Clinical predictors of persistent temporomandibular disorder in people with first-onset temporomandibular disorder: A prospective case-control study

Background: When patients first develop a painful temporomandibular disorder (TMD) and seek care, 1 priority for clinicians is to assess prognosis. The authors aimed to develop a predictive model by using biopsychosocial measures from the Diagnostic Criteria for Temporomandibular Disorders (DC-TMD) to predict risk of developing TMD symptom persistence.

Methods: At baseline, trained examiners identified 260 participants with first-onset TMD classified by using DC-TMD-compliant protocols. After follow-up at least 6 months later, 72 (49%) had examiner-classified TMD (persistent cases), and 75 (51%) no longer had examiner-classified TMD (transient cases). For multivariable logistic regression analysis, the authors used blocks of variables selected using minimum redundancy maximum relevance to construct a model to predict the odds of TMD persistence.

Results: At onset, persistent cases had multiple worse TMD clinical measures and, among Axis II measures, only greater baseline pain intensity (odds ratio [OR], 1.5; 95% confidence interval [CI], 1.04 to 2.2; P = .030) and more physical symptoms (OR, 1.8; 95% CI, 1.2 to 2.9; P = .004) than did transient cases. A multivariable model using TMD clinical measures showed greater discriminative capacity (area under the receiver operating characteristic curve, 0.74; 95% CI, 0.73 to 0.75) than did a model involving psychosocial measures (area under the receiver operating characteristic curve, 0.63; 95% CI, 0.62 to 0.64).

Conclusions: Clinical measures that clinicians can assess readily when TMD first develops are useful in predicting the risk of developing persistent TMD. Psychosocial measures are important predictors of onset but do not add meaningfully to the predictive capacity of clinical measures.

Practical implications: When TMD first develops, clinicians usefully can identify patients at higher risk of developing persistence by using clinical measures that they logically also could use in treatment planning and for monitoring outcomes of intervention.