The purpose of this study is to determine whether the combination of self-management support, pain science education, and individualized, goal-oriented exercises helps people with chronic pain to increase their function.

Primary objective: This study will test the hypothesis that participants with chronic pain experience greater change in function with ChrOnic pain self-ManageMent support with pain science EducatioN and exerCisE (COMMENCE) in comparison to a wait-list control.

Secondary objective #1: This study will test the hypotheses that people with chronic pain experience greater change in pain intensity, pain interference, self-efficacy, catastrophic thinking, fear of movement/re-injury, pain neurophysiology knowledge, how much participants are bothered by difficulty with functional activities, fatigue, depressive symptoms, health care utilization and work status with COMMENCE in comparison to a wait-list control.

Secondary objective #2: This study will identify demographic, psychological, or psychophysical variables that are predictive of treatment response.

Design: This study will be a randomized trial with parallel groups, a wait-list control and balanced randomization. A cross-over will take place after a 12-week follow-up assessment (ie. the group initially receiving treatment will receive no treatment and the group initially on the wait-list will receive treatment for 6 weeks). The between group comparison will be performed using only the data up to the 12-week follow-up due to anticipated carry-over effects with self-management support. Data collected after the cross-over will be used to help identify factors that predict response to the intervention.

Blinding: Due to the nature of the treatment and comparison, participants and the treating physiotherapist will not be blinded. The assessor who is completing the two objective measures will be blind to the treatment allocation. Data analysis will be performed by someone who is not blinded to the group allocation.

Setting: 110 participants with chronic pain will be recruited at Woodstock and Area Community Health Centre (WACHC) in Woodstock, Ontario, Canada. WACHC has an interdisciplinary team of health care providers that provide care to priority populations in Oxford County, Ontario, Canada. The priority populations include people with the following barriers to accessing health care: addictions concerns, mental health challenges, low incomes, lack of health insurance, and isolated seniors. The investigators expect this sample to represent a marginalized population of people with chronic often excluded from chronic pain treatment and research by barriers to accessing health care.

Sample Size: The sample size necessary for a randomized controlled trial with three repeated measures at 0, 7 and 18 weeks was calculated using online sample size software (GLIMMPSE 2.0). The calculation was performed using a significance level of 0.05, a power of 0.8, a minimum detectable mean difference between groups of 10 points on the Short Musculoskeletal Function Assessment Dysfunction Index (SMFA-DI) at both 7 weeks and 18 weeks, and a standard deviation of 23 points on the SMFA-DI based on clinical data at WACHC. The needed sample size calculated was 88 participants. To account for a potential 20% drop-out rate, 110 participants will be recruited

Allocation: The allocation sequence will be generated by a study investigator who is not involved in the enrolment of participants or assigning interventions. A computer-generated blocked random number schedule will be used to determine allocation sequence. The block size will be unknown to the other study investigators. The allocation sequence will be concealed through the use of sequentially numbered, opaque envelopes, opened after the initial assessment.

Co-intervention: Participants will be free to continue with other treatments. Other treatments will be recorded through self-report at each assessment time-point and analyzed for between group differences.

Withdrawing Participants from this Study: Participants may withdraw from the treatment at any time. Participants who choose to withdraw will be documented and data will be analyzed as a member of the group to which they were randomly assigned (intention-to-treat)

Potential prognostic indicators or predictors of response: Potential predictors of response for secondary objective #2 will include baseline measures for each of the outcome measures listed under outcomes as well as post-traumatic stress measured by the Post-traumatic Stress Disorder Checklist, sense of perceived injustice measured by the Injustice Experience Questionnaire, number of medications, comorbidities measured by disease count, and expectations for recovery. Expectations for recovery will be assessed with two questions: i) Do you think your pain will improve? ii) Do you think your functional abilities will improve?

Demographic information: The following information will be collected at the initial assessment and analyzed as potential covariates and predictors of response: age, sex, work status prior to symptom onset, length of time since symptom onset in months, diagnosis provided by a medical professional as reported by the patient, medication use, previous treatment received, and expectations for recovery.

Treatment fidelity: Treatment fidelity will be assessed through a combination of attendance (categorized as <25%, 25-49%, 50-74%, ≥75% of visits) and adherence to self-management strategies. This will be measured by clinician report when the clinician reviews the workbook at each individual treatment session.

Data monitoring and auditing: This trial will not include a data monitoring committee and will not be audited outside of the study investigators

Analysis: Statistical analysis will be conducted using Stata software, version 13 (StataCorp, College Station, Texas, USA). Baseline characteristics for treatment and wait-list groups will be compared using a Student's t-test for continuous data and Chi squared or Fisher's exact test for categorical variables. Analysis for the primary objective and secondary objective #1 will be performed using linear mixed-effects modelling with data from three repeated measures (baseline, 7 weeks, 18 weeks). A p value of less than 0.05 will be considered indicative of statistical significance. The primary analysis will use intention to treat principles, but a secondary analysis to investigate efficacy for those that complete 75% of visits will be performed as well (per protocol analysis). Predictors of response (change in SMFA-DI) will be determined using a series of multiple regression models.

Inclusion Criteria:

• participants will all be adults who have been experiencing non-cancer related pain for at least 12 weeks

Exclusion Criteria:

• cancer related pain
• medical "red flags" suggestive of a non-neuromusculoskeletal etiology of symptoms
• casted fracture within the last 12 weeks
• related surgery within the last 26 weeks
• evidence of upper motor neuron lesion