In 2015, a National Academy of Medicine (NAM) committee recommended a set of 15 core metrics (“Vital Signs”) to help streamline quality reporting. That report continues to drive discussions about reducing the burden of quality reporting as more voices, including CMS, call for a parsimonious set of measures. On Oct. 26, NAM hosted a discussion on quantifying the problem and next steps.
The session started with presentations by researchers who conducted quantitative and qualitative studies on measurement burden.
Nancy Dunlap, M.D., Ph.D., M.B.A. professor emerita at the University of Alabama-Birmingham, led a 2016 study that surveyed leaders from more than 20 health systems of varying size. The number of mandatory quality metrics they have to report ranged from 284 to more than 500.
In addition to reporting for outside entities, they also participate in collecting metrics internally for quality improvement, she noted. The respondents said that the complexity of reporting is increasing, requiring increasing staff and that many of the metrics change annually, with slight definition differences in metrics required by different groups.
“Although all the providers use EHRs, the large majority responded that only a portion of the metric reporting is automated,” she said. Documentation by the physician is crucial for accurate attribution and capture of information, and often physicians must re-examine their records to clarify terms and ensure that the relevant wording is used to describe the care delivered.
Standardizing the documentation input into EHRs can be helpful, but with changes in definitions, Dunlap said, the data fields with the EHR must be changed, and training and document must be updated. Health systems report that on average 50 to 100 individuals are involved in this process. The range of cost of personnel was estimated at from $3.5 million to $12 million per year, with most in the range of $5 million to $10 million annually. In addition, institutions may spend a substantial sum to recruit and train these individuals.
David Gans, senior fellow for industry affairs at the Medical Group Management Association, followed by describing a 2014 web-based survey that sought to measure the cost from the physician perspective.
The survey, conducted by researchers from Weill Cornell Medical College and MGMA, focused on cardiology, orthopedics, primary care and multispecialty practices.
It collected time estimates for physicians and staff on six categories of activity related to external quality measures. They converted the time estimates into estimates of the cost to the practices of dealing with external quality measures.
“We found 15 hours per week estimated in total effort per physician dealing with external quality measures,” Gans said. The cost varies by specialty, but on average they reported spending about $40,000 per doctor annually. And although the physicians spent a substantial amount of time on collecting data, there was relatively little time spent using the information.” When asked if they use the measures for quality improvement, less than a third said yes.
“Unfortunately, what we found is that practices have poor opinions overall about quality reporting,” Gans said. Only 28 percent of respondents answered that the measures actually represent quality, and there were consistent concerns expressed that the measures are not relevant to their specialty.
• 81 percent said they spend substantial time and effort validating the information.
• 46 percent said there was a substantial burden due to multiple metrics measuring the same elements or different payers looking at slightly different aspects or time elements.
Gans summarized their sentiments this way:
• Quality measures do not adequately represent quality of care.
• Entering quality data decreases clinician’s productivity.
• Providing quality data to external entities is very expensive.
• Quality measures, methods of reporting, and reporting periods should be standardized.
• It should be possible for an EHR to automatically collect and report quality measures.
• Measures should be specialty-specific. Orthopedists, in particular, said that current measures are not suitable for them.
In summarizing the findings of both studies, Dunlap said that the respondents believe that:
• Externally reported measures should be kept to a manageable level.
• Measures should be regularly evaluated to ensure that they drive actual improvements in care outcome.
• Alignment and standardization of definitions among groups requesting metrics are needed.
• Metrics should be piloted and definitions finalized prior to widespread dissemination.
• EHRs should be designed to more easily collect and report metrics and we should move away from quality metrics derived from billing and administrative systems. Clinical metrics would be more useful.
Reaction From a Panel
In a reactor panel discussion, John Bernot, M.D., senior director of quality measurement at the National Quality Forum, began by noting that he is a practicing family medicine physician. “I get to experience quality measurement on the front lines. but also in the community working to reduce the burden,” he said. “I want to reiterate that the burden of measurement is real. It is a challenge to everyone in the field of quality measurement.”
Bernot said there are now some promising initiatives. “Progress is being made at the level of awareness, but a lot of work still needs to be done,” he added. “We are trying to take a purposeful approach to measurement burden.” Over a year ago, NQF tried to change the dialog to the prioritization of the right measures and reducing unnecessary measures. “We are trying to take a very scientific approach to identifying a prioritized set of measures that should be used. We want to highlight gaps in needed measures and remove ineffective or low-priority measures from public- and private-sector programs.”
On the feedback side, he said, NQF and others need to learn more about measures after they have been implemented. “This has forced us to change our measure endorsement process to require feedback on measure use for maintenance of endorsement. So we are taking concrete steps toward a parsimonious set of measures and getting to measures that matter the most.”
Nancy Foster, vice president for quality and patient safety policy at the American Hospital Association, said it is noteworthy that Vital Signs provided a real foundation for discussion. “From my perspective, the dialog has indeed changed around the need to get to a parsimonious set of measures. A few years ago we felt like we were the only ones calling for a parsimonious set. Now there are many activities all designed to help identify the most meaningful measures to drive quality forward. Unfortunately they are not all on the same page. So we have a ways to go.”
She noted that hospital administrators do not feel that most measures for which they are collecting data are having the desired impact on quality because they don't’ feel like they are the right measures. “We stay focused on a parsimonious set that will actually improve care,” Foster said. The latest wrinkle for most hospitals, as they seek to provide more integrated care, is that they are engaging with other organizations in post-acute care and other settings. They employ or contract with 270,000 physicians. “Physicians have a large number of metrics for which they are held accountable. That has added to the burden of data collection for hospitals,” Foster said. “It has added enormously to the confusion and dizzying array of data coming at people. Vital Signs laid out incredibly important framework for this work.”
Finally, Lewis Sandy, M.D., UnitedHealth’s executive vice president for clinical advancement, said that the current state of measurement is too complex, too manual, and it is really not supporting enough improvement. “Given how much we spend and how much technology we apply, I was struck by how much manual processing is going on. On the issue of measurement burden, that is an area for significant improvement. I wince when I see how much physician time is devoted to data entry. I think we all agree we can that makes no sense in 2017 and we can do better.”
But Sandy, who served on the Vital Signs panel, stressed that there are some important efforts emerging around addressing measurement burden. “There is a multi-stakeholder group across payers and providers called the Core Measure Collaborative aimed squarely at this,” he said.
“NQF has a new strategic plan focused on priority measures and increasing the meaningful use of measures.” He mentioned a Physician Consortium for Performance Improvement, a multi-stakeholder effort focused on driving improvement through measurement. There are initiatives beginning, but we have more work to do.”
