Failure to recognize potential drug–laboratory test interactions can lead to misinterpretation of results, misdiagnosis, and unneeded tests or therapies. This review examines the evidence on this issue and suggests that decision support technologies can help avoid these errors.