Published online by Cambridge University Press: 04 August 2005
Objectives: The aim of this investigation was to study the effect of general practitioners' (GP's) characteristics on two decisions: whether to have the Helicobacter pylori rapid test (HPRT) in the office laboratory and whether to use this test or a similar hospital-based serological test in a typical clinical situation described in a vignette.
Methods: Discrete choice analysis with binary logit models were used to predict the probability that a general practice has the HPRT, as well as the GP's probability of using the HPRT or a similar test in this clinical situation.
Results: We found that the number of consultations per week has a positive effect on the probability of having the HPRT, indicating that the size of the practice affects the decision to have such a test in the repertoire. Furthermore, four variables significantly increased the probability of using one of the lab tests: more if located in urban practices, more by solo practitioners, more when the GP stated a high probability for H. pylori associated disease, and more when the GP had the rapid test available in the practice. In our analysis, the remuneration system is endogenous and does not have a significant effect on the two decisions.
Conclusions: Our study demonstrates that characteristics of the GP affect the availability and use of a specific laboratory analysis.