Avoidable mortality and life expectancy in New Zealand.

The effect of avoidable mortality on the life expectancy of Maori and non-Maori New Zea-landers has been examined using a method similar to that of Benavides et al' for the Valencia region of Spain. Deaths that are avoidable by primary and secondary prevention are considered separately and the potential number of years of life gained by their prevention is compared. Methods: Sex specific abridged life tables were constructed for the Maori and non-Maori populations using data from the Departments of Health and Statistics. Further tables were then constructed with: (1) primary avoidable, (2) secondary avoidable, and (3) all avoidable deaths omitted from the numerator of the age specific mortality rates. Primary avoidable deaths2 (at the specified ages) were: lung cancer (5-64); liver cirrhosis (15-74); and motor vehicle accidents. Secondary avoidable deaths2 were: respiratory diseases, measles (1-14); whooping cough (0-14); tuberculosis , Hodgkin's disease, appendicitis, hernia, cholelithiasis and cholecystitis, typhoid (5-64); cervical cancer (15-64); chronic rheumatic heart disease, asthma (5-44); tetanus (0-64); osteo-myelitis, hypertensive and cerebrovascular diseases (35-64); perinatal and maternal deaths. Potential gains in life expectancy are expressed as years of life gained (YLG) and as a percentage of the sex and race specific "true" life expectancy-that is, that calculated with all causes of death included. As in Benavides et al,l competitive mortality is not accounted for, but it is assumed that people who do not die if avoidable deaths are prevented will not die at the same age from other causes. Results: Table 2 shows the main results. The difference between males and females in the relative importance of primary and secondary prevention is a result of the greater male mortality from motor vehicle accidents. Life tables constructed with these deaths not included in the primary prevention group produce a male/ female pattern of YLG by primary prevention that is similar to that for secondary prevention. Tables constructed without perinatal mortality in the secondary preventable group of causes of death (to enable cornparison with Benavides et all) show that this group is similarly dominated by the perinatal deaths. Excluding perinatal mortality, YLG by secondary prevention would be 0 3 for non-Maori males and 0 42 for non-Maori females, compared with 0-38 and 0-34 for males and females respectively in Valencia. For primary preventable causes YLG in Valencia were 1 59 and 046, compared with 1-28 and 0 63 for (non-Maori) males and females respectively. Discussion: The accuracy of Maori …