5 Days Left – Save up to 50% + FREE Code Books with Select Training Packages |  Register Now

Proper Use of Modifier 91

Proper Use of Modifier 91

Modifier 91 Repeat clinical diagnostic laboratory test is used to report the same lab test when performed on the same patient, on the same day, to obtain subsequent test results.
Modifier 91 causes a lot of confusion when differentiating its use from that of modifier 59 Distinct procedural service. When reporting lab procedures, modifier 59 is used when the same lab procedure is done, but different specimens are obtained, or the cultures are obtained from different sites.
The June 2002 CPT® Assistant provides the following example showing the correct application of modifier 59:

For example, if multiple bacterial blood cultures are tested, including isolation and presumptive identification of isolates, then code 87040, Culture, bacterial; blood, with isolation and presumptive identification of isolates (includes anaerobic culture, if appropriate), should be used to identify each culture procedure performed. Modifier -59 should be appended to the additional procedures performed to identify each additional culture performed as a distinct service.

A contrasting example shows correct application of modifier 91:

A 65-year-old male patient with diabetic ketoacidosis had multiple blood tests performed to check the potassium level following subsequent potassium replacement and low-dose insulin therapy. After the initial potassium value, three subsequent blood tests were ordered and performed on the same date following the administration of potassium to correct the patient’s hypokalemic state.
Coding for this scenario is:
84132 Potassium; serum, plasma or whole blood

Per CPT® guidelines, you should not append modifier 91 for lab tests:

  • That are repeated to confirm the initial results;
  • That are repeated due to malfunctions of either the testing equipment or the specimen; or
  • When another appropriate one-time code is all that is needed to report the service.

If the test is repeated to confirm the initial results or because of a malfunction of the equipment, the service cannot be coded and modifier 91 would not apply.
If multiple tests are run but a single code describes the test, you should report only one code and modifier 91 would not apply. For example, 82951 Glucose; tolerance test (GTT), 3 specimens (includes glucose) includes three specimens; therefore, if three specimens were obtained during the encounter, you’d report only 82951.

John Verhovshek
Latest posts by John Verhovshek (see all)

About Has 577 Posts

John Verhovshek, MA, CPC, is a contributing editor at AAPC. He has been covering medical coding and billing, healthcare policy, and the business of medicine since 1999. He is an alumnus of York College of Pennsylvania and Clemson University.

5 Responses to “Proper Use of Modifier 91”

  1. Helen Bair says:

    I always like an update on modifier use. I was curious about the new modifiers that can be used instead of 59. Can you use them on lab and do you have any information on this?

  2. Sherry says:

    What about CT testing (CPT 87491)with specimens collected from multiple sites such as throat, rectal, vaginal. Some payers are requesting 59 modifier, but others want a 91.

  3. Patricia Thompson says:

    Anybody using this in Pharmacogenomics? PGx testing?

  4. CasSandra Mason says:

    If we have a patient that we are billing for an office visit by one provider at one location and billing for a lab at the different location by a different provider the same day, is there a modifier needed? If so which code and and what modifier? For example 99214 (e/m) and 80306 (lab)

  5. Rajeev Bhatt says:

    Patient has a multi-day test but Day 1 the test doesn’t start until after midnight when the patient falls asleep and then Day 2 the patient falls asleep at 10pm- Both Dos are technically the same since the DOS is when the test starts from what Jennifer has told me. So what modifier gets added to what day so that we are telling the payers that these are the same code, but actually separate tests due to the start dates?