error, IMR and U5MR in 2005 had been estimated about 23.5 and 30 per thousand, respectively. These estimates are close to estimated rate by international agencies, but indicate an considerable underestimation in calculated rate based on registered deaths and estimation of DHS. By accepting the logic of the approach used in this study, vital registration system of the country has registered under 54% of child death under five and estimation of IMR and U5MR by DHS had a 15-25% underestimation, respectively.