Background: Ecological studies have identified associations between various contextual factors and infant mortality rates but studies examining these relationships at the individual level are rare. We aim to elucidate the relationship between county-level social economic, clinical care, and health behavioral factors and infant mortality risk.

Methods: Data are from US Vital Statistics 2010 Cohort Linked Birth and Infant Death (LBID) records and the County Health Rankings dataset. We fit multilevel logistic regression models to test whether US county characteristics were associated with the likelihood of infant mortality, while adjusting for individual-level, and county-level confounders. Social and economic factors included neighborhood safety (Violent crime rate), education (Proportion of adults with a college degree), and social capital (Social Capital Index). Clinical care factors included Percent Uninsured and Patient-to-Physician Ratio. The only health behavioral factor we used was Sexually Transmitted Infections (STI) rate. All county-level measures were standardized using Z-transformation.

Results: The infant mortality rate in 2010 was 5.9 deaths per 1,000 live births. Adjusted analysis reveals that an increase in standard deviation of Primary Care Physician per 100,000 was significantly associated with a decreased odds in infant mortality (OR=0.90, 95% CI=0.87,0.93). An increase in standard deviation of violent crime rate (OR=1.07, 95% CI=1.03,1.10) and sexually transmitted infection rate (OR=1.04, 95% CI=1.01,1.07) were each significantly related to an increased odds of infant death.

Conclusions: Increasing access to health services and creating safer environments (i.e. considering the root causes of the violent crime rate) may help to decrease the odds for infant mortality.

Background: Research has indicated that income inequality (the gap between rich and poor) within a residential area, is associated with infant mortality. One proposed way to decrease income inequality is to increase minimum wage. Ecological studies have identified a relationship between increasing state-level minimum wage and decreased infant mortality rates. We aim to elucidate the relationship between state-level income inequality and individual-level infant mortality risk.

Methods: Data are from US Vital Statistics 2010 Cohort Linked Birth and Infant Death (LBID) records and the 2010 US Bureau of Labor Statistics. We fit multilevel logistic regression models to test whether US state minimum wage was associated with the likelihood of infant mortality (death before the first birthday), while adjusting for individual-level, and state-level confounders. Minimum wage was standardized using the z-transformation and was dichotomized high vs. low using the 75th percentile as a threshold. Analyses were also stratified by mother's race (black vs. white).

Results: The average state-level minimum wage was $7.46, and ranged from $6.15 to $8.15. No significant relationship was observed when minimum wage z-score was tested. High minimum wage (OR=0.93, 95%CI=0.83,1.03) was associated with decreased odds of infant mortality, but was not significant. High minimum wage was significantly associated with reduced infant mortality among black infants (OR=0.80, 95%CI=0.68,0.94), but not among white infants (OR=1.04, 95%CI=0.92,1.17).

Conclusions: Increasing minimum wage might be beneficial to infant health, especially among black infants and can help decrease the racial disparity in infant mortality. Future research should examine whether this association is causal.

Background: Poverty and birth order are well established predictors of child maltreatment (CM); however, if, and to what extent, these factors, may contribute to maltreatment among immigrant families remain inconclusive. Objective: The aim of our study was to assess whether socioeconomic and child birth order gradients in CM vary by immigrant status. Methods: We linked birth, hospitalization, emergency department visits, small-area income, and death records with an official Canadian Immigration Database to create a retrospective cohort of all 1,240,874 children born from 2002 to 2012 in Ontario, Canada followed from birth to age 5. We used multivariable regression models to estimate rate ratios of maltreatment injuries across neighborhood income quintiles and birth order separately for children of immigrant versus non-immigrant mothers. Results: CM increased with decreasing neighborhood income quintiles, particularly among non-immigrants. The adjusted rate ratio (ARR) of CM in the highest neighborhood income quintile (Q1) versus the lowest quintile (Q5) was 1.33 (95% CI: 1.15, 1.54) for immigrants and 1.57 (95% CI: 1.49, 1.66) for non-immigrants. Compared to a first child, the ARR of CM for a fourth child was 0.57 (95% CI: 0.44, 0.74) among immigrants, but was 1.75 (95% CI: 1.63, 1.89) among non-immigrants. Conclusion: Immigrants exhibited lower CM rates than non-immigrants across all neighborhood income quintiles and birth orders but the differences deepened with decreasing neighborhood income and increasing birth order. The contrasting birth order gradients between immigrants and non-immigrants require further investigation.

Background: The association between income and health is well-established. Unfortunately, many data sources either do not have income information or have income variables with many missing responses. One method for dealing with this challenge is to use proxy measures for income.

Objectives: This study sought to assess the use of occupation as a proxy for income in health studies.

Methods: Data from the 2015 wave of the 1997 National Longitudinal Survey of Youth was used for this analysis. Each participant was put into one of 5 income categories based on their self-reported annual income. The mean and median incomes for twenty five occupation groups were used to assign occupation-based income for each participants, which were used to put participants into the same five income categories described above. We assessed the prevalence of income category misclassification by comparing the participants’ self-reported income category to the occupation-based income proxy measures and also how using the proxy measures impacted the observed association between income categories and self-reported health.

Results: When occupation-based median and mean income were used to classify participants, 62.6% and 75.2% of participants were misclassified, respectively. Both the median and mean income based proxy measures tended to underestimate the relationship between income category and self-rated health. The degree of underestimation was greater when using the median based proxy measurement.

Conclusion: Further research should examine the utility of occupation as a proxy for income in other populations. Occupation as a proxy should also be compared to other commonly used proxy measures.