Study Finds Evidence-Based Interventions Added to Screening Did Not Improve Cancer Pain

Get Permission

When added to regular screening and guidelines, evidence-based interventions implemented at the patient, health professional, and service levels did not significantly improve cancer pain, according to research led by Melanie Lovell, MBBS, PhD, Medical Head of Palliative Care at HammondCare and Clinical Associate Professor at the University of Sydney, Australia. When compared with screening and guidelines alone, these interventions did not improve the proportion of patients with at least a 30% reduction in pain, she reported at the 2021 Multinational Association of Supportive Care in Cancer/International Society of Oral Oncology (MASCC/ISOO) Annual Meeting.1

Melanie Lovell, MBBS, PhD

Melanie Lovell, MBBS, PhD

“There is an ongoing evidence-practice gap in cancer pain management, which is causing morbidity and disability for people living with cancer worldwide,” she commented. “Giants in our profession have worked to bridge this gap, but success remains elusive in the ambulatory care setting.”

The Australian Cancer Pain Management Guideline provides a clinical pathway and evidence for the optimal management of cancer pain. However, as Dr. Lovell pointed out, guidelines require implementation strategies to ensure the uptake of recommendations and drive guideline-recommended practice change at the system, health professional, and patient levels.2 Interventions designed to implement clinical pathways, however, have been met with variable success.

Cluster Trial Design

In this study, Dr. Lovell and her colleagues sought to evaluate the efficacy and cost-effectiveness of a suite of strategies designed to implement the Australian Clinical Practice Guidelines and aimed at improving pain outcomes for patients in metropolitan, regional, and rural oncology as well as palliative care outpatient settings. They evaluated adult outpatients who had moderate unrelieved cancer pain, with scores of 5 or higher (out of 10) on the Numeric Rating Scale (NRS) for pain.

“We know this group has more interference with quality of life than those with lower pain scores,” Dr. Lovell noted.

The intervention comprised evidence-based components: health professional education was delivered virtually, patient-held resources were developed from a systematic review of randomized controlled trials of educational strategies, audit and feedback was conducted at the system level, and screening for pain was implemented as a clinical tool in both phases of the study (training and intervention). The primary outcome was the percentage of people with a 30% reduction in the NRS score for worst pain at 1 week.

Patients were recruited from six ambulatory oncology and palliative care services (“clusters”). The study was conducted from February 2016 to May 2019, and each site served as its own control. During the first phase, patients were routinely screened for pain; during the intervention phase, “clinical champions” and staff at each center were provided with tailored implementation strategies to use in accordance with their service’s needs.

A total of 8,099 patients were screened for pain. Of note, 25% had NRS scores ≥ 5 for worst pain. In the control phase, the final cohort consisted of 359 participants, of whom 280 had an NRS score ≥ 5; in the intervention phase, 264 of 329 patients had an NRS score ≥ 5.

No Significant Differences Observed

According to Dr. Lovell, the percentage of respondents with a reduction in NRS score for worst pain of more than 30% was not significantly different between the groups.

“About one-third of patients had improvements in worst pain NRS scores in both phases, and about half had improvements in average pain in both phases,” she said. Additionally, mean worst pain improved by ≥ 0.5 standard deviation at week 1 for those with an NRS score ≥ 5 but not at other time points. There were also no significant differences between groups in the secondary outcome measures of quality of life, patient empowerment, or caregiver experience.

The completion of health professional training varied among sites, from 12% to 74%. The proportion of patients receiving written information was 30% (20%–44%) in the intervention arm and 22% (2%–30%) in the control arm.


  • When added to screening and guidelines, evidence-based interventions did not significantly improve cancer pain when compared with screening and guidelines alone.
  • Even in centers with clinical champions using evidence-based strategies, achieving better pain management outcomes in cancer clinical practice remains a challenge.
  • Interventions should be adapted and tailored for each specific cancer population.

“We hypothesized reasons for these results,” she said. “The intervention dose may have been inadequate, fidelity to the intervention may have been insufficient, and the complexity of the intervention may have required a more intensive implementation approach than was achieved.”

According to Dr. Lovell, the reduction in pain seen in the control group may be attributed to factors such as the Hawthorne effect (alteration of behavior by subjects in a study due to their awareness of being observed) or good pain management at baseline. “It also may have been due to the impact of screening, although previous studies would not support that hypothesis or the fact that pain simply gets better over time,” she added. “Additionally, participating centers had a high baseline knowledge of pain guidelines.”

Lessons Learned and Future Steps

Dr. Lovell mentioned that major lessons were learned during the process of study implementation—notably, that health systems are dynamic—and this needs to be factored into study designs.

“The patient-reported outcome measure collection was seen as research rather than a clinical practice tool, partly because it focused on pain alone, and this resulted in considerable confusion,” she noted. The study investigators also concluded that the provision of patient self-management tools needs to be embedded into the workflow. “There was a lack of nurses in ambulatory care, and this was a barrier to patient-reported outcome collection and patient education,” Dr. Lovell said. “We learned that implementation research is messy, complicated, and requires perseverance. It’s very different from clinical trials of anticancer agents.”

According to the investigators, cancer pain screening and subsequent management will need to be highly valued across all levels of an organization, supported by adequate resources, embedded within the clinical workflow, and enabled by electronic data systems to be successful. Partnerships—among government, service providers, primary health, universities, trial groups, and consumers—are needed for robust evaluation of sustainable implementation. 

DISCLOSURE: Dr. Lovell reported no conflicts of interest.


1. Lovell MR, Phillips JL, Agar A, et al: A cluster randomised controlled trial of guidelines and screening with implementation strategies versus guidelines and screening alone to improve pain in adults with cancer. 2021 Multinational Association of Supportive Care in Cancer/International Society of Oral Oncology Annual Meeting. Presented June 24, 2021.

2. Jacobsen PB, Snyder CF: Improving pain assessment and management in routine oncology practice. J Clin Oncol 36:1272-1274, 2018.