|Home | About | Journals | Submit | Contact Us | Français|
The second round of data collection for ASCO's Quality Oncology Practice Initiative (QOPI) is underway this month. By now, QOPI should be a familiar concept, as the JOP has featured it on several occasions. The results of the pilot round of quality measures were published in the JCO in September 2005 (Neuss MN. J Clin Oncol 23:6233-6239, 2005); and additionally, QOPI was discussed at the 2005 and 2006 ASCO Annual Meeting Education Sessions.
QOPI measures processes of care—one of the three components of quality, along with environment of care and outcomes of care (Donabedian A. JAMA 260:1743-1748, 1988). With a potentially fatal illness, most would define the highest quality as that which achieves the best survival. For that reason, we should focus on outcomes such as survival. However, with cancer, in which improvements may not show up in the survival measurements for years, quality improvement efforts must focus on assessing and improving care processes that have been previously demonstrated (either in randomized clinical trials or other methods) to enhance survival.
The QOPI instrument is designed for use by nonphysician abstractors who condense charts from 80 or more patients. Measures focus on colorectal cancer, breast cancer, lymphoma, and end-of-life care, among other topics. Data are entered without patient identifiers into an ASCO-maintained, secure Web server. Physician-specific, office-specific, or practice-specific reports are available and are benchmarked against the other QOPI participants, as well as the results from previous QOPI data collections. Results and the reactions of some practices are the subject of the JOP “Focus on Quality” article on page 251 of this issue. In the March 2006 round, data for more than 9,300 patients were entered. Clearly, if participation is a measure, QOPI is a success.
Criticisms of the QOPI that have emerged include the measures chosen, the current lack of validation of the data, and the inability to link to practice electronic health record systems. Many revolve around the measures chosen. QOPI measures were developed by a working group of predominately office-based, community practitioners. Because existing quality measures (e.g., Health Plan Employer Data and Information Set [HEDIS]) were poorly applicable to oncology care, the steering group and measure work group looked elsewhere. Some measures are driven by decades of cooperative group clinical trial evidence (adjuvant hormone therapy for node-positive, estrogen-receptor–positive breast cancer; radiation therapy for rectal cancer). Some come from the NICCQ measures (Malin JL. J Clin Oncol 24:626-634, 2006), while others were consensus choices (use of flow sheets, numeric pain scoring) and lack an evidence base.
QOPI measures were developed to be practical and easily extractable from the medical record, and since the practices are not compensated for the data abstraction, time for abstraction should be minimized. However, just because an attribute is easily measured, doesn't mean a measurement is a valid indicator of quality. If applications beyond quality improvement are considered for QOPI, increased focus is needed for identifying the most salient measures of quality. If the measures are publicly reported, we run the risk of focusing on the reported attributes, to the exclusion of other perhaps more salient measures of quality (talking to patients, for instance). The fact that oncologists (particularly those in clinical practice) are driving the process should provide some measure of comfort to critics.
Furthermore, there is currently no audit similar to the clinical trial audit process of QOPI data abstraction, though an audit plan is being developed. Without an audit, there is an opportunity for bias and for mischief. Clearly both criticisms should be addressed before QOPI, or any other quality measurement system, is linked to reimbursement.
QOPI is a great start. It is oncologist-driven and adaptable; the measurement process is scalable and can be accomplished in private practices and in large cancer centers. By taking action, ASCO's leadership and QOPI's leadership have given oncologists a head start on defining our own measures of quality. Further refinements should add additional tumor types beyond breast cancer, colorectal cancer, and lymphoma; should implement an audit or certification process to assure that self-reported measures are valid; and should incorporate the electronic medical record to capture data (see Correspondence on page 262 of this issue). It is time to provide health care that is predictably safe and evidence-based. Take for example aviation, which transitioned from leadership by swaggering “Top Gun” test pilots (with their 25% mortality rate), to today's orderly system of quality measures, checklists, read-backs, and safety systems (which increase safety for both pilots and passengers; Wachter RM. JAMA 295:2780-2783, 2006). In this way, oncology systems, too, require artful virtuosos who love their work and render a great public service. We owe our patients this continuing transition.