Effect of Cancer Pain Guideline Implementation on Pain Outcomes Among Adult Outpatients With Cancer-Related Pain: A Stepped Wedge Cluster Randomized Trial

IMPORTANCE: An evidence-practice gap exists for cancer pain management, and cancer pain remains prevalent and disabling. OBJECTIVES: To evaluate the capacity of 3 cancer pain guideline implementation strategies to improve pain-related outcomes for patients attending oncology and palliative care outpatient services. DESIGN, SETTING, AND PARTICIPANTS: A pragmatic, stepped wedge, cluster-randomized, nonblinded, clinical trial was conducted between 2014 and 2019. The clusters were cancer centers in Australia providing oncology and palliative care outpatient clinics. Participants included a consecutive cohort of adult outpatients with advanced cancer and a worst pain severity score of 2 or more out of 10 on a numeric rating scale (NRS). Data were collected between August 2015 and May 2019. Data were analyzed July to October 2019 and reanalyzed November to December 2021. INTERVENTIONS: Guideline implementation strategies at the cluster, health professional, and patient levels introduced with the support of a clinical champion. MAIN OUTCOMES AND MEASURES: The primary measure of effect was the percentage of participants initially screened as having moderate to severe worst pain (NRS ≥ 5) who experienced a clinically important improvement of 30% or more 1 week later. Secondary outcomes included mean average pain, patient empowerment, fidelity to the intervention, and quality of life and were measured in all participants with a pain score of 2 or more 10 at weeks 1, 2, and 4. RESULTS: Of 8099 patients screened at 6 clusters, 1564 were eligible, and 359 were recruited during the control phase (mean [SD] age, 64.2 [12.1] years; 196 men [55%]) and 329 during the intervention phase (mean [SD] age, 63.6 [12.7] years; 155 men [47%]), with no significant differences between phases on baseline measures. The mean (SD) baseline worst pain scores were 5.0 (2.6) and 4.9 (2.6) for control and intervention phases, respectively. The mean (SD) baseline average pain scores were 3.5 (2.1) for both groups. For the primary outcome, the proportions of participants with a 30% or greater reduction in a pain score of 5 or more of 10 at baseline were similar in the control and intervention phases (31 of 280 participants [11.9%] vs 30 of 264 participants [11.8%]; OR, 1.12; 95% CI, 0.79-1.60; P = .51). No significant differences were found in secondary outcomes between phases. Fidelity to the intervention was low. CONCLUSIONS AND RELEVANCE: A suite of implementation strategies was insufficient to improve pain-related outcomes for outpatients with cancer-related pain. Further evaluation is needed to determine the required clinical resources needed to enable wide-scale uptake of the fundamental elements of cancer pain care. Ongoing quality improvement activities should be supported to improve sustainability.