Sensitivity, specificity and prevalences—comparison of four screening tests for temporomandibular disorders

2.2.1 Reference standard DC/TMD Axis I

For the purposes of this study, only Axis I, as measured by the DC/TMD Symptom Questionnaire and the DC/TMD Clinical Examination Protocol was considered. The DC/TMD procedure is described below based on the discussions of Schiffman and Ohrbach et al. [22].

In general, the right and left sides of the body were examined separately for corresponding findings.

The DC/TMD Symptom Questionnaire was utilized to document acute or past facial pain and discomfort experienced during chewing or opening the mouth. These symptoms were meticulously recorded in the medical history and assigned to one side, if feasible. Given the absence of a history of symptoms, it was not possible to localize any acute pain in the DC/TMD Examination Form.

Initially, the overjet, the overbite, and any midline shifts were measured with millimetric precision using a ruler. The mouth opening movement was described as straight, deviation or deflection.

The pain-free mouth opening, maximum active mouth opening, maximum passive mouth opening, laterotrusion right/left, and protrusion were then measured using a ruler. If the maximum assisted mouth opening had to be aborted, this was noted. Pain that occurred during active movements was differentiated according to its location (temporalis muscle/masseter muscle/joint/other masticatory muscles/non-chewing muscles) and quality (Pain/Familiar Pain). With regard to the temporalis muscle, pain could be differentiated into Pain, Familiar Pain, and Familiar Headache.

Active movements (mouth opening, mouth closing, lateral movements, and protrusion) were examined for joint noises (grinding and cracking). These sounds were classified according to whether they were perceived by the patient or the examiner. If cracking noises occurred together with pain, it was differentiated according to whether they were associated with Pain or Familiar Pain. If lockjaw occurred during mouth opening or wide opening position, this was noted, and a distinction was made as to whether the patient or the examiner was able to release it.

The temporalis muscle, with its associated temporalis tendon, the masseter muscle, the digastric venter anterior/posterior muscle, and the lateral pterygoid muscle, were each palpated in their entirety. The temporomandibular joints were palpated bilaterally and around the lateral condylar pole. The palpation pressure for the temporalis muscle, the masseter muscle, and for the region around the lateral condylar pole was 1 kg in each case. Palpation of the lateral pole of the condyle, the retro- and submandibular region, the temporalis tendon, and the lateral pterygoid muscle was performed with a pressure of 0.5 kg. Generally pain on palpation was categorized into: Pain, Familiar Pain, and Referent Pain, with the exception of palpation onto the temporal muscle with an added category of Familiar Headache [22].

2.2.2 CMD-Short Finding

The CMD-Short Finding procedure is briefly described based on the publications of Ahlers et al. [16, 17, 18] and the methods description partly reproduces their wording. A maximum of six non-side-specific parameters could be selected. A deviation or deflection greater than 2 mm was counted as a mouth-opening deviation. An active mouth opening of less than 38 mm was counted as a restriction of the mouth opening. The temporomandibular joints were examined for clicking or grinding noise when opening or closing the mouth. If the habitual occlusion made a dull or intermittent noise, this was counted as a positive finding. A non-physiological dynamic occlusion was present if occlusal abrasion, grinding facets, or balance/hyper-balance contacts that were not age-appropriate were present. Palpation of the masseter muscle pars superficialis, the temporalis muscle pars anterior, and the digastric muscle posterior belly were performed with a pressure of 0.5 kg [18]. In the CMD-Short Finding, the reliability of recognizing temporomandibular joint disorders was increased by summing the individual parameters [15, 16, 17, 18]. To maintain comparability across the TMD screening tests, parameters summation was not performed in this study. Further clinical functional analysis is recommended when at least two parameters are positive [16, 17, 18] (Fig. 2, Ref. [16, 17, 18]).

Fig. 2.Flow chart of the Craniomandibular Disorders Short Finding. Created by Maria Ertl based on the template in Ahlers et al.’s [16, 17, 18] publications. A checkmark is placed if a finding applies. All positive parameters are totaled. A clinical functional analysis is recommended for further diagnostics in cases with at least two positive parameters. No summation of parameters was performed in this study to ensure the comparability of the TMD screening tests. CMD: craniomandibular disorders.

2.2.3 CMD-Screening

The CMD-Screening procedure is briefly described based on a discussion by the German Society of Craniomandibular Function and Disorders and the methods description partly reproduces their wording [14]. Anamnestic questions and examination parameters were answered and noted on the findings form in a nonspecific manner. The survey asked about acute pain once or more per week in the temporomandibular joint in the temporal/facial/jaw area and about pain when opening the mouth or chewing. The questionnaire also asked about blockages or difficulties once or more per week when opening the mouth. Following the DC/TMD guidelines, the masticatory muscles were examined for pain, and the temporomandibular joints were assessed for both noise and pain. Palpation of the masseter muscle pars superficialis and the temporalis muscle pars anterior/media/posterior was performed with a pressure of 1 kg. An active mouth opening of less than 40 mm was counted as an RMO. Disorders of habitual occlusion were determined acoustically and visually with Shimstock foil. The temporomandibular joints were tested for clicking or grinding noise during mouth opening, mouth closing, protrusion and laterotrusion. The nine possible parameters were color-coded. In the medical history questions, pain in the temporomandibular joints and masticatory muscles and the RMO score were marked red, and occlusal disorders or temporomandibular joint noises were marked yellow [14]. This study did not use color coding to compare the TMD screening tests. Extended diagnostics are recommended for a yellow parameter and considered urgently indicated for a red one (Fig. 3, Ref. [14]).

Fig. 3.Flow chart of the Craniomandibular Disorders Screening. Created by Maria Ertl based on the template of the German Society of Craniomandibular Function and Disorders in the German Society of Dentistry and Oral Medicine (CMD-Screening) [14]. Extended diagnostics are recommended for a yellow parameter and considered urgently indicated for a red one [14]. This study did not use color coding to compare the TMD screening tests. CMD: craniomandibular disorders; DC/TMD: Diagnostic Criteria for Temporomandibular Disorders.

2.2.4 Preventive Manual Structural Analysis

The PMSA procedure is summarized according to the publications of Bumann et al. [19], with parts of the description adapted from the original wording. A side-specific examination of the patients was conducted. Pain, end feel, and muscle strength were assessed qualitatively. Active and passive mouth opening, lateral movements, protrusion, and retrusion were measured and analyzed for pain. The end feel was determined for pain-free passive mouth opening. If pain occurred, it was assigned to the masticatory muscle or temporomandibular joints by localization. Condylar mobility was described during mouth opening and protrusion. An active mouth opening of less than 43 mm was counted as an RMO. The temporomandibular joints were examined for crepitus and pain using dynamic compression and assessed for clicking noise using dynamic translation. The bilaminar zone was tested for pain using passive compression in the cranial, dorsal, dorsocranial, dorsolateral, and dorsolateral-cranial directions. The joint capsule was assessed for pain and mobility using ventral translation and caudal traction. Palpation of the masseter muscle pars superficialis/profunda and the temporalis muscle pars anterior/media/posterior was performed with a pressure of 1 kg, whereas for the digastric muscle anterior/posterior bellies it was only 0.5 kg. The isometric tests lasted 80 seconds each and were carried out for mouth opener muscles and mouth closer muscles. The suprahyoidal structures were tested for sagittal and vertical restrictions. Furthermore, static and dynamic occlusion, bruxism, and dysfunctions were analyzed concerning their influence on the existing findings. Finally, suspected diagnoses related to stress and restriction were summarized [19]. Physiological findings are highlighted in green, and compensated findings are highlighted in yellow [19]. The present study did not include end feel, occlusal vectors, pain quality, restrictions, condylar mobility, isometric tests, or the use of color coding because the objective was to compare TMD screening tests (Fig. 4, Ref. [19]).

Fig. 4.Flow chart of the Preventive Manual Structural Analysis. Created by Maria Ertl based on the template in the publication of Bumann et al. [19]. The red outline indicates parameters that can be assigned to the DC/TMD diagnoses. The other parameters were evaluated as additional information in this study, with no distinction being made between the quality of the pain and muscle strength and only whether the respective parameter was present.

2.2.5 Preventive Structural Stress Screening

The PSSS procedure is briefly described based on the publications of Bumann et al. [19, 20] and the methods description partly reproduces their wording. Up to six non-side-specific, color-coded parameters could be selected. In addition, neither the quality of pain nor the final Sensation was specified. The passive mouth opening was examined for pain in the masticatory muscles or temporomandibular joint. The active mouth opening was measured metrically and assessed for asymmetries, defined as deviation or deflection exceeding 2 mm. An active mouth opening of less than 43 mm was counted as an RMO. The temporomandibular joints were examined for crepitus using dynamic compression and for cracking noise using dynamic translation. Passive compression was performed in the dorsolateral and dorsolateral-cranial directions to assess the bilaminar zone. Palpation of the masseter muscle pars superficialis/profunda and the temporalis muscle pars anterior/media/posterior was performed with a pressure of 1 kg. The anterior and posterior bellies of the digastric muscle were palpated with a pressure of 0.5 kg. The dentition was examined for signs of attrition and for the presence of balance or hyperbalance contact. This study did not assign the parameters to a green, yellow, or red result field because they were examined individually. If two green parameters or one green and one yellow parameter are present, this results in a yellow outcome field. Two yellow parameters—or one yellow combined with one green—lead to a red outcome field. A yellow result suggests that further diagnostics are recommended, and if the result field is red, diagnostics are urgently indicated (Fig. 5, Ref. [19, 20]).

Fig. 5.Flow chart of the Preventive Structural Stress Screening. Created by Maria Ertl based on the template in the publications of Bumann et al. [19, 20]. Two green or the combination of a green and a yellow parameter can lead to a yellow result field. The presence of two yellow parameters or one yellow and one green parameter results in a red result field. If the result field is yellow, extended diagnostics are recommended, and if the result field is red, diagnostics are urgently indicated [19, 20]. This study did not assign the parameters to a green, yellow, or red result field because they were examined individually. DL: dorsolateral; DLC: dorsolateral-cranial.

2.2.6 Comparability of the screening tests

Five of the twelve DC/TMD diagnoses [22] were used in this study: myogenic pain, arthralgia, disc displacement (DD) with reduction, DD without reduction with limited opening (DD w/o R), and degenerative joint disease (DJD). Due to the absence of imaging and reported symptoms in the participants’ medical history, several modifications were made: (local) myalgia, myofascial pain with or without reduction, and TMD headache were combined under the term myogenic pain. For DD with reduction, intermittent jaw clenching was not included. In addition, DD w/o R without RMO was not included.

Since the screening tests vary significantly in their parameters, scale levels, and evaluation methods. The screening test parameters were assigned to each of the five DC/TMD diagnoses to compare the four TMD screening tests. A positive screening parameter was interpreted as evidence for the corresponding diagnosis (Fig. 6).

Fig. 6.Flow chart of the specific parameters of the four screening tests and for the five diagnoses. Created by Maria Ertl. Positive parameters were taken to indicate the presence of the diagnosis in question. PMSA: Preventive Manual Structural Analysis; PSSS: Preventive Structural Stress Screening; CMD: craniomandibular disorders; D: dorsal; DC: dorsocranial; DL: dorsolateral; DLC: dorsolateral-cranial; C: cranial; DD: disc displacement with reduction; DD w/o R: disc displacement without reduction with limited opening; DJD: degenerative joint disease.

2.2.7 Adapted DC/TMD diagnostic algorithm

DC/TMD diagnostic algorithm was modified for arthralgia, myogenic pain, and DJD regarding the required positive history [22], as the patients in this study had no history of symptoms.

In routine clinical practice, a significant proportion of patients exhibit no symptoms but demonstrate compensated findings during screening procedures. In order to ensure a realistic reflection of the condition, the study exclusively included patients who had not exhibited any symptoms. As the DC/TMD diagnostic algorithms for arthralgia, myogenic pain, and degenerative joint disease require symptoms, these were adjusted accordingly. For better differentiation of these three DC/TMD diagnoses, results were presented using modified and unmodified DC/TMD diagnostic algorithms. The results of unmodified diagnostic algorithms were described as decompensated diagnoses (e.g., decompensated arthralgia). The results of modified diagnostic algorithms were called compensated diagnoses (e.g., compensated arthralgia).

The flowchart of the five DC/TMD diagnoses, partially adapted for this study based on the discussion by Schiffman and Ohrbach et al. [22] is briefly described below:

Using the DC/TMD diagnostic algorithm, decompensated arthralgia could only be diagnosed if the patient had a positive pain history which was confirmed by the examiner [22]. Furthermore, only Familiar Pain during mouth opening, excursive movements, or palpation of the temporomandibular joint were considered [22]. In contrast, for compensated arthralgia findings, neither a pain history nor confirmation during examination was required; instead, unfamiliar (unknown) pain in the temporomandibular joint was included (Fig. 7, Ref. [22]).

Fig. 7.Flow chart of the DC/TMD diagnostic algorithm for arthralgia, myogenic pain, disc displacement with reduction, disc displacement without reduction with limited opening, and degenerative joint disease. Created by Maria Ertl based on the template of the German DC/TMD protocol of the International RDC/TMD Consortium Network [22] references to the DC/TMD Examination Form [E] and the DC/TMD Symptom Questionnaire [SQ] are in the square brackets. TMD: temporomandibular disorders; TMJ: temporomandibular joint; CT: computer tomography; MRI: magnetic resonance imaging.

Using the DC/TMD diagnostic algorithm, decompensated muscle findings could only be diagnosed if the patient had a positive pain history which was confirmed by the examiner [22]. Furthermore, only Familiar Pain during mouth opening or palpating the masseter/temporalis muscles were considered [22]. In contrast, for compensated muscle findings, both the patient’s pain history and its confirmation were disregarded. Any unfamiliar pain in the masticatory muscles (temporalis muscle, masseter muscle) was counted regardless of its distribution and pain transmission during mouth opening and palpation. Additionally, muscle findings were evaluated both with and without including pain on palpation of the digastric muscle. In the DC/TMD examination procedure, the retro- and submandibular structures, including the digastric muscle, are palpated. However, they are not included in the diagnostic algorithm for muscle pain. Reporting results both with and without the digastric muscle prevents loss of information and helps avoid misinterpretation. For example, it ensures that the screening tests are not misinterpreted as being inadequate for recording the mouth closers. For this reason, the results for myogenic pain are presented both ways (submandibular region, posterior mandibular region) (Fig. 7).

According to the diagnostic algorithm, DJD could be diagnosed in the absence of current temporomandibular joint (TMJ) noises by history, but examiners had to confirm the presence of crepitus [22]. For compensated DJD, the grinding noise perceived by the subjects was counted, regardless of their medical confirmation (Fig. 7).

No distinction was made between compensated and decompensated for DD and DD w/o R. A cracking noise was classified as DD if no previous cases of lockjaw were reported in the DC/TMD Symptom Questionnaire and if the noise was also perceived by the patient during the examination. Moreover, the examiner was required to DC/TMD Decision Tree identify any clicking noises observed during mouth opening and closing, mouth opening and exclusive movement, or mouth closing and exclusive movement. Both DD with reduction with intermitted locking and DD without reduction without RMO were not taken into consideration (Fig. 7).

In the event that a previous case of lockjaw was confirmed in the DC/TMD Symptom Questionnaire and a maximum assisted mouth opening of 40 mm or less was observed during the examination, the resulting DC/TMD diagnosis was DD w/o R. When taking the second part of the DC/TMD (if “No” at 9 and 10 in Symptom Questionnaire), no DD w/o R could be diagnosed. Questions 11 and 12 of the Symptom Questionnaire addressed current jaw lock episodes. A “Yes” response to either question led to exclusion from the study (Fig. 7).

The examination quality was ensured via eight years of professional experience and by completing the TMD curriculum [28] to learn the manual examination techniques of the PMSA and PSSS. The examination techniques of the CMD-Short Finding [16, 17, 18], CMD-Screening [14], and DC/TMD [21, 22] were learned by studying the literature and using the instruction video from the University of Leipzig [23]. Rauch et al. [23] demonstrated that DC/TMD can be mastered and applied without errors during self-study using this instructional video. The German assessment instruments were used to calibrate the examiner for performing the DC/TMD [22].

2.2.8 Statistical analysis

The target sample size of 192 to 216 subjects was confirmed by the Institute of Biometry and Clinical Epidemiology at Campus Charité Mitte. The number was determined using the function power.diagnostic.test from the R package MKmisc, assuming that ten examinations/day, two days per month, could be performed within a year as a part of the dissertation [27]. The power was set at 80%.

The prospective design ensured complete data collection; therefore, missing data were not an issue. The statistical analysis for age distribution was performed using the Shapiro and MannWhitney U tests. The scale levels of the parameters were nominal. Prevalence values of the five DC/TMD diagnoses were given for each screening. Differences were analyzed using the Cochran Q test. If a significant difference was found, pairwise comparisons were performed using the McNemar test. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated using a four-field table.

The PPV and NPV were each assigned with study prevalence and with age-specific weighted [29] reference prevalence [1, 30, 31, 32, 33]. This study demonstrated, among other findings, that binary test outcomes are strongly influenced by the underlying prevalence.

The 95% confidence intervals (CI) and value verifications were calculated using the MedCalc Software Ltd diagnostic test evaluation calculator v.20.112 (https://www.medcalc.org/calc/diagnostic_test.php). In the literature, a 70% sensitivity and a 95% specificity are acceptable values for TMD tests [34]; therefore, this study followed these guidelines. The described values were reported for each DC/TMD diagnosis.

Finally, statistically significant differences between DC/TMD and the respective TMD screening test were analyzed using McNemar tests. In addition, a side-specific comparison was made between the PMSA and DC/TMD, considering the right/left correlation using a general estimating equation and odds ratio (OR). The significance level was two-tailed and set at α = 0.05.

3.2.1 Decompensated findings

When using the unadapted DC/TMD diagnostic algorithm, zero values occurred for arthralgia, muscle, and degenerative joint findings. Sensitivity, PPV, and p-value could not be calculated for these diagnoses. In each case, the observed study prevalence was 0.00%, falling below the corresponding prevalences. The NPV of the four tests decreased significantly for these three diagnoses with a change to the reference prevalence. For arthralgia, the specificity values of the individual screening tests were high and the NPV very high (Table 3, Ref. [29, 31, 32]). For decompensated myogenic pain, the specificity values were insufficient for all four screenings and the NPV with study prevalence was very high (Table 4, Ref. [1, 29, 30, 31]). For the degenerative joint findings, the specificity values of all screening tests were insufficiently high (Table 5, Ref. [29, 31, 33]). For these three DC/TMD diagnoses, positive findings were rather unlikely to be detected with DC/TMD compared to PMSA (OR_DC/TMD <0.001, OR_PMSA = 1.000).

3.2.2 Compensated arthralgia findings using the adapted DC/TMD diagnostic algorithm

The PMSA, PSSS, and CMD-Screening fulfilled the requirement of at least a 70% sensitivity [34]. All screenings achieved a specificity of 95% [34]. The CMD-Short Finding showed a sensitivity of 0.0% [0.00; 97.50], and a p-value could not be calculated. The PMSA (p = 0.125), PSSS (p = 0.125), and CMD-Screening (p = 0.500) did not differ significantly from the DC/TMD. Considering the right/left correlation, the PMSA also did not differ significantly from the DC/TMD (p = 0.063, OR_PMSA = 1.000, OR_DC/TMD = 0.860) (Table 3).

The study prevalence (12.50%) for compensated arthralgia findings was 2.02 times higher than the reference prevalence (6.18%) for decompensated arthralgia findings. The PPVs of the PMSA and the PSSS decreased compared to the reference prevalence. The NPVs of the four screenings increased slightly with the reference prevalence. The different prevalences did not influence the PPV of the CMD-Screening.

3.2.3 Compensated myogenic pain findings without the digastric muscle using the adapted DC/TMD diagnostic algorithm

Unlike other tests, the CMD-Screening achieved a 95% specificity [34]. All four screening tests exceeded the required sensitivity of 70% [34]. The PMSA (p < 0.001), PSSS (p < 0.001), and CMD-Screening (p < 0.001) differed significantly from DC/TMD. The CMD-Short Finding demonstrated no significant difference from DC/TMD (p = 0.064). The PMSA also differed significantly from the DC/TMD (p = 0.003, OR_PMSA = 1.000, OR_DC/TMD = 0.808), considering the right/left correlation (Table 4).

The study prevalence with compensated muscle findings (41.0%) was 3.33 times higher than the reference prevalence with decompensated muscle findings (12.53%). The PPVs of the PMSA, PSSS, and CMD-Short Finding were lower with the reference prevalence than with study prevalence. The opposite was true for the NPV of the four tests. The PPV of the CMD-Screening remained unchanged with the change in prevalence.

3.2.4 Compensated myogenic pain findings with the digastric muscle using the adapted DC/TMD diagnostic algorithm

Except for the CMD-Screening, all screening tests achieved the required 70% sensitivity [34], and all four screening tests reached the required 95% specificity [34]. The CMD-Screening (p < 0.001) and CMD-Short Finding (p = 0.004) differed significantly from the DC/TMD. The PMSA (p = 0.500) and PSSS (p = 0.500) displayed no significant difference from the DC/TMD. There was also no significant difference between PMSA and DC/TMD (p = 0.055, OR_PMSA = 1.000; OR_DC/TMD = 1.064), considering the right/left correlation (Table 4).

The study prevalence (50.5%) was 4.10 times higher than the reference prevalence (12.53%) and higher than the study prevalence excluding the digastric muscle (41.0%). The lower the NPVs of the four screening tests, the more it increased with lower reference prevalence. The PPVs of the four tests did not change when the prevalences changed.

3.2.5 Compensated DJD findings using the adapted DC/TMD diagnostic algorithm

Except for the CMD-Short Finding, all screenings achieved the required 70% sensitivity and the 95% specificity [34]. The PMSA, PSSS, and CMD-Screening did not differ significantly from the DC/TMD (p = 1.000 each). The sensitivity of the CMD-Short Finding was 0.0% [0.00; 97.50], and no p-value could be determined. Considering the right/left correlation, no p-value could be calculated, the OR for comparing the PMSA with the DC/TMD could be calculated (OR_PMSA = 1.000, OR_DC/TMD = 1.000) (Table 5).

The study prevalence (0.50%) was significantly lower than the reference prevalence (27.94%). The NPV of the CMD-Short Finding was lower with reference prevalence than with study prevalence. The NPV and PPV of the other tests did not change with the change in prevalence.

3.2.6 DD w/o R using the unadapted DC/TMD diagnostic algorithm

All four screening tests exceeded the required 70% sensitivity and the 95% specificity [34] and did not differ significantly from the DC/TMD (p = 1.000 each). The side-specific comparison between the PMSA and DC/TMD was omitted (Table 6, Ref. [29, 31, 33]).

The study prevalence (0.50%) was lower than the reference prevalence (9.82%). The PPVs of the PMSA and the PSSS were higher with reference prevalence than with study prevalence. The PPVs of the CMD-Screening and the CMD-Short Finding, as well as the NPVs of all four tests, remained unchanged when the prevalences were changed.

3.2.7 DD using the unadapted DC/TMD diagnostic algorithm

All four screening tests exceeded the required 70% sensitivity [34]. The CMD-Short Finding was the only test to achieve 95% specificity [34]. All four screening tests differed significantly from the DC/TMD (PMSA: p < 0.001, PSSS: p < 0.001, CMD-Screening: p < 0.001, and CMD-Short Finding: p = 0.031). The PMSA also differed significantly from the DC/TMD in the side-specific comparison (p < 0.001, OR_PMSA = 1.000, OR_DC/TMD = 0.436) (Table 7, Ref. [29, 31, 33]).

The study prevalence of DD (11.50%) was 2.10 times lower than the reference prevalence (24.18%). The PPVs of the four screening tests were higher with reference prevalence than with study prevalence. The NPVs of the four tests remained unchanged when the prevalences were changed.

4.1.1 Compensated arthralgia

The CMD-Short Finding does not include a specific parameter that is capable of detecting arthralgia findings. Its prevalence (0.0%) differed significantly from that of the other tests (p < 0.001). Due to the zero values, no sensitivity, PPV, or p-value (comparison with reference standard) could be specified. The present study found that PMSA and PSSS (with 15.0% each) detected a greater number of compensated findings by passive compression than DC/TMD (12.5%) and CMD-Screening (11.5%) with palpation. This resulted in higher false-positive rates and lower PPVs (at study prevalence: 80.00% [64.47–89.81], at reference prevalence: 64.84% [45.55–80.25]). The broad CI of the sensitivity (96.00% [79.65–99.90]) could be reduced by an increased number of cases.

PMSA (p = 0.125, side-specific p = 0.063), PSSS (p = 0.125), and CMD-Screening (p = 0.500) did not differ significantly from the reference standard and hypothesis B was rejected for these screening tools. They also did not differ from each other (PMSA vs. PSSS: p = 1.000; CMD-Screening vs. PMSA p = 0.528, CMD-Screening vs. PSSS p = 0.528).

Hypotheses A and C were retained exclusively for the CMD-Short Finding, as it was the only one that didn’t meet the minimum diagnostic requirements and differed significantly from the other screening tools in terms of prevalence.

4.1.2 Compensated myogenic pain

For compensated myogenic pain, when using the DC/TMD without including the digastric muscle, the CMD-Screening represented the sole test that attained the required 95% specificity (100.00% [96.92–100.00]) [34]. This is attributable to the fact that the test exclusively examined the masseter muscle pars superficialis and temporalis muscle pars anterior/media/posterior. The prevalence of CMD-Screening was significantly found to be the lowest of the four screening tools (p < 0.001). The PMSA and PSSS exhibited the lowest specificity levels and differed significantly from the reference standard (each 83.90% [76.00–90.02], p < 0.001, PMSA side-specific p = 0.003). This is attributable to the fact that both tests palpate the digastricus venter anterior and posterior, in addition to the masseter and temporalis muscles and the examination of passive mouth opening. The CMD-Short Finding also demonstrated insufficient specificity yet exhibited no significant discrepancy from DC/TMD (85.59% [77.94–91.38], p = 0.064), as the digastricus pars posterior was the sole muscle examined, in conjunction with the masseter pars superficialis and temporalis pars anterior.

Hypotheses A, B and C were retained exclusively for the CMD-Screening, as it was the only one that met the minimum diagnostic requirements, showed no significant difference to the reference standard and differed significantly from the other screening tools in terms of prevalence.

When the reference standard included the digastric muscle anterior/posterior bellies, the opposite was true. The CMD-Screening was the only test that did not achieve the required 70% sensitivity [34] and, like the CMD-Short Finding (p = 0.004), differed significantly from the DC/TMD (p < 0.001). The PMSA and PSSS did not differ significantly from the reference standard (p = 0.500, PMSA side-specific p = 0.055). The different palpation combinations used in the tests led to varying prevalence rates and explain this observation.

Hypotheses A and C were retained exclusively for the CMD-Screening, as it was the only one that met not the minimum diagnostic requirements and differed significantly from the other screening tools in terms of prevalence. Hypothesis B were retained only for CMD-Short Finding and CMD-Screening.

4.1.3 Disc displacement

Concerning DD, the DC/TMD diagnostic algorithm required cracking noise to occur either during the opening and closing of the mouth or in combination with excursive movements [22]. In the CMD-Screening, the clicking noise was counted during active movements regardless of the combination [14, 22]. In the PMSA and the PSSS, the clicking noise was detected using dynamic translation [19, 20]. The CMD-Short Finding analyzed only cracking noise during mouth opening or closing [16, 17] and, therefore, had a higher specificity (96.61% [92.77–98.75]) and a significantly lower prevalence (14.5%, p < 0.001) than the other tests.

Hypotheses A and C were retained exclusively for the CMD-Short Finding, as it was the only one that met the minimum diagnostic requirements and differed significantly from the other screening tools in terms of prevalence. In conclusion, clicking noise is more common in asymptomatic individuals during lateral and protrusive movements than in mouth opening or closing.

All screening tools differed significantly from the DC/TMD (PMSA p < 0.001, PMSA side-specific p < 0.001, PSSS p < 0.001, CMD-Screening p < 0.001, CMD-Short Finding p = 0.031). Hypothesis B were retained for all screening tools.

4.1.4 Compensated degenerative joint disease

The CMD-Short Finding (0.0%) detected no compensated degenerative joint findings, unlike the other tests. Consequently, sensitivity, PPV, and p-value could not be calculated. This test assessed joint sounds only during mouth opening/closing, not during excursive movements like the other screening tools and DC/TMD. In one case, crepitus occurred during protrusion and was detected by both the reference standard (0.5%) and the other TMD screening tests (0.5%). Due to the extremely low prevalence, all tests indicated a wide CI for sensitivity and the results should be interpreted with caution. All screening tests, except for the CMD-Short Finding, met the required thresholds of at least a 70% sensitivity and a 95% specificity and had no significant difference from DC/TMD (each p = 1.000). Therefore, hypotheses A and B were rejected for all tests except the CMD-Short Finding.

The screening tools did not differ significantly in prevalence (p = 0.392), so hypothesis C was also rejected (for all screening tools).

4.1.5 Disc displacement without reduction with limited opening

In DD w/o R, no significant differences were detectable between the screenings (p = 0.392) and the DC/TMD (p = 1.000) despite the different thresholds for RMO (PMSA and PSSS <43 mm, DC/TMD and CMD-Screening <40 mm, CMD-Short Finding <38 mm). The PMSA and PSSS exhibited a false positive value, resulting in a lower PPV (each 50.00% [12.40–87.60] at study prevalence; each 95.59% [75.42–99.35] at reference prevalence). All tests showed wide CIs for sensitivity due to the low study prevalence, and the results should be interpreted with caution.

Hypotheses A, B and C were rejected for all tests, as they all met the minimum diagnostic requirements, showed no significant differences from the reference standard, and did not differ significantly in prevalence.

The low prevalence of compensated DJD and DD w/o R reveals that such findings are rarely unnoticed and play a subordinate role in patients with no history of symptoms.