This is one of a very few national surveys of physicians for which methods, response rates, etc., make for valid results that can be trusted by health care professionals. It is also likely the only national survey of physicians that has looked specifically at the issue of quality of care.

Quality and quality improvement is certainly a key priority in health policy these days, and one of the challenges is really in the adoption of a scientific method (quality improvement as developed by Deming and others over 30 years ago) by the medical profession—this has been very slow indeed, and the profession and other healthcare leaders continue to struggle with this challenge. So anyone studying, working, writing or publishing on issues that relate to quality, barriers to quality, and potential solutions, would hopefully find the results of the survey we conducted quite relevant.

The results that we report in the paper represent new information about physicians' attitudes, which reflects self-reported behavior with regard to their using quality improvement tools and techniques in their office practice. Such data, from a nationally representative sample of practicing physicians has not been reported (to my knowledge) before, and certainly not in a scientifically valid way.

Health care purchasers, accrediting organizations, and consumer advocates are among the stakeholders currently using quality improvement (QI) methods to improve patient care. But there is still one key group for whom the pursuit of QI has not become routine: physicians.

To date, QI has not permeated the culture of professional medicine. Only one-third of doctors have been involved in any redesign efforts aimed at improving quality-of-care in their practice. Just a third, moreover, have access to any data about the quality of their own clinical performance, while seven of ten physicians do not feel the public should have access to quality-of-care data.

The survey also revealed surprisingly low use of electronic medical records (EMRs): only about a quarter (27%) of doctors reported using an EMR routinely or occasionally.

No Data, No QI:
A large part of the lag in QI among physicians seems to be that they lack essential data about their own practices.

In the survey, less than half of doctors reported they could easily identify patients by age group (49%) or diagnosis (44%). And most said it would be difficult or impossible to generate data about patients with abnormal lab results (83%) or to identify patients taking high-risk medications that might require follow-up care (84%). Only 33% of physicians surveyed said they receive any data about the quality of care they provide (patient surveys were the data source doctors most commonly cited). Doctors also have problems getting performance information when they refer patients for specialized care: nearly two-thirds (64%) say they rarely or never have access to such information.

Reluctance to Share Information about their Quality of Care:
In addition to not using data themselves, physicians are generally reluctant to share performance-level data with others. While nearly three-quarters of physicians agreed that clinical performance data should be shared with the medical leadership of their health systems, they were not as likely to share information with patients or the general public. 55% of doctors felt such data should be shared with patients, and less than a third (29%) felt the general public should have access to quality-of-care data.

Practice Size Matters:
Practice size figured prominently throughout the survey. Physicians in large practices (i.e., 50 or more doctors) were more likely to be engaged in quality improvement. Why are these providers more up-to-speed? Collecting and analyzing data, and then using the information to implement change requires resources. Physicians in large group practices, as well as salaried physicians, might have more financial flexibility and access to capital and thus be in a better position to implement both the measurement and the improvement parts of the QI cycle. Organizational culture and management may also play a role.

The organization my colleagues and I work for, i.e., the Commonwealth Fund, is a private foundation that specializes in health care. Several of our programs target quality-of-care for various populations—children, elders in nursing homes, underserved populations.

In order to develop strategies to impact quality-of-care, it is essential that we identify the key levers that will lead to action and change. A great deal of knowledge that has been accumulated about quality has come from organizational settings—hospitals, nursing homes, integrated networks of care.

Very little is known about the professional’s perspective on quality, and even less about physicians vs. other health professions such as nurses, pharmacists, social workers, dieticians, etc. So we identified this gap in knowledge as one that needed to be explored. Indeed, if physicians do not apply quality improvement in their daily practice, whatever the setting may be, it is very unlikely that any progress will be made.

As to the problems we encountered, some are quite interesting indeed. First, getting an adequate response rate from physicians required careful planning (i.e., length of survey, framing of questions). Then we had to craft questions to avoid social desirability bias—since questions about attitudes or behaviors are self-reported. Those were not that difficult to address.

What was very surprising were the difficulties we encountered in the peer review process. The peer reviewers’ comments (as well as editors) were very questioning about quality improvement—and about the importance of our findings in terms of their implications for the practice of medicine (the physician audience), and for policy (the public and policy makers audience). These reservations supported our point even more strongly, i.e., that the science of QI, although well-described and established, has not diffused in health care yet.

So, after carefully and intently going through multiple rounds of reviews and revisions and resubmission, we finally were successful in publishing this paper, along with two others, one that focused on information technologies and their adoption by the medical profession for QI, and the other on patient-centered care practices. What is interesting is that we now hear that the IT paper is also one which is considered of strongest scientific validity by colleagues who need to use the results in their work—for example, the Office of the National Coordinator for Health Information Technology (ONCHIT) led by David Brailer, has been charged by the President to report on the progress towards establishing an IT infrastructure in the US. They will conduct a national survey, and are using results from The Commonwealth Fund 2003 National Survey of Physicians and Quality of Care as a baseline.

I think that the results we report do pose critical questions about the medical profession (physicians) and the practice of medicine. One of our findings was that practice size had a significant impact on physicians’ engagement in quality improvement. And we uncovered quite a significant gap between solo and small-group physicians and large-group physicians.

We also know that in the US, the majority of care is delivered by solo and small group MDs. So one cannot avoid the issue of what these results might mean in terms of changing physician practices. Diffusion of knowledge, of best practices, is a huge challenge when one thinks of affecting thousands of physicians, one at a time. So many are struggling with how to connect physicians in order to diffuse innovations more rapidly and have an impact on practice and quality.

Professional organizations will have to deal with this issue—and some are evolving from a role more akin to "trade or advocacy organizations" representing the interests of their members, to provide services such as technical assistance and interactive training.

The socialization of the medical profession will thus likely evolve and adapt to the reality of practice and to new models of improvement. The role of government and the private sector in fostering quality is another important issue that our results bring forth—not only in terms of technical assistance and training, but also in terms of investment in a quality infrastructure that will support physicians’ practice. Who should invest in tools that physicians need in order to provide better care—tools whose impact has been demonstrated by scientific evidence?

Payment policies are another area that our results suggest need to be reformed. In the US, we pay for care based on a service rendered, regardless of the quality of that service. And that has led to numerous problems, overuse of inappropriate services, etc. Physicians reported that the quality of care they provide was rarely a criteria on which they were paid, and that "productivity" (i.e., number of patients seen each day) was the dominant factor affecting their compensation. How physicians and other providers of care are paid is thus a crucial factor that is hotly debated.

Finally, the issue of competence and how we assess physician competence will be key. Physicians reported that they use personal information or professional "reputation" to select MDs when they refer their patients to other MDs. Also, if we are to pay for "performance," this means that we will need to have a valid, fair method to assess skills, a knowledge of physicians vs. relying on qualitative, non-validated, "impressions."

This has significant social implication for the role of the medical profession—it will have to evolve into a much more "transparent" profession, which shares information about the quality of their practice vs. a more "closed" guild that self assesses in a "black box."

This may mean some re-visiting of the Hippocratic Oath and its interpretation—the confidentiality of the patient-physician compact is still of utmost importance, and should not be violated, nor does it need to be, by a demand for public information about quality and outcomes of care.

Anne-Marie J Audet, MD, MSc
Vice President, Quality Improvement
Commonwealth Fund
New York, NY, USA