Patient-friendly summary
If you read nothing else
Bottom line
In a large, well-conducted Danish trial, the lay-led Chronic Pain Self-Management Programme did not reduce pain-related disability, pain, or health care costs, and the authors do not recommend it as a treatment for chronic pain-related disability.
Moderate evidencePublished
Evidence hierarchy
Study participants
Adults from 19 Danish municipalities with chronic pain of any etiology, pain duration >=3 months and self-rated pain intensity >=5/10 at enrollment; heterogeneous with multiple causes of pain and comorbidities
Study Summary
This pragmatic randomized controlled trial tested a Danish lay-led version of the Stanford Chronic Pain Self-Management Programme (CPSMP) — six weekly 2.5-hour group workshops led by trained laypeople, at least one of whom has a long-term pain condition — against treatment as usual in 424 adults with chronic pain of any cause. The program had no effect on the primary outcome of pain-related disability, nor on pain, self-efficacy, pain catastrophizing, physical symptoms, or health care expenditure at 3-month follow-up. Small positive effects on emotional distress and illness worry remained at 3 months but were judged clinically insignificant, and improvements seen immediately after the course faded quickly. The authors conclude the lay-led CPSMP is not recommended as a treatment to reduce chronic pain or disability.
Key Findings
| Finding | Detail | Impact |
|---|---|---|
| No effect on pain-related disability at 3 months | Random effect analyses found no statistically significant effect of the CPSMP on the primary outcome of pain-related disability (Modified Roland-Morris Disability Questionnaire) at 3-month follow-up (T3). | High |
| No effect on pain, self-efficacy, catastrophizing, or physical symptoms at follow-up | At T3 there were no statistically significant effects on the secondary outcomes of pain, self-efficacy, pain catastrophizing, or physical symptoms. | High |
| Small but clinically insignificant gains in emotional distress and illness worry | At 3 months, emotional distress decreased 13% in the intervention group vs 3% in controls and illness worry decreased 12% vs an increase of 1% in controls (P: 0.0043-0.0079); however, chi-square analyses showed no difference in the percentage of patients above clinical cut-offs, so these effects were judged clinically nonsignificant. | Medium |
| Immediate post-course benefits were transient | Immediately after the course (T2), the intervention group showed significantly greater reductions in disability (t=3.373, P=0.001), distress, somatic symptoms, pain catastrophizing, self-efficacy, and health worry than controls, but these effects largely disappeared by the 3-month follow-up. | Medium |
| No effect on health care expenditure | Using national registry data on costs for the 5-month intervention and follow-up period, cost ratios between groups showed no significant differences when controlling for baseline costs (P: 0.12-0.88). | High |
| Unexpected decrease in self-efficacy in the intervention group | Self-efficacy decreased more in the intervention group than in controls, but the authors caution this should be interpreted carefully because the test-retest reliability of the self-efficacy scale was too low to reliably interpret change over time. | Low |
| High satisfaction despite lack of measurable benefit | At follow-up, 55% reported applying skills learned, 39% said the skills helped them manage pain, and 79% would recommend the course to others, with average attendance of 4.9 of 6 sessions. | Low |
Random effect analyses found no statistically significant effect of the CPSMP on the primary outcome of pain-related disability (Modified Roland-Morris Disability Questionnaire) at 3-month follow-up (T3).
At T3 there were no statistically significant effects on the secondary outcomes of pain, self-efficacy, pain catastrophizing, or physical symptoms.
At 3 months, emotional distress decreased 13% in the intervention group vs 3% in controls and illness worry decreased 12% vs an increase of 1% in controls (P: 0.0043-0.0079); however, chi-square analyses showed no difference in the percentage of patients above clinical cut-offs, so these effects were judged clinically nonsignificant.
Immediately after the course (T2), the intervention group showed significantly greater reductions in disability (t=3.373, P=0.001), distress, somatic symptoms, pain catastrophizing, self-efficacy, and health worry than controls, but these effects largely disappeared by the 3-month follow-up.
Using national registry data on costs for the 5-month intervention and follow-up period, cost ratios between groups showed no significant differences when controlling for baseline costs (P: 0.12-0.88).
Self-efficacy decreased more in the intervention group than in controls, but the authors caution this should be interpreted carefully because the test-retest reliability of the self-efficacy scale was too low to reliably interpret change over time.
At follow-up, 55% reported applying skills learned, 39% said the skills helped them manage pain, and 79% would recommend the course to others, with average attendance of 4.9 of 6 sessions.
Strengths
- Pragmatic, real-world design conducted in municipal health support centers rather than a specialized hospital, increasing applicability to everyday practice
- Much larger sample (424 participants) than the three previously published RCTs of the CPSMP, with adequate statistical power
- Low attrition: follow-up data collected from 92% of participants, reducing dropout bias
- Intention-to-treat analysis with adjustment for age, sex, pain duration, and education, and accounting for clustering
- Objective health expenditure outcomes drawn from national patient registries rather than self-report
- Most outcome measures were previously validated in Danish populations; protocol and trial were ethics-approved and registered (NCT01306747); preceded by a feasibility study of 87 patients
Limitations
- The unrestricted control group could pursue other treatments and improved over time, potentially blurring any effect of the program; some controls knew they would soon start the course
- Highly heterogeneous sample (wide age range, varied pain types and durations) may have obscured benefits for any specific subgroup
- Test-retest reliability of the pain VAS and the self-efficacy scale was too low to reliably interpret change, so the apparent negative self-efficacy effect should be viewed cautiously
- Uncertain whether participants actually applied the self-management skills; qualitative interviews suggested application was highly variable and inconsistent
- Many workshop leaders were novice lay-leaders trained immediately before the course, and some participants felt leaders read from the manual rather than engaging with them
- Planned sample size was not fully reached during the inclusion period (though lower-than-expected attrition meant the analyzed sample still exceeded the planned number)
Key Takeaways for Patients
What This Means for You
- 01A six-week, group-based pain self-management course led by trained laypeople (at least one of whom lives with chronic pain) did not, in this large Danish study, reduce pain-related disability or pain over the following months.
- 02Some people felt better right after finishing the course, but those benefits faded by the 3-month mark.
- 03Small improvements in mood and worry about illness remained at 3 months, but they were too small to count as a meaningful clinical change.
- 04Most participants (79%) said they would still recommend the course to others, and many found parts of it helpful even though it did not measurably reduce their pain.
- 05The authors suggest that brief, lay-led programs may not be intensive enough to produce lasting change, and that more in-depth psychological pain treatment or longer programs led by health professionals may be more effective for chronic pain.