Why a BI-RADS 4 Mammogram Is Not a Diagnosis

A suspicious mammogram is a reason to look closer, not a conclusion

A BI-RADS 4 mammogram is a risk category, not a diagnosis. It tells you that a radiologist saw something suspicious enough to recommend a biopsy, not that you have cancer. In the largest national analysis of these assessments, roughly one in five category 4 findings turned out to be malignant, which means most did not. The letter and number are a structured estimate of probability that guides the next test, and the biopsy, not the mammogram, is what produces an actual diagnosis.

What BI-RADS actually encodes

BI-RADS, the Breast Imaging Reporting and Data System, is a standardized vocabulary created by the American College of Radiology so that a mammogram report carries the same meaning across facilities. Its final assessment categories run from 0 to 6. Category 1 is negative, category 2 is a benign finding, and category 3 is "probably benign," carrying under a 2 percent likelihood of malignancy. Category 5 is "highly suggestive of malignancy," at 95 percent or higher. Category 4, "suspicious," sits in between and is defined by the ACR lexicon as a finding with more than a 2 percent but less than a 95 percent chance of being cancer.

Read that range again. A single category is allowed to describe a finding that is 3 percent likely to be cancer and, at the other extreme, one that is 90 percent likely. That is an enormous spread, and it is the central reason a category 4 result should not be read as a verdict in either direction.

Why one category holds such a wide range

Because that span is so broad, the ACR subdivides category 4 into three tiers. Category 4A is low suspicion, defined as greater than 2 percent up to 10 percent. Category 4B is moderate suspicion, from more than 10 percent up to 50 percent. Category 4C is high suspicion, from more than 50 percent up to less than 95 percent. A 4A finding and a 4C finding both trigger a biopsy recommendation, yet they describe very different pre-test probabilities. Two people can each be told they have a "BI-RADS 4" mammogram and still be in genuinely different situations.

What the national numbers show

The value of these categories is that they can be checked against outcomes. Elezaby and colleagues did exactly that in the National Mammography Database, drawing on more than 1.3 million diagnostic mammograms in a study published in Radiology in 2018. About 125,000 of those examinations were assigned category 4, and roughly 91,000 had a biopsy result on record. Across those biopsied category 4 findings, the positive predictive value for biopsy, meaning the share that proved to be cancer, was 21.1 percent. Broken down by subdivision, it was 7.6 percent for 4A, 22.2 percent for 4B, and 69.3 percent for 4C.

Two things stand out. First, the observed rates line up closely with the probability ranges the lexicon assigns, which is evidence that the system is well calibrated. Second, most category 4 findings, and the large majority of 4A findings, are not cancer. A biopsy recommendation is a request for information, not a prediction that the information will be bad.

The figure that matters is your subcategory, not the headline

This is where reading the full report helps. A national average for "category 4" blends low, moderate, and high suspicion findings together, so it overstates the risk for someone with a 4A and understates it for someone with a 4C. The more informative number is the one attached to the subcategory. Broader benchmark work from the Breast Cancer Surveillance Consortium, also published in Radiology, puts the overall probability of cancer following a biopsy recommendation at roughly a quarter, consistent with the category 4 figure from the mammography database.

None of these numbers describe an individual. They are population averages, and a given finding's true probability depends on its specific imaging features, breast density, and personal history, all of which the interpreting radiologist weighs. What the data do establish is direction: a suspicious label is built to be sensitive, catching cancers early at the cost of sending many benign findings to biopsy.

Why the distinction matters

Treating a category 4 as a diagnosis carries real costs. It can turn the days before a biopsy into a period of avoidable distress, and it can distort decisions if a person assumes the outcome is already settled. Understanding that BI-RADS encodes a probability, and that category 4 in particular spans from low to high suspicion, reframes the result as what it is: a triage signal that has done its job by routing a finding to the test that can answer the question. The biopsy, and the pathology it produces, is the diagnosis.

This article is educational and not a substitute for individual medical advice. Anyone with questions about a specific mammogram report should discuss the findings, the assigned category, and the recommended next step with the clinicians managing their care.