Cummings, G.G., Olivo, S.A., Biondo, P.D., Stiles, C.R., Yurtseven, O., Fainsinger, R.L., & Hagen, N.A. (2011). Effectiveness of knowledge translation interventions to improve cancer pain management. Journal of Pain and Symptom Management, 41, 915–939.

To determine the effectiveness of interventions to facilitate uptake of new knowledge by healthcare practitioners, patients, and family caregivers to improve cancer pain management. Primary outcomes of interest were change in behavior or practice of healthcare professionals or patients and caregivers. Secondary outcomes included change in patient quality of life or satisfaction with treatment.

Databases searched were CINAHL, MEDLINE, EMBASE, AMED, Web of Science, Cochrane Database of Clinical Trials and Cochrane Database of Systematic Reviews, National Cancer Institute of Canada, Canadian Cancer Society, American Cancer Society, American Society of Clinical Oncology, European Association for Palliative Care, and other websites.

An extensive listing of specific search terms and operators is provided. Main search terms included pain, analgesia, cancer, practice guidelines, evidence based medicine, and organizational innovation.

Studies were included in the review if they

• Were randomized controlled trials, controlled clinical trials, time series, and pre-post studies that evaluated the effect of knowledge translation interventions on patient outcomes
• Included patients with cancer.

Case reports, cross-sectional studies, or noncontrolled pre-post studies were excluded.

• The initial search yielded 15,625 articles.
• A total of 308 articles were retrieved and reviewed according to inclusion criteria.
• Risk of bias assessment was done using methods of the Cochrane Effective Practice and Organization of Care Group.

A final sample of 26 studies was included: 16 examined interventions for patients only, 4 focused on family caregivers, and 5 investigated effects of interventions targeted to health professionals. One study compared a solo intervention for patients only, and a dyad intervention for patients and their significant others, within the same study.

• Of 26 studies, 25 were rated as having high risk of bias due to lack of blinding in the design, incomplete data, lack of allocation concealment, and other issues. Inter-rater reliability of this scoring showed only slight agreement between raters.

Interventions targeting health professionals (5 studies)

• Findings related to patient outcomes were mixed, with two studies showing decrease in pain intensity and two studies showing no change in patients’ pain.
• Two studies showed significant increases in healthcare professionals’ knowledge and attitudes.

Interventions targeting patients and caregivers (21 studies)

• Most interventions were face-to-face coaching sessions and distribution of printed material. In more than half of the studies, in-person or telephone follow-up was used to reinforce education.
• Eight studies showed significant decrease in patients’ pain intensity with the intervention, and five showed positive change in patient behavior regarding pain control strategies.
• Five studies targeting patients and family caregivers showed mixed results, with two showing improvement in pain intensity and adherence to therapy, and three showing no effect.

Meta-analysis results

     Interventions targeting patients and caregivers

• There was no significant effect of interventions on pain interference with activities.
• In six studies, the effect on usual/average pain showed a tendency favoring the intervention (SMD = 0.43, 95% CI 0.13 – 0.74, effect = 2.76, p = 0.006). These findings are limited by the high heterogeneity among studies.
• There was no significant overall effect of interventions on worst pain, current pain, or overall pain measures in meta-analysis.
• There was a significant effect of the intervention on least pain intensity (SMD = 0.93, 95% CI 0.44 – 0.142, p = 0.0002), but this was from only two studies.

Intervention dose and intensity

• It is noted that the “dose” and intensity of knowledge translation interventions varied greatly. Authors define high- and low-dose parameters and show that those studies with higher dosage of the intervention tended to report significant results on patient outcomes.

Knowledge translation interventions aimed at healthcare professionals have demonstrated change in practitioner behavior and attitude, but have not demonstrated an effect on patient pain outcomes measured. There is some evidence to suggest that higher intensity and “dose” of these interventions, indicating amount of time spent in education, inclusion of follow-up, patient/family contact, and longer duration of follow- up, tend to be more effective.

Knowledge translation interventions aimed at patients and caregivers have been associated with improvement in patients’ least and average pain scores, but not other pain measures. These findings need to be viewed with caution, since studies included showed high heterogeneity, and the timing, dosage, and intensity of the interventions varied.

Educational and knowledge translation interventions regarding pain management that are aimed at the patient and caregiver have been shown to be effective in improving some measures of pain intensity, and have often, but not always, resulted in improved knowledge, attitudes, and skills related to pain management. Elements of success included more intensive education such as involvement of a multidisciplinary team in face-to-face education, locally constructed education materials, standardized approaches, and patient follow-up to reinforce the education. Pain management educational programs should be constructed to include key elements that are likely to improve outcomes.