Appropriateness and necessity criteria developed based on the RAND expert panel method were employed to determine underuse and overuse of diagnostic procedures for patients presenting with chest pain not due to myocardial infarction in one of five urban hospital emergency departments in Los Angeles. The sample included a large number of HMO patients. For the purpose of this study, a diagnostic test was deemed inappropriate if "the benefits did not outweigh the inherent risks," and necessity was defined "as the obligation of a physician to recommend diagnostic testing for coronary artery disease because it has a high probability of significance." "Underuse" was therefore defined as the "failure to provide diagnostic testing for coronary artery disease for a patient with a necessary indication." "Overuse" was the "provision of diagnostic testing for coronary artery disease when the indication was inappropriate." Participants were mailed a questionnaire four weeks after their presentation at the hospital. The survey included questions on patient demographics and whether patients received diagnostic procedures for coronary artery disease, as well as the type of procedure received.

Among the 181 patients who met the necessity criteria for diagnostic cardiac testing, 40 patients (22%) did not receive the necessary testing. Lower patient education was found to be the only significant predictive variable for this underuse. However, race was also associated with underuse of diagnostic testing. The rates were higher among African- Americans and Latinos in comparison to whites but lower among Asians and Pacific-Islanders; these differences were not statistically significant.

The results of this study have important implications. Most studies to date had examined racial differences in therapeutic cardiac care. Assessing the disparities at the diagnostic stage would reveal a larger problem because it is at this stage that a “therapeutic cascade” can be triggered. Underuse of diagnostic procedures could lead to inaccurate assessment of severity of disease, which would then further contribute to the underuse of appropriate and necessary therapeutic procedures and surgery.

Furthermore, the findings that “lower educational level may be a proxy for the difficulty that patients have in paying for health care, including diagnostic testing” and that “it may also be a marker for other factors associated with testing, such as income, understanding of complicated medical conditions, and access to alternative provider opinions, legal interventions, or appeal processes” suggest “that universal health insurance may not eliminate disparities in use of health-care services related to socio-economic status.” In sum, the “findings underscore the failure of the US health-care system to ensure the provision of medical care for all patients who may need it.”

A potential limitation of this study is the high rate of loss of subjects in follow-up. Of those subjects able to be contacted, only 63% consented to participate in the study. Hence the final study sample of 356 may be unrepresentative.