The decision whether or not to test women for genital infection with Chlamydia trachomatis is an important clinical problem in primary care medicine. It is a common disease which is asymptomatic in a majority of women, yet may result in serious morbidity if not diagnosed and treated. Specific criteria for screening women in low prevalence populations have not been well-defined. The proposed research has the broad objective of developing and clinically evaluating a clinical decision aid (CDA) to assist physicians in deciding whether to test patients for chlamydia. The CDA will consist of two components. The first is a chlamydia predictive instrument (CPI) based on multivariate logistic regression analysis of risk factors which will calculate a probability of chlamydia infection for an individual patient. The second component will be a clinical decision model (CDM) which will use the probability generated by the CPI to prescribe a recommended course of action. The study will proceed in four phases. Phase I will simultaneously measure risk factors and test for chlamydia in a large number (2500) of women in a primary care gynecology practice. In Phase II, the CPI will be developed from the data collected during Phase I using multivariate logistic regression analysis. In Phase III, prospective validation of the CPI will be undertaken in the same clinical setting in which the original data will have been collected. In Phase IV, the data from the validation phase will be analyzed and two previously published predictive models for chlamydia will be tested using our data set. Also during both Phases I and III the unaided diagnostic behavior of clinicians will be studied. The data collected during Phases I and III, will be used with decision analysis to assess the potential improvement by the CDA of the selectivity of testing. The CDA will combine multivariate predictive model with a decision analytic model for real-time clinical use. It will have many potential benefits for future patients. The most important of these is the ability to select for testing, patients most likely to have chlamydia. This would reduce unnecessary treatment and lower overall costs of testing. It may thus allow testing to be applied in settings for which it would otherwise be deemed too costly. The proposed study will advance our knowledge of risk factors for chlamydia and the prevalence of chlamydia in a primary care population. It will also evaluate the performance of a rapid antigen test (Chlamydiazyme) in a low prevalence population. The study will also evaluate the extent to which application of the CDA can improve the test ordering decisions of clinicians.
| Sonnenberg, F A; Roberts, M S; Tsevat, J et al. (1994) Toward a peer review process for medical decision analysis models. Med Care 32:JS52-64 |
| Jubran, A; Gross, N; Ramsdell, J et al. (1993) Comparative cost-effectiveness analysis of theophylline and ipratropium bromide in chronic obstructive pulmonary disease. A three-center study. Chest 103:678-84 |
| Sonnenberg, F A; Beck, J R (1993) Markov models in medical decision making: a practical guide. Med Decis Making 13:322-38 |
