Temporal Changes in Socioeconomic Influences on Health: Maternal Education and Preterm Birth

Objectives. To consider how the relationships between social determinants and health indicators change over time, we assessed the time-varying influence of maternal education on risk of preterm birth (PTB) between 1989 and 2006.Methods. We used bivariate and multivariable Poisson regression models with robust variation estimates to examine (1) the association between maternal education and PTB risk by year; (2) the relationship between low maternal education and PTB, late PTB, and very PTB risk by year relative to 1989; and (3) the relationship between high maternal education and PTB, late PTB, and very PTB risk by year relative to 1989.Results. After adjustment, PTB risk increased among the most educated and did not change among the least educated women over time. Risk of PTB among the least educated relative to the most educated women decreased with time. Late PTB risk increased among both the most and the least educated groups but more among the most educated.Conclusions. Maternal education may be becoming less protective against PTB. The influence of the social determinants of health is dynamic, warranting revisions of our understanding of their roles over time.The association between lower socioeconomic position and poor health is among the most robust epidemiological relationships documented. The association between socioeconomic metrics such as education,1,2 household income,1,3 occupational status,1,4 and neighborhood deprivation5,6 and adverse health has been described in the context of mortality,7 cancer,8 cardiovascular disease,9,10 diabetes,11 obesity,12 mental health,2,5,13 and several other adverse health indicators.In their work on fundamental cause theory, Link and Phelan14 suggested that higher social status is always associated with better health simply because social status—through access to more knowledge, money, power, social connectedness, and prestige—affords access to resources that can optimize health even if these resources change over time. In that way, it is entirely plausible that the salutary resources associated with a particular social factor may change with time. For example, it is likely that the extent to which high education affords relatively better access to health information changes over time as health information infiltrates all levels of social status. This would result in differing temporal relationships between social factors and health indicators and would suggest that we need to consider and recognize that the links between social factors and health indicators likely change over time. In this way a dynamic, rather than a static, approach to the study of social determinants is indicated. However, with very few exceptions, there is scant literature that has considered how the relationship between social determinants and health indicators changes over time in this way.Of interest here is the association between maternal education and the risk of preterm birth (PTB), which has been well documented.15–18 PTB is defined as birth before 37 completed weeks of gestation or birth before 259 days since the first day of the mother’s last menstrual period.19 Although only 6% to 10% of all births are preterm in wealthy countries, preterm neonatal deaths account for more than two thirds of all neonatal deaths in this context.19 PTB is an important determinant of serious neonatal morbidity, moderate to severe childhood disability, and perinatal and neonatal mortality in wealthy countries.18,19Given the potential for changes in the relationships between social determinants and health indicators, we assessed how the association between maternal education and PTB changed over time to understand how secular changes in social factors may influence their roles as health determinants. We assessed the time-varying influence of maternal education on the risk of PTB from 1989 through 2006.