The purpose of this study is to consider whether technological change in medicine has been a factor that has contributed to rising health expenditures.
The first Chapter is concerned with describing, and evaluating, the various approaches that economists have employed to address this issue. The existing literature is categorised into four approaches, viz. econometric modelling of hospitals, three residual approaches, the cost-of-illness approach and the analysis of supplier interests. All approaches can be subjected to specific criticisms, but a general criticism is that no distinction is made between product and process innovations, a categorisation that is long standing in the economic literature on technological change.
This general distinction between product and process innovations is applicable in the health sector, and it is argued that unambiguous process innovations, being essentially different ways of doing particular things, are substitutes and can be identified by the positive sign of the cross-price elasticity of demand. The conditions under which substitutable process innovations can lead to an increase or decrease in health expenditures are examined in Chapter 2.
The theoretical conception advocated in this study requires a case-by-case approach, and the particular innovations considered here, diagnostic tests of the gastrointestinal tract, are then described. Being diagnostic tests, attention is then directed to the economic meaning of medical diagnosis, per se. First, the phenomenon of medical diagnosis in the context of discrete or dichotomous choice is described. This discussion provides answers to the following two inter-related questions. First, what are the outputs of a diagnostic test in medicine? Second, how are the outputs to be measured? From this descriptive analysis of medical diagnosis it is shown that an economic analysis of the phenomenon can proceed by a combination of traditional (Hicksian) demand theory and Lancastian characteristics theory of consumer demand. These theoretical considerations lead to several particular empirical applications.
First, the concept of outputs of diagnostic tests is employed to determine, empirically, if barium meal radiology and fibre optic endoscopy are technically different. The procedures employed are in sharp contrast to the evaluative technique that has been employed by clinicians to evaluate these technologies. The statistical results indicate that for most diseases/conditions of the upper gastrointestinal tract, fibre optic endoscopy provides statistically significantly more accurate diagnostic information than does barium meal radiology. From an economic point of view, these results inter alia, indicate that there is an a priori reason to regard the technologies as imperfect substitutes.
The available data on fee-for-service medical practice in Australia are then described. Although this data set has been affected by various changes to health insurance arrangements through time, it is shown in Chapter 6 that statistical analysis of Australian time series data on fee-for-service medicine is possible.
These data are used in Chapter 7 to estimate demand equations, for three distinct groups of patients, relating to the diagnostic technologies that are the subject matter of this study. The purpose to this econometric work is to determine the sign of the crossprice elasticity of demand. The results provide no evidence that these medical technologies have been employed by Australian medical practitioners, operating on a fee-for-service basis, as substitutes. The econometric results indicate that these technologies are unrelated in demand, or are complements. Thus there is no evidence to indicate that the process innovations associated with fibre optics have decreased health expenditures of diagnosing diseases/conditions of the gastrointestinal tract.
There are several distinctive contributions in this study. First, the issue of medical innovation is placed in the general economic literature on innovations and/or technological change. Second, an economic theory of medical diagnosis is developed. This theory leads to a testable hypothesis concerning the substitutability of diagnostic tests. Third, it is shown, empirically, that available time series data on fee-for-service medicine in Australia can be statistically analysed despite changes to health insurance arrangements. Fourth, econometric results on diagnostic tests for the gastrointestinal tract provide no evidence of substitutability of fibre optic techniques for existing technologies for diagnosis of the gastrointestinal tract. Thus the results indicate that these innovations will have increased health expenditures associated with diagnosis of the gastrointestinal tract.