The Medicare data release rule called the “Availability of Medicare Data for Performance Measures” or rule CMS 5059-P has been in a comment period for a few weeks. The Wisconsin Health Information Organization commented on the rule this week and these comments are included at the bottom of this blog.
There are a number of issues with the rule as presently structured but we consider three big ones in this simple summary,please read the entire comment for further important details.
1. Cost. Medicare proposes to charge 80-100,000 dollars per year for entities such as WHIO to access the data. Much of this cost is justified by the cumbersome audit process CMS has proposed for the Qualified Entities that accept the data. WHIO describes how this audit process could be redesigned in a way to reduce the cost. Most all payor claims data base organizations such as WHIO have extremely limited funding and the proposed cost to access the data would limit the number of organizations that can access it. The other way to off set the cost is to allow Qualified Entities to charge for access to the data. It’s unclear if CMS will allow this but WHIO is suggesting that should be an option as well.
2. Timeliness. As proposed in 2012 Qualified Entities would be able to have access to Medicare data from 2008, 2009, and 2010. This data is too old and doesn’t match with the commercial and medicaid data WHIO receives every 6 months. Every 6 month updates from the previous 6 months of experience is what WHIO believes is required.
3. Integration. Integration of the commercial,medicaid, and medicare data is critical to being able to access provider specific data. As it stands this will not be possible in the proposed rule.
It’s good news this data will be released. The Center supports WHIO’s proposed changes to the CMS rule which will allow the data to be much more useful to providers who are trying to improve their cost and quality performance.
As the founding Chairman of the Wisconsin Health Information Organization (WHIO) it’s encouraging that we may be getting closer to adding Medicare data to the all payer data bases being created in the U.S.Two significant problems still exist however.
The problems are 1.The cost of accessing the data 2.The timeliness of the data
As the following article in Modern Healthcare points out organizations such as WHIO must purchase the data from Medicare, a cost of up $270000. WHIO and other all payer claims databases across the country are not for profit organizations held together with bailing wire and goodwill. This type of financial burden could tip some in the wrong direction. We need just the opposite, we need them to the thrive and grow and we need more of them. It’s important for CMS to recognize that getting this data to the providers through trusted sources such as WHIO and other qualified reporting entities will have a huge impact on cost improvement. Making the financial burden for access steep is counterproductive to CMS’ goals.
Timeliness? For 200K Qualified Entities get 2008,2009, and 2010 data. That’s old news by 2012 which is when entities such as WHIO get access. In Wisconsin we developed a process which delivers the data as a 27 month rolling average updated every six months.If we can do it with Medicaid and Commercial data why can’t CMS do the same with Medicare claims data?
These suggestions are written with appreciation that CMS is now working on this very important issue. We need to go a few steps further. The Center has been working with a handful of CMS officials on this and remains willing to share any and all expertise gleaned from our experience in Wisconsin.
The following is the Modern Healthcare piece.
By Maureen McKinney
Posted: June 13, 2011 – 12:01 am ET
Medicare is poised to open its cache of claims data to organizations that produce healthcare quality reports, a long-awaited move that promises more accurate and complete views of provider performance.
But many of the regional groups that want the data say the proposed price tag may stand in the way of their participation.
“The costs are prohibitive,” said Julie Bartels, executive director of the Wisconsin Health Information Organization, a De Pere-based not-for-profit organization that has amassed claims data for more than 3.7 million of the state’s residents. “Most of the organizations that would participate are non-profits that are running very lean operations, and those fees would be hard for them to consider.”
Placing the burden of those fees squarely on the shoulders of small, not-for-profit organizations is not sitting well with many groups like Bartels’. They argue that all stakeholders, including providers, patients and the CMS, will benefit from the reports they produce and the quality-improvement and cost-cutting opportunities they identify.
Mandated by the Patient Protection and Affordable Care Act, the proposed rule would require the CMS to provide extracts of Medicare claims data to qualified entities for reports detailing the performance of hospitals and physicians.
For the program’s first year, the CMS estimates that about 35 organizations will apply and approximately 25 will make it through the approval process. The majority of those entities, the CMS said, will probably be existing not-for-profit community collaboratives that already use private-payer and Medicaid claims data to evaluate providers.
Many of those organizations, such as the Wisconsin Health Information Organization, are multistakeholder and incorporate feedback from providers, payers, employers and consumers.
But a major roadblock for interested groups could be the fees required to obtain Medicare’s data set. In the proposed rule, published in the June 8 Federal Register, the CMS acknowledged that costs would vary based on the amount of information requested, but it estimated the fee—which covers the cost of data, technical assistance, application processing and monitoring—to be roughly $200,000 for three years of data for 2.5 million beneficiaries. Fees for three years of data for 5 million beneficiaries could run as high as $275,000, the agency said.
“That amount would not be feasible for us without going back to our members and subscribers and raising costs,” Bartels said. “That might stop them from participating altogether.”
Costs aside, the release of Medicare data for public reporting is a big step in the right direction, according to officials from many of the regional collaboratives. Organizations such as Bartels’ have significant data from the commercial market, but a complete lack of Medicare data.
Access to that data set will enable those groups to produce reports that are much more reflective of total patient populations, Bartels said. But only if they can afford it.
“At this point, it appears that CMS is looking to all of the other entities to support the cost of this program,” Bartels said, adding that her organization is tentatively planning to apply for a spot, based on the provisions of the finalized rule. “We’ll offer our comments and see what the response is.”
Nancy Foster, the American Hospital Association’s vice president for quality and patient safety, said the AHA’s members have spoken highly of many regional quality organizations. Those organizations have been working at a big disadvantage, she said, because without Medicare information, they have been blind to a huge amount of data.
Granting access to that information could help those groups find opportunities for improvement, and it could also help them to more accurately gauge the frequency of very rare adverse events, such as wrong-site surgery. “You need a fairly large data set to do that,” Foster said.
Foster also expressed concern that the proposed costs seemed high and the application process looked cumbersome, barriers that she said could prove difficult for some organizations to overcome.
The Pacific Business Group on Health, a San Francisco-based business coalition, praised the proposed rule as a means of achieving greater transparency and bolstering public-private cooperation, said Jennifer Eames Huff, director of the PBGH’s consumer-purchaser disclosure project. That public-private partnership allows for more “reliable, granular-level reporting,” she said.
Huff also said the PBGH is considering applying to be a qualified entity, but has not yet made a decision.
For many of these groups, however, the high data fees seemed especially exorbitant because the resulting reports will be useful for everyone, including the government.
“This data will produce value for CMS because it will help healthcare providers see where there are opportunities for cost reduction, and it will help them participate in accountable care organizations and other delivery models,” said Harold Miller, president and CEO of the Network for Regional Healthcare Improvement, Pittsburgh, a national membership association for regional health improvement collaboratives.
In a September letter to the CMS, NRHI and its members provided a list of recommendations for making Medicare claims data available. The letter urged the CMS to keep data fees as low as possible, “recognizing that collaboratives are non-profit agencies which are using the data to help improve the quality and lower the cost of healthcare for Medicare beneficiaries.”
If the proposed fee structure remains unchanged in the final rule, Miller said, the regional collaboratives that have more resources and stronger community support may be able to come up with the necessary cash. But others—particularly those whose member providers have tighter margins—may be unable to participate.
“CMS may be creating harm by setting the cost bar so high because communities will be busy trying to find money instead of setting improvement priorities,” Miller said, adding that the cost of data is only one of several big expenses. “You have to spend money on analysis and you have to invest money in hiring staff who know how to process the data.”
Oregon Healthcare Quality Corp., Portland, will probably apply to qualify for the data, despite the cost burden, said Mylia Christensen, the group’s executive director. Medicare claims data has been on the organization’s wish list for some time, Christensen said, and she predicted they would do what they needed to do to get it.
Still, she said she hoped the CMS would realize the overall benefits of releasing the data and relax on the fees.
“We see the audience for this information as a cross section of stakeholders,” Christensen said. “I’m assuming CMS has the same interests and would benefit. I hope that over time, as they see the value, those costs might be reduced or eliminated.”
Other concerns about the rule include the timeliness of the data. The CMS proposed providing qualified entities with the three most recent years of Medicare data available at the time an organization is approved for participation.
In 2012, for instance, an approved organization would receive data for 2008, 2009 and 2010. After that first year, qualified organizations would receive an additional year of data on an annual basis.
But as more hospitals and physicians enact big changes and the evolution of the healthcare delivery system accelerates, 3-year-old data just won’t cut it, said Miller, of the Network for Regional Healthcare Improvement.
“Data from three years ago won’t help providers understand opportunities for improvement today,” he said. “When you’re trying to form an ACO, you need the most accurate and current information.”
Another concern is the proposed rule’s restrictions regarding performance measures, said Christopher Queram, president and CEO of the Wisconsin Collaborative for Healthcare Quality, Middleton. Applicants must prospectively identify the measures they plan to use, and if they want to make changes, they must submit additional information to the CMS.
“Most people would acknowledge that performance measures—and particularly measures that use large databases—are still in their infancy, and some degree of experimentation is important,” Queram said. “There needs to be a balance between using established measures and allowing for that flexibility. That would be another area where I would expect to see a lot of comments.”
A number of famous writers have been touting the use of checklists as a way to create patient safety. Here’s the problem with this thinking told in the form of a real life story.
Years ago when I was still CEO I was alerted by the president of the hospital that a wrong sided surgery had occurred. She assured me a root cause had been started and we would know shortly the cause. Two days later another wrong sided surgery occurred. The President called me again this time noticeably shaken, and said she had no idea what was going on and she had consulted her Toyota Production System sensei and he had recommended a containment process that essentially meant we had to shut down all the operating rooms until the problem was isolated and “contained”. The purpose of containment is to stop the production process and solve the problem so the same problem can’t be passed on to the next customer. Despite a lot of political push back from doctors who didn’t think this drastic action was necessary she shut down all the ORs with my full support.
What does this have to do with checklists? The analysis of the above real life story showed that part of a checklist called the “timeout” was not being followed. The timeout is the few seconds before surgery is started during which all members of the surgical team stop and a series of questions are asked regarding the surgery including the patient’s name,type of procedure, and where the procedure is going to be done,right side vs. left side for example. There had not been a wrong sided surgery in the organization for nine years so why all of a sudden were there two? We found the surgical teams were not following their documented timeout procedure? The question was why?
This is where it becomes more than just a checklist. A process such as a timeout is what lean leaders and practitioners know as standard work which checklists can be considered a form of. But the key to standard work is actually doing it. How do we know if the standard work is being followed? The only way to know is to actually observe the work and understand whether standard work is being followed or not. So a checklist is great if it’s being followed. What we found was that we didn’t know whether the timeout procedure was being followed by every team in this critical time just prior to surgery.
Process observation is an important management tool which involves a person in the organization,manager, supervisor, or front line staff member, actually observing other staff do work. This is not done for punitive purposes but for understanding where improvement can occur. If, for example, a process observation shows that no staff member is actually following the established standard work a number of conclusions can be drawn. In our experience the most common reason is the standard work does not work. In other words, the process established either is missing critical steps or there has been a change in the patient environment making the standard work obsolete. This requires the staff members to re-write standard work with the new appropriate steps and then determine if this standard work delivers the desired outcome. In the example above the outcome is zero wrong sided surgeries. Another common reason we have found for standard work not to be followed is the staff have not been trained. In healthcare there is a lot of turnover and part time players. Sometimes there are gaps in education and process observation can identify these gaps in.
This is a management process called Kamishibai and is an important component to improving performance. This is not the only important management process leading to 100% reliability in patient safety outcomes but it is one critical component.
So what happened in our story. Once the management team realized standard work was not being followed a team of doctors and technicians quickly came together and redesigned the initial timeout procedure from nine years prior and then at an emergency meeting of the medical executive committee of the medical staff it was approved. The medical staff leadership voted to make this new timeout procedure part of the requirement of doing surgery in our surgery centers(which it had been before but they wanted to reinforce the point). They also agreed to update the standard work as new ideas were surfaced in the ORs for improvement but made it clear this was not negotiable, all surgeons would be performing the timeout procedure. 99.9% of surgeons were on board and agreed, a couple didn’t and left. The ORs were reopened and 100% of all surgeries for the next several months had an independent process observer to determine whether standard work was being followed. Once there was 100% compliance the audit process became random but still part of a standard work process observation calender led by front line managers and audited several times a month. The timeout process has changed slightly over the years but essentially is quite similar to the process established by surgeons nine years before the two wrong sided surgeries.
As you can see, delivering zero defects in patient safety is about more than simply checklists. The lean management system is required to actually sustain results and identify improvement. If you are interested in more details about this management system and other real life examples of patient safety problems I refer you to our book “On the Mend”.
Recently the Center received a call from Dr. Berwick to send results from Healthcare Value Network members that show what he called double wins, both cost and quality improvements together. In 24 hours network members came up with this remarkable series of improvements derived from their lean work.
Does Lean work in healthcare? The following reports of improvements from around the country, some in peer reviewed journals and others published by the individual organizations show the answer is a resounding yes!
Here are a few of the highlights but dig deeper into the work in this attachment and be shocked at why more healthcare organizations aren’t transforming themselves using this clearly defined methodology.
The following results are only a few of thousands of improvements made by these organizations and others who are committed to the use of the manufacturing derived quality improvement methodology described as lean. For more results please consult the appendix attached to this document.
Akron Children’s Hospital: Has applied Lean Six Sigma to enhance the voice of theirpatients and families, empowering their staff and clinicians to eliminate activities that do not add value to the patient experience, allowing for a greater focus on delivering high quality patient care. From January 2009 to March 2011, ACH has reduced costs by over $8 million and reduced appointment access waiting times by a total of 74,608 days.
Group Health of Puget Sound: Developed a patient centered medical home applying lean principals to create the new process of care. They achieved 29% fewer emergency room visits,6% fewer hospitalizations and a $10.3 per member, per month savings after 21 months.
Gundersen Lutheran: Improved their breast cancer screening, diagnosis, biopsy and treatment. Their new interdisciplinary approach resulted in a reduced patient call back rate for unnecessary biopsy from 10% to 5% and a 35% reduction in cost in patients requiring breast biopsy.
Henry Ford Health System: Hospitals’ aggregate inpatient harm rate has dropped almost 25% since the start of the No Harm Campaign. Identified measures of harm include but are not limited to patient falls, pressure ulcers, medication harm, procedure harm, and employee injuries, Using standard improvement tools Henry Ford has decreased the harm rate by 90 events/month even while adding a new hospital and increasing the total number of patient days. The total estimated cost of harm in 2009 was $39,910,375 or $916 per admission. This represents 8.7% of all costs associated with treating inpatients in 2009. Through the intense improvement efforts of the No Harm campaign in 2010, total costs were reduced to $34,465,612, a $4.4 million dollar improvement with a cost-savings of $85 per patient.
Inova: Using employee driven kaizen (change) in nine Inova Inc. emergency rooms throughout Northern Virginia, patient’s who left without being seen dropped from 2% to .48%, waiting time decreased by 31%, time to see the doctor dropped from 55 minutes to 22 minutes. This reduction in waste led to dramatic improvement in patient satisfaction (see appendix) and ten million dollars in improved operating margin.
Mercy Hospital in Mason City Iowa: A lean implementation team redesigned the lab process, consolidated redundant equipment, and redesigned lab personnel’s roles achieving a 53% faster turnaround time for patient blood test results, $470,954 in annual cost savings and $70,000 in construction avoidance.
Seattle Children’s Hospital: Has used their Continuous Process Improvement (CPI) methodology, based on Lean and Toyota methods, to reduce waste in a scientifically rigorous manner, improve service quality, clinical access, patient safety, staff engagement and financial results. Quality improvements include a 66% reduction in TPN medication error rates, a 50% reduction in ICU bloodstream infections, and 20% fewer ventilator days for patients. They have avoided $200 million in capital costs by using CPI to increase capacity in lieu of building additional patient rooms and ambulatory services space. They have also reduced patient costs by 3.7% and supply expenses by $2.5 million.
ThedaCare: Using lean tools and principals redesigned inpatient care and created a multidisciplinary team which has resulted in zero medication reconciliation errors for 4 years in a row and a 21% reduction in cost ($1200 per case) of inpatient care while at the same time reducing readmission rates to fewer than 12%.
University of Michigan: Using rapid improvement teams to reduce waste and delays in reduce waste and delays in red blood cell (RBC) dispensing and administration the reduced expense associated with RBC administration by $200,000 per month while reducing unnecessary transfusions for patients.
In this blog piece John Torinus describes how Serigraph continues to reduce utilization and costs keep going down,now one third the average cost for the nation.
Torinus has been a real pioneer in what employers should do to control healthcare costs. This report again shows what can happen when employers get engaged and have a strategy to manage healthcare costs.
The Partnership for Patients Initiative was recently announced by CMS. It aims to reduce patient harm and readmission rates.
CMS kicked off its Partnership for Patients webinar series with John Toussaint and Gary Kaplan presenting. This is a serious statement if not an outright endorsement by CMS that the Toyota Production System is an important methodology for all hospitals in America and the world to consider when trying to create 100% reliable care for patients. The CMS initiative aims to keep patients from getting injured. The goal is to reduce hospital required conditions by 40% and to do it by 2013. The other key goal is to help patients heal without complication. By 2013 the target is a 20% reduction in preventable hospital readmissions.
Hats off to CMS for getting patient focused with this initiative. These goals can lead to major care improvement for patients at the same time significantly reducing cost.
Today is the final day for submission of comments regarding the proposed ACO rules. The Center has had significant concerns regarding the initial rules and has submitted a series of suggested changes to the rules.
The Center has recently published concerns from several of the participants of the Physician Group Practice Demo as well as others regarding the existing ACO rules.
Many problems have been identified with the rules which we comment on with our submission to CMS. In the following comment we review the issues of retrospective patient attribution, risk corridors, size of ACOs, risk adjustments, quality reporting and much more. We believe if CMS would adopt the changes we suggest the ACO concept will have a chance to deliver what was intended: better patient value.
Former Treasury Secretary Paul O’Neill along with Toyota researcher Steven Spear and Richard Shannon M.D.,Chair of the University of Pennsylvania Medicine Department spent the last two days at the front line observing ThedaCare’s lean transformation work.
Paul O’Neill is a stickler on quality. In fact, during his tenure as CEO of Alcoa he made it the safest company in the world by applying continuous improvement principles. Specifically, by using the Toyota Production System to continually improve.
It was an honor and a privilege to have him on site at ThedaCare for 2 days going to the floors and observing the work of ThedaCare’s nurses and doctors as they apply the quality improvement principles they have been learning using lean methodology. He was energized by what he saw and he was interviewed in the following article by a local newspaper reporter about his reactions.
The following editorial written by a nurse describes perfectly what we documented in our book “On the Mend”. The shame and blame existing in healthcare today must be rooted out before improvement will happen.
Lucian Leap has recently written recommendations for medical education which explicitly lay out that the behavior exhibited by the physician in the following article should be eliminated if patient safety is to improve. I agree with him completely that physician behavior drives the quality of the team approach to medicine. By the way,medicine is a team sport and each player has a critical role to play to improve care. This doctor apparently has been somewhere else as the new environment has unfolded.
Organizations participating in the Physician Group Demonstration Program from CMS outline why they won’t participate in the ACO as structured. And the Center for Medicare and Medicaid Innovation introduces a potentially different approach to ACOs.
CMS is trying to calm the raging waters of the proposed ACO rules by writing editorials in national newspapers and, most recently a webinar from Dr. Berwick and Dr. Gilfillan that suggest things will change in the proposed rules.
In the meantime organizations who have been working with CMS on a shared savings model have suggested that it’s time to throw in the towel on the existing rules. Subsequently now CMMI has suggested there may be other ways to participate as an ACO which begin to address the concerns by the PGP demo group and others.
Below read the written summary of the subsequent webinar referenced above and a link to the letter sent to Dr. Berwick by the PGP demonstration program participants.
“Earlier today the Center for Medicare & Medicaid Innovation (CMMI) hosted a teleconference to announce three new initiatives related to the development of accountable care organizations (ACOs). Dr. Berwick opened the call restating the importance of the Three Part Aim: better care, better health, and lower cost. Dr. Berwick stated that he believes that ACOs are one important route to that vision. He explained that there are many competing considerations in developing the ACO rules, such as providing incentives for providers to achieve savings but also ensuring that ACOs do not stint on care, providing data to ACOs but guaranteeing that beneficiary privacy is protected, and ensuring coordination of care but also ensuring that beneficiaries retain the freedom of choice among Medicare providers. Dr. Berwick emphasized the importance of striking the right balance across these competing interests. He then announced three new initiatives:
Pioneer ACO program – an accelerated pathway to forming an ACO for providers that already have the infrastructure and care coordination models in place. This program would allow more sophisticated ACOs to move rapidly from shared savings to a population-based payment model. The Pioneer program would have synergies with the Medicare Shared Savings Program and would be designed to work in coordination with private payers.
Request for comments – on a proposal to provide up-front payments to providers who want to form ACOs. Dr. Berwick stated that some of the early comments on the proposed rule suggest that providers lack access to capital they need to invest in ACOs. The CMMI requests comments on potentially providing early access to shared savings that would allow these ACOs to make such necessary investments.
ACO Learning Sessions – the CMMI will offer four ACO learning sessions to help providers learn to build ACOs. Participation will not be a factor in selecting providers for the ACO programs, but these sessions are designed to give providers access to necessary information.
The call was then opened for questions and answers. Dr. Berwick, Dr. Rick Gilfillan, Peter Lee and Jonathan Blum responded to questions from the audience.
Q: The request for application calls for 50% of total revenues to come from outcomes-based contracts by the end of the second performance year. What does this include, and what is an outcomes-based contract?
A: Pioneer ACOs will be responsible for care coordination and management across their patient panel, including both Medicare and commercial business. For example, commercial payers might account for 25% and Medicare might account for 25%. Outcomes-based means that whether it is shared savings or some other approach, the organization is focusing on outcomes rather than volume of services provided. Pioneer ACOs are encouraged to think about how their contracts will result in achieving the Three Part Aim.
Q: How many quality measures will be required for this program and will they be different from the shared savings proposed rule?
A: We have to look closely to ensure that quality of care is improving. We intend to align the quality metrics with the Shared Savings Program metrics. We will also seek alignment with private payers and look at how this alignment works to achieve the Three Part Aim.
Q: Will there be flexibility on the 50% mandatory EHR requirement in the Pioneer program?
A: In the Pioneer Program, 50% must meet the requirement by 2012. In the Shared Savings Program, the agency is taking comments on the proposed requirement.
Q: Will CMMI look at a methodology where they can meet and help promote ACOs?
A: We realize that there exists a spectrum of care delivery today. The CMMI is offering a range of products that provide opportunities along the spectrum. The pioneer model is intended to move rapidly and show the country what is possible in terms of FFS Medicare. For those who are not as far along on the spectrum, the CMMI is requesting comments on an assistance program and an advance payment program so that advance payment can be made available to establish those capabilities.
Q: Can you go into more detail on the methodology for population-based payments?
A: There is a model approach in the request for applications but the idea is that there would be a reduction in FFS payments and a per-beneficiary population-based payment.
Q: Are there any planned discussions about inclusion of sectors that were excluded from participating in the Shared Savings program? In particular, home care and mental health.
A: Although the statutory model is a primary care-based model, we believe that the overall ACO effort will encourage the development of integrated delivery systems that better manage and coordinate care. ACOs that are serious about the Triple Aim will be reaching out to behavioral health and other providers because that is how you improve population health and shared savings.
In addition, the ACO effort is part of a broader set of activities to support delivery system as it transforms into the future of a seamless coordinated care system. One is the Medicaid health home initiative under which states can apply for reimbursement at 90% for health home services. CMS is working with states to develop programs with a focus on behavioral health and home health services.”
