The impact of income definitions on mortality inequalities

• Main Authors: Jiaxin Shi (Author), Lasse Tarkiainen (Author), Pekka Martikainen (Author), Alyson van Raalte (Author)
• Format: Book
• Published: Elsevier, 2021-09-01T00:00:00Z.
• Subjects: article
• Online Access: Connect to this object online.
• Notes: 2352-8273; 10.1016/j.ssmph.2021.100915
• Keywords: Health and mortality inequality; Measurement; Data quality; Life expectancy; Administrative data; Finland; Public aspects of medicine; RA1-1270; Social sciences (General); H1-99
• Source: SSM: Population Health, Vol 15, Iss , Pp 100915- (2021)

Income is a strong predictor of adult mortality. Measuring income is not as simple as it may sound. It can be conceptualized at the individual or the household level, with the former better reflecting an individual's earning ability, and the latter better capturing living standards. Furthermore, respondents are often grouped into income categories based on their positions in the income distribution, and this operationalization can be done on the basis of age-specific or total population income distributions. In this study, we look at how four combinations of different conceptualizations (individual vs. household) and operationalizations (age-specific vs. total population) of income can affect mortality inequality estimates. Using Finnish registry data, we constructed period life tables for ages 25+ from 1996 to 2017 by gender and for four income definitions. The results indicated that the slope index of inequality for life expectancy varied by 1.1-5.7 years between income definitions, with larger differences observed for women than for men. The overall age patterns of relative index of inequality for mortality rates yielded by the four definitions were similar, but the levels differed. The period trends across income definitions were consistent for men, but not for women. We conclude that researchers should pay particular attention to the choice of the income definitions when analyzing the association between income and mortality, and when comparing the magnitude of inequality across studies and over time.