Skip Navigation U.S. Department of Health and Human Services
Agency for Healthcare Research Quality
Archive print banner

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

Ambulatory Care Quality Alliance: Invitational Meeting

Report of the Reporting Workgroup

Randy Johnson, Motorola
Nancy Nielsen, American Medical Association

Principles for Public Reports on Health Care

Randy Johnson stressed that employers want information about quality and said that the workgroup has made progress. Others are focused on cost, quality, and access, he said, while we're focused directly on reporting of quality and efficiency—both of which will affect cost. As we make improvements, he said, we will make progress on making data more available.

Johnson provided an overview on the status of the Principles for Public Reports on Health Care and the Principles for Reporting to Physicians and Hospitals.1 He said the goal is for data to be available that are meaningful and useful to patients. He added that the end users for the public reports include both patients and physicians contemplating referrals.

More broadly, Johnson said that the aim of the reports is to be comprehensive and to provide the greatest return on investment for making care safe and equitable. He said the workgroup's goals for 2006 include testing the reporting principles in the selected pilots and other ongoing local coalitions and discussing potential designs for a template for reporting information.

Regarding the format of the reports, Johnson said it was important that people focus on how patients think. The reports should address hospitals, physicians, and integrated delivery systems, he said. Johnson added that it was also important to take into account cultural factors and patient literacy.

Johnson stressed that the use of public reporting should support informed choice. These reports should be continually improved so they are increasingly effective, he said. They should also be as transparent as possible and allow for timely results. He also noted that the reports should employ standard measures so they can be used nationwide. And he said that when portraying performance differences, the methods should show significant differences.

Finally, Johnson said he hoped participants could come to consensus on the two sets of revised principles and he asked for feedback on the first two principles (regarding comprehensive reporting and consumer-friendly formats).


The discussion opened with a question about comprehensive reporting and the language that says the reports "should address" a list of items. Does this mean they should report in these areas, asked the participant, who expressed concern that the wording did not encompass everything that needed to be reported. He recommended instead that the language read that the reports "should address these aspects for the broad health care system."

A second participant warned that reporting could have adverse consequences and suggested that the principles be worded to make clear that there be no unintended consequences. Johnson noted that such language had been in an earlier draft of the principles and he asked for more feedback from those in the room.

I disagree with the comment, said one participant. She also noted difficulty understanding the intent of the language about addressing "these areas for hospitals, physicians and physician groups…" In response, Johnson said the language was intended to mean treatment by the whole system, adding that perhaps more overview language would make the intent more clear.

Continuing the discussion on comprehensive reporting, another participant said a sentence needed to be added after "reports should focus on areas that have the greatest opportunities in making care safe, timely, effective, efficient, equitable, and patient centered" to make clear that good contextual data was needed. Consumers are interested, she said, and these data need to include what we're measuring and why—especially when using clinically important measures. She noted that while measures may already sing to physicians and nurses, it was important to make them resonate with consumers as well.

Another participant asked for clarification about the sentence that reads "Information reported should include both information that consumers want based on the literature as well as information that is important for consumers." Are these two types of information? he asked. In response, Johnson noted that this is new language added since the last AQA meeting. The sentence means that there would potentially be literature consumers would want, and we need to look at that as well.

I agree that this is awkwardly worded, commented another participant. Research shows consumers want certain information. While the statement is an effort to express that thought, it's not clear, she said.

Does part of the workgroup's process involve studying existing reporting systems? asked one participant. If not, can we try to capture this? In response, Johnson said that he did not recall looking at existing reporting systems. Carolyn Clancy stepped in and suggested including language in the principles that makes clear the intent to add to them when everyone gets smarter about collecting information. We can get a lot better at making information useful to others, she said.

Another participant recommended adding a principle around going back and designing the reports for learning. This would address unintended consequences and contextual understanding, she said. She added that the principle would have pieces of contextual understanding within it regarding continual improvement. Someone else pointed out that continual improvement was addressed under "use for public reporting." I know, said the first participant, but I'd like to see this as a separate bullet.

In additional to the portrayal of performance differences as articulated here, observed one participant, it is important that we explain small versus large data to portray data fairly. Another suggested rolling out the reports after the data are brought together and aggregated.

Finally, one participant recommended not using the word "standard" (regarding "Reports should rely on standard measures when available."), noting that many measures are standard but not standardized.

Principles for Reporting to Physicians and Hospitals

Nancy Nielsen discussed the Principles for Reporting to Physicians and Hospitals and highlighted changes made since the beta set was endorsed by AQA last April. She stressed that the principles were general and not intended to cover every eventuality.

Under design, she stressed that it was important for physicians to be involved in designing the performance reporting system. She said that the language that says that the performance measures "should be stable over time, unless there is compelling evidence or a justifiable reason not to be," may need review. The rest, said Nielsen, should be pretty clear and people ought to know what's being measured.

Under data collection, continued Nielsen, there are some key issues. The overview is that administrative data shouldn't be the only data used, yet recognizes that going beyond that could be a burden for everyone. Regarding data accuracy, she said the intent has not changed significantly. Nielsen stressed that there needs to be a way to correct inaccuracies, particularly as patients move from one health plan to another.

Regarding data aggregation, Nielsen stressed that the workgroup felt that it was important to recognize that the more comprehensive the approach to how a physician practices, the more accurate the snapshot.

Regarding the report format section, Nielsen highlighted the final bullet, which says that:

  • Results of individual provider performance should be displayed relative to peers. Any reported differences between individual providers should include the statistical significance of the differences and relevancy. Reports should focus on meaningful and actionable differences in performance.

This is important, she said, and comes from the idea that an individual should be displayed relative to his or her peers. The consensus of the workgroup, Nielsen continued, was that providers would work harder if they could see where they stand compared to their peers. She added that it was also important for the reports to focus on providing information that is meaningful and can be acted on.

Regarding report purpose and frequency, Nielsen stressed the need for collaboration to advance quality. She also highlighted new language:

Performance data should, when available, reflect trend data over time (run charts and control charts) rather than periodic snapshots to optimize data use for quality improvement.

Finally, Nielsen stressed that the final principle, regarding review period, was intended to make sure that physicians have an opportunity to review performance results prior to their release to the public.


The discussion opened with a question about the difference between physician and provider reports. In response, Nielsen said that the reports to physicians are intended to contain information meaningful to improving care. She said, for example, that it is important for a physician to know if a patient doesn't have a mammogram so that the physician can reach out to that patient.

One participant noted that a footnote to the principles addresses reporting physician-specific information to hospitals. He pointed out that a physician's relationship to a hospital could be minimal, and noted that if hospitals receive information in advance other parties might want it as well.

Another participant suggested that both sets of reporting principles needed brief preambles that made clear that the measures used in public reports would be attuned to AQA's performance measures. The preamble, he said, should refer people to more detailed information on measures. The participant also offered a second, unrelated comment: that a lot of employers and vendors will look to use these reports to see if they are doing the right thing.

A participant asked about the language under design that said that performance measures should be "stable over time." In response, Nielsen stressed that this language isn't meant to preclude the development of other performance measures.

What's your definition of display? asked another participant (referring to the language that "Results of individual provider performance should be displayed relative to peers"). Who will individual physician reports be displayed to? he asked, thinking of trial lawyers or The Washington Post.

In response, Nielsen said that the display would include points on a graph. The question is, she said, should we identify physicians by name as points on the graph so you know who the high and low performers are?

Another participant also took issue with displaying relative to peers. Does that mean individuals or an aggregate? she asked. In response, Nielsen said that the reports are intended to provide a comparison to other individual providers doing the same work.

What was the upshot of past AQA discussions about high and low providers? asked another participant. The question was should each provider be identified, or only you, replied Nielsen, who stressed that the principles leave it open for those designing the report to answer the question.

One participant offered general comments on assessing effectiveness and the potential for unintended consequences regarding public reports. He gave an example related to beta blockers in which one hospital uses them at a rate of 95 percent while a second uses them at a rate of 40 percent. The second hospital failed on reporting but was running a clinical trial. The participant also said that there would be concern in the physician community regarding discoverability and legal implications. The principles should reflect these concerns, he said.

A participant noted that a sole practitioner would get a report and asked whether someone part of a team would get an individual report or a team report. Another participant suggested that the language under the review period should specifically say that a provider has an opportunity to respond to the report. In response to the latter comment, Carolyn Clancy noted that such a provision was specifically left out so as not to get into a debate with the provider who comes out not looking good.

Another participant suggested that the two reports—public and practitioner—should be released together and should say they are from AQA. A second person wondered, however, how to do that since the end users for the reports are different. A third person said the two reports should flow together and that there needed to be language that explains why we're talking about different groups for each document. A fourth participant stressed that it was important for physicians to know how results are being reported to the public.

Regarding the use of non-administrative data, a participant asked for clarification. Do you mean claims data submitted for billing purposes? he asked. He noted that many people say administrative when they mean electronic data. We need to define the difference between claims and clinical data, he said, and show that we want to move toward electronic data.

One participant asked about the bullet under report format that says that:

  • Justification and explanation of the rationale for setting specific targets for physician performance should be disclosed publicly to consumers, physicians, and hospitals.

Does this mean that if 100 hospitals have normal quality then their scores won't be reported? he asked. No, replied Nielsen, who explained that there would be some explanation provided.

Given unintended consequences, said one participant, do the principles consider the same quality improvement process in the future? Yes, said Nielsen, who pointed to the language on continual improvement incorporated into the Principles for Public Reports on Health Care.

Noting that a majority of the principles overlap, one participant suggested combining the two sets into one document that highlights the differences. I'm personally opposed to doing that, said Nielsen, because the reports are intended for different purposes. Johnson added that the purpose of the reports for consumers are to guide decision making, whereas the reports for providers are intended for quality improvement. He asked for purchasers and consumers to comment on whether the principles for providers would be of use to them.

I don't understand why the public reporting principles are so different from those for providers, said one participant. All of the provider principles under design, collection, and accuracy would apply to anyone creating a public report. Which wouldn't apply to public reports? he asked. In response, Nielsen noted that the sample size and data collection methodology might be explained more to those whose data are being analyzed—as opposed to consumers.

As a consumer-oriented publisher, said one participant, I would want to know that I'm providing consumers information about sample size. He added that he would rather have a system that includes reported methodological strengths and weaknesses. He and a second participant also noted that the methodology provides context, and helps to make the reports as transparent as possible.

One participant wondered who in the consumer community would get the reports and noted that there needs to be language that makes clear that these reports are for public consumption. Another suggested including a public comment period, noting that the main audience is plans and vendors—not individual consumers.

I would prefer not to get into the level of detail that says, for example, that there would be a 30-day comment period, said Nielsen. Another participant agreed.

When a plan analyzes data on an obstetrician, asked one participant, would this imply that my health plan would send letters to all the physicians and allow them to respond? He noted that this would add a lot of bureaucracy. This question goes to organization, said another person. Having people review the information and having a conversation with the decision maker does go toward providing accountability. As long as there's the corollary that a physician cannot share the document with his attorney, it's okay, he said.

Carolyn Clancy stepped into the discussion, noting that she saw the review as more a matter of seeing whether something was missed or that the statistics were inaccurate.

One participant suggested that language about the comments be embedded in the principles. The workgroup should work with physician organizations to define "unintended consequences," she said.

Next, Randy Johnson asked how to proceed. He laid out two alternatives:

  1. That the principles be finalized when the workgroup considers them okay, or
  2. That the workgroup bring comments back to the full AQA body.

He asked what participants preferred. "We tend to tweak and change the principles whenever we look at them."

There was discussion about the two alternatives. A few voiced interest in reviewing the principles again over time (and noted that while people were comfortable with the thrust of the principles the wording really matters), a process Nielsen pointed out could go on forever. Others suggested approving the principles and letting the workgroup move ahead.

Carolyn Clancy noted the need to make a decision. A show of hands around the room indicated:

  • A preference for recommended universal acceptance of the reporting principles.
  • A direction to the workgroup to consider integration.

1. These principles and other information presented by the Reporting Workgroup are available at

Previous Section Previous Section         Contents         Next Section Next Section


The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care