Management of painful temporomandibular disorders

Abstract

Background.

Patients often seek consultation with dentists for temporomandibular disorders (TMDs). The objectives of this article were to describe the methods of a large prospective cohort study of painful TMD management, practitioners’ and patients’ characteristics, and practitioners’ initial treatment recommendations conducted by The National Dental Practice-Based Research Network (the “network”).

Methods.

Participating dentists recruited into this study treated patients seeking treatment for painful TMDs. The authors developed self-report instruments based on well-accepted instruments. The authors collected demographics, biopsychosocial characteristics, TMD symptoms, diagnoses, treatments, treatment adherence, and painful TMDs and jaw function outcomes through 6 months.

Results.

Participating dentists were predominately White (76.8%) and male (62.2%), had a mean age of 52 years, and were general practitioners (73.5%) with 23.8% having completed an orofacial pain residency. Of the 1,901 patients with painful TMDs recruited, the predominant demographics were White (84.3%) and female (83.3%). Patients’ mean age was 44 years, 88.8% self-reported good to excellent health, and 85.9% had education beyond high school. Eighty-two percent had pain or stiffness of the jaw on awakening, and 40.3% had low-intensity pain. The most frequent diagnoses were myalgia (72.4%) and headache attributed to TMDs (51.0%). Self-care instruction (89.4%), intraoral appliances (75.4%), and medications (57.6%) were recommended frequently.

Conclusions.

The characteristics of this TMD cohort include those typical of US patients with painful TMDs. Network practitioners typically managed TMDs using conservative treatments.

Practical Implications.

This study provides credible data regarding painful TMDs and TMD management provided by network practitioners across the United States. Knowledge acquired of treatment recommendations and patient reports may support future research and improve dental school curricula.

Temporomandibular disorders (TMDs) as a group are the second most common musculoskeletal disorder (after chronic back pain) resulting in pain and disability.¹ Patients often seek consultation with dentists for TMDs.^2,3 However, few studies have reported from varied community dental practices what treatments are provided or the impact associated with these treatments.

In 2013, 5 of the authors of this article (A.M.V., E.S., B.R., G.H.G., J.F.) worked with the National Institute of Dental and Craniofacial Research Collaboration on Networked Dental and Oral Research to identify the initial treatments being implemented by practitioners for patients with painful TMDs.⁴ A total of 80.3% (525/654) of the dental practitioners reported treating patients with TMDs, with an average of 3 patients per month per dentist. The survey did not query practitioners’ reports of treatment outcomes, adverse effects, or patient adherence to treatment recommendations. To our knowledge, there are no studies in the TMD literature that have assessed comprehensively all these issues.

Therefore, the National Institute of Dental and Craniofacial Research—funded The National Dental Practice-Based Research Network (the “network”)⁵ supported our prospective cohort study, with the primary aim to identify the factors that contribute to dental practitioners’ treatment decisions for patients with painful TMDs. The 2 secondary aims were to identify factors that contribute to patients’ adherence to treatment and describe observed changes from baseline at 1-, 3-, and 6-month follow-ups in pain intensity and jaw function associated with treatments. The results for these secondary aims will be presented in future articles.

The specific objectives of this article were to

• present an overview of the methods for the entire study

• describe sociodemographic characteristics of the study practitioners and the patients with painful TMDs

• describe the biopsychosocial characteristics and TMD diagnoses of the patients

• describe the initial treatment recommendations.

METHODS

Overview

This was a 6-month, prospective cohort study of dental practitioners who treat patients with painful TMDs in the network, which is composed of 6 regions in the United States. An overview of the methods is provided in Figure 1. After explaining the nature of the study procedures, the practitioners obtained informed consent from all patients who were recruited as participants. Activities for this investigation were approved by the institutional review boards that govern each of the network’s regions.

Figure 1.

Participant flow diagram and study overview.

Study participants and recruitment

Practitioner Enrollment

All network practitioners who reported treating patients with painful TMDs were invited to participate. Additional recruitment occurred at national dental meetings and dental study clubs and via advertisements. Practitioner eligibility criteria are provided in Table 1.

Table 1.

Eligibility criteria for practitioner and patient.

ELIGIBILITY CRITERIA TYPE	CRITERIA
Practitioner Inclusion Criteria	Currently practicing licensed dentist Enrollment in The National Dental Practice-Based Research Network Completion of study training Provision of informed consent according to regionally approved procedures Treatment of TMD* pain patients during the past year Has email to receive notification of online surveys
Patient Inclusion Criteria	Report of jaw or temple pain occurring in the last month Diagnosis of painful TMD by practitioner Age ≥ 18 y, except in Nebraska where age of consent is ≥ 19 y Seeking TMD treatment and accepts proposed treatment Provides informed consent according to regionally approved procedures Willing to comply with study procedures and available for 6 months of follow-up Able to receive emails and telephone calls and able to access online surveys Willing to be contacted as needed by the practice, regional coordinator, and the HealthPartners Survey Research Center Willing to provide contact information for another person who would know the patient’s whereabouts if the patient could not be reached
Patient Exclusion Criteria	Presence of painful dental pathology Inability to understand study procedures or provide informed consent in English or Spanish

^*TMD: Temporomandibular disorder.

The goal was to recruit 200 practitioners, with at least 180 practitioners completing the study. Practitioner recruitment and training were implemented from May 2016 through May 2018. Practitioners were provided at no cost a $300 tablet computer for data collection. In addition, practitioners were compensated $50 for obtaining informed consent from and completing the baseline questionnaire for each patient beginning with their seventh patient. Practitioners received $20 for each 6-month follow-up questionnaire they completed.

Patient Enrollment

Patients were screened for this study when they sought treatment from a participating practitioner for painful TMDs. Patient eligibility criteria are provided in Table 1. Each practitioner’s target recruitment was 11 consecutive consenting eligible patients, with a maximum of 20. Figure 1 summarizes the process for patient screening, enrollment, and baseline assessments. The planned enrollment was for 1,980 patients with at least an 80% retention rate, allowing for 1,584 patients at study closure. Patient recruitment started in October 2016 and was completed in July 2018. Patients were mailed gift cards for $25 on completion of questionnaires at baseline and 1-month and 3-month follow-ups and $50 for completion of the 6-month follow-up questionnaire.

Data collection procedures

We developed the questionnaires on the basis of well-accepted instruments to obtain data from the practitioners and patients. Questionnaire development focused on minimizing participant burden in a practice setting, with a completion time of 10 minutes or less. Before the formal study, an iterative process of pilot testing and refining allowed for prioritization of content as well as ascertainment of question clarity, comprehension, and respondent burden. Table 2 shows the study aims linked to data collections for outcomes, predictors, putative confounding variables, and the questionnaires used.⁶ The forms can be found on the network’s website at: https://www.nationaldentalpbrn.org/study-results/#1589312894305-232c429d-1724.

Table 2.

Study outcomes, study predictors of outcome, and putative confounding variables.^*

FACTORS ASSESSED	PRIMARY AIM: TREATMENT SELECTION	SECONDARY AIM 1: TREATMENT ADHERENCE	SECONDARY AIM 2: TMD† PAIN AND FUNCTION OUTCOMES
Study Outcomes	BL: initial treatment plan recommendations: self-care‡ medications‡ intraoral appliance therapy‡ referred treatments (allied care)‡	M1, M3, M6: patient-reported treatment adherence§	M1, M3, M6: patient-reported outcomes: jaw or temple pain intensity§ pain disability§ jaw functional limitation§ headache§ overall change in jaw or temple pain§ overall change in jaw function§
	M6: change in treatment plan at follow-up: revised treatments‡	M6: practitioner-reported treatment adherence‡	M6: dentist-reported outcomes change in jaw or temple pain‡
Study Predictors at Baseline	patient complaints§ aggravating and alleviating factors‡ oral examination data‡ TMD diagnoses‡ practitioner’s expectation of treatment efficacy‡ practitioner’s experience with painful TMD treatment‡ practitioner’s expectation of treatment adherence‡ practitioner’s anticipation of difficulties with treatment‡	Practitioner’s estimate of acceptability of initial treatment plan: patient’s satisfaction‡ ease in compliance‡ level of understanding‡	Follow-up outcomes scores adjusted for baseline Characteristic Pain Intensity score§ baseline pain disability§ baseline Jaw Function Limitation Scale score§ baseline headache§ Patient’s anticipation of improvement at baseline: jaw or temple pain§ jaw function§ Patient’s view of treatment plan at baseline: satisfaction with treatment plan‡ ease in following treatment plan‡ clarity of treatment plan explanation‡
Treatment-Related Predictors at Follow-up: M1-1 m; M3-3 m; M6-6 m	M6: practitioner’s view of treatment: difficulty in treatment implementation‡ inadequate pain management‡ patient’s treatment satisfaction‡ treatment side effects‡	M1, M3, M6: patient’s report on management of adverse effects§	M1, M3, M6: patient’s view of treatment: satisfaction§ ease in implementation§ treatment plan understanding§ cost of treatment§
Putative Confounding Variables	BL: baseline TMD pain chronicity§ BL: oral parafunctional habits (Oral Behavioral Checklist)§ BL: psychological distress (Patient Health Questionnaire-4)§ BL: widespread pain (areas of pain)§	BL: sleep quality (Grade Chronic Pain Scale questions 1-11)§ BL: general health (12-item Short Form Health Survey question 1)§ BL: number of professionals seen for TMD before the study§ M1: patient self-efficacy assessed§

^*Time points were baseline (BL), 1-month follow-up (M1), 3-month follow-up (M3), and 6-month follow-up (M6). Forms can be found on The National Practice-Based Research Network’s website.⁶

^†Temporomandibular disorders.

^‡Practitioner-provided data from dentist questionnaires.

^§Patient-provided data from patient questionnaires.

Practitioner assessment

Practitioner Baseline Assessment

The practitioners’ baseline questionnaire documented each patient’s symptoms, aggravating and alleviating factors, examination findings, diagnostic tests, TMD diagnoses, and treatment recommendations. Practitioners also indicated their expectations for treatment to relieve jaw pain and improve jaw function, whether the patient understood and was satisfied with treatment recommendations, how easy the treatment would be to complete, and any anticipated difficulties. The practitioners completed this baseline questionnaire after they explained the diagnosis or diagnoses and treatment recommendations to each patient.

Practitioner Follow-up Assessment at 6 Months

The practitioner questionnaire was used to report updated symptoms and diagnosis or diagnoses at 6 months. In addition, it was used to report any new or altered treatment provided with the reasons for the change.

Patient assessment

Patient Baseline Assessment

At the initial visit, patients completed questionnaires providing their contact information and demographic information. Patients completed their baseline questionnaires after the practitioners presented their TMD-related diagnosis or diagnoses and recommendation for treatment to the patients. If patients could not complete this questionnaire while at the dental office, they were asked to complete it within 2 weeks of their baseline visit, either online or via a telephone interview with the HealthPartners Data Coordinating Center.

The baseline questionnaire included 6 questions from the TMD screener⁷ to corroborate the practitioner-rendered painful TMD diagnosis. It also included components of the Diagnostic Criteria for Temporomandibular Disorders Axis II protocol³: Graded Chronic Pain Scale (GCPS)⁸ (pain grades I, IIa, IIb, III, IV^9,10), Jaw Function Limitation Scale (JFLS-8),¹¹ modified Oral Behavioral Checklist (OBC) assessing 15 behaviors,¹² and Patient Health Questionnaire-4.¹³ We used these instruments to measure pain intensity, pain-related disability, jaw functional limitation, frequency of oral parafunctional habits, and the presence and degree of psychological distress related to anxiety or depression. We calculated Characteristic Pain Intensity (CPI), from the GCPS, as the mean of 0 through 10 ratings for current, worst, and average pain during the prior month, multiplied by 10. Another GCPS measure, the Disability Score, was the mean of 3 0 through 10 interference ratings: daily activities, work, and social or family activities multiplied by 10. Additional biopsychosocial assessment included the first question of the Short Form Health Survey¹⁴ and question 11 from the Pittsburgh Sleep Quality Index Questionnaire.¹⁵ The modified Widespread Pain Index allowed patients to report any pain in 19 body areas in the past 7 days¹⁶; the only modification was that shoulder girdle was called shoulder. In addition, there was 1 question regarding widespread pain chronicity in these areas (that is, painful for at least 3 months). Finally, the baseline questionnaire asked patients about the number of previous providers seen for their pain as well as their understanding of and satisfaction with the current proposed treatment plan.

Patient Follow-up Assessments at 1, 3, and 6 Months

We sent patients an email to complete follow-up questionnaires at 1, 3, and 6 months online or via telephone contact. We emailed, called, or texted nonresponders to encourage them to complete the surveys. These questionnaires assessed change from baseline regarding painful TMD and jaw function, patient perspective of treatment, and patient adherence to treatment. Patient self-efficacy for symptom self-management was queried only at 1 month.

When patients dropped out or were withdrawn from the study, we recorded only the date and reason for withdrawal for reporting purposes. Patients who dropped out continued to receive the usual oral health care without prejudice from their participating practitioner. Recruitment of another patient was allowed for replacement of patients who left the study.

Sample size and power analysis

We considered study outcome data to be correlated within practices or clusters. Thus, statistical power estimation required assessment of the effective sample size (ESS). ESS is the total sample size divided by the variance inflation factor that is based on the number of observations per cluster and the intraclass correlation coefficient (ICC) between observations within each cluster. Because there were few data in the TMD literature for expected treatment frequencies and no data for within-cluster ICCs, we performed ESS estimation using a range of outcome frequencies (5%-50%) and ICCs (0.00-0.05). Within these ranges, and with α set at 5%, our power analysis showed that a projected sample of 1,584 patients would provide 90% power for detecting conservative odds ratios (ORs) ranging from 1.23 (based on an ESS of 1,980) through 1.78 (based on an ESS of 1,343) as measures of association between a categorical outcome and a categorical predictor. For continuous outcomes, we projected the sample size of 1,584 to have more than 80% power to detect mean treatment-group differences of 20 points for pain intensity (CPI, 0-100) and mean differences ranging from 1 through 4 for jaw function (JFLS-8, 0-10), even in the presence of large standard deviations (SDs) for CPI (40) and JFLS-8 (5).

We implemented the calculations using the Power and Sample Size Calculation software (Vanderbilt University Department of Biostatistics) to estimate the detectable alternative effects (OR and mean difference measures).

Statistical analysis

We performed descriptive statistical analyses to describe the practitioners and the cohort with painful TMD. We used frequency and percentage to describe the binary and categorical variables. We used means, SD, standard error (SE), median, and range for the continuous variables. Also, for this study, we used a linear mixed-model accounting for within-cluster correlation among observations to estimate adjusted CPI and JFLS means at baseline. We performed statistical analysis using SAS Version 9.4 (SAS Institute).

RESULTS

Of the original 201 practitioners who agreed to participate in the study, 185 actively recruited at least 1 study patient with painful TMDs. Practitioner characteristics are provided in Table 3, including the number by network region. The practitioners were predominantly White (76.8%), non-Hispanic (94.6%), and male (62.2%) with an average age of 52 years. Regarding years of experience, 95.6% of the practitioners graduated in 2011 or earlier, 136 (73.5%) were general practitioners, and 44 (23.8%) completed a residency or training programs in orofacial pain (Table 3). One-hundred sixty-eight (90.8%) had treated TMDs for 2 through 45 years, with a mean of 17.9 years. More than one-half of these dentists (86, 51%) had treated TMDs for 17 through 45 years. Even though practitioners were recruited from different regions, the participants’ practices may be different from those of the average American dentists.

Table 3.

Study participants demographic data.

CHARACTERISTIC	DATA
Dentist Practitioners (n = 185)
Sex, no. (%)
Male	115 (62.2)
Female	69 (37.3)
Missing	1 (0.5)
Age
Mean (standard deviation), range, y	52 (12), 29-76
Missing, no, (%)	11 (5.9)
Ethnicity, no. (%)
Non-Hispanic	175 (94.6)
Hispanic, Latino, of Spanish origin	7 (3.8)
Missing	3 (1.6)
Race, no. (%)
White	142 (76.8)
Asian	25 (13.5)
Other	16 (8.6)
Missing	2 (1.1)
Practice, no. (%)
Full time (≥ 32 h/wk)	145 (78.4)
Part time (< 32 h/wk)	35 (18.9)
Missing	5 (2.7)
Practice type, no. (%)
No advanced training in an orofacial pain residency	136 (73.5)
Advanced training in an orofacial pain residency	44 (23.8)
Missing	5 (2.7)
Region, no. (%)
Western	29 (15.7)
Midwest	39 (21.1)
Southwest	35 (18.9)
South Central	23 (12.4)
South Atlantic	27 (14.6)
Northeast	32 (17.3)
Patients (n = 1,901)
Sex, no. (%)*
Male	316 (16.6)
Female	1,584 (83.3)
Missing	1 (0.05)
Age
Mean (standard deviation), range, y	44.1 (15.6), 18-86
Missing, no. (%)	18 (0.009)
Ethnicity, no. (%)
Non-Hispanic	1672 (87.9)
Hispanic, Latino, of Spanish origin	212 (11.2)
Missing	17 (0.9)
Race, no. (%)†
White	1603 (84.3)
Non-White	282 (14.8)
Missing	16 (0.8)
Education, no. (%)
< High school diploma	28 (1.5)
High school graduate	229 (12.1)
Some college	686 (36.1)
Bachelor’s degree	569 (29.9)
Graduate degree	375 (19.7)
Missing	14 (0.7)
Annual household income, no. (%)
Decline to answer	238 (12.5)
< $40,000	404 (21.3)
$40,000-$79,999	514 (27.0)
≥ 80,000	732 (38.5)
Missing	13 (0.7)

^*Due to rounding, the sum of percentages is 99.95%.

^†Due to rounding, the sum of percentages is 99.9%.

Figure 1 shows that of 6,772 patients with TMDs screened as potential participants, 1,901 fulfilled all eligibility criteria and agreed to participate in the study. Patient demographic characteristics are provided in Table 3. A total of 84.3% of the patients were White, 88% were non-Hispanic, and 83.3% were female, with a mean age of 44 years.

Table 4 provides baseline patient biopsychosocial characteristics. Most participants described good to excellent general health status, although 1 in 5 (372/1,901) reported moderate or severe psychological distress, and 842 (44.3%) responded feeling refreshed or rested on awakening none or a little of the time. Seventy-two percent (1,360/1,901) of the patients reported intermittent (“comes and goes”) painful TMDs, 82.0% (1,558/1,901) reported pain or stiffness of the jaw on awakening, 40.3% had low-intensity pain with no to low disability (GCPS grade I), and 27.7% had high-intensity pain with no disability (grade IIa). The observed mean (SD) CPI and JFLS scores were 49.8 (19.2) and 2.1 (1.9), respectively. The CPI and JFLS means (SE) adjusted for within-cluster correlation between observations were 49.7 (0.6) and 2.0 (0.05), respectively. A summary statistic for Widespread Pain Index data from 1,842 patients was a mean (SD) of 5.2 (3.4) areas of pain in their body in the past 7 days. Forty-seven percent of painful TMDs began at least 3 years before patients’ enrollment into the study.

Table 4.

Patient biopsychosocial characteristics self-reported at baseline.

VARIABLE	RESULT
General Health Status, Frequency, No. (%)
Excellent	270 (14.2)
Very good	767 (40.3)
Good	652 (34.3)
Fair	173 (9.1)
Poor	22 (1.2)
Missing	17 (0.9)
Psychological Distress, Frequency, No. (%)
No clinical distress	978 (51.4)
Mild distress	498 (26.2)
Moderate distress	253 (13.3)
Severe distress	119 (6.3)
Missing	53 (2.8)
Sleep Quality, Frequency, No. (%)
All of the time	23 (1.2)
Most of the time	332 (17.5)
Some of the time	692 (36.4)
A little of the time	534 (28.1)
None of the time	308 (16.2)
Missing	12 (0.6)
GCPS* Characteristic Pain Intensity. Over Last 30 Days, Mean (SD†), Median, Range (n = 1,901)‡
Characteristic Pain Intensity baseline§	49.8 (19.2), 50.0, 3.3-100.0
Present pain	3.5 (2.4), 3.0, 0.0-10.0
Worst pain	6.8 (2.2), 7.0, 1.0-10.0
Average pain	4.7 (2.0), 5.0, 3.0-10.0
GCPS Pain-Related Disability Points Over the Last 30 Days, No., Mean (SD), Median, Range
Disability points based on disability days	1,864, 0.7 (1.2), 0, 0.0-3.0
Disability points based on interference score	1,873, 0.5 (0.9), 0, 0.0-3.0
Total disability points at baseline	1,843, 1.2 (1.8), 0, 0.0-6.0
GCPS at Baseline, Frequency, No. (%)
Grade I, low pain intensity—none to low disability	766 (40.3)
Grade IIa, high intensity pain—no disability	526 (27.7)
Grade IIb, high intensity pain—low disability	130 (6.8)
Grade III, high disability—moderately limiting	254 (13.4)
Grade IV, high disability—severely limiting	167 (8.8)
Missing	58 (3.1)
JFLS¶ at Baseline
JFLS, no., mean (SD), median, range	1,688, 2.1 (1.9), 1.5, 0.0-10.0
JFLS mean adjusted for within-cluster correlation between observations, no. (standard error)	2.0 (0.05)
Missing, no. (%)	213 (11.2)
Widespread Pain Index
Widespread Pain Index, no., mean (SD), median, range	1842, 5.2 (3.4), 5.0, 1.0-19.0
Missing, no. (%)	59 (3.1)
Temporomandibular Disorder Pain Duration, Frequency, No. (%)
< 1 m	121 (6.4)
≥ 1 m but < 3 m	216 (11.4)
≥ 3 m but < 6 m	158 (8.3)
≥ 6 m but < 1 y	170 (8.9)
≥ 1 y but < 3 y	334 (17.5)
≥ 3 y	891 (46.9)
Missing	11 (0.6)

^*GCPS: Graded Chronic Pain Scale.

^†SD: Standard deviation.

^‡Percentages were calculated by dividing the frequency of each category by the total sample (1,901). Mean, standard deviation, and standard error were based on the number of responders.

^§CPI baseline mean (standard error) adjusted for within-cluster correlation between observations, 49.7 (0.6).

^¶JFLS: Jaw Function Limitation Scale.

Figure 2 details the symptoms and diagnoses of the 1,901 patients. The most common painful TMD diagnoses were myalgia (1,376, 72.4%), arthralgia (1,228 [64.6%]: unilateral, 579 [30.5%]; bilateral, 649 [34.1%]), TMD-related headache (970, 51.0%), and myofascial pain with referral (768, 40.4%). Eighty percent (1,522) had a muscle diagnosis of myalgia or myofascial pain with referral, and 27.0% (514) had only a muscle pain diagnosis (no joint pain), 11.6% (220) had only a joint pain diagnosis (arthralgia), 53% (1,008) had both muscle and joint pain diagnoses, and 8.4% did not have a specific TMD diagnosis beyond painful TMD because a specific diagnosis was not required for study inclusion.

Figure 2.

Temporomandibular disorder (TMD) symptoms and diagnoses for study patients. Percentage of total sample (n = 1,901 patients).

Oral behaviors were reported by 44.8% of patients, occurring some of the time (33.6%) or all the time (20.0%). The overall mean (SD) OBC was 0.75 (0.30) (median, 0.73). The most frequent oral behaviors were sleeping in a position that puts pressure on the jaw; clenching and grinding teeth during sleep or when awake; chewing food on 1 side; holding, tightening, or tensing muscles (no clenching); or pressing teeth together (Figure 3).

Figure 3.

Oral behaviors checklist for study patients. Percentage of total sample (n = 1,901 patients). *: 0.9%. Missing: < 2.3%.

Table 5 lists all treatments recommended by the practitioners. Self-care instruction (89.4%), intraoral appliances (75.4%), and medications (57.6%) were the most frequently recommended treatments, and 66.9% of the patients were referred to other health care professionals, including 27.7% to physical therapy and 8.4% to a mental health care specialist.

Table 5.

Painful TMD^* treatment recommendations.^†

TREATMENT	YES, NO. (%)	NO, NO. (%)	MISSING, NO. (%)
Self-Care Recommended for Treatment of TMD Pain
≥ 1 self-care activities recommended‡	1,700 (89.4)	151 (7.9)	50 (2.6)
Individual self-care recommendations
Avoid oral habits	1,418 (74.6)	337 (17.7)	146 (7.7)
Relax your jaw muscles‡	1,392 (73.2)	362 (19.0)	147 (7.7)
Apply heat or ice	1,343 (70.6)	433 (22. 8)	125 (6.6)
Eat a soft diet	1,338 (70.4)	435 (22.9)	128 (6.7)
Avoid chewing gum	1,304 (68.6)	444 (23.4)	153 (8.0)
Keep your teeth apart	1,083 (57.0)	596 (31.3)	222 (11.7)
Chew food on both sides	838 (44.1)	832 (43.8)	231 (12.1)
Reduce caffeine intake	507 (26.7)	1,135 (59.7)	259 (13.6)
Other	182 (9.6)	1042 (54.8)	677 (35.6)
Medication for Treatment of TMD Pain
≥ 1 medications recommended	1,094 (57.6)	725 (38.1)	82 (4.3)
Individual medication recommendations
Over-the-counter analgesic	828 (43.6)	979 (51.5)	94 (4.9)
Prescription nonsteroidal antiinflammatory drug	305 (16.0)	1,397 (73.5)	199 (10.5)
Muscle relaxant	268 (14.1)	1,410 (74.2)	223 (11.7)
Tricyclic antidepressant	26 (1.4)	1,623 (85.4)	252 (13.2)
Prescription narcotic	19 (1.0)	1,639 (86.2)	243 (12.8)
Prescription cannabinoid	2 (0.1)	1,641 (86.3)	258 (13.6)
Other	86 (4.5)	1,362 (71.7)	453 (23.8)
Intraoral Appliance
Recommended to be part of treatment plan	1,434 (75.4)	428 (22.5)	39 (2.1)
Insertion when sleeping	1,378 (72.5)	29 (1.5)	494 (26.0)
Insertion when awake (except during eating)	391 (20.5)	718 (37.8)	792 (41.7)
Insertion when eating	93 (4.9)	938 (49.3)	870 (45.8)
No specific insertion time recommended	35 (1.8)	NA§	NA
Implementation of treatment plan with the fabrication of recommended mouthguard	1,068 (56.2)	NA	NA
Referral for Treatment by Allied Care Providers
≥ 1 treatment referrals recommended	1,271 (66.9)	572 (30.1)	58 (3.0)
Physical or Biomedical Treatment Referrals
Any physical or biomedical treatment referral	1,126 (59.2)	677 (35.6)	98 (5.2)
Individual physical or biomedical treatment referrals
Self-massage of jaw or temples	773 (40.7)	942 (49.5)	186 (9.8)
Jaw exercises	644 (33.9)	1,057 (55.6)	200 (10.5)
Physical therapy	527 (27.7)	1,144 (60.2)	230 (12.1)
Massage therapy	355 (18.7)	1,303 (68.5)	243 (12.8)
Trigger-point injections	133 (7.0)	1,480 (77.9)	288 (15.1)
Chiropractic therapy	107 (5.6)	1,503 (79.1)	291 (15.3)
Occlusal Stabilization Treatment Referrals
Any occlusal stabilization treatment referral	276 (14.5)	1,366 (71.9)	259 (13.6)
Individual occlusal stabilization treatment referrals
Occlusal equilibration	167 (8.8)	1,462 (76.9)	272 (14.3)
Orthodontic treatment	113 (5.9)	1,505 (79.2)	283 (14.9)
Restorative dentistry, for example, crown	68 (3.6)	1,534 (80.7)	299 (15.7)
Full-mouth reconstruction	28 (1.5)	1,571 (82.6)	302 (15.9)
Psychological Treatment Referrals
Any psychological treatment referral	159 (8.4)	1,453 (76.4)	289 (15.2)
Surgical Intervention Treatment Referrals
Any surgical intervention treatment referral‡	56 (2.9)	1,531 (80.5)	314 (16.5)
Individual surgical intervention treatment referrals
Temporomandibular joint arthrocentesis	40 (2.1)	1,559 (82.0)	302 (15.9)
Orthognathic surgery	11 (0.6)	1,591 (83.7)	299 (15.7)
Arthroscopic surgery	12 (0.6)	1,583 (83.3)	306 (16.1)
Temporomandibular joint open-joint surgery	3 (0.2)	1,588 (83.5)	310 (16.3)
Additional Treatment Referrals
Any of the following treatments	262 (13.8)	1,176 (61.9)	463 (24.3)
Individual treatment referrals
Low-level laser therapy	60 (3.2)	1,548 (81.4)	293 (15.4)
Botulinum toxin type A (Botox) injections	58 (3.0)	1,554 (81.8)	289 (15.2)
Acupuncture	34 (1.8)	1,566 (82.4)	301 (15.8)
Herbs or supplements	28 (1.5)	1,578 (83.0)	295 (15.5)
Other	130 (6.8)	1,269 (66.8)	502 (26.4)

^*TMD: Temporomandibular disorder.

^†Percentages were calculated by dividing the frequency of each category by the total sample (1,901).

^‡Owing to rounding, the sum of percentages is 99.9%.

^§NA: Not applicable.

DISCUSSION

To our knowledge, this is the first study to provide a comprehensive description of a large cohort of patients seeking treatment for painful TMDs in US-based network community settings (Tables 3 and 4). The large sample size provides a credible description of the diagnoses and treatment recommendations being rendered by practitioners in these community settings. In this prospective cohort study, we investigated factors that contribute to practitioners’ treatment decisions, patient’s adherence to treatment, and observed changes from baseline at 1-, 3-, and 6-month follow-ups in pain intensity and jaw function associated with treatments.

Most dentist practitioners were White, non-Hispanic, male, generalists, practicing full-time, and graduated more than 10 years ago. This was also the case in the previously reported study of the Collaboration on Networked Dental and Oral Research practitioners.^4,5,17-19 Similar to other previously reported TMD studies, female,^10,20-26 White,^25,26 and non-Hispanic patients with TMD^25,26 were predominant, and the age-specific prevalence was greatest among people in their 40s.²⁷

Pain intensity as measured with the CPI (mean, 49.7) was similar to that reported among Orofacial Pain Prospective Evaluation and Risk Assessment (OPPERA) study’s TMD chronic cases (mean, 51.8).²⁸ GCPS grade I percentage (40.3%) was similar to that found by Von Korff and colleagues⁸ but lower than that found in other studies (50.3%,²⁵ 46%²⁹). The percentages of GCPS grade IIa (27.7%) and grade IIb-IV (29%) were a little higher than those found in OPPERA (24.3%, 25.4%, respectively).²⁸ This was not unexpected, given that patients with TMDs in our study were seeking TMD treatment. Seventy-two of the TMD participants in our study and 70.5% of the chronic TMD cases from OPPERA reported intermittent painful TMDs.²⁵ One study found a higher percentage of poor or fair self-rated general health (7.1%) and lower JFLS (mean, 1.7)²⁸ than our study (10.3%; mean, 2.1).

Common diagnoses were muscle related, which is consistent with other studies of TMD diagnostic prevalence in patient populations.³⁰ The TMD diagnoses were established by the practitioners on the basis of their background and experience. Although they were trained with regard to study enrollment procedures and data collection protocol, practitioners intentionally were not calibrated with regard to TMD diagnosis and treatment, so that the outcomes would reflect current practice in community practice settings. In the context of the broad geographic scope of the network, these findings suggest that the sample provides a credible representation of the network practice in the United States.^4,5,17-19

We found that practitioners most frequently used reversible treatments for painful TMDs, including patient education with self-care, intraoral appliances, and pharmacologic pain control as recommended by the National Academies of Sciences, Engineering, and Medicine’s report on TMDs.² This finding is consistent with our previous network survey results.⁴ Occlusal adjustments were used even less frequently by practitioners in our current study (8.8%) than those in our previous survey (57.6%).⁴ This finding is also consistent with the National Academies of Sciences, Engineering, and Medicine’s report on management of painful TMDs.² Limiting the use of this irreversible treatment is recommended, as a 2018 review found no evidence for its effectiveness.³¹ Our findings, therefore, suggest that current network practices are using more reversible treatments than reported in the past.

This study has several notable strengths. First, we recruited 185 practitioners broadly distributed in the United States, with a high participation rate of 92% (185/201). This high participation rate decreases the chance of selection bias and increases the external validity and generalizability of the study’s findings regarding what practitioners do in diagnosis and treatment of painful TMDs. Second, the patients were seeking treatment for painful TMDs, which reduces selection bias and makes the results generalizable to patients seeking treatment in the United States. Third, we used well-accepted and validated assessment instruments to reduce information bias. Fourth, patients completed their questions at baseline and all follow-ups with the knowledge that their practitioners would not see their responses. Patients were also instructed to answer the questions without discussing them with anyone. We believe these conditions allowed them to provide responses largely free of bias. Fifth, the HealthPartners Data Coordinating Center developed a data management plan with detailed quality management procedures, including the development of comprehensive data quality checks in the database system. Sixth, we were able to standardize outcome measures and eligibility criteria for practitioners and patients. Finally, this investigation reflects current network-based clinical practice, and the results will be applicable to that clinical setting. Network studies are practical science conducted about, in, and for the benefit of real-world, everyday clinical practice.³²

This study has limitations. First, missing data are an inevitable outcome when self-report questionnaires do not have required fields. Second, there is a chance of diagnosis misclassification that could affect the treatment selection, because we decided not to calibrate the practitioners to reflect current practice in community practice settings. Third, although practitioners were recruited from different regions, the participants’ practices may be different from those of the average US dentist.

Future reports will provide findings pertaining to factors that contribute to practitioners’ treatment decisions, patients’ adherence to treatment, and treatment outcomes at 6-month follow-up.

CONCLUSIONS

Most of the characteristics of this TMD cohort include those typical of patients with painful TMDs in the US-based population. Our results may be the best evidence that network practitioners usually select reversible treatments to manage painful TMDs in patients. This large cohort provides a credible description of the diagnoses and treatment recommendations being rendered by practitioners in their community settings.

ABBREVIATION KEY

CPI

Characteristic Pain Intensity

GCPS

Graded Chronic Pain Scale

JFLS-8

Jaw Function Limitation Scale

OBC

Oral Behavioral Checklist

OPPERA

Orofacial Pain Prospective Evaluation and Risk Assessment

TMD

Temporomandibular disorders

TMJ

Temporomandibular joint