Skip to main content

Table 2 Logistic regression results for main effects and interactions of marital status, neighborhood poverty and primary health insurers: Predictors of chemotherapy receipt

From: Multiplicative disadvantage of being an unmarried and inadequately insured woman living in poverty with colon cancer: historical cohort exploration in California

Predictors

Odds ratio

95% Confidence interval

Nonsignificant main effects and interactions

 Unmarried

0.72

0.36, 1.44

 Neighborhood poverty prevalence ≥ 30%

0.60

0.31, 1.15

 Primary health insurer was private

1.06

0.59, 1.89

 Neighborhood poverty by primary health insurer

Pinteraction = .511

Significant interaction effects

 Unmarried by neighborhood poverty

Pinteraction = .030

 Unmarried by primary health insurer

Pinteraction = .047

 Unmarried by health insurer by poverty

Pinteraction = .012

  1. Notes. All effects were adjusted for age, stage of disease at diagnosis, tumor grade and each other. The fit of the regression model with the 3-way interaction was significantly better than the model without it: likelihood ratio test, χ2 (1) = 6.33, p < .05.