Diagnostics, over-testing, and patient-centered care

June 22, 2016

Patient-centered care, a term coined in 1988, is part of the movement away from the old paternalistic medical model—“doctor knows best.” It is a heartening, natural development, as the complexity of evaluation and treatment options expand with the advancement of medical science.1

Patient-centered care is the right care, for the right person, in the right place, in the right manner, at the right time, and for the right reason. The Institute of Medicine defines patient-centered care as “providing care that is respectful of and responsive to individual patient preferences, needs, and values, and ensuring that patient values guide all clinical decisions.”2 As Atul Gawande eloquently states in Being Mortal, patient centeredness, to discern what’s important to each individual patient, is labor intensive.3

As the complexity of healthcare increases, with many practitioners involved in the care of the patient—subspecialist physicians, physical therapists, dietitians, pharmacists, advanced practice nurses, and the like—the opportunity for adverse effects of medical diagnosis and treatment is magnified. Such effects are more likely to occur when diagnostic treatment and technology take precedent over patient-centered care.

The test result is not the diagnostic

I was taught, as all MDs are, to take a comprehensive history of the distressed patient and perform a physical examination. From that a list of possible diagnoses is then formulated. Testing is done to confirm or refute the diagnoses on the list. That is, the laboratory results are used as an adjunct to diagnosis, not as a diagnostic itself. Forty-five years later, appropriate laboratory test interpretation is more important than ever. There are many ways that harm can be done due to inappropriate laboratory testing and test result interpretation.

A modern example of inappropriate laboratory test interpretation is the use of brain natriuretic peptide (BNP) to diagnose acute heart failure. The utility of this test is in its negativity. A normal test result rules out a cardiac cause of the patient’s problem. A test result out of the normal range is often deemed diagnostic of acute heart failure—when in fact no conclusion can be drawn about the cause of the abnormal result. An erroneous interpretation can lead to inappropriate therapy, with patient harm and an unwarranted delay in pursuing and making the correct diagnosis. Another example of inappropriate testing is getting a digoxin blood level just because the patient takes the drug; the patient’s clinical presentation may indicate a problem unrelated to heart disease.

Inappropriate testing and erroneous diagnosis often occur in the emergency department (ED). Pressed for time and with an increasing workload, ED physicians too-frequently diagnose and treat the laboratory test result(s). The test result becomes the diagnosis. Little thought is given to the cause of the abnormal result—the true diagnosis. Critical thinking about the patient cannot be squeezed into the time needed to deal with the workload and documentation requirements of various stakeholders—hospital, insurers, and local, state, and federal regulators. All of this detracts from patient-centered care.

Over-testing is not the answer

Additionally, the emphasis on screening well patients who do not have known risk factors can lead to some with abnormal results, and an abnormal result yields more and more testing and unwarranted management, sometimes including dangerous over-prescription. Too often, the physician forgets that the normal range of test results encompasses the middle 95 percent of the bell-shaped curve; 2.5 percent of patients will have a test result that is high (or low), but normal. The antidote for this unfortunate tendency: patient-centered care. Does the test result confirm a pre-test concern for disease? After careful explanation of what the significance of an abnormal result might be, was the patient asked before the sample was obtained what an abnormal result might mean for him?

There is an inordinate amount of time pressure placed upon physicians caring for hospitalized patients. Fixed payment for service—linking the volume of patients moved through the hospital to financial viability—discourages one of the best aids in diagnosis and patient care: time. Time to simply observe patients in non-critical situations is not used. Instead, testing is done to get the diagnosis more quickly so that the patient can be discharged and the bed filled with a new patient. Fear of our tort legal system can also encourage inappropriate testing: The patient who feels unwell calls his physician; to avoid liability, the physician tells the patient to go to the Emergency Room, where testing is assured.

Laboratory testing is ordered excessively in hospitalized patients, especially for patients in a critical care unit. A major reason is to “just to be sure.” Serial, daily testing might show a falling white blood cell count, perhaps indicating an improving patient. That does not help the patient’s worsening chest x-ray and progressive respiratory failure from pneumonia. Daily complete blood counts and basic metabolic profiles that are stable day after day might reassure the physician but are of no help to the patient. Hospital policy that allows nurses to order tests can compound the problem of unnecessary testing.

Appropriate testing

Of course, appropriate laboratory testing is invaluable; the practice of medicine would be impossible without it. And certainly there are situations in which insufficient testing sometimes occurs; prescription drug monitoring and illicit drug testing are two that come to mind. Laboratory testing is a necessity for patient care. It is most effective when its use is guided by patients in partnership with their physician: patient-centered care.


  1. Barry MJ, Edgman-Levitan S. Shared decision making: the pinnacle of patient centered care. N Eng J Med. 2012;366:780-781.
  2. National Research Council. Crossing the quality chasm: a new health system for the 21st century. Washington, DC: National Academies Press. 2001.
  3. Gawande A. Being Mortal. New York, NY: Metropolitan Books; 2014.