![[ Visit Client Website ]](images/banner.png)
BACKGROUND: Maternal race/ethnicity and maternal education are traditionally used in maternal and child health (MCH) to describe socioeconomic position (SEP). However, they are limited in their ability to describe MCH disparities. As women become more educated, jobs will fill with women of similar education, increasing the likelihood that more educated women will live at lower socioeconomic levels. Therefore, education may become less informative and occupation more informative in describing disparities. An assessment of SEP measures would inform current knowledge and facilitate monitoring of MCH disparities. This study examined the association between preterm birth and socio-demographic and occupation-based indicators.
METHODS: Geocoded Massachusetts Birth Certificate data from 2008-2010 were analyzed to examine the association between gestational age (preterm vs. term) and SEP measures. Study population included women ≥18 years who delivered their infant in Massachusetts (N=172,658). SEP measures included: maternal education, maternal education and paternal occupation, Goldethorpe classification of maternal occupation, and Nam-Powers-Boyd Occupation Status Scale (OSS) classification of maternal occupation. SEP measures and maternal race/ethnicity were fit to separate logistic regression models that adjusted for maternal age, marital status, insurance at delivery, the Kotelchuck index, and smoking during pregnancy. Changes in maternal race/ethnicity adjusted odds ratios (aORs) and R2 values calculated from log likelihood estimates were used to compare models. The maternal education only model was considered the null model.
RESULTS: The 2008-2010 prevalence of preterm birth in Massachusetts was 8.3%. Women with only a high school diploma or GED were more likely to have a preterm birth compared to women with a bachelor’s degree or higher (aOR: 1.19, 95% confidence interval [CI]: 1.13-1.25). Hispanic women were less likely than non-Hispanic White women to have a preterm birth (aOR: 0.93, 95%CI: 0.87-1.00). Adjusting for paternal occupation, Hispanic women were not significantly different non-Hispanic White women (aOR: 0.93, 95%CI: 0.83-1.05). The R2 value was 0.693. R2 values comparing Goldethorpe and Nam-Powers-Boyd OSS to maternal education only were low at 0.254 and 0.156, respectively.
CONCLUSIONS: In 2008-2010, inclusion of paternal occupation improved the maternal education’s ability to describe disparities in preterm birth but decreased precision of maternal race/ethnicity estimates. The Goldethorpe and Nam-Powers-Boyd OSS models did not meaningfully describe more of the disparities in preterm birth than maternal education. R2 estimates in logistic regression are sensitive to additional covariates so R2 value for paternal occupation may be falsely inflated. However, including occupation in future analyses may facilitate understanding of disparities in infant outcomes.