11.3 Assessment Categories, Breast Density, and Recommendations

Assessment Is a Physician Task With a Federal Word Requirement

BI-RADS assessment categories 0 through 6 communicate likelihood of malignancy and the next management step. The mammographer does not assign the category. The interpreting physician does. Your role is to recognize incomplete work-ups that need extra views or priors, to understand what each category implies so you sequence diagnostic imaging correctly, and to never tell a patient “you are a 4” as if you issued the report.

ACR’s BI-RADS v2025 manual is an extension of the 5th edition. The category numbers, the classic likelihood bands, and the four composition (density) phrases below are the established clinical framework the ARRT outline still tests. MQSA, however, does not treat a numeral as the legal assessment. The Mammography Quality Standards Act requires an overall worded assessment in the report (for example, “Suspicious”), not a number alone. Optional 4A/4B/4C letters may appear in addition to that word; they do not replace it.

FDA’s amended MQSA rules (effective September 10, 2024) also require an overall breast-density statement using the same four phrases used in BI-RADS composition, and they require specific dense versus not-dense language in the patient lay summary. MQSA also lists Post-Procedure Mammogram for Marker Placement as a permitted worded assessment; that phrase is a federal report option, not a seventh BI-RADS cancer-risk category you assign in the room.

Categories 0–6

Category 0 is the screening recall code: need additional mammographic views, ultrasound, and/or prior films. After those images exist, the study is no longer incomplete; it receives a final 1–6. Do not use 0 because magnetic resonance imaging (MRI) might be interesting later if the mammogram already has a suspicious finding — the mammogram still needs a final assessment. Category 0 also is not a hiding place for a finding that is already clearly a 4 or 5; if the lesion is suspicious on the images you have, the assessment is suspicious, not incomplete.

Category 1 means there is nothing to comment on: symmetric tissue, no mass, no distortion, no suspicious calcifications. If a benign finding is described (cyst, intramammary node, typical secretory calcifications, implants, fat-containing hamartoma), the assessment is 2, not 1. Both 1 and 2 return the patient to routine screening; the difference is whether a benign finding is named in the report.

Category 3 is not a parking lot for indecision. Classic 5th-edition examples include a noncalcified circumscribed mass on a baseline mammogram, a focal asymmetry that thins on spot compression, or a solitary group of punctate calcifications — each with ≤ 2% chance of malignancy. Management is short-interval follow-up, classically a unilateral diagnostic mammogram at about 6 months, then bilateral imaging at 12 months if stable, with later conversion to category 2 if unchanged. Growth or increasing calcifications moves the finding to 4 or 5. Category 3 is a final assessment after a complete diagnostic work-up, not a screening incomplete.

Category 4 covers a wide band, which is why 4A/4B/4C exist. MQSA still wants the word “Suspicious.” Category 5 is reserved for classic cancers (for example, an irregular spiculated high-density mass, or fine linear calcifications in a segmental distribution). A benign needle biopsy of a category 5 finding is discordant until proven otherwise. Category 6 is used after a known malignant biopsy, including while monitoring response to neoadjuvant chemotherapy. A new suspicious finding in the other breast is still 4 or 5, not 6.

The chart below uses classic 5th-edition likelihood numbers for teaching. Category 0 is plotted at 0 because it is incomplete, not because the recall is proven benign. Category 4 is plotted near the top of its wide band so you can see that “suspicious” reaches just below category 5; the table, not the bar, is the place to remember 4A/4B/4C splits.

Breast Density (Composition)

Density is reported with four statements. These are the same four used in federal density reporting:

• a. The breasts are almost entirely fatty. Mammography is highly sensitive in this setting.
• b. There are scattered areas of fibroglandular density.
• c. The breasts are heterogeneously dense, which may obscure small masses.
• d. The breasts are extremely dense, which lowers the sensitivity of mammography.

Categories c and d are “dense.” Categories a and b are “not dense” in the federal lay-summary split. Assignment is based on masking risk, not a strict percentage of white tissue. A breast with less than half fibroglandular volume can still be heterogeneously dense if a region is thick enough to hide a small mass. The 5th edition discourages the old 25/50/75% quartile labels as the sole rule.

Population teaching figures from the National Cancer Institute are about 10% almost entirely fatty, 40% scattered, 40% heterogeneously dense, and 10% extremely dense. Dense tissue both masks masses (both appear white) and is a moderate independent risk factor. Density is judged by the radiologist on the mammogram; it cannot be assigned by palpation or by cup size.

Recommendations Tied to Category

• 1 or 2: routine screening interval (annual or the interval the referring clinician uses).
• 0: do not send the patient home as “normal.” Complete extra views, ultrasound, or prior comparison.
• 3: schedule short-interval diagnostic follow-up (about 6 months), not a new screening-only visit that ignores the finding.
• 4 or 5: tissue diagnosis should be obtained; the mammographer’s next images are often biopsy guidance, not “wait a year.”
• 6: imaging follows oncology, not a screening “all clear” letter.

If a patient asks what her number means, point her to the lay letter and the interpreting physician. Explaining that category 0 means more pictures, not a cancer diagnosis, is patient care. Assigning 4C yourself is not.

What the Exam Will Try to Mix Up

Numeric codes are not a substitute for MQSA words. 4A is still “Suspicious.” Density letters a–d are composition, not assessment categories 0–6. Category 3 is follow-up, not biopsy-by-default and not a screening incomplete. Category 6 is known cancer, not “probably cancer.” Category 0 is incomplete work-up, not a lifetime risk score. And the person who types the assessment on the report is the interpreting physician, not the mammographer who positioned the paddle.