Preventive guidance from the USPSTF
An independent volunteer panel of 16 national experts in prevention and evidence-based medicine develop recommendations for the US Preventive Services Task Force (USPSTF).9 These guidelines are based on evidence and are updated as new evidence surfaces. Thirteen recommendations, some of which advise avoiding preventive procedures that could cause harm to patients, cover laboratory tests used in screening for conditions such as hyperlipidemia10 and prostate cancer.11 We review the ones pertinent to our patient later at the end of the Case.
While the target audience for USPSTF recommendations is clinicians who provide preventive care, the recommendations are widely followed by policymakers, managed care organizations, public and private payers, quality improvement organizations, research institutions, and patients.
Take a critical look at how you approach the diagnostic evaluation
To reduce unnecessary testing in the diagnostic evaluation of patients, first consider pretest probability, test sensitivity and specificity, narrowly out-of-range tests, habitually paired tests, and repetitive laboratory testing.
Pretest probability, and test sensitivity and specificity. Pretest probability is the estimated chance that the patient has the disease before the test result is known. In a patient with low pretest probability of a disease, the ability to conclusively arrive at the diagnosis with one positive result is limited. Similarly, for tests in patients with high pretest probability of disease, a negative test cannot be used to firmly rule out a diagnosis.12
Reliability also depends on test sensitivity (the proportion of true positive results) and specificity (the proportion of true negative results). A test with high sensitivity but low specificity will generate more false-positive results, with potential harm to patients who do not have a disease.
The pretest probability along with test sensitivity and specificity help a clinician to interpret a test result, and even decide whether to order the test at all. For example, the anti-nuclear antibody (ANA) test for systemic lupus erythematosus (SLE) has a sensitivity of 100% and a specificity of 86%13; it will always be positive in a patient with SLE. But when applied to individuals with low likelihood of SLE, false-positives are more common; the ANA is falsely positive in up to 14% of healthy individuals, depending on the population studied.13
Ordering a test may be unnecessary if the results will not change the treatment plan. For example, in a female patient with classic symptoms of an uncomplicated urinary tract infection, a urine culture and even a urinalysis may not change treatment.
Continue to: Narrowly out-of-range tests