Forensic Dentistry, Second Edition

Forensic dental identification 175

odds of the occurrence and allowance that a duplicate exists. This calculation

assumes that each tooth has the same likelihood to be missing or filled.

If an examined body showed that seven teeth were missing (antemortem,

not peri- or postmortem) and sixteen teeth were filled, the odds of two people

having the exact same configuration of those features would be

K 327

32

1

32

2

32 71

7

, 3 365 856

()

=××→ ,,

−−

=

multiplied by

K32 16

32

1

32

2

32 16 1

16

, 601 080 390

()

=××→ ,,

−−

=

3,365,856 × 601,080,390 = 2,023,150,037,163,840

The likelihood then is 1 in 2 × 10^5 , a figure that is more than the all-time
world population estimated by Haub to be approximately 106 billion.^12
Keiser-Nielsen suggested taking the analysis even further, by considering
the different combinations of restorations in each of the five restorable surfaces
of each tooth. If other features such as endodontic treatment and the types of
restorative material used were also considered, the numbers of possible combi-
nations of features become enormous.
Keiser-Nielsen also suggested that forensic odontological identification data
should be quantifiable, like in fingerprint analysis, and that forensic dentists
should follow comparable standards. He suggested that twelve points of com-
parison be recorded before establishing certainty in identification.
Drs. Soren Keiser-Nielsen and Riedar Sognnaes were dedicated and
visionary practitioners. Their contributions to forensic odontology were con-
siderable and perhaps unsurpassed by other forensic dentists.11,13–15 In a sincere
effort to establish and promote scientific and mathematical bases for forensic
dental identification, they made assumptions that failed to carefully consider
the nature of the dental features they analyzed for their statistical conclusions.
They assumed that these characteristics occurred independently and that
their resulting values could be multiplied to produce expected frequencies.
In a 2003 paper discussing Keiser-Neilsen’s and Sognnaes’ statistical con-
clusions, Adams stated, “This type of statistical assessment suggests that
all of the various combinations of missing and filled teeth occur randomly
and that they are equally probable in the population, an assumption that is
not valid.” He further stated, “Statistical arguments of this type regarding
the possible number of combinations of missing and filled teeth have been
especially stressed by Keiser-Nielsen and Sognnaes for forensic purposes;