Clinlab Navigator

Critical Values

The concept of critical values for laboratory results was first introduced by Lundberg in 1972. He defined critical laboratory values as "values which reflect pathophysiological derangements at such variance with normal as to be life-threatening if therapy is not instituted immediately". Critical values have also been called panic or alert values, but the use of these names is discouraged.

A College of American Pathologists (CAP) Q-PROBES study of critical value practices in the United States provided evidence that calling of critical values improves patient care. Review of more than 5000 patient charts revealed that 45 percent of the critical values reported were unexpected and 65 percent led to a change in treatment.

Critical values policies and procedures are required by many licensing and accrediting organizations including the College of American Pathologists (CAP), the Joint Commission, and Centers for Medicare & Medicaid Services (CMS). The laboratory is required to have a written policy for critical values with the following attributes:

• Written critical value policy and procedure
• Policy established in conjunction with medical staff
• Policy communicated to healthcare providers
• Policy followed by all laboratory staff
• Inclusion of both low and high critical levels, when appropriate
• Inclusion of all analytes meeting the definition of “critical value”
• Exclusion of abnormal results
• Critical results called to someone authorized by the medical staff to take immediate action
• Read-back required by healthcare providers
• Documentation of each call including date, time, caller's identification & receiver's identification
• Contingency plan for unsuccessful calls
• Periodic review of critical value list
• Assessment of critical value plan effectiveness
• Remedial action taken when necessary

In 2004, the JCAHO expanded their critical value reporting requirements to include a “read-back” of direct and telephone verbal reports. Read-back means that the health care provider who receives the critical value report must repeat the results back to the medical laboratory scientist. This requirement is intended to further reduce medical errors. The relatively simple process of verifying a verbal communication by having the recipient repeat it has been shown to be effective in the airline industry.

 In the laboratory, a prospective review of more than 800 critical microbiologic results reported by telephone at three healthcare organizations found a 3.5% error rate.The three most common errors were incorrect patient name (34 percent), incorrect test result (31 percent), and incorrect specimen (21 percent). Calls to physicians had the highest error rate (5 percent) followed by nurses (3.4 percent) and ward clerks (zero errors).

A suggested approach for verbally reporting critical laboratory results is:

1. State the patient’s name, test name, time of draw, critical lab results, units of measure, and reference range.

2. Ask the recipient to repeat back the patient’s name and the critical lab result.

3. Verbally correct any errors and repeat the request for a “read-back”.

4. Document the date, time, test results, and person to whom the test results were reported.

Regulatory agencies require timely reporting of critical results by laboratory personnel but do not provide a definition of timely. Each laboratory must establish its own expectations for timely reporting and then audit compliance with this goal.  Many laboratories specify that critical results should be called to health care provider within 10 minutes after the result is available.

A College of American Pathologist Q-Probes study in 2008 reported that a median of 5 minutes was required for staff to notify someone about a critical result once testing was complete. A median of 56 minutes elapsed from the time a specimen was collected until the critical result was called to a licensed caregiver.

Some laboratories may choose to verbally report all critical results, regardless of whether or not similar results have been previously reported on the same patient during that admission. Others may choose to only report the first critical result during a hospital stay or the first critical result within a defined time interval such as 24 hours. The approach should be determined with input of the medical staff served by the laboratory and explained in the written laboratory policy.

Many laboratories have delta checks to automatically compare the difference between two subsequent results on the same patient. By incorporating delta check limits in the call back procedure, repeat calls can be avoided in situations where subsequent results are abnormal but not significantly different from the previously reported result.

Most clinical laboratories do not allow physicians to opt-out of critical value notifications. Some specialists request a different set of critical values for their unique patient population. For example, nephrologists at a dialysis clinic may be concerned about a different range of electrolyte values than hospitalists. The majority of clinical laboratories discourage this practice because maintaining separate lists is challenging and it increases the likelihood of reporting errors.

When the laboratory is unable to reach a healthcare provider by telephone to report a critical value, the laboratory may try other means of communication such as fax, voicemail, or email. However, these alternatives do not provide assurance the message was received by the right person in a timely manner and do not fulfill the read-back requirement.

Some laboratory information systems provide electronic result notification, as well as a mechanism that enables the recipient to acknowledge that the message was received. Even when these transactions are automated, a process must be in place to follow-up those calls that are not acknowledged. Automated systems must comply with patient privacy requirements.

The laboratory that performs the test is responsible for reporting critical results. If a test is collected at one site and sent to a reference laboratory for testing, the reference laboratory is responsible for calling critical results. For this reason, it is important for the sending laboratory to include patient contact information to the reference laboratory.

Although regulatory agencies mandate reporting of critical values, they do not state which laboratory tests require critical value limits and notification. Individual clinical laboratories are challenged with creating a list of critical values that reflect institutional organization, clinical demand, patient population, laboratory instrumentation and staffing. Such variations have hindered the development of universal standards for critical value reporting. The medical literature contains very little outcomes-based data on critical value thresholds. The medical director of the laboratory is responsible for establishing the critical value list. Cutoff levels should be established in consultation with the medical staff served by the laboratory.

The following table provides an example of a critical value list and cutoff values.

Anatomic pathology and cytology also have critical values. Examples include unexpected malignancy, fungi in a specimen from an immunocompromised patient and transplant rejection.

A small subset of clinical laboratory tests comprises almost 99% of the critical values encountered in an acute care hospital. These tests include potassium, glucose, calcium, troponin, white blood cell count, hemoglobin, hematocrit, platelet count, magnesium, phosphorus and therapeutic drugs.

Not every laboratory test should have a critical value associated with it. The critical value policy is intended to address results that are unexpected and potentially life-threatening. It is not intended to communicate results that are merely abnormal. Creation of too many critical values overburdens laboratory personnel, creates alert fatigue, annoys physicians and nurses and increases dissatisfaction with laboratory services.

For example, troponin is only ordered in the context of suspected myocardial injury. Therefore, abnormal elevated troponin values are never unexpected. Any detectable level of troponin above the upper limit of the reference range is medically significant. No one has clearly defined what level of troponin is critical or life threatening. The medical significance of a particular troponin value depends on the individual patient. For these reasons, troponin should probably not be included in the hospital’s critical value list.

Laboratories have different policies regarding repeating critical lab result prior to reporting. A recent Q-Probes study by the College of American Pathologists reported that 61% of clinical laboratories delay communication of critical values by as much as 20 minutes because they repeat the test to confirm the result before reporting (Lehman C, Howanitz P, Souers R et al. Arch Pathol Lab Med 2014; 138: 788-93). More than 99% of initial and repeat critical results were deemed to not be significantly different. These results confirmed the findings of a previous study which showed that repeat testing of critical hemoglobin, platelet count, WBC count, prothrombin time and activated partial thromboplastin time results did not offer any advantage over singlet testing (Toll, AD, et al. Arch Pathol Lab Med 2011;135:440-444). Even though critical values occur at the upper and lower ends of the analytical range, these results are just as reliable as normal results in a well-managed laboratory. Repeat testing may do more harm than good by delaying reporting of critical results.

References

Barenfanger J, Sautter RL, Lang DL, Collins SM, Hacek DM, Peterson LR. Improving patient safety by repeating (read-back) telephone reports of critical information. Amer J Clin Pathol 2004;121:801-803.

Emancipator K. Critical values: ASCP practice parameter. Am J Clin Pathol. 1997;108:247-253.

Genzen JR and Tormey CA. Pathology consultation on reporting of critical values. Amer J Clin Pathol. 2011;135:505-13. 

Howanitz PJ, Steindel SJ, Heard NV. Laboratory critical values policies and procedures: College of American Pathologists Q-PROBES study of 623 institutions. Arch Path Lab Med. 2002;126:663–669.

Lundberg GD. When to panic over abnormal values. MLO.1972; 4(2):47-54.

Lundberg GD. Critical (panic) value notification: an established laboratory practice policy (parameter). JAMA. 1990;263:709.