QOPI Measures Identify Areas for Quality Improvement in Oncology Practice
Los Angeles, CA—Value is the preeminent goal in cancer care, because it unites all of the interests of the stakeholders, including patients, said Douglas W. Blayney, MD, Medical Director, Stanford Cancer Institute, CA, at the Fourth Annual Conference of the Association for Value-Based Cancer Care. American Society of Clinical Oncology (ASCO)’s Quality Oncology Practice Initiative (QOPI) is providing oncology practices with the means to improve quality care through the use of quality measures toward the improvement of patient outcomes.
The 3 elements of quality are structure, process, and outcome. Structure is the site of delivery, which includes the information infrastructure. Uniform processes, such as those for ordering chemotherapy, are important to quality care, because they aid efficiency, said Dr Blayney. In addition to survival, appropriate symptom relief and palliation are important outcomes to patients.
Because testing the effect of a cancer intervention on the most meaningful outcome—survival—may take years, quality of cancer care should focus on the processes, Dr Blayney said. “In oncology, we’re fortunate enough to have a rich evidence base upon which to make guidelines. We can measure adherence to guidelines, and then we can target interventions to improve adherence,” he said, pointing to the many evidence-based guidelines issued by the National Comprehensive Cancer Network.
Poor quality processes can be defined by:
- Too little care (underuse)—the failure to provide an effective service when it would have provided a favorable outcome
- Too much care (overuse)—the provision of care when its risk of harm exceeds the potential benefit
- The delivery of wrong care (misuse).
- Core measures—care documentation, chemotherapy administration, pain management, smoking cessation, and psychosocial support
- Disease-specific module
- Domain-specific module—end-of-life care, symptom/toxicity management.
US Oncology measured the cumulative costs of care for non–small-cell lung cancer (NSCLC) by pathway status and found that over 12 months, “on-pathway” care costs slightly more than 50% of care that is “offpathway.” “It looks like a measure of overuse,” said Dr Blayney, “because the survival was exactly the same whether a patient was treated on or off pathway.”
QOPI was developed by ASCO to give clinicians a role in defining, measuring, and improving the quality of care. QOPI measures are oncologist-developed, evidence-based, guideline-based, and consensus-based, and are continually reviewed and updated by experts. They are grouped into several domains, including:
Since the spring of 2014, more than 25,000 patient charts have been voluntarily submitted to QOPI for analysis. From 2006 to 2010, 308 unique practice groups participated in at least 1 of 10 possible rounds of data collection, and QOPI shows that most practices “have maxed out on treatment process measures,” noted Dr Blayney. For example, adherence to breast cancer process measures has been approximately 95% over the past 5 years.
The rates of mutation testing in breast cancer, colorectal cancer, and NSCLC, as well as the appropriate use of anti-EGFR monoclonal antibodies have been improving, but gaps persist, Dr Blayney said.
QOPI data show a decline from 50% in 2006 to approximately 15% in 2010 in the use of chemotherapy in the last 2 weeks of life at the University of Michigan after a grand rounds presentation at the cancer center (Figure). The 15% rate is more in line with national aggregates, said Dr Blayney.
At Stanford University, using QOPI measures identified a deficiency in obtaining patient consent for chemotherapy.
The Michigan Oncology Quality Consortium is a quality collaborative funded by Blue Cross Blue Shield of Michigan, with the goal of improving the care of patients with cancer in Michigan by using data gathered as part of QOPI.
The consortium practices are small; most practices see less than 500 new patients annually and have fewer than 4 oncologists. Despite the small size of most participating practices, small practice groups could collaborate to measure adherence to the processes.
A compilation of the results over 5 QOPI data collection rounds found that, in general, adherence to the cancer treatment process for breast cancer, colorectal cancer, and NSCLC is generally good, but adherence to other care processes, such as symptom and toxicity management and end-of-life care, is not as good and did not change with measure sharing.
Much unwanted variation in measured adherence to many processes of care was discovered. Because it measures processes, participation in the Michigan consortium does not necessarily improve outcome, but compliance with supportive care was improved when practices were asked to use the Edmonton Symptom Assessment Scale during the patient encounter. “This shows what can be done in a statewide collaborative,” Dr Blayney said.
“The secret sauce here is that Blue Cross Blue Shield helped fund some of the infrastructure to do data extraction and also to hold quarterly meetings with these doctors in the state,” he concluded.