BENEFITS, COSTS AND RISKS

When used appropriately, diagnostic tests can be of great assistance to the clinician. Tests can be used for screening, ie, to identify risk factors for disease and to detect occult disease in asymptomatic persons. Identification of risk factors may allow early intervention to prevent disease occurrence, and early detection of occult disease may reduce disease morbidity and mortality through early treatment. Blood pressure measurement is recommended for preventive care of asymptomatic low risk adults. Screening for breast, cervix, colon, and lung cancer is also recommended, whereas screening for prostate cancer remains controversial. Screening without demonstrated benefits should be avoided. Optimal screening tests should meet the criteria listed in Table 1–1. Some screening test results (eg, rapid HIV Ab tests) require confirmatory testing.

Criteria for use of screening procedures.
Characteristics of Population
  1. Sufficiently high prevalence of disease.
  2. Likely to be compliant with subsequent tests and treatments.
Characteristics of Disease
  1. Significant morbidity and mortality.
  2. Effective and acceptable treatment available.
  3. Presymptomatic period detectable.
  4. Improved outcome from early treatment.
Characteristics of Test
  1. Good sensitivity and specificity.
  2. Low cost and risk.
  3. Confirmatory test available and practical.

Tests can be used for diagnosis, ie, to help establish or exclude the presence of disease in symptomatic persons. Some tests assist in early diagnosis after onset of symptoms and signs; others assist in developing a differential diagnosis; others help determine the stage or activity of disease.

Tests can also be used in patient management. They can help (1) evaluate the severity of disease, (2) estimate prognosis, (3) monitor the course of disease (progression, stability, or resolution), (4) detect disease recurrence, and (5) select drugs and adjust therapy.

One evolving field of medicine is personalized medicine, which involves tailoring treatment to the individual patient. A companion diagnostic test may be used to identify which patients could benefit from a drug and which patients would not benefit or even be harmed. As an example, only patients with breast cancer that shows overexpression of HER2 protein or extra copies of the HER2 gene or both could benefit from trastuzumab treatment.

When ordering diagnostic tests, clinicians should weigh the potential benefits against the potential costs and adverse effects. Some tests carry a risk of morbidity or mortality—eg, cerebral angiogram leads to stroke in 0.5% of cases. The potential discomfort associated with tests such as colonoscopy may deter some patients from completing a diagnostic workup. The result of a diagnostic test may mandate additional testing or frequent follow-up, and the patient may incur significant cost, risk, and discomfort during follow-up procedures.

Furthermore, a false-positive test may lead to incorrect diagnosis or further unnecessary testing. Classifying a healthy patient as diseased based on a falsely positive diagnostic test can cause psychological distress and may lead to risks from unnecessary or inappropriate therapy. A screening test may identify disease that would not otherwise have been recognized and that would not have affected the patient. For example, early-stage prostate cancer detected by prostate-specific antigen (PSA) screening in a 76-year-old man with known heart failure will probably not become symptomatic during his lifetime, and aggressive treatment may result in net harm.

The costs of diagnostic testing must also be understood and considered. Total costs may be high, patient out-of-pocket costs may be prohibitive, or cost-effectiveness may be unfavorable. Even relatively inexpensive tests may have poor cost-effectiveness if they produce very small health benefits. Factors adversely affecting cost-effectiveness include ordering a panel of tests when one test would suffice, ordering a test more frequently than necessary, ordering an inappropriate test, and ordering tests for medical record documentation only. The value-based, operative question for test ordering is, “Will the test result help establish a diagnosis, affect a treatment decision, or help predict a prognosis?” If the answer is “no,” then the test is not justified. Unnecessary tests generate unnecessary labor, reagent and equipment costs, and lead to high health care expenditures. Molecular and genetic testing is readily available, and genome-scale and high-throughput DNA sequencing technology is increasingly being applied in the clinical diagnostic realm. However, their cost-effectiveness and health outcome benefits need to be carefully examined. Diagnostic genetic testing based on symptoms (eg, testing for fragile X in a boy with mental retardation) differs from predictive genetic testing (eg, evaluating a healthy person with a family history of Huntington disease) and from predisposition genetic testing, which may indicate relative susceptibility to certain conditions or response to certain drug treatment (eg, BRCA1/BRCA2 or HER2 testing for breast cancer). The outcome benefits of many new pharmacogenetic tests have not yet been established by prospective clinical studies; eg, there is insufficient evidence that genotypic testing for warfarin dosing leads to outcomes that are superior to those using conventional dosing algorithms, in terms of reduction of out-of-range INRs. Other testing (eg, testing for inherited causes of thrombophilia, such as factor V Leiden, prothrombin gene mutation, etc) has only limited value for treating patients, since knowing whether a patient has inherited thrombophilia generally does not change the intensity or duration of anticoagulation treatment. Carrier testing (eg, for cystic fibrosis) and prenatal fetal testing (eg, for Down syndrome) often requires counseling of patients so that there is adequate understanding of the clinical, social, ethical, and sometimes legal impact of the results.

Clinicians order and interpret large numbers of laboratory tests every day, and the complexity of these tests continues to increase. The large and growing test menu and the inconsistencies in nomenclature for many tests have introduced significant challenges for clinicians, eg, selecting the correct laboratory test and correctly interpreting the test results. Errors in test selection and test results interpretation are common and could impact patient safety but are often difficult to detect. Using evidence-based testing algorithms that provide guidance for test selection in specific disorders and expert-driven test interpretation (eg, reports and interpretative comments generated by clinical pathologists) can help decrease such errors. Consultation and collaboration with laboratory professionals (ie, pathologists, medical technologists) can also help improve the timeliness of diagnostic testing and optimize laboratory test utilization.

There's more to see -- the rest of this topic is available only to subscribers.