Developing evidence-based guidelines? Difficult. Launching a sophisticated software platform? Challenging. Teaching 35,000 EPs to comply with ACEP’s new CEDR regulations? Nearly impossible.
I recently read with interest about the new ACEP endeavor to create a Clinical Emergency Data Registry (CEDR). As I understand it, ACEP has contracted with a company, FIGmd, Inc., and this company has developed the ability to suck data from more than 50 EMRs as well as practice management, billing and coding systems. The goal of the project seems to be to measure ED throughput and clinical quality down to the provider level. It just so happens that this is the basis for Medicare physician payment beginning in 2019.
With the repeal of the “sustainable growth rate” method for determining Medicare payment each year, a new process has been adopted – the Merit-Based Incentive Payment System (MIPS). The system will be phased in over five years beginning in 2019 and bases clinician payment on the achievement of Medicare-approved quality measures -– ranging from payment cuts or gains of 4% the first year (2019), 5% in 2020, 7% in 2021 and 9% in 2022. In addition, the law allocates $500 million a year to be paid as bonuses to the top 75% of professionals who perform above the quality threshold baseline.
To develop specialty-specific quality measures, $15 million per year has been allocated for deriving and vetting new quality metrics. Although payment adjustments will be at the individual provider level, congress has allowed for self-identified groups to work together as an accountable unit for the assessment of performance and payment purposes (and here’s where CEDR comes in).
So, ACEP will establish quality metrics (to be approved by CMS) that relate to individual provider behavior and depending on performance, Medicare payment will be adjusted for each of the participating providers. In the roll out of the testing phase of CEDR, ACEP will provide the service free of charge to 1,000 providers. As the program expands, the next 1,000 will be charged $195 each and subsequently, the annual fee will be $295 per provider (about $10 million a year if all ACEP members participate – though what percentage ACEP will get, and what percentage the software company gets, is not generally known).
To date, an ACEP committee of 25 individuals (mostly EPs) has developed 18 quality measure that will be tested in 2015 along with 8 pre-existing measures (from the Physician Quality Reporting System – PQRS). Anyone interested in this project (and it seems that most EPs will need to be) can go to the ACEP website (www.acep.org/CEDR) and check out the specifics.
The bottom line – the data registry was designed to provide CMS and potentially any other regulator with all manner of information and, in the process, eliminate the need for individual providers to participate in separate entities measuring EP quality metrics as it relates to Medicare payment.
Like every endeavor in healthcare, whether the CEDR program will provide truly meaningful quality measures depends on nuances. For example, all of the current CMS “time” measures are included.
Finally somebody who pays the bills is caring about ED throughput. I know a municipal hospital that received a Joint Commission commendation for its ED, but the average patient wait to be seen was well in excess of four hours! Totally nutty – a grossly dysfunctional department by the crudest measure getting an award?
I believe that among the time measures adopted by the CEDR program, the “door to diagnostic evaluation by a qualified medical personnel” (#9) is among the most important. Yet when you read the fine print, what is being asked is the “median time between patient presentation to the ED and the ‘first moment’ the patient is seen by a qualified medical person for evaluation and management.”
What is not specified is, “Who is a qualified medical person?” I think it is implied that it is not a triage nurse (at least hopefully so). To my mind, the definition of a qualified medical person is essential. Hopefully they mean a physician, PA or NP.
And what does the “first moment” mean? All patients have a more or less accurate “clock in” time when they arrive at the ED. However, there is no automated way of capturing “the first moment.” Will it be required that when a physician goes to document in their EMR that they will specifically be required to remember and document the time of the “first moment?” Sounds like a great opportunity for garbage in, garbage out. I know a hospital where they calculate this time as the time from “clock in” to the time an order is entered into their CPOE system (because this latter time is tracked by the system). So here is an example of a mechanized process to capture an important time that is entirely bogus.
Unfortunately, there are no time requirements specified for admitted patients. These are the patients who often languish in the hallways of the ED waiting for an inpatient bed. The literature on all the negative things that occur when admitted patients are held in the ED, especially ICU patients, is very clear. EDs desperately need this data to help make the compelling case with administration to fix the problem (we got men to walk on the moon 40 years ago, but it seems that clearing out the ED of admitted patients is just an insolvable problem).
Here’s a favorite – CEDR #4 – “Appropriate Utilization of CT for Pulmonary Embolism.” The literature on this topic is truly embarrassing. There are at least four papers specifically looking at this topic in the EMA database and they show that EPs are all over the dartboard when it comes to following the most simple of diagnostic algorithms concerning the assessment of possible PE. These papers have been reported in detail in prior EPM columns. The recommended diagnostic approach to assessing a patient with a suspected PE is very straightforward. Risk stratification is first (low risk gets a d-dimer and if positive gets a CTPA); intermediate or high risk gets no d-dimer and a CTA. Seems pretty straightforward – but not according to what the literature actually says happens in real life.
Seems that if CEDR #4 is to be measured, EPs will need to be taught to consistently and specifically state in the record, in a manner than can be sucked into the CEDR database, the perceived risk of a PE being present (using gestalt or Wells or modified Geneva scores or something else). Then, only if the patient is documented to be low-risk, is it allowed that a d-dimer can be ordered and then, only if the d-dimer is elevated (and what threshold should be used? And what specific test?) should a CTPA be ordered. All moderate and high-risk patients (as documented in the record) get no d-dimer and do get a CTPA.
Will the CEDR software be smart enough to figure these permutations out? Will it be able to decipher free text dictation to ascertain clinical risk stratification? Will EMRs require completion of a Wells score whenever a d-dimer is about to be ordered? Sounds clinically simple – but it may require substantial definitive additional documentation by providers that is now often absent – and some clever CEDR programming.
Here’s another “simple” one – CEDR #18 – no antibiotics for those with a diagnosis of bronchitis. Would seem that all the CEDR software would need would be the ICD-10 for bronchitis or the word “bronchitis” in the diagnosis and determine that no antibiotics were given in the ED or by prescription. But here’s the kicker – you’re allowed to prescribe antibiotics if the symptoms have lasted more than three days! Huh? Where is the data to support this practice? The literature consistently demonstrates that routine bronchitis does not need antibiotics and that it typically last up to three weeks – so where the “3-day rule” came from is unknown to me and is certainly not well defended in the literature. In addition, you are also allowed to give antibiotics for bronchitis if:
- There is also another infection present
- Cystic fibrosis
- Disorders of the immune system
- Malignancy
- COPD, asthma or chronic bronchitis
- TB
- Other disorders of the respiratory system
It would seem that to have a chart in which it is documented as acceptable practice to give antibiotics for bronchitis, the infection must be at least four days in duration OR one of the seven other exclusion (some of which are quite vague) need to be specifically documented. How will EPs remember this list – and the three day rule? Will this information be integrated into the 50 EMRs as decision support components? Unlikely.
We have just looked at just a few of the 26 items that are going to be scrutinized in the initial phase of the ACEP CEDR project. Although all of the items that are being assessed seem perfectly reasonable, getting physicians to document all that is required will be an enormous challenge as will teaching them the exceptions to the measures being assessed (no antibiotics for sinusitis of less than ten days duration, but OK if more than 10 days / EPs need to know the PECARN rules for minor head trauma / pregnancy tests must be ordered for female abdominal pain (14-50 Y/O) unless it is documented that the patient is pregnant or status post-hysterectomy – and it goes on and on and on).
So, ACEP has a ton of work to do teaching physicians what the expectations are regarding the current 26 measures and both their clinical and documentation requirements. Sure, it is great that ACEP is taking up the initiative of assessing ED care, but this is going to be a very, very tough job with lots of opportunities to screw it up.
Photo by Dean Hochman.
2 Comments
We need to implement what the British call – a SNAP prescription. It stands for Safety Net Antibiotic Prescription. It is ideal for patients that do not meet criteria for antibiotics, but appear clinically Ill. It decreases antibiotic usage, and serves as a safety net for the patient. Simply telling the patient to call your physician if it lasts 4 days or more is disingenuous. Do you know any patient that can get an appointment with their PCP in 3 days or sooner? They return to the ER where they can ill afford another $ 1,000 dollar bill for the second visit. Only 30 percent of diseases present in the classical fashion.We need to instill common sense back into the algorithm. THANK YOU FOR ALL THE GOOD WORK/WRITING THAT YOU DO.
Joseph Soler MD
Heisenberg’s Uncertainty Principle applies as much to social mechanics as it does to quantum mechanics: the more accurately you try to measure one property of an object, the more you affect another property of the object. In this case, using data extraction from EHR’s for quality metrics (and therefore payment) will suck more of an EP’s energy-matter into the black hole singularity we call medical documentation. It is a black hole because it seems that we EP’s can’t escape it. The added documentation burden will increase inefficiencies at multiple levels of the ED experience, more than negating any potential benefit of many (though probably not all) CEDR measures.
I highly recommend “The Digital Doctor” by Robert Wachter, to those who still adhere to the premise/ promise that the EHR can both work for the doctor and benefit the patient. Sadly, at the current time, the opposite is true: the doctor works for the document, and because of this, the patient suffers.
I have no doubt that FIGmd, Inc. will spin-off (perhaps jointly with ACEP) an Organization-Guised-as-CME to educate EP’s (and their employers) on correctly documenting enounters involving a CEDR, using evidence-based criteria.
Lo, an epiphany ! Perhaps if expanded to, say, 100+ CEDR’s, FIGmd, Inc. could obtain enough performance data on, say myself, that I could actually show ABEM that recertifying for the 5th time would be superfluous – as would annual LLSAs. And therefore, I would save thousands of dollars over a recertification cycle, thus negatively affecting the bottom line of both ACEP and ABEM. I’m certain this is not what these organizations intended.
Another example of Heisenberg’s brilliance !