Begging for Improvement: IMEs & Impairment Ratings
A recent expose by the New York Times about workers' compensation Independent Medical Examinations in the state of New York, coupled with public debate about the inaccuracy and inadequacy of methods for evaluating (rating the extent of) Permanent Impairment in both California and New York have raised the visibility of these issues. On one of the professional list-servs I follow, someone asked today whether anyone has had experience with quality improvement initiatives in these areas. You might be interested in my response (see below):
The State of Washington 's Department of Labor & Industries (L&I) has done two projects on IMEs, both of which I was involved with, the first in 2001 and another in 2004-5 . Our company, Webility, was a sub-contractor in both of those projects.
L&I was exploring ways of improving the quality of independent medical examinations (IMEs) procured by the agency. Dissatisfaction by both labor and employers in Washington led to the commissioning of these projects. By the time the second project was complete, that furor had died down. (I suspect both sides were appeased by the fact that L&I was studying the matter, and their fickle attention turned to other issues.) To my knowledge, neither report has never been made public, but they should be available upon request. From the two projects, L&I received much input and many recommendations, some of which they have implemented.
In the 2001 project, the IME Improvement Project, MedFX was the prime contractor, Dr. Jeffrey Harris of MedFx was Project Director and I was Clinical Director. (Dr. Harris led the development of ACOEM’s 1st Edition of its evidence-based practice guidelines.) This project researched existing literature on IMEs, explored best practices around the country in IMEs, analyzed L&I’s IME procurement processes, interviewed workers, treating physicians and IME physicians, and audited a sample of IME reports. The final work product laid out findings and made recommendations for improvement.
Among other things we discovered during the first project was the remarkable dearth of literature on this topic. About the only literature that exists would be classified as "exhortative" in which one practitioner advises others how to write a good report. One tidbit we uncovered was that in the late 1990's, only 9 (or 19, don't recall now) doctors in the entire state of New Jersey were doing impairment ratings. We were able to find only one little comparative study (long forgotten now). The Lax article didn't appear until 2004. David Sikberg and I wrote an article about the IME procurement process for the Journal of Workers' Compensation entitled "Time to Release the Untapped Power of IMEs" which appeared in the Summer 2002 issue.
In the 2004-05 project, entitled the Independent Medical Examination (IME) Quality Assurance Pilot Project, Expert Clinical Benchmarks (ECB), a subsidiary of MedRisk, was the prime contractor, with Webility and IMX Medical Management Corporation as sub-contractors. ECB provided Ruth Estrich as liaison with L&I, while Webility’s principals led the project. I was Project Director and David Siktberg stepped up as Technology Lead when the technological demands of the project exceeded ECB’s capability. Dr. Elizabeth Genovese from IMX served as Associate Project Director for Peer Review and Operations.
I drove the second project from the beginning, envisioning it as an opportunity to make a significant contribution both to Washington and to a whole industry. It was essentially a feasibility project to answer the question whether auditing IMEs is feasible and whether providing feedback of IME authors improves the subsequent quality of their reports. The work involved (a) developing a tool and method for auditing and then scoring IME reports, (b) assembling and training a team of 19 peer reviewers to do the audits, (c) reviewing and scoring several hundred IME reports, (d) providing feedback to half of the IME report authors, and (e) monitoring for changes in audit scores over time. As part of that effort, I wrote two web-based training courses hosted on Webility’s training system. The first one trained the 19 peer reviewers/auditors on how to evaluate IMEs and use of the auditing instrument. The second one was entitled “Writing Even Better IMEs” and was offered to half of the 42 orthopedists whose IME reports were being steadily audited over a six month period. Only 2 of them started the course but both found it excellent. (Interestingly, they had both produced high-scoring IME reports.)
The project demanded very intensive technology support. In addition to implementing the two web-based courses, software had to be custom-developed to capture audit data, deploy an complex automated scoring algorithm involving variable weighting of IME report components, and produce feedback reports to the IME authors. The final deliverable was a report of findings and recommendations, in which we pronounced that IME auditing was feasible and produced valid enough results to justify business decisions (procurement preferences) -- and that changes to the procurement and payment process could be used to drive up the typical quality of IME reports obtained.
Warning: Do not read beyond this point if strongly-worded opinions offend you.
In doing these and other related projects, I have come to see that claims adjusters have only a few arrows in their quiver for use when the claim situation seems to be veering off in an unexpected or obviously wrong direction. They use all four arrows very freely, especially the middle two. The more inexperienced or overwhelmed the examiner is, the more seductive these options look. Those arrows are:
1. Nurse case management (positive approach)
2. IME (usually seen as hostile by claimant)
3. Surveillance (usually not seen by claimant but when detected, is seen as very hostile)
4. Litigation
One of the most stunning features of both of the Washington projects was L&I's lack of curiosity about (or lack of willingness to explore) the impact of the IME on the overall claim situation as well as a rather profound failure to grasp the potential impact of low quality IMEs and the other arrows in the claims quiver on claim outcomes. This view is NOT unique to L&I -- it pervades the claims industry. No claims payer that I have personally encountered has been willing to explore the question of return on investment in IMEs. I heard ONE presentation at a conference in which the claims payer tried to document the AGGREGATE quantatitve / objective impact of case management on claim outcomes. I've never personally heard any discussion about the QUANTITATIVE AGGREGATE benefit of surveillance. On the other hand, I am aware that a few claim operations have quietly examined the impact of litigation on claim outcomes. One friend's internal company survey showed that litigating a claim added a median of $30,000 to its ultimate cost.
Why should L&I buy $30 million worth of independent opinions each year -- if they have no impact on the outcomes? If they have no impact, why buy any at all? Would it make sense to pay $40 million instead IF those IMEs would help resolve stuck claims by clarifying issues and getting services to workers who need them so they can get better and close their claim more quickly thus shortening claim duration? Or, IMEs that would stand up in court and help reduce costs by stopping inappropriate care or continuing work disabiltiy -- to the tune of, say, $200 million in reserve reductions? In my opinion, claims operations are afraid that an inquiry into these matters might embarrass them (the claims operation) and they prefer to keep the "embarrassment" gun aimed at the independent examiners.
And yet, current IME procurement practices are designed to deliver EXACTLY the low quality crap they do today. In the absence of clear indications for use, documented outcome expectations and verification, quality standards and report specifications that provide a basis for refusal to accept work that does not measure up, and either positive or negative consequences for comparative quality, physician examiners are rewarded for the lowest quality stuff they can get away with.
Q: When are claims shops going to start tackling these questions?
A. When claims operations become less primitive, and more willing to take a hard look at the effectiveness of their processes and start taking a more systematic approach rather than seat-of-the-pants approach, that's when. Today, they are like giant factories with many assembly lines and complex machines, each operated by an "artiste". None of the machines have dials or indicators of how well the process of managing the injury episode is working (its ACCURACY or EFFECTIVENESS ). The only thing they have been monitoring and trying to improve is SPEED, EFFICIENCY, and COMPLIANCE with the law. For one specfic example with respect to IMEs, most claim operations don't collect and track information about which doctor provides the IMEs they need, nor volume/quality/nature of the work they do. All that data is locked inside individual claim files. So, until they start to pay attention, many insurers (like L&I) may think they have 550 doctors on their "approved examiner list" and not even realize that 70% of their IMEs are being done by a relative handful of doctors, half of whom are spending 15 minutes with the patient and churning out reports that would be consistently scored as "serviceable", "fair" or "poor" by our grading criteria.
Q. And when will that be?
A. When the people who run claims operations are people with curiosity and training in the use of quantitative / objective evidence to support making changes in the management of the claims process. I bet the people who procure copy machines and paper and pencils at major insurance companies use a more rigorous and quality oriented process than the people handling their IME proccurement process do. The procurement people know the difference between buying only on price vs. optimizing both quality and price. I'm quite sure the procurement people at Airbus and Boeing do.
Q. Aren't claim operations moving in that direction?
A: Not really. They have had efficiency (administrative cost reduction) as the goal, not improved decision-making or claims outcomes. In my opinion, the net result of the HUGE changes the claims management industry has been through in the last decade basically amount to heroic efforts to be able to make stupid decisions more quickly. Many people with years of experience in claims have said that today's much-more-efficient claims adjusters now have no time to get familiar with their claims and simply cannot make good decisions because they don't know what the heck is going on. That blinking "EASY" button labeled IME is very appealing.
End of outburst. Back to work.
