Quantile and IV Regression Analysis of Dental Care Expenditures and Utilization
Dental care is necessary for all individuals as oral health is very much related to general physical health. Medical studies have shown links between periodontal disease and diabetes, heart disease and obesity among other diseases (American Dental Association Oral health topics 2007).
This paper analyses the determinants of dental expenditures and the determinants of the probability that an individual has dental insurance. The data are drawn from the Medical Expenditure Panel Survey (MEPS). The MEPS Household component and Dental Visits data are used for survey years 1996 through 2011. A Linear Probability Model and a Probit model each measure the probability of dental insurance purchase. We then examine dental care expenditures using quantile regression and an instrumental variables approach. Preliminary results indicate that dental coverage is more likely for those with higher levels of education, females, and those with higher incomes. Hispanics are least likely to have dental insurance. Predicted dental expenditures increase with age and with higher education levels. Expenditures are highest for whites; and blacks are predicted to spend the least on dental care. In particular, the presence of dental insurance has a great impact on dental expenditures. For whites, having dental insurance triples expenditures. Both blacks and Hispanics are predicted to have zero dental expenditure when they are without coverage. Clearly the policy implications are especially pertinent for black and Hispanic groups. Since dental health can have a large impact on overall physical health, it is important to extend public coverage to these vulnerable groups so that they may maintain their oral health as well as their physical health.
Key terms: dental insurance, dental expenditure, health care, health disparities, public policy
JEL: I1, I10, I14