Controversies in Laboratory Medicine: A Series From the Institute for Quality in Laboratory Medicine

John R. Butterly, MD; Richard E. Horowitz, MD

In This Article

Top 5 Issues That Irritate Physicians About the Laboratory vs Top 5 Issues That Irritate the Laboratory About Physicians

The appropriate ordering and interpreting of laboratory tests is an essential element of a physician's clinical skills. Along with history taking, physical examination, and the thoughtful use of imaging techniques, the clinical laboratory is a major tool in the clinician's armamentarium.

The introduction of sophisticated quality improvement techniques into the clinical arena has evolved substantially in the past decade. It makes sense to integrate the changes that we make in our daily practice of medicine with quality improvement changes in the clinical laboratory in order to maximize the functionality of both areas for the safety and quality of care for our patients.

For that reason, I was delighted to be asked to contribute to this controversial topic as an opportunity to improve communication between the disciplines of the practice of clinical medicine and clinical laboratory medicine.

The more that I thought about the topic, "the top 5 issues that most irritate physicians about the clinical laboratory," the more I was reminded of an apocryphal story about Albert Einstein as a child. Apparently, Albert had not spoken a word up to age 3, when one morning his mother served him a bowl of oatmeal for breakfast. Albert took 1 spoonful, grimaced, and said to his mother (in German, of course), "This stuff is horrible." His mother was shocked. "Albert, you haven't said a word in all of your 3 years, and now you speak in a complete sentence! Why is this?" And Albert replied, "Well, up until now, everything has been fine." That is kind of how I feel about the clinical laboratory, and, frankly, how the many colleagues who I consulted prior to writing this essay also feel: Everything has been fine. One of the reasons that the clinical laboratory has been so important in the practice of medicine is the very fact that the "service" is, for the most part, timely and user-friendly, and, most importantly, the results are relevant and reliable.

Having said that, improvement is always possible, so I will discuss below some issues that my clinician colleagues and I believe have some room for improvement.

Failure to Provide Useful Information to Help Interpret a Test Result. Although clinical laboratories provide normal ranges for test results, they do not provide likelihood ratios of the test results. These would be very valuable for the ordering physician in enabling her/him to know how much the odds of disease change with those specific results. Other valuable information that the laboratory could provide would be feedback as to thoughtless test-ordering practices, such as ordering multiple tests when 1 test would suffice.

Standardization of Reference Ranges and Units of Measurement. Because tests may be done with different techniques and/or different reagents, depending on the institution, reference ranges for "normals" can differ substantially. Although this may not be a problem for physicians who only practice in 1 setting, many physicians practice in multiple settings with multiple clinical laboratories. This, of course, is a setup for a poor quality and safety environment. An excellent example of how this was changed in a way that has improved the quality and safety of patient care is in the use of the international normalized ratio (INR), created by the World Health Organization, in reporting the prothrombin time (PT) in anticoagulated patients. This has become the widely accepted standard of care because clinicians and pathologists recognized the marked improvement over the widely variable PT.

Similarly, when laboratories report results with different metrics, it can be confusing to clinicians and potentially dangerous for patients. For example, some laboratories report in millimoles per liter, whereas others report in milligrams per deciliter. Clinicians tend to get comfortable with particular measurements and the normal ranges associated with them, and it can become difficult to translate meaningfully from one unit to another. For example, as a cardiologist, a cholesterol level of 270 mg/dL catches my eye, whereas a level of 7.0 mmol/L doesn't. Yet, I quickly got used to using the INR over the PT, so if clinical laboratories agreed to standardize measurements for quality or safety reasons, I could get over it!

Laboratory Policies That Interfere With Patient Care. This is a potentially sensitive topic, and one that has right and wrong on both sides of the fence. One colleague tells me that his single most irritating issue is the page that he gets about "critical results" when the information is on a patient with whom he has had no contact, ie, his name was mistakenly put on a requisition. When this occurs, the laboratory personnel will frequently tell the clinician that it is now his/her responsibility to follow up on the results, ie, the laboratory was only responsible for informing the identified clinician, even though it was the wrong person. Of course, this is incorrect. Quality and safety are everybody's responsibility, and both the clinician and the laboratory should work together to maximize patient safety. Although I can think of a number of ways that this may be accomplished, the details are beyond the scope of this article. I would just leave it to say that a policy developed jointly by laboratory medicine and the clinicians, tailored to the specific structure of each particular organization, is most likely to be functional. Other examples under this heading include ordered tests being canceled because they appear to be duplicate or not clinically indicated, but which actually are appropriate in a particular case; disposal of specimens that appear to be incorrectly labeled, which may have been obtained at some pain and risk for the patient -- direct communication should take place before the specimen is disposed; timely notification when a specimen is "QNS" (quantity not sufficient); and a general sense of inflexibility of the rules or policies without adequate explanation to the clinician. It will be obvious to the most casual reader that these are problems that are all symptomatic of inadequate or poor communication (on both sides).

Performance of the Wrong Test. This can often be the result of an unclear order, but if the lab guesses wrong, it can be very frustrating for the clinician or patient. Should the lab guess or contact the ordering physician to clarify the order? Again, clear communication on both ends would minimize this. Computer physician (provider) order entry (CPOE), although it may have its own inherent problems, should eliminate this particular problem.

Miscellaneous. Other minor irritating issues may include changes in reference range, assay methodology, or specimen requirements without notification or explanation; turnaround times that are perceived as too slow; and failure to recognize that a test was ordered stat.

As should be obvious from the last grouping, the list is short, and it becomes increasingly difficult to think of significant things about the laboratory that are irritating to the physician. As stated above, the clinical laboratory occupies an important place in the practice of medicine, and is viewed by physicians as a valued and reliable partner in the delivery of safe, quality care to our patients. The few problems that we do encounter can be obviated by more frequent and open communication -- as is frequently established by inclusive quality improvement techniques in the clinical setting. Also, as pointed out by multiple quality-directed organizations, CPOE would be a major step in systems improvement that would obviate many of these problems -- not to mention the quality and safety issues surrounding the pharmacy, medication errors, and associated patient safety.

I am looking forward to Dr. Horowitz's list of the 5 issues that most irritate the laboratory about physicians, although I personally cannot think of much about us that would irritate anybody.

Before I list the top 5 things that irritate the laboratory about clinicians, I must emphasize that most of the time the clinician-laboratory relationship is just fine: Tests are ordered appropriately; specimens are handled correctly; test results are received and interpreted properly; and the demeanor is gracious and amiable.

It is useful, however, to address those situations when the relationship is not ideal so that we can all work together with a greater sense of collegiality, with resultant reduction in our mutual frustrations and improvement in the care of our patients.

Rather than merely presenting my own thoughts, I asked several colleagues for their ideas. They were from all parts of the country and practiced in varying settings, including rural community hospitals, large tertiary medical centers, university teaching hospitals, Veterans Affairs medical centers, and commercial laboratories. Here is what I learned.

Critical Values Reporting (Hey Doc, Where Are You?). The most exasperating problem faced by the laboratory is being unable to find the ordering clinician when a highly abnormal test result, in the critical or panic value range, has been detected. This is particularly vexing when specimens arrive from a doctor's office or from an outpatient facility and the doctor can't be found. The laboratory technologist has an obligation to notify the ordering physician when there is a potentially life-threatening test result, but that obligation is often difficult to fulfill. The page goes unanswered; the office receptionist fails to put the call through; the doctor is "off call"; and the physician "covering" the practice claims ignorance and rejects responsibility. In hospitalized patients, the same problem exists. Getting an answer to a page in a timely fashion is rare, and if the test result is called to the ward after the doctor has completed rounds, the nurses or ward clerks, particularly on the short-staffed evening and graveyard shifts, hesitate to take responsibility. It's not much better in hospitals with interns and residents; they have the same reluctance to accept a critical value result, especially when covering a service for a colleague.

Inappropriate Ordering of STATs (Not Another STAT!). When the lab receives a request for a STAT, it is treated as an emergency, and special effort is expended in specimen collection, testing, and result reporting. It may take lab personnel away from other duties, thereby delaying routine work, but laboratories must respond to STATs immediately. Unfortunately, STATs are often not ordered appropriately. Some surgeons request a STAT frozen section, then leave the operating room -- obviously the STAT was not needed for determining the diagnosis or adequacy of the surgery. Similarly, STATs are often ordered for the clinicians' convenience but without impact on diagnosis or patient management. The most frustrating STATs are those that are actually repeats of a previously performed test whose result had not been looked at or cannot be found.

Missing History (What's the Story?). Pathologists should feel complimented because clinicians so often believe that pathologists can make diagnoses without a patient's history. It is critical for pathologists to have information about the patient in order to provide the best diagnosis; however, it is common for surgical pathology specimens to arrive in the laboratory without any indication of what the surgeon wishes to know, ie, adequacy of margins or hormone receptor analysis or DNA analysis? The diagnosis of skin biopsies is often dependent on the history and clinical appearance of the lesion; similarly, the x-ray provides essential information in the diagnosis of a brain or bone tumor. Additionally, requests for consultation in clinical pathology, eg, hematology or immunology, hardly ever come with any relevant history, eg, drug administration, when that may be crucial to test selection and test interpretation.

Inappropriate Test Ordering (What Do You Really Want, Doc?). Laboratories have become so complex and sophisticated in the past decade that what most doctors learned in medical school is no longer valid, and thus they are not able to use the lab in the most efficient or effective way. They may not know what tests are available and order outdated tests, eg, protein-bound iodine or phenolsulfonphthalein, or they may order the "latest and greatest" test that is not yet available. Few clinicians have a clear understanding of test strategy or the use of testing algorithms to solve clinical problems resulting in "shotgun" ordering, ie, "Do everything!" Conversely, the result may be insufficient ordering, perhaps prolonging length of stay. Another frequent annoyance is redundant ordering, usually when there is more than 1 physician caring for a patient and both order the same tests without the other's knowledge.

Illegible or inexact orders or the use of abbreviations often leads to confusion. Being unaware of test or specimen requirements, eg, which tests require fasting or which need to be drawn at a specific time, or being unaware that the lab needs to prepare for the receipt of certain types of specimens, such as muscle biopsies, which require special processing, can result in improper or degraded specimens and can delay result reporting.

Surgery, the emergency department, and private offices frequently neglect strict patient and specimen identification requirements. The improper labeling of a Pap smear or worse, an improperly labeled specimen for crossmatch, can lead to tragedy. Physicians are generally unaware of specimen collection, storage, and transportation variables, which can affect test results. None of the above, so-called "preanalytic" phases of testing are under the control of the laboratory, but if something were to go wrong, the lab always gets the blame. That is irritating!

Finally, it is unfortunate that so few clinicians have a clear understanding of the limits inherent in quantitative measurements or the concepts of sensitivity, specificity, and predictive value. Whenever a test result does not fit the clinician's preconceived diagnosis, it is invariably blamed on "lab error."

Rudeness (We Get No Respect!). It is remarkable how many synonyms there are for rudeness -- discourteous, abusive, disrespectful, arrogant, grouchy, crude, raging, and furious. All of these terms, and many others, are often used by laboratory personnel to describe clinicians. Physicians who call the laboratory with a problem regularly berate the receptionist or technologist to whom they are speaking and shift blame for their own quandary on laboratory personnel. Unjustified complaints or vague complaints without specifics, eg, "Your lab tests are never of any help," or complaining about a problem that occurred several weeks or months ago are disturbing and can embitter lab personnel.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.