Study on Physician Ordering Behavior on Hypercoaguable Screening Tests

We can now identify hereditary and acquired risk factors in patients with a venous thrombotic event. Hereditary factors include factor V Leiden; prothrombin G 20210A mutation; or deficiencies of antithrombin, protein C, or protein S. But considerable uncertainty exists in hypercoagulable testing. Clinical criteria (Bauer, 2002) guidelines are available and laboratory evaluation can confirm the diagnosis. But we know of anecdotal stories where physicians ignored the testing pathways. Even when correctly utilized, testing was inappropriately timed after the thrombotic event. There are many scenarios for thrombosis with no single test to identify these risk factors. We need to re-examine the use of these screening tests for inherited and acquired thrombosis syndromes.

Retrospectively, 200 patient charts were reviewed on the use of hypercoagulable screening panels, patient characteristics, and physician characteristics. The hypercoaguable screening tests contain prothrombin time, partial thromboplastin time, thrombin time, fibrinogen, antithrombin, plasminogen, activated protein C resistance, protein C, protein S, and lupus anticoagulant. Factor V Leiden and prothrombin G 20210A mutation results were reviewed if available. Data reviewed included age, gender, location of thrombosis (arterial vs. venous), malignancy, connective tissue disorder, diabetes, hypertension, nephrotic syndrome, liver disease, active infections, recent surgery, trauma, anticoagulation medications, obstetric history, family history, hypercoaguable screening test results, physician specialty, training level, and indications for the tests.

Patient age range was 18–91; 79 males and 12 females. Among 200 cases, 23 were positive from the hypercoaguable screening tests but only 4 were true positive for hereditary thrombophilia (Factor V Leiden, prothrombin G 20210A, antithrombin deficiency and protein C and S deficiency, respectively). False positive results (low levels of antithrombin, protein C, protein S) were due to coumadin. Ordering physicians were diverse (internal medicine, general medicine, family medicine, hematology/oncology, cardiology, pulmonary, rheumatology, nephrology, neurology, general surgery, vascular surgery, and pediatrics). Reason for ordering tests were varied: family history of thrombosis, recurrent deep vein thrombosis, myocardial infarction, pulmonary embolus, stroke, malignancy, myeloproliferative disorders, connective tissue disorders, inflammatory bowel disease, liver disease, diabetes, nephrotic syndrome, arteriovenous shunt operation, and fetal loss. Few cases fit the clinical criteria. Ordering tests were inconsistent with the indications for hereditary thrombophilia, with no relation to the clinical history, physician training level, or specialty.

Why these irregularities in behavior exist, we are not sure. From our literature review, there were few papers available on discrepancies in physician reasoning for utilizing laboratory tests. Wertman (1980) identified no single reason for test ordering behavior of physicians. Axt-Adams (1993) suggested that motivators, other than physician education, had a higher correlation of influencing physician ordering behavior when over-utilizing laboratory tests. These motivators included 1) fear of failure to diagnose, 2) fear of criticism, 3) inability to cope with diagnostic uncertainty, 4) eagerness to complete the screening evaluation while in the hospital, 5) desire to be complete in evaluation, 6) hope that additional follow-up testing provides the correct diagnosis, 7) provide reassurance for patient, 8) collective ordering, and 9) ignorance of costs and diagnostic significance of tests and their sensitivity, specificity, and predictability. In many cases, the diagnostic criteria for ordering hypercoagulable screenings were not followed. We did not have the opportunity to interview the ordering physicians about their rationale which may have provided more insight. More education is necessary on hereditary thrombophilia, limitations of coagulation tests, acquired conditions for thrombosis, and the costs of these specialized tests. More studies are necessary to understand physician behaviors in ordering these expensive tests.

No relevant conflicts of interest to declare.