A recently published study analyzes the reasons for inconsistency in doctors’ ordering of lab tests to monitor patients’ drug levels. I’d like to use this space to tell you a bit more about this interesting article, which is also a troubling one—because the failure to monitor such medications must surely be putting some patients in danger.
The article, “Factors Associated with Ordering Laboratory Monitoring of High-Risk Medications” is written by six scholars affiliated with the Division of Clinical Informatics, Beth Israel Deaconess Medical Center, Brookline, MA; Meyers Primary Care Institute, Worcester, MA; and University of Massachusetts Medical School, Worcester, MA. (See also this issue’s “Observatory.”)
The authors begin by noting that “one major challenge to appropriate laboratory monitoring by health care providers is the lack of national guidelines and lack of expert agreement on appropriate monitoring standards.” They explain that even when guidelines have been promulgated, they have not brought about the kind of improvement in monitoring that one might expect: “The recommendations that do exist, whether from expert guidelines or product inserts, are not routinely followed” by physicians.
Why not? Why do physicians sometimes, but not consistently, order lab tests to monitor medications? What can we learn from studying the patterns of ordering or not ordering such tests? Most of the article seeks to answer those questions. Electronic medical data collected from a large multispecialty group practice that serves an (unidentified) New England-based health plan were used for the study, which focused specifically on tests ordered, as opposed to tests completed.
The 34 medications covered by the study included some names you would probably predict—Allopurinol, Cyclosporine, Digoxin, Lithium, Metformin, Methotrexate, potassium, Statin, thyroid replacement, etc.—and they were linked to appropriate tests: AST, ALT, BMP, CBC, creatinine, K+, CrCL, TSH, others. The researchers presented the percentage of times the tests were in fact ordered when clinically indicated; the numbers ranged from fewer than 10% ordering a CBC to monitor Primidone, to more than 90% ordering the appropriate tests to monitor Metformin, Allopurinol, and ACE/ARB, among others.
The wide range presumably reflects physician perceptions of the relative importance of monitoring particular drugs; in fact, drugs with black box warnings tended to get the most respect, if you will. But what may be more revealing are other factors that seemed to affect whether tests were ordered. A few of the findings:
Tests were ordered more often for male patients, for older patients, for patients who visited their doctors more frequently, and for patients who took multiple medications.
Tests were ordered more often if the medications required less frequent testing (e.g, annually), as opposed to more frequent testing; and if a single test, rather than multiple tests, were required.
Tests were ordered more often by specialists, by physicians who in general were more inclined to adhere to clinical guidelines, and by younger physicians (some interesting food for thought there).
In discussing why clinicians failed to order tests, the authors suggest that among the reasons are “lack of familiarity with clinical guidelines,” “lack of agreement/expectation of outcomes or benefits,” “inertia,” and “concerns about patient adherence and preferences.” “In some cases, providers may doubt the credibility or applicability of guidelines,” and “the appropriate test frequency is often unknown.” Clearly, better guidelines are needed, and with them, perhaps better directions in how to apply them. In any case, “the implications of the lack of good guidelines are concerning and need remedy.”
How can laboratorians be part of the remedy? How can they effectively communicate their comprehensive knowledge of these important tests to physicians whose own knowledge may not be as complete? I’d like to hear your thoughts.