Aims: To determine, in subjects with painful temporomandibular disorders (TMD), whether baseline SF-12 Physical Component Summary (PCS), SF-12 Mental Component Summary (MCS) and Jaw Functional Limitation Scale (JFLS) predict persistent TMD pain measured by Characteristic Pain Intensity (CPI) scores after 9 years. Methods: 258 subjects with painful TMD diagnoses and CPI>0 completed baseline SF-12 Health-Related Quality of Life (HRQoL) and JFLS questionnaires. After 9 years, they were reevaluated for painful TMD diagnoses and completed the CPI questionnaire. Univariable and multivariable linear regression adjusted for age and sex examined the relationship between baseline predictors and follow-up CPI. Results: After 9 years, 186 (72%) had persistent TMD pain. Baseline PCS and JFLS, but not MCS, presented weak linear relationships with follow-up CPI. One SD (9.0) increase in baseline PCS was associated with a 4.9-point decrease in follow-up CPI (SE=1.2, p<0.001), or 5.7% of the follow-up CPI score range. One SD (1.4) increase in baseline JFLS was associated with a 5.0-point increase in follow-up CPI scores (SE=1.2, p<0.001), or 5.8% of the follow-up CPI score range. In the 3-predictor multivariable model, follow-up CPI change predicted by 1 SD increase in baseline scores was of -4.7 (SE=1.3, p<0.001) for PCS, and 3.9 points (SE=1.2, p=0.002) for JFLS. Conclusions: In subjects with TMD pain, baseline PCS and JFLS, but not MCS, were statistically significant predictors of CPI at a 9-year follow-up. However, the magnitude of the effects is small and below the 10-20% minimum change in pain intensity recommended for clinical significance.