The impact of diagnostic testing on therapeutic interventions

Objective.-To gain insight into the usefulness of managing the earlier phase of decision making, we examined the relationship between selected diagnostic tests and the therapeutic interventions they might trigger. Design.-Medicare's National Claims History and Part B files were used to obtain summary information on 100% of all physician claims submitted from 1987 through 1993. We regressed the annual rates of selected therapeutic interventions on selected diagnostic tests, which had been previously paired based on the clinical expectation that the test might drive subsequent intervention. Population and Setting.-These data represent the physician services received by approximately 30 million elderly Americans in each of 7 years. Main Outcome Measure.-Coefficient of determination (R(2)). Results.-The annual rate for the diagnostic tests increased rapidly during the period (range, 1.4- to 3.0-fold increase) and accounted for the bulk of the variance in therapeutic intervention rates (R(2)>0.80, P<.01) for five diagnostic-therapeutic pairs: cardiac catheterization with cardiac revascularization procedures, imaging of the spine with back surgery, swallowing studies with percutaneous gastrostomy, mammography with breast biopsy and excision, and prostate biopsy with prostatectomy. Although the rate of abdominal ultrasound increased during the period, it was not related with either cholecystectomy or abdominal aortic aneurysm repair. Conclusion.-There has been a substantial increase in diagnostic testing in the United States that closely tracked the increase of clinically relevant downstream procedures. Managing diagnostic testing could be an important strategy for controlling the increase of therapeutic interventions.