Clinical Lab Overutilization, Utility & Reimbursement

Vol. 20 • Issue 10 • Page 41

Legal / Regulatory Issues

While no reference point denoting when an excessive number of tests has been performed has been established, hospital administrations are concerned about the costs of any tests, since they are viewed as costs when hospitals are reimbursed for patients’ care via the Medicare and Medicaid Diagnosis-related Groups (DRG) and negotiated contracts from insurance companies. Thus, overutilization is primarily a cost issue.


A small number of studies have shown that overutilization is not associated with improved outcomes for patients.1 In 1998, van Walraven and Naylor asked, “Do we know what inappropriate laboratory utilization is?” and tried to answer the question by conducting a systematic review of laboratory clinical audits.2 They concluded that few studies used appropriate criteria to identify inappropriate use and suggested the need for researchers to develop appropriate standards that would do this. To date this has not happened.

Although there likely is overuse and misuse of testing on outpatients, these tests generate revenue for a clinical laboratory so there has been no effort in the U.S. to reduce testing. In some respects, laboratory testing has supplanted clinical judgment. The pressure on physicians to see an ever-increasing number of patients in a short period of time has lessened the time a physician has to take a comprehensive history, which is the most useful tool for making a diagnosis.


The only appropriate way to identify overutilization seems to be to compare the number of tests ordered on a large number of patients with the same severity of the same disease or condition with similar outcomes. Since many patients are treated according to institutional guidelines, often by the same team of physicians, comparisons should be between hospitals rather than within a single hospital.

Table 1 illustrates laboratory usage for a small number of conditions within the three hospitals of the University of Pennsylvania Health System. Teaching hospitals are typically where overutilization is most probable since much of the test ordering is done by physicians in training who have rarely been taught by laboratory physicians about proper resource utilization and who may be supervised by attending physicians who have a philosophy that trainees must have the opportunity to learn from their mistakes. Many attending physicians are also ignorant of the extent of their trainees’ test ordering, as their trainees may only relay abnormal results to them while neglecting to mention a possible plethora of normal test results.

In the teaching hospital environment, many tests that are not strictly needed for patient management may be ordered according to research protocols. While some of these tests may be clinically irrelevant, the hospital’s costs should be covered by the sponsors of the research studies and, thus, should not be of concern.

Archive Image

Click to view larger Table.

Test Selection

Studies have identified the non-utility of many tests ordered on admission of patients to a hospital.3 An authoritative committee within a hospital can establish testing criteria for admission testing on different patient populations. When these have been developed, they need to be endorsed by the hospital’s medical board to establish a local standard of practice.

Tests should only be ordered selectively, for example, when they contribute to making a diagnosis, but most diagnoses are made in the reimbursable outpatient setting, except for those patients admitted through an emergency department. If not used to assist making a diagnosis, a test should play an important role in the decision-making process, for example, contributing to the monitoring of a patient’s condition or influencing a therapeutic decision.


Overutilization must be considered in light of the relatively low cost of commonly requested laboratory tests compared with the overall costs of hospitalization. If additional tests can lead to a reduced length of stay, there is a real financial benefit. Pressure to reduce patient stays sometimes leads to premature discharges and, ultimately, expensive readmissions.

One example is the patient whose warfarin therapy has not been stabilized. The laboratory director must take the initiative in detecting overutilization. This requires review of a large number of patient records. Repetitive ordering of tests when results are unlikely to influence patient care wastes resources. Follow-up testing to the slightly abnormal test result is frequently unproductive. To change this, a starting point is to look at the number of tests repeated following previous normal values and develop a computerized approach to track the number of consecutive normal values on each patient.

Repeat Ordering

In stable patients admitted to hospitals, there frequently is no need to repeat previously ordered tests with normal results obtained as long as six months previously. The laboratory director working with the appropriate clinicians can also develop guidelines for the number and type of tests to be performed on a daily basis. However, these clinical pathways tend to be of greater value for surgical patients for whom conditions have similar treatments than for medical patients whose conditions may differ markedly in terms of acuity and for whom intensity of testing could justifiably be different.

There is no need to repeat the ordering of tests that unequivocally have established a diagnosis, e.g., troponin I or T once a patient has been proven to have had a myocardial infarction, or repeat antibody measurements once these have been confirmed in a patient. Many physicians are unaware of the half-lives of the analytes that they request to be measured. A general guide is not to perform the same test more frequently than once every three half-lives.

When a patient is under the care of several physicians with different specialties, there is risk of the same test being ordered with inappropriate frequency. Most laboratory computer systems are programmable to reject repeated testing of the same analyte within a certain time frame. The laboratory director and physician users should jointly agree such time frames.

Stat Testing

Tests ordered stat can be exempted from such rules to minimize the need for telephone calls in emergency situations to enable bypassing of the rules. Any test ordered on an inpatient that does not influence the management of the patient while he or she is hospitalized must be considered inappropriate since it becomes an unnecessary cost to be charged against the DRG through which the hospital is paid. Such tests, if performed more than three days prior to a patient’s hospitalization or after the patient is discharged, are fully reimbursable.

Meetings of the laboratory director with clinical residency program directors and their chief residents serve as opportunities to highlight and address overutilization issues. Yet, it is well-recognized that overutilization has to be addressed on a continuing basis and that administrative measures may be more effective than education alone.

Dr. Young is former vice chair for Laboratory Medicine, Department of Pathology & Laboratory Medicine Hospital of the University of Pennsylvania.


1. Axt-Adam P, vab der Wouden JC, van der Does E. Influencing behavior of physicians ordering laboratory tests. Med Care 1993;31:784-94.

2. Van Walraven C, Naylor CD. Do we know what inappropriate laboratory utilization is? JAMA 1998;280:550-8.

3. Kaplan EB, Sheiner LB, Boeckmann AJ, Roizen MF, Beal SL, et al. The usefulness of preoperative laboratory screening. JAMA 1985;253:3576-81.

About The Author