Medicaid Policies Increased Preventive Services for Racial/Ethnic Minority Children
Objectives: Racial/ethnic minority children experience more tooth decay than White children, but are less likely to receive dental visits. This study examined the impact of state Medicaid policies allowing delivery of preventive oral health services (POHS) in medical offices on receipt of POHS among young children by race/ethnicity. Methods: We used 2006-2014 Medicaid data for children aged 6 months to 6 years in 38 states. We constructed a categorical variable indicating the length of the state’s medical POHS policy (no policy, <1, 1, 2, 3, 4, 5+ years) and interacted it with race/ethnicity (White, Black, Hispanic, and Other/Multi-racial). Logistic regression models estimated the odds that a child received any POHS in a medical office or in medical or dental offices in a given year, stratified by the presence of a state requirement that medical providers obtain training prior to delivering POHS. Models also included state and year fixed effects, state-level clustered standard errors, and adjusted for child- and county-level characteristics. Results are presented as adjusted predicted probabilities. Results: Among 8,711,192 child/year observations, 2.8% received POHS in medical offices and 38.3% received POHS in medical or dental offices. Longer enacted medical POHS policies significantly increased the probability of a child receiving any medical POHS and any medical or dental POHS, with the greatest benefit observed for Hispanic and Other/Multi-racial children (see Table). A larger effect was observed in states with training requirements. For example, after 5 years of enactment in states with training requirements, Hispanic children had a 51.1% predicted probability of receiving any POHS compared to 40.3% for White children (P<0.05). Conclusions: Longer enacted state Medicaid policies allowing delivery of POHS in medical offices increased receipt of POHS among Hispanic and Other/Multi-racial children enrolled in Medicaid at higher rates than White children, which may help to reduce inequities.
Division:IADR/AADR/CADR General Session
Meeting:2020 IADR/AADR/CADR General Session (Washington, D.C., USA) Location:Washington, D.C., USA
Year: 2020 Final Presentation ID:0174 Abstract Category|Abstract Category(s):Behavioral, Epidemiologic and Health Services Research
Authors
Kranz, Ashley
( RAND
, Arlington
, Virginia
, United States
)
Stein, Bradley
( RAND
, Arlington
, Virginia
, United States
)
Dick, Andrew
( RAND
, Arlington
, Virginia
, United States
)
Oral Session
Keynote Address; Oral Health Decision Science & Policy
TABLES
Predicted probabilities of receipt of POHS by race/ethnicity and length of policy enactment
Predicted probability of any medical POHS
Predicted probability of any medical or dental POHS
Among states requiring training
Among states not requiring training
Among states requiring training
Among states not requiring training
Predicted probability
SE
Predicted probability
SE
Predicted probability
SE
Predicted probability
SE
White
No active policy
+
+
+
+
35.8%
1.0%
29.3%
0.9%
Policy enacted 4 to <5 years
3.6%
0.4%
2.5%
0.1%
36.7%
1.3%
33.3%
1.7%
Policy enacted 5+ years
8.7%
1.2%
2.4%
0.7%
40.3%
2.0%
28.4%
2.6%
Black or African American
No active policy
+
+
+
+
37.9%
1.1%
31.0%
1.0%
Policy enacted 4 to <5 years
3.3%
0.4%
2.1%*
0.1%
37.6%
1.3%
32.5%
2.0%
Policy enacted 5+ years
10.1%
1.0%
3.1%*
0.9%
40.0%
1.7%
30.4%
2.8%
Hispanic or Latinx
No active policy
+
+
+
+
43.5%
1.4%
37.1%
1.7%
Policy enacted 4 to <5 years
4.9%*
0.4%
2.6%
0.2%
47.5%*
1.2%
39.6%*
1.7%
Policy enacted 5+ years
10.3%
1.7%
3.7%*
1.0%
51.1%*
2.2%
40.6%*
2.7%
Other or multiracial
No active policy
+
+
+
+
40.3%
1.1%
33.7%
1.4%
Policy enacted 4 to <5 years
4.3%*
0.6%
4.2%*
0.5%
41.8%*
1.5%
36.8%*
1.6%
Policy enacted 5+ years
9.9%*
1.0%
3.3%*
0.9%
44.9%*
2.4%
34.1%*
2.8%
Unweighted N
4,536,630
1,715,651
6,995,541++
4,174,562++
Predicted probabilities were generated from four logistic regression models estimating the odds of POHS in states with and without training requirements. For ease of interpretation, results are presented for only the following years of policy enactment: no policy; policy enacted 4 to <5 years; and policy enacted 5 or more years.
*Indicates the predicted probability is significantly different (P<0.05) from the prediction for a White child in a state with the same number of years since enactment.
+ Models estimating any medical POHS include only state/years with policies enacted.
++ Both models include children in states without policies enacted, which is why the sum of children in these models in greater than our total sample.
POHS, preventive oral health services. SE, standard error.