Skip to main content

Table 4 Odds ratios for predictors of child Medicaid disenrollment from best-fitting GEE models

From: Maternal and child patterns of Medicaid retention: a prospective cohort study

Variable Any gap > 14-day gap > 60-day gap
Odds Ratio 95% confidence interval P-value Odds Ratio 95% confidence interval P-value Odds Ratio 95% confidence interval P-value
Maternal disenrollment
 Under Medicaid coverage Ref. Ref. Ref.
 Disenrolled from Medicaid (any gap) 12.60 (8.11, 19.58) < 0.001 11.78 (7.38, 18.82) < 0.001 16.75 (9.67, 29.02) < 0.001
Maternal knowledge that Medicaid and TANF eligibility criteria differ
 Yes Ref. Ref.
 No 2.01 (1.21, 3.35) 0.007 1.81 (1.05, 3.13) 0.03
Combined cash assistance status
 Neither had cash assistance Ref. Ref. Ref.
 Only mother had cash assistance 1.85 (0.99,3.41) 0.05 2.11 (1.11, 3.99) 0.02 2.15 (0.95, 4.86) 0.07
 Only child had cash assistance 0.38 (0.19, 0.76) 0.006 0.48 (0.23, 0.98) 0.04 0.50 (0.22, 1.14) 0.10
 Both had cash assistance 0.48 (0.29, 0.82) 0.007 0.59 (0.34, 1.04) 0.07 0.78 (0.41, 1.46) 0.43
Household income*
  < $1000/month Ref. Ref.
 $1000 or more /month 0.59 (0.36, 0.96) 0.03 0.57 (0.34, 0.94) 0.03
Family housing situation*
 Lives in own housing Ref.  
 Rents or lives with relatives/friends 2.08 (1.21, 3.56) 0.008
  1. Note: Maternal disenrollment was defined as having any gap in Medicaid coverage. All models were based on 604 dyads. The best fitting model was selected based on the QIC and the QICu. All final models were checked to ensure that adding another variable did not significantly change the QIC or the QICu
  2. *Results from 10 imputed datasets
  3. Entries in boldface have p-values less than 0.05