Question:In adults with chronic pain (excluding headache), what are the benefits and harms of psychological therapies?

Clinical Answer:

In adults with non-headache related chronic pain, randomized controlled trials including a total of around 2,500 participants suggest that cognitive-behavioral therapy (CBT) may improve disability at post treatment and at 6-12 months compared with active control or treatment as usual. CBT appeared to have no impact on pain, mood or catastrophizing in these studies. More limited data from randomized controlled trials including a total of around 300 patients suggest that behavioral therapy may reduce catastrophizing immediately post treatment compared with treatment as usual but effects on disability, pain or mood are uncertain. It is not possible to apply the results of these studies in clinical practice because the comparator interventions (active control intervention/treatment as usual) were poorly defined. Most importantly it was unclear whether interventions were given alone or in addition to other pain medication.

Full outcome data is detailed below:

1.Cognitive behavioral therapy versus active control

  • Population, Intervention, Comparator

    Population:

    Adults (age range 43-80 years) with pain for more than 3 months, caused by a range of conditions including low back pain, osteoarthritis of the knee or rheumatoid arthritis

    Intervention:

    Cognitive behavioral therapy involving a psychologist in delivery (no further details available)

    Comparator:

    Active control defined as a protocolized treatment which engaged the patient, such as an exercise program, a medical procedure, an education program, a support group or a self-instruction booklet

  • OUTCOME 1.1: Pain at post-treatment

    Risk of bias of studies:

    The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

    Narrative result:

    13 RCTs with 1258 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

    Relative effect or mean difference:

    There was no statistically significant difference between groups (standardized mean difference -0.10, 95% CI -0.24 to 0.04).

    Forest plot from Cochrane Review

    Reference:
    Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
    • OUTCOME 1.2: Pain at 6 to 12 months

      Risk of bias of studies:

      The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

      Narrative result:

      11 RCTs with 1261 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

      Relative effect or mean difference:

      There was no statistically significant difference between groups (standardized mean difference -0.08, 95% CI -0.23 to 0.06).

      Forest plot from Cochrane Review

      Reference:
      Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
      • OUTCOME 1.3: Disability at post-treatment

        Risk of bias of studies:

        The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

        Narrative result:

        12 RCTs with 1130 participants found that CBT reduced the proportion of people with disability compared with active treatment control. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

        Relative effect or mean difference:

        There was a statistically significant difference between groups, in favor of CBT (standardized mean difference -0.19, 95% CI -0.33 to -0.05).

        Forest plot from Cochrane Review

        Reference:
        Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
        • OUTCOME 1.4: Disability at 6 to 12 months

          Risk of bias of studies:

          The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

          Narrative result:

          12 RCTs with 1295 participants found that CBT reduced the proportion of people with disability compared with active control. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

          Relative effect or mean difference:

          There was a statistically significant difference between groups, in favor of CBT (standardized mean difference -0.15, 95% CI -0.28 to -0.02).

          Forest plot from Cochrane Review

          Reference:
          Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
          • OUTCOME 1.5: Mood at post-treatment

            Risk of bias of studies:

            The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

            Narrative result:

            13 RCTs with 1256 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

            Relative effect or mean difference:

            There was no statistically significant difference between groups (standardized mean difference -0.05, 95% CI -0.19 to 0.09).

            Forest plot from Cochrane Review

            Reference:
            Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
            • OUTCOME 1.6: Mood at 6 to 12 months

              Risk of bias of studies:

              The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

              Narrative result:

              11 RCTs with 1261 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

              Relative effect or mean difference:

              There was no statistically significant difference between groups (standardized mean difference -0.07, 95% CI -0.18 to 0.05).

              Forest plot from Cochrane Review

              Reference:
              Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
              • OUTCOME 1.7: Catastrophizing at post-treatment

                Risk of bias of studies:

                The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                Narrative result:

                Six RCTs with 735 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                Relative effect or mean difference:

                There was no statistically significant difference between groups (standardized mean difference -0.18, 95% CI -0.36 to 0.00).

                Forest plot from Cochrane Review

                Reference:
                Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                • OUTCOME 1.8: Catastrophizing at 6 to 12 months

                  Risk of bias of studies:

                  The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                  Narrative result:

                  Two RCTs with 282 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                  Relative effect or mean difference:

                  There was no statistically significant difference between groups (standardized mean difference 0.06, 95% CI -0.18 to 0.29).

                  Forest plot from Cochrane Review

                  Reference:
                  Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]

                  2.Cognitive behavioral versus treatment as usual

                  • Population, Intervention, Comparator

                    Population:

                    Adults (age range 40-64 years) with pain for more than 3 months, caused by a range of conditions including low back pain, osteoarthritis of the knee or rheumatoid arthritis

                    Intervention:

                    Cognitive behavioral therapy involving a psychologist in delivery (no further details available)

                    Comparator:

                    Waiting list or standard care (no active treatment such as those described in previous comparison)

                  • OUTCOME 2.1: Pain at post-treatment

                    Risk of bias of studies:

                    The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                    Narrative result:

                    16 RCTs with 1148 participants found that CBT reduced the proportion of people with pain compared with usual treatment/waitlist. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                    Relative effect or mean difference:

                    There was a statistically significant difference between groups, in favor of CBT (standardized mean difference -0.21, 95% CI -0.37 to -0.05).

                    Forest plot from Cochrane Review

                    Reference:
                    Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                    • OUTCOME 2.2: Pain at 6 to 12 months

                      Risk of bias of studies:

                      The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                      Narrative result:

                      Seven RCTs with 635 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                      Relative effect or mean difference:

                      There was no statistically significant difference between groups (standardized mean difference -0.09, 95% CI -0.25 to 0.08).

                      Forest plot from Cochrane Review

                      Reference:
                      Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                      • OUTCOME 2.3: Disability at post-treatment

                        Risk of bias of studies:

                        The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                        Narrative result:

                        15 RCTs with 1105 participants found that CBT reduced the proportion of people with disability compared with usual treatment/waitlist. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                        Relative effect or mean difference:

                        There was a statistically significant difference between groups, in favor of CBT (standardized mean difference -0.26, 95% CI -0.47 to -0.04).

                        Forest plot from Cochrane Review

                        Reference:
                        Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                        • OUTCOME 2.4: Disability at 6 to 12 months

                          Risk of bias of studies:

                          The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                          Narrative result:

                          Six RCTs with 450 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                          Relative effect or mean difference:

                          There was no statistically significant difference between groups (standardized mean difference -0.13, 95% CI -0.51 to 0.25).

                          Forest plot from Cochrane Review

                          Reference:
                          Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                          • OUTCOME 2.5: Mood at post-treatment

                            Risk of bias of studies:

                            The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                            Narrative result:

                            12 RCTs with 899 participants found that CBT reduced the proportion of people with negative mood compared with usual treatment/waitlist. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                            Relative effect or mean difference:

                            There was a statistically significant difference between groups, in favor of CBT (standardized mean difference -0.38, 95% CI -0.57 to -0.18).

                            Forest plot from Cochrane Review

                            Reference:
                            Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                            • OUTCOME 2.6: Mood at 6 to 12 months

                              Risk of bias of studies:

                              The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                              Narrative result:

                              Seven RCTs with 637 participants found that CBT reduced the proportion of people with negative mood compared with usual treatment/waitlist. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                              Relative effect or mean difference:

                              There was a statistically significant difference between groups, in favor of CBT (standardized mean difference -0.26, 95% CI -0.51 to -0.00).

                              Forest plot from Cochrane Review

                              Reference:
                              Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                              • OUTCOME 2.7: Catastrophizing at post-treatment

                                Risk of bias of studies:

                                The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                Narrative result:

                                Five RCTs with 308 participants found that CBT reduced the proportion of people with catastrophic thinking compared with usual treatment/waitlist. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                Relative effect or mean difference:

                                There was a statistically significant difference between groups, in favor of CBT (standardized mean difference -0.53, 95% CI -0.76 to -0.31).

                                Forest plot from Cochrane Review

                                Reference:
                                Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                • OUTCOME 2.8: Catastrophizing at 6 to 12 months

                                  Risk of bias of studies:

                                  The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                  Narrative result:

                                  One RCT with 59 participants found no statistically significant difference between groups. The reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                  Relative effect or mean difference:

                                  There was no statistically significant difference between groups (standardized mean difference -0.22, 95% CI -0.73 to 0.29).

                                  Forest plot from Cochrane Review

                                  Reference:
                                  Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]

                                  3.Behavioral therapy versus active control

                                  • Population, Intervention, Comparator

                                    Population:

                                    Adults (age range 45-53 years) with pain for more than 3 months, caused by a range of conditions including low back pain, osteoarthritis of the knee or rheumatoid arthritis

                                    Intervention:

                                    Behavioral therapy involving a psychologist in delivery (no further details given)

                                    Comparator:

                                    Active control defined as a protocolized treatment which engaged the patient, such as an exercise program, a medical procedure, an education program, a support group or a self-instruction booklet

                                  • OUTCOME 3.1: Pain at post-treatment

                                    Risk of bias of studies:

                                    The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                    Narrative result:

                                    One RCT with 77 participants found no statistically significant difference between groups. The reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                    Relative effect or mean difference:

                                    There was no statistically significant difference between groups (standardized mean difference -0.02, 95% CI -0.47 to 0.43).

                                    Forest plot from Cochrane Review

                                    Reference:
                                    Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                    • OUTCOME 3.2: Pain at 6 to 12 months

                                      Risk of bias of studies:

                                      The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                      Narrative result:

                                      One RCT with 73 participants found no statistically significant difference between groups. The reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                      Relative effect or mean difference:

                                      There was no statistically significant difference between groups (standardized mean difference 0.03, 95% CI -0.43 to 0.49).

                                      Forest plot from Cochrane Review

                                      Reference:
                                      Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                      • OUTCOME 3.3: Disability at post-treatment

                                        Risk of bias of studies:

                                        The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                        Narrative result:

                                        Two RCTs with 148 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                        Relative effect or mean difference:

                                        There was no statistically significant difference between groups (standardized mean difference -0.26, 95% CI -0.58 to 0.07).

                                        Forest plot from Cochrane Review

                                        Reference:
                                        Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                        • OUTCOME 3.4: Disability at 6 to 12 months

                                          Risk of bias of studies:

                                          The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                          Narrative result:

                                          Two RCTs with 144 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                          Relative effect or mean difference:

                                          There was no statistically significant difference between groups (standardized mean difference -0.17, 95% CI -0.50 to 0.16).

                                          Forest plot from Cochrane Review

                                          Reference:
                                          Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                          • OUTCOME 3.5: Mood at post-treatment

                                            Risk of bias of studies:

                                            The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                            Narrative result:

                                            One RCT with 71 participants found no statistically significant difference between groups. The reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                            Relative effect or mean difference:

                                            There was no statistically significant difference between groups (standardized mean difference -0.47, 95% CI -0.94 to 0.00).

                                            Forest plot from Cochrane Review

                                            Reference:
                                            Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                            • OUTCOME 3.6: Mood at 6 to 12 months

                                              Risk of bias of studies:

                                              The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                              Narrative result:

                                              One RCT with 71 participants found no statistically significant difference between groups. The reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                              Relative effect or mean difference:

                                              There was no statistically significant difference between groups (standardized mean difference -0.37, 95% CI -0.84 to 0.10).

                                              Forest plot from Cochrane Review

                                              Reference:
                                              Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                              • OUTCOME 3.7: Catastrophizing at post-treatment

                                                Risk of bias of studies:

                                                The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                                Narrative result:

                                                Two RCTs with 146 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                                Relative effect or mean difference:

                                                There was no statistically significant difference between groups (standardized mean difference -0.28, 95% CI -0.60 to 0.05).

                                                Forest plot from Cochrane Review

                                                Reference:
                                                Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                                • OUTCOME 3.8: Catastrophizing at 6 to 12 months

                                                  Risk of bias of studies:

                                                  The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                                  Narrative result:

                                                  One RCT with 73 participants found no statistically significant difference between groups. The reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                                  Relative effect or mean difference:

                                                  There was no statistically significant difference between groups (standardized mean difference -0.06, 95% CI -0.52 to 0.40).

                                                  Forest plot from Cochrane Review

                                                  Reference:
                                                  Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]

                                                  4.Behavioral therapy versus treatment as usual

                                                  • Population, Intervention, Comparator

                                                    Population:

                                                    Adults (age range 35-52 years) with pain for more than 3 months, caused by a range of conditions including low back pain, osteoarthritis of the knee or rheumatoid arthritis

                                                    Intervention:

                                                    Behavioral therapy involving a psychologist in delivery (no further details available)

                                                    Comparator:

                                                    Waiting list or standard care (no active treatment such as those described in previous comparison)

                                                  • OUTCOME 4.1: Pain at post-treatment

                                                    Risk of bias of studies:

                                                    The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                                    Narrative result:

                                                    Five RCTs with 484 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                                    Relative effect or mean difference:

                                                    There was no statistically significant difference between groups (standardized mean difference -0.27, 95% CI -0.79 to 0.24).

                                                    Forest plot from Cochrane Review

                                                    Reference:
                                                    Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                                    • OUTCOME 4.2: Pain at 6 to 12 months

                                                      Risk of bias of studies:

                                                      The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                                      Narrative result:

                                                      Two RCTs with 182 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                                      Relative effect or mean difference:

                                                      There was no statistically significant difference between groups (standardized mean difference -0.03, 95% CI -0.32 to 0.26).

                                                      Forest plot from Cochrane Review

                                                      Reference:
                                                      Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                                      • OUTCOME 4.3: Disability at post-treatment

                                                        Risk of bias of studies:

                                                        The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                                        Narrative result:

                                                        Five RCTs with 504 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                                        Relative effect or mean difference:

                                                        There was no statistically significant difference between groups (standardized mean difference -0.41, 95% CI -0.98 to 0.16).

                                                        Forest plot from Cochrane Review

                                                        Reference:
                                                        Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                                        • OUTCOME 4.4: Disability at 6 to 12 months

                                                          Risk of bias of studies:

                                                          The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                                          Narrative result:

                                                          Three RCTs with 336 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                                          Relative effect or mean difference:

                                                          There was no statistically significant difference between groups (standardized mean difference -0.54, 95% CI -1.51 to 0.44).

                                                          Forest plot from Cochrane Review

                                                          Reference:
                                                          Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                                          • OUTCOME 4.5: Mood at post-treatment

                                                            Risk of bias of studies:

                                                            The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                                            Narrative result:

                                                            Three RCTs with 278 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                                            Relative effect or mean difference:

                                                            There was no statistically significant difference between groups (standardized mean difference -0.53, 95% CI -1.42 to 0.35).

                                                            Forest plot from Cochrane Review

                                                            Reference:
                                                            Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                                            • OUTCOME 4.6: Mood at 6 to 12 months

                                                              Risk of bias of studies:

                                                              The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                                              Narrative result:

                                                              Two RCTs with 160 participants found no statistically significant difference between groups. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                                              Relative effect or mean difference:

                                                              There was no statistically significant difference between groups (standardized mean difference -0.65, 95% CI -2.07 to 0.77).

                                                              Forest plot from Cochrane Review

                                                              Reference:
                                                              Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]
                                                              • OUTCOME 4.7: Catastrophizing at post-treatment

                                                                Risk of bias of studies:

                                                                The reviewers did not perform a GRADE assessment of the quality of the evidence. Overall around 60% of the studies were at unclear risk of selection and detection bias and incomplete outcome data was properly addressed in only 25% of the studies.

                                                                Narrative result:

                                                                Three RCTs with 269 participants found that behavioral therapy reduced the proportion of people with catastrophic thinking compared with control. As the studies used different scales to assess this outcome, the reviewers calculated a standardized mean difference. These are hard to interpret clinically but rules of thumb in their interpretation suggest that 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect (Cohen J. Statistical Power Analysis in the Behavioral Sciences (2nd edition). Hillsdale (NJ): Lawrence Erlbaum Associates, Inc., 1988).

                                                                Relative effect or mean difference:

                                                                There was a statistically significant difference between groups, in favor of behavioral therapy (standardized mean difference -0.72, 95% CI -1.43 to -0.01).

                                                                Forest plot from Cochrane Review

                                                                Reference:
                                                                Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]

                                                                Additional Information:

                                                                DOI

                                                                10.1002/cca.318

                                                                Publication Dates

                                                                1. Published Online: 19 MAR 2014

                                                                CCA derived from

                                                                Williams ACDC, Eccleston C, Morley S. Psychological therapies for the management of chronic pain (excluding headache) in adults. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD007407. DOI: 10.1002/14651858.CD007407.pub3. [Review search date: September 2011]

                                                                How to Cite

                                                                In adults with chronic pain (excluding headache), what are the benefits and harms of psychological therapies? Adarsh Gupta (DO, MS, FACOFP) (on behalf of Cochrane Clinical Answers Editors). Cochrane Clinical Answers 2012. DOI: 10.1002/cca.318.

                                                                Further Information

                                                                • CCA Associate editor: Adarsh Gupta (DO, MS, FACOFP), Associate Professor, ROWAN-SOM, Stratford, NJ, USA.
                                                                • CCA Editor: Karen Pettersen. Correspondence to kpettersen@wiley.com.