Development and internal validation of prediction models for persistence of self-reported orofacial pain in the follow-up of patients with myofascial pain

OBJECTIVES: To identify predictors in patient profiles, and to develop, internally validate, and calibrate prediction models for the persistence of self-reported orofacial pain at the 6-month and 12-month follow-up in patients with myofascial pain. MATERIALS AND METHODS: A cohort of 63 adult patients with moderate to severe chronic myofascial pain was included. Patient and disease characteristics at baseline were recorded as potential predictors. Patients` presence or absence of improvement of orofacial pain at follow-up was considered the outcome. Binary logistic regression analyses were used to develop the models. The performance and clinical values of the models were determined. RESULTS: Forty-three percent and 30% of the patients had persistence of orofacial pain at 6-month and 12-month follow-up, respectively. Pain elsewhere, depression, parafunctional activities, and mandibular function impairment (MFI) were significantly associated with persistence of the pain at 6-month follow-up, whereas depression, parafunctional activities, and MFI were significantly associated with persistence of the pain at 12-month follow-up. Both of the models showed good calibration and discrimination, with shrunken area under the curve (AUC) values of 0.73 and 0.76, respectively. The clinical added predictive values for ruling in the risk of the persistence were 0.30 and 0.31, respectively, and those for ruling it out were 0.25 and 0.20, respectively. CONCLUSIONS: Potential predictors for prediction of the persistence of self-reported orofacial pain at follow-up were identified. The calibration, discrimination, and clinical values of the models were acceptable. CLINICAL RELEVANCE: The models may assist clinicians in decision-making regarding the improvement of orofacial pain of individual patients during follow-up in clinical settings.