Answers are grounded in this paper and cite their sources.
Some things never change; sometimes a pity. Appropriate (and inappropriate) laboratory utilization was a recurring theme in the Journal of the American Medical Association (JAMA), of which I was privileged to serve as the editor-in-chief from 1982 to 1999. Some things do change; also sometimes a pity. My successor JAMA editors had other compelling interests, so JAMA’s leadership in the field waned. Unfortunately, other leading journal voices seemed less enamored of the topic, so its coverage also lagged. The newly reintroduced concepts of “less is more” and “choosing wisely” in medicine have inevitably brought new attention to this age-old, always “appropriate” topic. Laboratory tests that matter are those that produce actionable results that bring a positive outcome benefit for the patient. Thus, to assess the appropriateness of laboratory test use, assessing the benefit (perhaps vs cost, risk, and potential harm) as an outcome would seem to be a central component, even a “no brainer.”1 In the most modern parlance of advanced laboratory testing, molecular oncology stands as a potential model for doing it right from the get-go. Until recently, there have been no truly beneficial therapeutic options for metastatic malignant melanoma. Then came the process of genomic study of individual solid tumors for cancer mutations. A BRAF mutation has been found in around 50% of metastatic melanomas of cutaneous origin.2 Successful clinical trials of targeted therapy demonstrated a substantial positive effect on tumor size, progression-free survival, and total survival.3 Not cure, but major benefit, at high monetary cost but low cost of therapy-induced collateral damage to quality of life. So, testing for BRAF mutations in metastatic melanoma with following therapy has become a “poster child” for outcomes-based appropriate laboratory utilization.4 Meanwhile, back in the nonmolecular world, past and present, Misbah et al5 have recently described three levels of outcomes of laboratory testing, emphasizing a positive role for laboratory physicians to improve clinical outcomes with a series of illustrative case studies. They also call laboratory clinical audits a powerful tool to assess the appropriateness of laboratory use. Fifteen years ago, Van Walraven and Naylor6 asked the question—“Do we know what inappropriate laboratory utilization is?”—of the world literature base from 1966 to 1997. They answered their own question with a “no” and suggested ways the field could progress. Elsewhere in this issue, Hauser and Shirts7 ask (and answer) almost the same question: “Do we now know what inappropriate laboratory utilization is?” They had help from “a half century of work by hundreds of individuals” and decided that the 2014 answer is still “no” and that even developing a consensus among interested parties on a way to study the question is “an elusive goal.” I am pleased that the authors tried to bring some kind of new order to this traditionally chaotic situation. Cheers! But so disappointing. Lessons from history. In the foreword of my book entitled Managing the Patient-Focused Laboratory, published in 1973, I wrote, “What laboratory tests should be done, and when? Who should do the tests and where? How much should the tests cost and why? Should quality and turnaround time be assured and how?”8 All are still proper questions to be faced by every stakeholder in the field—test performers, requestors, supervisors, owners, payers, providers, regulators, patients, consumers, and the media, among others. Also in 1973, retired US Army Colonel Robert Moser, fresh from being Chief of Medicine at Walter Reed General Hospital, was named the editor-in-chief of JAMA. In 90 years, JAMA had never had an editorial board. Bob wanted one, the American Medical Association said okay, and I was asked to be the pathologist on that editorial board. My boss at the University of Southern California (USC) said okay, so I joined it. Bob asked me to begin a column to guide doctors on how to use the laboratory appropriately. I said fine and went off to Burlington, Vermont, to ask Larry Weed, of the then recent problem-oriented medical record fame, to help me. Larry taught me about algorithms, decision trees, and decision tables. We set out together to teach the physicians of the world how to use that approach to guide appropriate laboratory test selection and branching logic decisions toward the best actions based on the test results. To test this approach in 2014, click on therapy.collabrx.com/lung to find an open-access live web app in the spirit of “toward optimal laboratory use.” I named that column “Toward Optimal Laboratory Use” (TOLU), commissioned authors to get started, and kicked it off in 1975 by writing the following in the initial series article: “The huge variety and volume of available laboratory tests confronts the clinician with a major dilemma. What tests should be ordered on what patients? When, how often, at what cost, grouped or individually, and in what sequence? What is the interpretation of the results and what steps should then be taken? The frequency with which a laboratory test may be ordered varies from none to constant monitoring of all variables. Should the physician order one of each available laboratory test on each patient once a lifetime, once a year, once an hour, or never? Many forces influence these decisions. Among them are availability of laboratory service, policies of technical instrument and reagent manufacturers, what the physician knows about the medical need for the test, and economic considerations, such as who makes the profit or takes the loss. Laboratory tests should not be performed without a plan for using the information gained. What will be done if the test result is normal? High? Low?”9 In 1969, responding to the death of a young man from unrecognized severe hypoglycemia, we invented the “Panic (Critical) Laboratory Value” result reporting system.10 This system soon became the standard of practice everywhere and defines the quintessential essence of appropriate laboratory use. By studying this event and the subsequent endeavor, we recognized that a “laboratory test” is best thought of as many things in addition to the analysis itself.11 Thus, we conceptualized the “brain-to-brain loop” in laboratory testing as a continuum from step one, when the physician’s brain decides on the need for a laboratory test, through ordering, collection, identification, transportation, preparation, analysis, reporting, interpretation, and action. Anything that interferes with any step in the process will be at least a waste and at most a potential tragedy. From this loop springs the understanding of the importance of both preanalytic (front-end) and postanalytic (back-end) laboratory errors. There is now a substantial literature documenting that the most frequent locations for laboratory errors are pre- and postanalytic.12 Both the ordering and the interpretation/action ends of the loop define whether a laboratory test is appropriate or inappropriate.13 We first published this loop concept in JAMA in 1981 in an editorial entitled “Acting on Significant Laboratory Results.”11 This brain-to-brain loop concept has been updated “40 years later” in this journal.14 Bob Moser’s editorship of JAMA lasted only 18 months. The TOLU column lasted more than 2 decades, and in 1983, the ASCP Press published 35 of the columns as a collection in a book entitled Using the Clinical Laboratory in Medical Decision-Making.15 Later in JAMA, when I was editor-in-chief, we published many other articles dealing with appropriate laboratory utilization. One of the last examples under my editorship lamented the dearth of results, even the dearth of a research agenda attempting to define the outcomes of specific laboratory testing.1 Why do physicians order laboratory tests? We asked that question about specific laboratory tests ordered by hundreds of physicians (largely residents) at the Los Angeles County–USC Medical Center in 1973. The results were as follows: diagnosis, 37%; monitoring, 33%; screening, 32%; prognosis, 7%; education, 2%; and medicolegal, 1%.16 Later, in Socratic teaching sessions on laboratory utilization in multiple states and several countries, I asked the same question of audiences of medical professionals. These audiences told me that there were also many other reasons, including the following: to confirm a clinical opinion, to both establish and complete a database, insecurity, curiosity, public relations, patient pressure, family pressure, peer pressure, pressure from the recent literature, concern for liability, documentation, question of accuracy of a prior result, unavailability of a prior result, personal education, to show to an attending physician, hospital policy, state legal requirement, personal or hospital profit, fraud, kickbacks, hunting or fishing expeditions, frustration at nothing better to do (“I don’t know what is wrong with this patient—better get some laboratory tests”), to buy time (“maybe by the time the laboratory results come back, I will have a better idea what is wrong with this patient or the patient may cure herself or himself”), simple availability, ease of doing, or to win a bet. Perhaps the most important is “what I learned in residency” or pure habit.1 How much of that is appropriate? Inappropriate? Who knows? So much for “appropriate” laboratory testing. Most clinical laboratory tests (and other diagnostic tests such as various imaging) do not need to be done. The results are normal, are negative, or show no change from prior tests. Then, why are they done? See the paragraph above for the psychologic, economic, and other explanations. Laboratory tests that matter are those that lead to actionable results and help the patient. In 1975, I wrote, “Laboratory performance can be judged by the indexes of its products—data. Laboratory utilization performance is far more difficult to judge. Standards have not been established, but are urgently needed. The final questions include these: Were the laboratory tests appropriate? Were they beneficial? Were they economical? Were all loops closed? Was every laboratory test maximally useful?”9 Thirty-nine years later, these questions, plus the big new one—what was the outcome that resulted from the test?—still beg to be answered.
No takes yet. Share an insight, caveat, or question.
George D. Lundberg (2014) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: