Notifying clients of life-threatening results.

Regardless of what you call 'panic values,' your lab's actions should include establishing discrete limits and a practical plan for telephoning alerts.

THE CLINICAL Laboratory Improvement Amendments of 1988 (CLIA '88) require laboratories to:

"... develop and follow written procedures for reporting imminent life-threatening laboratory results or panic values. In addition, the laboratory must immediately alert the individual or entity requesting the test or the individual responsible for utilizing the test results when any test result indicates an imminent life-threatening condition." (493.1107f)[1]

Our lab, the Boston office of a national referral lab, has a policy that predates these Federal regulations. Based on our experience, we would like to offer several suggestions that may benefit you.

* What to call the policy. A common term used is "physician panic values." Some labs use the term "imminent life-threatening values." But such titles misrepresent the lab's knowledge and the actions it takes.

First, inclusion of the word "physician" in the title is too restrictive. CLIA allows non-physicians to accept lab results; nurses, secretaries, and other non-clinicians, for example, may receive the information. There is no need to include the word "physician" or even the broader term "health care provider."

Second, the CLIA term "imminent life-threatening" leaves out the important adjective "potentially." Clinical laboratory results are only potentially imminently life-threatening; they must be interpreted in the full context of the patient's condition.

Third, the all too commonly used word "panic" suggests "something setting off a precipitous emergency response."[2] Similarly, the alternative term "critical" suggests a point beyond which "some quality, property, or phenomenon suffers a definite change." Both of these terms imply an urgency that may or may not exist.

* Alert limit policy. In our experience, laboratory results that exceed the limits defined in the policy most often do not require urgent medical intervention. The problem with such terms as "panic" and "critical" is that they suggest that laboratorians are in a position to know the medical context of results. They do not have this information, however. No medical assessment is possible in a vacuum.

The best terminology reflects the action the lab takes. Words that imply such activity include "alert," as used in CLIA. The word "notification" is rather weak. "Alert" is stronger and is, therefore, what we use. The policy establishes discrete limits, so we also use the word "limits" in the title of the policy, making the whole name "alert limit policy."

Our lab has companions to its alert limit notifications; we refer to them as "call backs." They are courtesy calls made to clients concerning results that exceed more narrowly defined limits. These calls are made during "working hours" 6 days a week as a service to clients. They provide an early warning of out-of-range results that are not as drastic as those that qualify for alert limit policy tactics.

* How alert values are created. We derived our limits from several sources but prefer using literature references when they are available. We might alert a client, for example, that bleeding may occur because his patient's platelet count is below 40,000.[3] In other cases (such as with therapeutic drug levels), we use the lower limit of the toxic range.

We employ medical judgment when sources are not available. When setting limits, we may ask clients for their opinions. After we set a limit and have the occasion to call on it, we inquire if the result led to medical intervention. We want to know if the alert was valuable.

* Screaming doctors. A 3 am call from the laboratory is not always well received, and we have had our share of screaming doctors. The following morning, we call the physician back to explain the policy. When we need to, we state that the Federal Government requires all clinical labs to have such guidelines. Typically, the physician will apologize.

In certain cases, our lab has different limits for different patients. We have a lower limit for a newborn's bilirubin, for example, but no limit for an adult's reading on that analyte. We have no limit for urea nitrogen with dialysis patients. The list is an evolving one, with changes made after considering all of the input from the physicians who are the end-users of the policy.

* What constitutes contact. The physician's office line is busy. Does making a call that did not get through constitute an adequate attempt at notification?

Our policy requires a minimum of three tries at phoning; the attempts are spaced from 3 to 10 minutes apart. When no one answers, the phone must ring a minimum of 10 times. For each call, we document:

* Person placing call

* Telephone number

* Date and times called

* Outcome

After three unsuccessful attempts, a supervisor or other manager may get involved to let the laboratorian get back to his or her most pressing work. The supervisor may try to reach the physician through a partner, if any, or the hospital. It often happens that an answering service is covering for the physician. In that case, we ask the agency to page the doctor. Nearly all attempts at reaching the clinician are ultimately successful.

If all reasonable attempts fail, however, we note it in the report. When we do finally have success in contacting the physician, we discuss several ways to avoid repeating this particular problem.

* What is immediate? When we say "immediate," we mean the next reasonable moment that communication can be made without compromising the integrity of the lab or other specimens. If our policy is to repeat the analysis, then alerting the physician must occur only after the result is repeated and verified. Other specimens on the analyzer must not be compromised. The laboratorian must walk, not run, to a telephone.

Generally, we do not set up new runs or begin new tasks before calling the physician. If, however, the laboratorian wants to complete his or her review of the list of tests run on the specimen in question in order to see how other values turn out, that is acceptable. If, for instance, the patient's sodium is low and the technologist knows the chloride value (which may also be low) will become available in 2 minutes, he or she can wait for this result.

* Why we're here. In all circumstances, however, once a result is verified and ready for release, the laboratorian must make the initial attempt to notify the physician within 30 minutes. In reality, the majority of these calls are made within 10 minutes.

One of our quality assurance measures is assessing our compliance with the alert limit policy. Excuses for failure to comply are unacceptable. A laboratorian may not sit on a potentially life-threatening value in order to meet other deadlines. A life may be at stake. After all, isn't this why we're here?

References

1. Medicare, Medicaid, and CLIA Programs: Regulations Implementing the Clinical Laboratory Improvement Amendments of 1988 (CLIA). Department of Health and Human Services, Health Care Financing Administration. Federal Register. Feb. 28, 1992; .57: 7001-7186.

2. Webster's Ninth New Collegiate Dictionary. Springfield, Mass: Merriam-Webster; 1988.

3. Isselbacher KJ, Brunwald E, Wilson JD, Martin JB, et al. Harrison's Principles of Internal Medicine. 13th ed. New York, NY; McGraw-Hill: 1994.

Kaufman is regional medical director and Collins is laboratory manager of MetPath New England, Boston, Mass.