Estimating life-tables for very small areas in a national context: an analysis of Israel statistical areas

Jon Anson, Ben Gurion University of the Negev

As mortality declines, so the shape of the mortality curve necessarily changes, with deaths being more and more concentrated at older ages. Historical evidence, based on national data, has shown variations in the ages at which major declines have occurred, by time period and the level of mortality, but there has been little attention paid to variations, by level of mortality, within one national system. For this we need estimates of local area mortality curves, by age, from which life tables may be derived, but these are problematic given the paucity of deaths by age and sex in each local area and the consequent large estimation errors. A possible solution to this problem is to estimate age specific mortality rates simultaneously for all the subunits of a particular country, using the reported number of deaths, by age and sex, for each unit as the input data, together with information on local social conditions and an estimate of overall mortality. We use population and social data for 1337 statistical areas (enumeration districts) from the 1995 Israeli census, together with numbers of deaths over the five years 1993 to 1997, and compare the results obtained with those from a model which makes no adjustment for the change in the shape of the mortality curve. The social data for each unit include estimates of the average standard of living, a measure of traditionality of the family structure and population group (Jewish mainstream, Ultra-Orthodox or Palestinian-Arab). We show that as mortality declines, the major structural change is in mid-adulthood, with an extension of the period during which mortality is at a minimum, before beginning to rise into old age.