A study of the production technology of general dental practices in the U.S.

Date of Completion

January 2010


Economics, General|Health Sciences, Dentistry




This dissertation studies the production technology of the U.S. dental care industry using practice level data from a 2006 survey of general dental practices conducted by the American Dental Association (ADA). In particular, two main questions about the economic aspects of dental care industry are addressed: (i) How do we empirically construct the production frontier of the U.S. dental practices and measure their technical efficiency? (ii) How do we measure the cost efficiency of dental practices if we consider multiple outputs? ^ The empirical methodologies consist primarily of Data Envelopment Analysis (DEA) and Stochastic Frontier Analysis (SFA). DEA is a non-parametric method that measures firm-specific efficiency without assuming any explicit form of the relationship between inputs and outputs. SFA is a parametric method that can be employed to investigate a firm's technical or cost inefficiency taking explicit account of noise or random shocks. ^ We first employ DEA to evaluate the technical efficiency of each dental practice in the sample. Both radial and non-radial DEA models are applied. This is followed by a second-stage regression that reveals how the organizational characteristics and socioeconomic environmental factors influence the technical efficiency of the practice. ^ Next, the SFA method is applied to estimate the technical efficiency of dental practices in the same sample. We specify a model where the technical efficiency of a practice follows a truncated normal distribution with a changing mean. This permits us to assess how internal and external factors affect the technical efficiency. ^ Finally, a dual cost analysis is conducted and the cost efficiency of each practice is estimated by both DEA and SFA methods. An average cost curve is drawn to identify existence of economies of scale. A multiple output DEA cost analysis allows us to measure the marginal cost of different outputs. ^