Journal of Oral Rehabilitation

Journal of Oral Rehabilitation 2017 44; 493–499

Diagnostic criteria for temporomandibular disorders

(DC/TMD): interexaminer reliability of the Finnish version
of Axis I clinical diagnoses
J. LESKINEN*, T. SUVINEN*, T. TEERIJOKI-OKSA†, P. KEMPPAINEN‡§,
€ AN
R . N AP € KANGAS¶**, P. ALSTERGREN††, Y. LE BELL‡‡, H. FORSSELL†‡‡,
€ ¶**§§
R . M Y L L Y K A N G A S * § § , M . T O L V A N E N ‡ ‡ , M . D O E P E L ‡ ‡ & K . S I P I L A* *Institute
of Dentistry, University of Eastern Finland, Kuopio, †Department of Oral Diseases, Turku University Central Hospital, Turku, ‡Institute of
Dentistry, University of Helsinki, Helsinki, §Helsinki University Central Hospital, Helsinki, ¶Research Unit of Oral Health Sciences, University
of Oulu, **Oral and Maxillofacial Department, Medical Research Center Oulu, Oulu University Hospital, Oulu, Finland, ††Department of Oro-
facial Pain and Jaw Function, Malm€o University, Malm€o, Sweden, ‡‡Institute of Dentistry, University of Turku, Turku, and §§Oral and Max-
illofacial Department, Kuopio University Hospital, Kuopio, Finland

SUMMARY Recently, updated diagnostic criteria for
temporomandibular disorders (DC/TMD) were
published to assess TMD in a standardised way in
clinical and research settings. The DC/TMD
protocol has been translated into Finnish using
specific cultural equivalency procedures. To assess
the interexaminer reliability using the Finnish
translations of the DC/TMD-FIN Axis I clinical
diagnostic assessment instruments. Reliability
assessment data were collected during a 1-day DC/
TMD Examiner Training Course at the University
of Turku, Finland, in collaboration with the
International DC/TMD Training and Calibration
Center in Malm€ o University. Clinical TMD
examinations according to the Finnish pre-final
version of the DC/TMD Axis I assessment protocol
were performed by four experienced TMD
specialists on altogether 16 models. Kappa
coefficient, overall percentage agreement (%A) as
well as positive (PA) and negative (NA)
agreements were used to define the reliability.
Myofascial pain with referral, headache attributed
to TMD and disc displacement (DD) without
reduction without limited opening showed
excellent kappa values (range 0
87–1
00). Fair-to-
good reliability was observed for diagnoses of
myalgia (k = 0
67), arthralgia (k = 0
71) and DD
with reduction (k = 0
64). The PA was high for all
pain-related diagnoses and DD without reduction
without limited opening (medians ≥83%), and
acceptable for DD with reduction (median 67%).
The NA was high (medians ≥87%) for all DC/TMD
diagnoses, except for myalgia which showed
acceptable NA (median 75%). The %A was high for
all assessed diagnoses (medians >85%). The
findings of this study showed DC/TMD-FIN Axis I
to demonstrate sufficiently high reliability for
pain-related TMD diagnoses.
KEYWORDS: DC/TMD, diagnostic criteria, TMD,
reliability, axis I translations
Accepted for publication 10 April 2017

temporomandibular joints (TMJs) and related

Background
Temporomandibular disorders (TMD) represent
pain-related, functional and intra-articular clinical
problems that involve the masticatory muscles, the
structures (1). One of the most widely used diagnostic
protocols for standardised TMD assessment was the
Research Diagnostic Criteria for Temporomandibular
Disorders (RDC/TMD) (2). Recently, updated diagnostic

© 2017 John Wiley & Sons Ltd doi: 10.1111/joor.12516

494 J . L E S K I N E N et al.
criteria for TMD (DC/TMD) were published by the RDC/
TMD Consortium Network of the International Associa-
tion of Dental Research (IADR) and the Special Interest
Group on Oro-facial Pain of the International Associa-
tion for the Study of Pain (IASP) to assess TMD in a stan-
dardised way in clinical and research settings (1).
Similar to RDC/TMD, the DC/TMD consists cur-
rently of a dual axis system that is based on the
biopsychosocial model of pain. The Axis I diagnostic
protocol is based on patient self-report and symptom
history and valid diagnostic criteria for the most
common pain-related TMDs and for intra-articular
disorders. The Axis II assessment protocol is
extended by new biobehavioural instruments that
assess disease-specific physical functioning as well as
psychological status and pain-related disability, suit-
able for screening and comprehensive assessment
levels (1). The new evidence-based DC/TMD assess-
ment protocol will contribute to the development of
more personalised care for patients with TMD and
for patients with other conditions related to oro-
facial pain (3).
The DC/TMD self-report instruments and clinical
examination protocol have been translated into Fin-
nish using specific cultural equivalency procedures
developed by the International RDC/TMD Consor-
tium (4). The DC/TMD was published in English in
2014 and since then, translations into approximately
30 languages have been in progress worldwide,
including Finnish as one of the first five lan-
guages. The translation process includes forward
translations, synthesis translations, back translations,
independent external review of back translations,
revision related to discrepancies, consolidation and
review activities into a single instrument appropriate
for expert committee review and cultural validity
revision, construction of a pre-final instrument,
administrative independent review of the translation
process by the International RDC/TMD Consortium
and documentation, for example for field testing
purposes. Currently, the final administrative review
of the Finnish version of the DC/TMD (DC/
TMD_FIN) has been completed, and it is currently
field tested (DC/TMD_FIN pre-final consolidated
translations) (5).
The aim of this study was to estimate the interex-
aminer reliability of the Axis I clinical diagnoses of
the DC/TMD-FIN.
Methods
Setting
The reliability assessment data were collected during a
1-day DC/TMD-FIN Examiner Training Course at the
University of Turku, Finland, as part of a 3-day DC/
TMD training course by the DC/TMD Training and
Calibration Center in Malm€ o University.
Models and examiners
The study sample consisted of 13 cases with TMD (pa-
tients of the Department of Oral Diseases of Turku
University Hospital) and three non-cases. The inclu-
sion criteria were age ≥18 years, and for cases,
included pain in the facial area for at least 5 days in
the previous 30 days and pain that had persisted for
at least 3 months; and for non-cases, no facial pain
over the prior 3 months. The exclusion criteria were
insufficient verbal fluency in the Finnish language
and inability to tolerate multiple examinations. Mod-
els were given oral and written information about the
study protocol and they signed an informed consent
form.
Methods and Translation procedures
Before the DC/TMD examination, the models com-
pleted the Finnish pre-final version of the DC/TMD
Symptom questionnaire [DC/TMD_FIN-SQ (5)] of
the Axis I protocol. To ensure the validity, the Fin-
nish translations of the Axis I DC/TMD-FIN SQ and
verbal commands were performed according to the
Guidelines for Establishing Cultural equivalency of
Instruments; www.rdc-tmdinternational.org (4–6).
Accordingly, the team leaders and an external
expert review panel, including four oro-facial pain/
TMD experts, reviewed the pre-final translations
and the Finnish Axis I examination form and proto-
col and made recommendations based on the inde-
pendent external back-translation reviews. The
consolidated and corrected translations were
reviewed by the RDC/TMD Consortium prior to this
study.
Clinical examinations according to the Finnish pre-
final version of the DC/TMD Axis I protocol were per-
formed on altogether 16 models by four experienced
© 2017 John Wiley & Sons Ltd
 RELIABILITY OF FINNISH DC/TMD AXIS I 495

TMD specialists (KS, PK, RN and TTO) using a strict Table 1. Computation of agreement
and specific procedure by the RDC/TMD Consortium
Examiner A diagnosis
Network (7). One of the four examiners (KS) who
Yes No Total
had earlier participated in the DC/TMD Examiner Examiner B diagnosis Yes a b f₁
Training Course in Malm€ o University, Sweden, acted No c d f₂
as an accredited Reference Standard Examiner (RSE). Total n₁ n₂ N
Furthermore, an oro-facial and TMD specialist (PA) Percentage agreement = (a+d)/N.
from the DC/TMD Training and Calibration Center Positive agreement = 2a/(f₁+n₁).
in Malm€ o University acted as the clinical protocol Negative agreement = 2d/(f₂+n₂).
supervisor, and the translation team leaders (KS, TS)
of the Finnish pre-final version of DC/TMD-FIN
absent. Statistics were performed using IBM SPSS
acted as the accredited Finnish protocol supervisors.
Statistics 21*.
After clinical examinations, the findings of the three
examiners were compared to those obtained by
the RSE. Results

Translation and cultural equivalency

Reliability assessment
There were no important discrepancies between the
To control rater examination order effects, a Latin square
host (Finnish) and source (English) languages regard-
design was used. Each clinical examination took a maxi-
ing the verbal commands and the clinical examination
mum of 20 min; between each clinical examination, the
protocol. In the DC/TMD-FIN SQ, some minor dis-
participants had an opportunity to rest for 10 min. The
crepancies concerned the translation of the following
Axis I diagnoses were based on the DC/TMD-FIN SQ and
questions:
the clinical Axis I examination protocol. More details,
Question 3: ‘Pain comes and goes’ and ‘Pain is
including the standards and the requirements for the relia-
always present’ were revised to be linguistically iden-
bility training procedures used in this study, are provided
tical to the source. Questions 9,11,12: The expression
in the Guidelines for DC/TMD Training and Calibration:
‘all the way’ was replaced with the linguistically more
www.rdc-tmdinternational.org (7).
appropriate synonym ‘fully’. In addition, some minor
linguistic modifications were required to alter the
Data analysis order of words into more fluent Finnish in SQ ques-
tions 3&5 and 4&7.
In the analysis kappa, overall percentage agreement
Detailed documentation of all the Finnish transla-
(%A), as well as positive (PA) and negative agree-
tion Logs and linguistic and culturally relevant
ments (NA) were used to evaluate the data from mul-
amendments are provided in the translation summary
tiple aspects (8). The reliability of all assessed DC/TMD
logs of the DC/TMD-FIN-SQ and the DC/TMD-FIN-
Axis I diagnoses was defined by computing kappa val-
clinical examination protocol (5,6).
ues for each diagnosis. Kappa coefficients were based
on the following: >0
75 indicating excellent reliability,
0
40–0
75 indicating fair-to-good reliability and <0
4 Distributions of the DC/TMD Axis I diagnoses in the study
indicating poor reliability (9,10). The reliability was sample
first determined by comparing each of the three exam-
The most common DC/TMD diagnosis in this study
iners to the RSE and then summarised as the medians,
sample was myalgia (n = 12), based on the findings of
lowest and highest values. The %A, PA and NA as well
the RSE (Tables 2,3). Joint-related pain diagnoses,
as prevalence of the diagnoses derived by RSE were
that is arthralgia (n = 6), were found more rarely in
calculated based on a 2 9 2 table (Table 1). %A indi-
comparison to muscle-related pain diagnoses. Among
cates the total number of agreements (10). PA indi-
cates the overall proportion when the RSE and the
other examiner agreed on an existing diagnosis. NA *IBM Corp. Released 2012. IBM SPSS Statistics for Windows, Ver-
indicates the agreement when the diagnosis was sion 21.0. Armonk, NY: IBM Corp.

© 2017 John Wiley & Sons Ltd

 496
J . L E S K I N E N et al.

Table 2. Reliability, agreement and distribution of Axis I DC/TMD diagnoses

Overall percentage
Kappa Positive agreement Negative agreement agreement
Distribution
Lowest, Median Lowest, Median Lowest, Median Lowest,
DC/TMD diagnosis Median highest (%) highest (%) highest (%) highest n (%)

Myalgia 0
67 0
54,0
85 0
92 (91
7%) 0
87, 0
96 0
75 (75
0%) 0
67, 0
89 0
88 (87
5%) 0
81, 0
94 12 (75%)
Myofascial pain with referral 0
87 0
49,0
87 0
92 (92
3%) 0
71, 0
92 0
95 (94
7%) 0
78, 0
95 0
94 (93
8%) 0
75, 0
94 7 (43
8%)
Arthralgia 0
71 0
61,0
84 0
83 (83%) 0
77, 0
89 0
88 (87
5%) 0
84, 0
95 0
86 (85
7%) 0
81, 0
93 6 (37
5%)
Headache attributed to TMD 1
00 1
00 1
00 (100
0%) 1
00 1
00 (100
0%) 1
00 1
00 (100
0%) 1
00 6 (37
5%)
DD** with reduction 0
64 0
64, 1
00 0
67 (66
7 %) 0
67, 1
00 0
97 (96
6%) 0
97, 1
00 0
94 (93
8%) 0
94, 1
00 1 (6
3%)
DD with reduction with intermittent NC* NC NC NC 1
00 (100
0%) 1
00 1
00 (100
0%) 1
00 0 (0
0%)
locking
DD without reduction with limited NC NC NC NC 1
00 (100
0%) 1
00 1
00 (100
0%) 1
00 0 (0
0%)
opening
DD without reduction without limited 1
00 1
00 1
00 (100
0.%) 1
00 1
00 (100
0%) 1
00 1
00 (100
0%) 1
00 5 (31
3%)
opening
Degenerative joint disease*** 0 0 0 (0
0 %) 0 1
00 (100
0%) 0
97, 1
00 1
00 (100
0%) 0
93, 1
00 0 (0
0%)

*no cases.
**disc displacement.
***no cases but one discrete finding.

© 2017 John Wiley & Sons Ltd

 RELIABILITY OF FINNISH DC/TMD AXIS I
intra-articular diagnoses, the most prevalent diagnosis
was disc displacement (DD) without reduction with-
out limited opening (n = 5). Of the 16 participants,
multiple DC/TMD diagnoses were observed in 10 par-
ticipants. There were no cases having diagnoses of DD
with reduction with intermittent locking or DD with-
out reduction with limited opening. For degenerative
joint disease, there was one discrete finding by one
examiner.
Diagnostic reliability
The kappa coefficients showed excellent values for
myofascial pain with referral (median k = 0
87),
headache attributed to TMD (k = 1
00) and DD with-
out reduction without limited opening (k = 1
00). The
reliability was fair to good for myalgia (k = 0
67),
arthralgia (k = 0
71) and DD with reduction
(k = 0
64). The reliability was poor for degenerative
joint disease (k = 0). As there were no cases of DD
with reduction with intermittent locking and DD
without reduction with limited opening, kappa values
were not computed (Table 2).
The overall %A was high for the assessed Axis I
diagnoses as shown by medians ≥0
86 (85
7%). The
median of PA was high for the diagnoses of myalgia
(91
7%), myofascial pain with referral (92
3%),
arthralgia (83
0%), headache attributed to TMD
(100%) and DD without reduction without limited
opening (100%). The median PA was acceptable for
DD with reduction (66
7%). For DD with reduction
with intermittent locking and DD without reduction
with limited opening, PA was not computed due to
the low prevalence. For degenerative joint disorder,
PA was 0. The median NA was high for myofascial
pain with referral (94
7%), arthralgia (87
5%), head-
ache attributed to TMD (100%) and intra-articular
disorders (≥96
6%). The NA was acceptable for myal-
gia (75
0%) (Table 2).
Discussion
According to this study, the overall interexaminer
reliability was found to be from acceptable to high for
TMD pain-related Axis I diagnoses (based on kappa
estimates, overall %A, PA and NA), whilst more vari-
ability was found for joint-related diagnoses. Thus,
the DC/TMD-FIN Axis I seems to demonstrate suffi-
ciently high reliability for the pain-related TMD
© 2017 John Wiley & Sons Ltd
497
diagnoses. High examiner reliability is important to
avoid misclassification of patients and provide person-
alised treatment for patients with TMD (3,11,12).
Partly based on this study, the Finnish versions of the
SQ and pain-related interview and verbal commands
can be implemented into clinical and research use in
the Axis I clinical diagnostics of TMD (5,6).
The new evidence-based DC/TMD Axis I protocol
includes additions, deletions and modifications in
comparison to the RDC/TMD (1). To achieve estab-
lished clinical assessment sensitivity and specificity,
the DC/TMD procedure includes strict commands to
the patient and clinical and diagnostic procedures.
In the new DC/TMD diagnostics, the concept of ‘fa-
miliar pain’ was introduced, indicating pain similar
to patient’s previous pain within the previous
30 days, as a criterion for diagnosing pain-related
TMD in order to minimise false-positive findings. In
addition, for all pain-related diagnoses, the patient
must have reported pain modified by jaw function
or movement in the SQ assessment. The new DC/
TMD criteria also include a new diagnosis, that is
headache attributed to TMD, as there is increasing
evidence that some forms of headache can be
related to TMD (13,14). This addition to DC/TMD
diagnostics is important and may also contribute to
future interdisciplinary research in this field of
study. Furthermore, the concept of ‘referred pain’
and the new diagnosis of myofascial pain with
referral are also included.
Kappa estimates, %A as well as PA and NA were
computed to assess interrater agreement in order to
evaluate the data from multiple aspects (8). The
kappa statistics observes the chance, but is sensitive to
prevalence, which is why kappa statistics may not be
useful for samples having very high or very low num-
ber of cases versus non-cases (8). For degenerative
joint disease, the kappa value manifested poor relia-
bility, indicating that only one discrete finding by one
examiner may have a huge impact on kappa values
when there are no findings detected by the other
examiners. The reliability study of this type tests the
operationalised criteria. It should be noted that the
findings by RSE (as presented in Table 3) are simply a
reference for pair-wise comparisons of each examiner,
and not ‘the truth’. Therefore, one subject with
degenerative joint disease is a misclassification, and
the Kappa estimate for the disorder could be com-
puted. The low kappa values may be due to the low
498 J . L E S K I N E N et al.

Table 3. Distribution of the DC/TMD Axis I diagnoses between the participants (n = 16), based on the Reference Standard Examiner

DD DD
DD with without without
reduction reduction reduction
DC/TMD Myofascial Headache with with without
diagnosisPatient pain with attributed DD* with intermittent limited limited Degenerative
ID Myalgia referral Arthralgia to TMD reduction locking opening opening joint disease

1 x x x x x
2 x x
3 x
4
5 x x x
6 x x
7 x x x x x
8 x x
9 x x x
10 x x x x
11
12 x x
13
14 x x x
15 x
16 x x x x

*disc displacement.

prevalence and even only one model exhibiting
examiner agreement, as was in the diagnosis of DD
with reduction that manifested fair-to-good reliability.
This diagnosis was based on TMJ clicking based on
symptom history and examination. As a limitation,
not all the DD with reduction manifest clinically
detected noises, which may be related to false nega-
tives (1,15). In the present study, there were no cases
having diagnosis of DD with reduction with intermit-
tent locking or DD without reduction with limited
opening. This is why kappa values could not be com-
puted. The original DC/TMD article (1) describes that
the detection of intra-capsular diagnoses was low for
most reliability estimates. Further, the validity of disc
displacements (except for DD without reduction with
limited opening) was inadequate.
The present study sample had non-optimal ratio of
cases (13 subjects) versus non-cases (three subjects),
which is a limitation of the study. The optimal size
for reliability testing includes a ratio ranging from
1:1 to no more than 2:1 (7). Consequently, further
studies on DC/TMD Axis I protocol with more opti-
mal study design are needed. As the present case
versus non-case ratio was not optimal for kappa
statistics, which may underestimate the agreement
on rare category, also %A, PA and NA were com-
puted. The %A was reported as it is the most obvi-
ous, straightforward and easiest-to-understand index
for agreement (10). The %A was demonstrated to be
high for all assessed diagnoses. As a disadvantage, %
A does not indicate how the agreements and dis-
agreements are distributed (10). This is why the PA
and NA were also computed. PA was high for the
pain-related TMD diagnoses and more variable for
disc-related diagnoses. As there were no cases of DD
with reduction with intermittent locking and DD
without reduction with limited opening, the denomi-
nator in PA was 0, and PA could not be computed.
This happens with rarely occurring diagnoses when
the RSE and the other examiner agree on the
absence of the diagnoses. As there was one discrete
finding for degenerative joint disease, PA was 0.
The reliability assessment of the new evidence-
based DC/TMD protocol is highly important in terms
of future research projects. Thus far, only two studies
considering the reliability of the new DC/TMD have
been published. Vilanova et al. (12). evaluated the dif-
ferences in diagnostic reliability of the DC/TMD
between groups of self-instructed examiners and
examiners taught in training and calibration course.

© 2017 John Wiley & Sons Ltd

 RELIABILITY OF FINNISH DC/TMD AXIS I
The results indicated that diagnostic reliability was
high after both self-instruction and course participa-
tion. Even though the self-instruction group improved
their reliability after participating in the course, self-
instruction was demonstrated to be sufficient in order
to diagnose the most common TMD diagnoses. Schiff-
man et al. (1). reported the reliability of pain-related
TMD to be excellent. Reliability coefficients were
above our estimates for myalgia and arthralgia, similar
to our results for myofascial pain with referral and
lower than our findings for DD with reduction and
DD without reduction without limited opening
(k = 0
84). Other intra-articular diagnoses could not
be computed accurately due to the low examiner
detection rates in the study sample.
The linguistically valid translations of the DC/TMD-
FIN Axis I assessment instruments were the first step
in the necessary cultural adaptation process. No
important discrepancies were found between the
source and the target document prior to this reliability
study. All the Finnish Axis I assessment instruments
have recently been published by the International
RDC/TMD Consortium on the website. The next step
in the cultural adaptation process will be the evalua-
tion of the validity of the Finnish versions of DC/TMD
diagnostics together with the Axis II instruments,
based on the currently ongoing field testing in TMD
pain patients.
In conclusion, the findings of this study showed
DC/TMD-FIN Axis I to demonstrate sufficiently high
reliability for pain-related TMD diagnoses. The DC/
TMD-FIN Axis I can be used to assess TMD in a stan-
dardised way in clinical and research settings.
Disclosure
Turku University Central Hospital approved the study
protocol. There is no conflict of interest and no source
of funding.
References
1. Schiffman E, Ohrbach R, Truelove E, Look J, Anderson G,
Goulet JP et al. Diagnostic criteria for temporomandibular
disorders (DC/TMD) for clinical and research applications:
recommendations of the international RDC/TMD consor-
tium network* and orofacial pain special interest group†. J
Oral Facial Pain Headache. 2014;28:6–27.
2. Dworkin SF, LeResche L. Research diagnostic criteria for
temporomandibular disorders: review criteria, examinations
© 2017 John Wiley & Sons Ltd
499
and specifications, critique. J Craniomandib Disord.
1992;6:301–355.
3. Sessle BJ. Editorial:Two new entities: newly developed
diagnostic criteria for temporomandibular disorders, and the
journal of oral & facial pain and headache. J Oral Facial
Pain Headache. 2014;28:5.
4. Ohrbach R, Bjorner J, Metric Q, Jezewski M, John M,
Lobbezoo F. Guidelines for Establishing Cultural equivalency
of Instruments: Updated May 11, 2013. www.rdc-tmdinte
rnational.org, Accessed on January 2, 2016.
5. Sipil€a K, Suvinen T. DC/TMD Symptom Questionnaire (DC/
TMD-FIN Oirekysely) and Pain-related interview and Exam-
iner Commands (DC/TMD-FIN Kliininen protokolla, osa 5).
In: Ohrbach R, ed. Diagnostic Criteria for Temporomandibu-
lar Disorders: Assessment Instruments. Version 15May2016.
[Diagnostiset Kriteerit Purentaelimist€ on Kivuille ja Toim-
intah€airi€
oille (DC/TMD-FIN): Tutkimusinstrumentit: Finnish
Version 25May2016] Sipil€a K, Suvinen T, Trans. www.rdc-
tmdinternational.org, Accessed on January 2, 2016.
6. Ohrbach R, Gonzalez Y, List T, Michelotti A, Schiffman E.
Diagnostic Criteria for Temporomandibular Disorders (DC/
TMD) Clinical Examination Protocol: Version 02June2013.
www.rdc-tmdinternational.org, Accessed on March 18, 2015.
7. Alstergren P, Gonzalez-Stucker Y, Castrillon E, Peck C, Gou-
let J-C, Koutris M. Guidelines for DC/TMD Training and
Calibration; www.rdc-tmdinternational.org, Accessed on
January 2, 2016.
8. Chen G, Faris P, Hemmelgarn B, Walker RL, Quan H. Mea-
suring agreement of administrative data with chart data
using prevalence unadjusted and adjusted kappa. BMC Med
Res Methodol. 2009;9:5.
9. Fleiss JL, Levin B, Paik MC. Statistical methods for rates and
proportions. Hoboken, NJ: Wiley-Interscience; 2003:598–626.
10. Feinstein AR. Principles of medical statistics. Boca Raton,
FL: Chapman & Hall/CRC; 2002:413–419.
11. John MT, Dworkin SF, Mancl LA. Reliability of clinical tem-
poromandibular disorder diagnoses. Pain. 2005;118:61–69.
12. Vilanova LS, Garcia RC, List T, Alstergren P. Diagnostic cri-
teria for temporomandibular disorders: self-instruction or
formal training and calibration? J Headache Pain.
2015;16:505.
13. Schiffman E, Ohrbach R, List T, Anderson G, Jensen R,
John MT et al. Diagnostic criteria for headache attributed to
temporomandibular disorders. Cephalalgia. 2012;32:683–
692.
14. Headache Classification Subcommittee of the International
Headache Society (IHS). The international classification of
headache disorders 3rd ed. Cephalalgia. 2013;33:629–808.
15. Okeson JP. Critical commentary 1: evaluation of the
research diagnostic criteria for temporomandibular disorders
for the recognition of an anterior disc displacement with
reduction. J Orofac Pain. 2009;23:312–315.
Correspondence: Kirsi Sipil€a, Institute of Dentistry, University of
Eastern Finland, Kuopio Campus, Box 1627, FIN-70211 Kuopio,
Finland.
E-mail: kirsi.sipila@uef.fi