I've been struggling in the quagmire of performance measures. It is both a complex and emotional issue that threatens our sense of who we are and what we do as physicians. We want the best for our patients and try to provide it within the constraints of our practices. We know we need to improve care and must measure what we do in order to bring about change. Before measuring anything, however, we must ask why, what, how, who and what to do with the results.
The “why” is easy. We don't do as well as we think we do. Some years ago, I checked how well my practice was controlling blood pressure in clinic patients. I found that not only were we not achieving the desired goals, we overestimated our individual and group performance. Our current knowledge about errors, less than optimal care and misuse contributing to excess morbidity and mortality has only become apparent because of measurement.
The “what” should also be simple. We should measure what is important. But important to whom? Physicians want good outcomes for patients. Patients want a good experience as well as good outcomes. Health systems want lower costs while demonstrating quality and safety.
Then, there is the dilemma of what is important and measurable. It is very easy to choose things that are easily measurable and give them an importance greater than their intrinsic value. It is also easy to ignore the factors that are most valuable because they are so hard to measure. Effective screening programs use measures that are accurate, affordable, and relevant to conditions with high prevalence, morbidity or mortality and reflect a component of the disorder that can be modified to promote benefit. For improved patient outcomes, patient experience and population health, we need measures with all these qualities but also ones that are patient-centered and respect patient values.
For information to be used effectively, it must be collected, tabulated and analyzed, making the value of an electronic database self-evident. However, the database in most of the available electronic health record (EHR) systems is not structured for easy entry, analysis or retrieval. Important information is often entered as text, which is unavailable for group reporting or simply excluded because it does not fit into the EHR structure. We need better ways to document what is important in a format that can be analyzed.
Another question is who should be doing the measuring and who should be accountable for improving outcomes. As a profession we have the responsibility for maintaining our own standards, so we must be involved in not just choosing what to measure but also how to institute changes. However, health and patient care are not the purview of physicians alone. As we collaborate in shared decision making and care delivery with other health care professionals, payers and patients, so must all be accountable for achieving the better health outcomes and patient experience we all desire.
Public reporting of the performance of individual physicians and institutions has yet to be evaluated. There is little evidence that paying for performance efforts improves quality. There is great concern about the inherent undesirable consequences, such as overemphasis on measured values to the detriment of others and de-selection of patients with less than optimal outcomes.
On the positive side, feeding back information about findings to those who can effect change has proven value. In our clinic, the knowledge of our less-than-optimal performance and a few tweaks to our system did improve blood pressure control for a large number of patients, and we were able to institutionalize some of the changes.
The College formed the Performance Measurement Committee in 2004 to ensure that evidence-based standards are used to develop nationally endorsed performance measures. The original policy paper on the subject “The Use of Performance Measurements to Improve Physician Quality of Care,” provides guidance for the research, development and implementation of performance measures as well as support for demonstration projects that reimburse physicians and other health professionals for reporting on quality measures.
Subsequent papers provide guidance for implementation of pay-for-performance programs, physician ratings related to performance measurement and publication of price and performance measure data. In March 2010, ACP published a paper in Annals of Internal Medicine on “Pay for Performance through the Eyes of Medical Professionalism” that examined the extent to which these approaches are compatible or in conflict with one another.
Now, in 2011, the ACP Board of Regents believes this issue needs a new look. College governance has suggested that all measures should be patient-centered and evidence-based; should be regularly reevaluated and revised; should focus on outcomes, not processes; should promote value, not volume; and should allow for shared accountability as models of health care delivery evolve.
Despite the daunting nature of this issue, I am persuaded that if we don't evaluate our performance we won't improve. But to evaluate means not only to measure what is measurable but also to reflect on what is important for patients.
Take this example. A study conducted by the Maori/Indigenous Health Institute (MIHI) in Otago, New Zealand, examined Maori patients' perceptions of primary care. Maori people, the original inhabitants of the land, have a rich language (te reo) that is not pronounced as it is written and thus frequently not pronounced correctly.
In the study, the participants were asked, “How do you know you are getting good health care?” Without hesitation, four focus groups answered consistently, “That is when the receptionist pronounces my name correctly.” The leader of the focus group was concerned that the patients had not understood that the issue being addressed was whether they thought they were getting appropriate tests and treatments, and asked the question again. The answer was, “When the medical receptionist, the nurse and the doctor pronounce my name correctly.” The project also addressed other issues relating to language and culture, but one clear conclusion, as Suzanne Pitama wrote in the February 2010 MIHI magazine, was that for many Maori patients “[T]he use of and respect for te reo was seen as an indicator of quality health service provision.”
We know that respect is essential for connecting with our patients, and connection is more than half the challenge we face in gaining insight into their problems and helping them adhere to treatment recommendations and participate in programs to improve care. Even with shorter and shorter scheduled visit times, making these connections is one of the most useful things any of us can do for our patients and may have the most lasting effect on their health outcomes and our satisfaction. Now if only we could measure it.