Bitemark analysis: the missing measurements behind a claimed match
A patterned injury can look compatible with someone's teeth without establishing who made it. Forensic bitemark analysis depends on more than a resemblance: the relevant teeth must supply distinguishing information, skin must retain it, and an examiner must interpret it reliably. In its October 2022 draft scientific foundation review, the National Institute of Standards and Technology found insufficient support for all three premises. The agency published its final report on 7 March 2023. The review concerned injuries on human skin, where a bite captures only part of the dentition under conditions that can alter its appearance.
The missing population evidence is particularly consequential for bitemark analysis. NIST's October 2022 announcement of the draft findings reported no studies establishing individual uniqueness in the patterns of the front-most biting teeth. It also reported no population studies identifying distinguishing features of those surfaces and measuring their frequencies. A full set of teeth contains information that a bite injury may never receive. Calling someone's entire dentition distinctive therefore does not establish the rarity of the much smaller pattern available for comparison on skin.
That distinction in bitemark analysis also appears in the 2016 review by Michael Saks and colleagues in the Journal of Law and the Biosciences. They separated two ideas often compressed into a claim of uniqueness. One concerns whether complete dentitions differ from one another. The stronger claim concerns whether every injury made by those teeth can be associated only with its maker. The second proposition requires information about what the biting surfaces actually leave behind, as well as information about the teeth themselves.
The same review discussed research comparing dental arrangements at a finer resolution than skin impressions can normally retain. As the reference database grew, the chance of finding similar arrangements increased. Reducing the resolution of measurement increased the number of dentitions treated as indistinguishable. Orthodontic treatment complicated the assumption further because it deliberately makes dental arrangements more alike. These results address the practical comparison problem: even an apparently unusual arrangement needs a measured frequency and a mark detailed enough to preserve the features being counted. An examiner's experience alone cannot supply that population measurement.
Skin introduces a separate source of variation. In their 2016 review in the Journal of Law and the Biosciences, Michael Saks and colleagues described experiments by the University at Buffalo research group using a mechanical biting apparatus and human cadaver skin. Cast dentitions were applied to different body areas, with multiple marks made by the same and different sets of teeth. Marks produced by a single dentition did not all look alike. Skin's direction-dependent properties changed the resulting shapes, and patterns could appear compatible with several dentitions in the researchers' collections. The review presents these as controlled experimental findings, rather than measurements taken during actual assaults.
Saks and colleagues likewise warned that cadaver experiments use a more stable surface than injuries on living people. Cadaver skin does not undergo the same inflammatory changes as living tissue. Movement, swelling, bruising and healing can affect an injury on a living person. Elasticity, bite location and the uneven biting surface can also alter the transfer. Even a highly detailed photograph cannot recover dental characteristics that the skin never recorded faithfully. This is why bitemark analysis has a transfer problem independent of whether the original teeth possess distinctive features.
Interpretation has been tested separately. NIST's 2022 summary described a 2016 study in which practitioners received images of patterned injuries. They were asked whether the injuries were bitemarks and, if so, whether adults, children or animals had made them. Examiners disagreed even about whether some injuries were bites. Those are classification decisions made before assigning a particular person's teeth to an injury. Agreement on a suspect cannot compensate for uncertainty about what kind of mark the photograph contains. The reported disagreements concerned the starting material for the comparison itself.
Professional terminology cannot supply the missing validation. NIST's 2022 explanation described American Board of Forensic Odontology guidance permitting exclusion, non-exclusion or an inconclusive finding. Non-exclusion means that a person's teeth could have caused a mark; it does not identify that person as its maker. The draft review found inadequate support even for reliable exclusion and non-exclusion through the technique. Its scope expressly left out identification of human remains through dental records, a different practice using information about the full dentition.
Saks and colleagues also separated resemblance from its statistical weight. Without information about how many other dentitions could produce a similar pattern, a comparison supplies no dependable measure of rarity. Their discussion contrasted this gap with DNA comparison, where population-frequency information helps assess a shared profile. NIST, a non-regulatory agency, assessed the scientific foundations rather than issuing a courtroom ban. Its findings place the limitation at the measurement itself: a compatible-looking injury has no established population framework that turns it into an individual identification.
- NIST, Forensic Bitemark Analysis Not Supported by Sufficient Data, draft findings, October 11, 2022 — www.nist.gov
- NIST, Bitemark Analysis: A NIST Scientific Foundation Review, publication record, March 7, 2023 — www.nist.gov
- Saks and colleagues, Forensic bitemark identification: weak foundations, exaggerated claims, Journal of Law and the Biosciences, 2016 — pmc.ncbi.nlm.nih.gov