By Christopher Guadagnino, Ph.D.
This month marks a watershed moment for health care’s pay-for-performance (P4P) movement, as the Centers for Medicare & Medicaid Services (CMS) launches a nationwide program that pays financial bonuses to physicians and other health care providers who report quality metrics on their Medicare fee-for-service patients. Until this month, physicians had limited opportunities to participate in P4P initiatives, including quality incentive programs and pilot projects offered by some commercial health plans, and a CMS demonstration project limited to ten large multispecialty physician groups across the country.
Under the Physician Quality Reporting Initiative (PQRI), CMS is offering physicians a bonus payment of 1.5 percent based on their total Medicare Part B payments for the second half of 2007 if they select at least three quality measures from a set 74 and report data for those measures on at least 80 percent of applicable patient encounters. New codes have been developed to capture the data and transmit it to CMS via existing claims processing systems, while physician use of electronic health records (EHRs) can facilitate the process, but is not required.
Many physicians may find the PQRI appealing, as they stand to gain extra reimbursement for reporting data that they may already capture in their patients’ clinical charts. Practices with EHRs already up and running may be able to tweak their system’s data templates to do the additional reporting with very little additional effort. PQRI may be a catalyst for large practices with only partial EHR adoption to ramp-up implementation across their entire network, while practices thinking about investing in EHRs may see this as the perfect time to take the plunge. Paper-based practices may be willing to absorb the additional data capture and coding burden now, with the expectation that quality reporting will become routine for them and prepare them to participate in future P4P initiatives with higher stakes and more detailed reporting.
PQRI participation is not a no-brainer, however. Capturing and coding quality data will likely entail workflow changes for most physicians and their billing staff, as new codes must be mastered by the practice. Few expect the 1.5 percent bonus to cover a practice’s additional cost for capturing and reporting the data. Some physician groups have logistical reservations about PQRI’s capacity to process a huge amount of new data timely and accurately, and they note that CMS has not set a specific date for issuing bonus payments – other than some time in mid-2008. Physicians, many of whom rely on third-party entities to submit their claims, are afforded no interim feedback from CMS on the accuracy, completeness or timeliness of their data submission, and would discover disqualifying irregularities in the form of bonus payment reductions or denials a year from now. And there are some who reject the premises of insurer-driven P4P programs, maintaining that payors’ economic interests may conflict with what is best for patients.
CMS describes PQRI as a first step toward linking Medicare health professionals’ payments to quality, which it says is consistent with Medicare’s ongoing transformation from passive payor to active purchaser of high-value health care. Physicians and other eligible professionals who choose to participate in PQRI in 2007, CMS says, will gain experience in capturing quality of care data and submitting it through the claims process.
PQRI runs from July 1 to Dec. 31 and replaces a similar, voluntary physician quality reporting program which did not offer financial incentives to participants. Physicians need not register for PQRI, but must have a National Provider Identifier (NPI) number on claims filed – which will be used to earmark bonus payments to individual physicians, even when a patient is seen by multiple physicians in the same practice – and must report on at least three evidence-based clinical performance measures. There are 74 process and outcome measures to choose from, developed by consensus organizations including the Physician Consortium for Performance Improvement – a coalition convened by the American Medical Association – and addressing various aspects of care, such as prevention, chronic and acute care management, surgical care, resource utilization and care coordination.
Sample measures include the following:
· Percentage of patients aged 18-75 years with diabetes type 1 or type 2 who had most recent LDL-C level in control (less than 100 mg/dl).
· Percentage of patients aged 18 and older with a diagnosis of coronary artery disease who were prescribed oral antiplatelet therapy.
· Percentage of patients aged 18 and older with a diagnosis of cataracts who were assessed for visual function status during one or more office visits within 12 months.
· Percentage of female patients aged 65 years and older who were assessed for the presence or absence of urinary incontinence within 12 months.
· Percentage of stage III colon cancer patients aged 18 to 80 years who were prescribed chemotherapy.
CMS recommends that physicians consider several factors when selecting measures for reporting, including conditions usually treated; types of care typically provided to Medicare patients – e.g., preventive, chronic or acute; settings where care is usually delivered – e.g., office, ED or surgical suite; and quality improvement goals for 2007.
Different measures have different reporting timeframes. Clinical action for some measures need be performed only once during a 12-month period for each patient seen during the reporting period, while the quality code need be reported only one time for each patient by each physician caring for the patient. If the measure calls for a clinical test result, then the most recent test result only needs to be obtained, assessed and reported one time per reporting period. A test does not need to have been performed within the reporting period, nor does it need to have been performed by the same physician.
For some measures, performance timeframes may also be tied to a specific clinical event that requires reporting each time the event occurs within the reporting period, for example:
· Procedure-related measures require reporting each time the procedure is performed and have distinct performance timeframes tied to them, such as perioperative care or imaging measures.
· Chronic care measures, such as those that call for prescribing a medication, require the physician to verify whether the medication is current and being taken by the patient. A new prescription is not required to meet the measure requirement unless it is clinically indicated.
· Acute care measures are tied to specific episodes of acute care and require reporting each time an acute event occurs, such as measures related to hospitalizations, fractures and osteoporosis management, or stroke measures.
According to CMS, the 1.5 percent bonus payment is not contingent on meeting any specific level of performance on the measures, and a physician can still earn the full bonus, for example, even if many of his or her diabetic patients didn’t control their blood pressures.
To earn the bonus, at least three measures – chosen by the physician – must be reported for at least 80 percent of the physician’s fee-for-service Medicare patients for whom those measures are applicable. Using claims data, CMS will track the percentage of a physician’s patients for whom a chosen measure was reportable, e.g., diabetes, CAD or colon cancer patients.
CMS has also established a formula for a bonus payment cap to physicians who choose measures that apply to few of their patients – which would satisfy the 80 percent reporting threshold with only a few patient encounters and make it unfair to receive the entire bonus payment for reporting such a limited set of data.
CMS encourages physicians to report on as many quality measures as practical, in order to increase their likelihood of meeting or exceeding the 80 percent successful reporting rate threshold for at least three. CMS will allow a two-month claims submission period following the end of the reporting period to successfully submit data, so physicians must submit claims for all services provided between July 1 and December 31, 2007 to their Medicare Part B carrier by February 29, 2008, in order to be eligible for the bonus payment.
Physicians can begin reporting PQRI data after the initiative’s start date of July 1, but CMS warns that the reporting period applies to all relevant covered professional services furnished between July 1 and Dec. 31 this year, and that delaying the date on which a physician begins to submit PQRI quality-data codes – which must be included on the same claim that is submitted for payment at the time the claim is initially submitted – may make it more difficult to achieve the 80 percent reporting rate threshold.
While CMS will judge eligibility for bonus payments at the individual level using a physician’s NPI, a group practice that bills under a single Taxpayer Identification Number (TIN) will receive one check that aggregates all of the bonuses earned by the practice’s individual members.
PQRI quality information will not be publicly reported at the individual professional, group practice, or TIN level, but some information about participation, reporting and performance rates aggregated at state-, regional-, or national-average levels may be made available to the public. CMS will provide confidential feedback reports to participating eligible professionals at or near the time that the lump sum bonus payments are made in mid-2008, with content and format yet to be determined. Because of the short lead time for implementation of the PQRI, CMS said it will not be able to provide participants with interim feedback reports of their reporting or performance rates during the reporting period, and there will be no process for physicians to appeal CMS’s determination of whether a practice has successfully submitted sufficient data to be eligible for a full bonus payment.
Physician enthusiasm toward PQRI ranges from enthusiastic to muted, while some remain doubtful as to whether now is the best time to get on board the quality reporting bandwagon, and others reject the concept althogether.
The Medical Group Management Association (MGMA) acknowledges that PQRI is a clear indication that the administration and the previous Congress have moved the Medicare program to a “value based purchaser,” and their first steps are paying for reporting, according to MGMA Government Affairs Representative Robert Bennett, who was formerly a member of CMS’ Physician Regulatory Issues Team.
MGMA is educating its members on PQRI, but is leaving it up to practices to decide if PQRI participation is worth the financial burdens. MGMA does have several concerns about the program. The 1.5 percent financial bonus payment is not likely to cover the added administrative burden on physician practices. No one knows precisely when the lump bonus payment will be made. Physicians have to submit data on at least three quality measures for at least 80 percent of applicable patient encounters, but participating practices have no clear way to check how close they are to that reporting threshold, and failure to meet this reporting requirement could subject the practice to a payment cap. Notwithstanding the Feb. 29, 2008 deadline for CMS to receive all claims, there are variables out of the practices’ control – such as claims getting held up by a carrier or claims clearinghouse. Legislation creating and funding the PQRI passed in late Dec. 2006, leaving CMS a short implementation timeline – only six months – to roll out and test the program’s system infrastructure. Meanwhile, future legislation could alter the program just as rapidly. While this pay-for-reporting program is currently voluntary, the stakes could be raised significantly if Congress associates “fixes” to the -9.9 percent conversion factor (Part B reimbursement formula) with reporting on quality measures.
Some physicians see PQRI as a low-stakes trial to see if their practice is up to the task of efficiently reporting, says Bennett, who notes that if a practice is a “well-oiled machine,” they may easily adapt to the changes required.
PQRI’s financial bonus is not large enough an incentive to nudge physicians to participate, as the average family physician with a 10 to 15 percent Medicare patient population could earn about $400 to $1,400 for six months of reporting, according to Bruce Bagley, M.D., medical director of quality improvement, American Academy of Family Physicians (AAFP).
A program that measures clinical performance and uses clinical data instead of administrative (billing) data is a step in the right direction of improving clinical performance, and the current PQRI is a relatively simple and straightforward way of getting in a game that is only going to get more complicated and have higher stakes in the future, says Bagley, who predicts that as much as 30 percent of total physician compensation could be related to performance measures during the next five years.
Physicians who participate most likely would see improved patient care and gain valuable experience in how to handle quality measure reporting, as the care team uses a checklist or EHR template as the patient goes through the office, and prospective data collection becomes a byproduct of the care process, Bagley says. AAFP recommends that family physicians pick three measures that apply to the same subset of patients – for example, hemoglobin A1c, low-density lipoprotein and blood pressure levels of diabetics – as the least burdensome way for family physicians – who might have 40 to 50 such patients – to participate in PQRI.
Several physician specialty societies have posted advice and shortcuts for PQRI participation, such as the American Academy of Orthopaedic Surgeons, which has flagged seven of PQRI’s 74 measures that potentially apply to its members’ patients, notes Helen Oscislawski, Esq., corporate health care associate in the Princeton office of Fox Rothschild LLP. Some subspecialties may find it difficult to find three applicable measures and, while they can report fewer than three, if CMS decides that more measures were applicable, bonus payments to those physicians can be jeopardized, warns Oscislawski.
PQRI presents paper-based physician practices with an opportunity to explore EHRs that have flexibility in their quality reporting templates, while the Internal Revenue Service has recently released a memorandum opening the door to nonprofit hospital EHR donations to physicians with staff privileges at the donating hospitals, if the donation is permitted under Department of Health and Human Services exceptions (created last year) to the Stark and antikickback statutes for donating EHR software, says Oscislawski.
Physicians choosing to participate in PQRI span various specialties and practice types, while some have EHRs and some do not.
Preferred Primary Care Physicians, a 35-physician private practice concentrated in the Pittsburgh region, started implementing EHRs in 2004 and now has them up and running at all 18 of its practice sites for internal quality reporting initiatives, while the system’s vendor offers a national reporting database that allows specialty-specific national benchmarking, according to Louis Civitarese, D.O., a board member of the group and chair of its IT committee.
The system uses clinical health maintenance flow sheets that already capture data being called for by PQRI, and all Preferred physicians will participate in diabetes measure reporting for the initiative, while the group’s IT department is working on tweaking its EHR system so that the PQRI database can be populated with the appropriate measures from patient charts at the click of a button, Civitarese notes.
Civitarese doesn’t know what it will cost to tweak his group’s EHR system for PQRI, or whether incentive payments will cover those costs. “We’re in this for the long haul. Looking down the road, whether it’s CMS or a large health plan, this is just the beginning. We want to expand this concept and report many more measures, consistent with our mission to improve quality,” says Civitarese.
The group hopes to apply this type of quality reporting in other venues, and is currently in discussions with commercial health plans in the region to develop P4P pilot projects in which the group’s physicians would export their clinical quality data directly to the health plans, says Civitarese. The group is also using the reporting system to apply for National Committee for Quality Assurance (NCQA) certification for diabetes, and hopes to use it for diabetes-specific certification from Bridges to Excellence, an employer-driven program that provides financial rewards for data reporting and quality improvement.
Successful PQRI participation does not require an electronic health record system, however.
All 200 or so physicians in University of Pennsylvania Health System’s Clinical Care Associates (CCA) will participate in PQRI, while only about five percent have EHRs, according to Ronald Barg, M.D., CCA’s executive director. Although the incentive bonus payment may or may not cover the additional physician and administrative staff time required to capture and report the new codes, Barg believes that “This is the way the future will be, and we might as well learn it now.” Each physician will report on four measures as a cushion to ensure reaching the required 80 percent reporting threshold for at least three measures, says Barg.
The workflow changes are not that complicated, says Barg, and CCA has created a single additional sheet for the front of each Medicare patient’s chart to collect data for the appropriate measures, which will get entered with the billing slip and sent electronically to CMS. The forms would appear on about 20 percent of a physician’s patient charts, representing a practice’s typical Medicare patient population, and most data would only have to be submitted once for each patient during the six-month reporting interval – a 30-second task for the physician, plus staff time to code the additional data on the bills, he says.
Practices using paper-based patient charts should be able to submit the new quality codes to CMS accurately by the required deadline, depending on the size and reliability of their billing system or company. “We all live in a structure where we rely on others to process our billing. If you already have problems with your billing company, PQRI data submission will add more problems. There are solo practices and large health systems that have problems with their billing – the only difference is that solo physicians would have no one to blame but themselves,” says Barg.
For some physician groups, PQRI will provide an incentive for more widespread adoption of EHR systems.
In New Jersey, Robert Wood Johnson University Medical Group’s 500 faculty physicians, representing about 40 different specialty practices, will participate in PQRI by reporting many measures they already routinely capture, according to Frank Sonnenberg, M.D., professor of medicine and medical director of the group’s clinical information systems. The oncology group, for example, will report eight measures to PQRI, he notes.
While all of the group’s physicians have access to an EHR system, about 30 percent are currently using it, and Sonnenberg expects the system’s facilitated data capture and coding functions to help the group meet its goal of bringing EHR use to 100 percent within a year. Through its clinical decision support feature, an EHR system can automatically identify patients who are eligible for a reportable quality measure, such as a diabetic, and prompt a physician at the time of the encounter to check for the appropriate measure, says Sonnenberg.
Although some RWJ faculty had reservations over whether 1.5 percent was enough of a bonus payment incentive to justify participating in PQRI, quality reporting is something that University Medical Group intended to do anyway, positioning itself for future P4P programs from payors, notes Sonnenberg. The group’s physicians are working on earning a Physician Practice Patient Safety Assessment designation, promoted by MGMA, which entails reporting a variety of process measures, such as checking for drug interactions, Sonnenberg adds.
Half of the 14 physician groups employed by Pittsburgh’s Western Pennsylvania Hospital will participate in PQRI, including anesthesiology, oncology, internal medicine, cardiology and cardiothoracic surgery, according to Steve Brown, director of specialty practices for the hospital. Of those 75 physicians, most have concluded that the program’s economic incentive is not significant, but that the program will support quality of care and patient satisfaction, says Brown. Some physicians have already been reporting quality measures to their specialty societies and receiving benchmarked data, he notes.
None of the physicians currently have EHRs, although the West Penn Allegheny Health System is beginning to implement them and has asked that all of West Penn’s employed physician groups adopt EHRs, and participate in PQRI, within a year, says Brown.
Not all physicians are enthusiastic about participating in PQRI, and they have principled reservations.
“We have not told any client to go forward with PQRI on the basis of current financial calculations. At best, they would break even,” says David Glusman, CPA, co-director, Healthcare Services Group for Margolis & Company P.C. Mega-practices that bill millions of dollars for Medicare may bring in tens of thousands of dollars in bonus payments, while smaller practices may find it less economical to make the necessary changes in their billing and practice management systems to process data that only a small percentage of practices are now capturing, Glusman notes.
Some practices may be able to justify the cost of participating more easily than others, particularly those that have a state-of-the-art EHR system whose vendor has a reasonably priced “patch” for accommodating PQRI reporting requirements, and whose billing system is reliable and flexible, says Glusman. Investing in those alterations may yield a return on investment only if PQRI continues beyond its six-month trial period, and if the government doesn’t dramatically change it, Glusman adds.
Glusman believes that physicians have little to lose by waiting for CMS to smooth out any “burrs” during PQRI’s initial run, to ensure that claims data will be processed cleanly and smoothly once physicians have made the commitment and investment to participate. He notes that CMS is not providing interim feedback about possible data capture glitches, and physicians will not know until it is too late to correct them.
The EHR market has still not matured enough for a paper-based physician office to purchase a system with assurance that it will continue to be supported and be interoperable with billing companies, P4P programs and other important interfaces in the future, says Mark Berger, M.D., Chief of Pennsylvania Hospital’s Section on Cardiovascular Diseases Division, and member of Pennsylvania Cardiology Associates, Ltd., a 13-physician group in Philadelphia. The scheduling, documentation and billing systems in Berger’s practice don’t communicate with one another, while he says that PQRI paperwork would require additional documenting, reviewing and coding in busy physician practices. For these reasons, Berger says his group will not participate, at least for the 1.5 percent bonus. “Clearly there is a threshold, and there are benefits of such a program other than financial, which we will consider down the line,” notes Berger. “But to be sure that our physicians achieve the 80 percent reporting requirements, we would need to invest in technology we don’t currently have. What I’m trying to do is drag my feet on an EHR system purchase until various systems sort themselves out and I can have confidence investing in a user-friendly, integrated infrastructure – which I expect will happen within two years or so,” he adds.
“The whole concept of pay-for-performance turns me off completely,” says George Isajiw, M.D., a solo internist in Delaware Co., who maintains that its premises have built-in conflicts of interest: insurance companies reward physicians to save money, and create perverse incentives to “game the system,” for example, by sending the sickest congestive heart failure patients to other physicians, or by reporting only four diagnoses when a patient actually has five, to achieve the best P4P profile. In the worst case, a cottage industry could develop whereby larger physician groups could hire someone to manipulate the data to maximize their income, Isajiw believes.
The only legitimate alternative, Isajiw maintains, is for a peer-reviewed system in which physicians who understand local demographic and epidemiological factors would set performance criteria and judge the care quality of local physicians.
Isajiw willingly says no to extra income offered by P4P programs, which he believes interfere with patient care, and he even pays a $1 penalty for each Medicare bill he submits to CMS, as his small practice is not compliant with the Health Insurance Portability and Accountability Act of 1996 (HIPAA), which requires Medicare claims to be submitted electronically, except for small providers that are permitted to continue to submit their claims on paper and pay the dollar-per-claim penalty. He says he sees between 70 and 100 patients a week, and estimates that 60 to 70 percent of them are Medicare beneficiaries. Isajiw would have to acquire new software to become HIPAA-compliant, but does not wish to, as he believes that the electronic reporting decreases the privacy of his patients’ medical data, despite HIPAA’s intent to increase privacy.
If Congress raises the stakes and requires P4P participation for a physician to receive their maximum Medicare Part B reimbursement – i.e., by associating “fixes” to the current -9.9 percent conversion factor – then Isajiw says he may drop Medicare, which he says many physicians have already done, and some have improved their bottom line by doing so. “Being principled hopefully won’t put me in the poorhouse,” he quips.