HSA’s first annual Survey on Utilization Review in Workers Compensation. is underway and we’ve got a few preliminary findings to report. (click on the link to fill out the survey)
We are surveying C level execs as well as desk-level folks (claims adjusters, claims execs) for their opinions concerning and results of UR; 80 responses so far and here are a few of the highlights.
– The vast majority of respondents utilize UM/UR to control medical costs and ensure appropriate medical treatment, while for a minority it is a revenue generator.
– Among desk-level respondents whose firms internalize UM/UR, over a third cite UM as a core competency as a key reason for performing UR in-house. 40% cite controlling UM costs, while almost half note UM is integrated with other medical management programs. This is in contrast to executives’ responses; only one in five claims UM is a core competency…
– There’s a big gap between desk level folks and execs when it comes to integration of UM with other systems, with execs twice as likely as front-line staff to report they are integrated. Interestingly, this is similar to the difference between execs and desk level staff we found in our bill review survey; most execs thought BR was integrated with UR but most desk folks did not.
There’s lots more to come, as we’re collecting a wealth of data on utilization review trends, services, vendors, costs, outcomes, and functions. Respondents receive a detailed version of the survey report; the public version is limited to highlights and major findings.
We welcome your participation in the survey. The On-Line Survey should take 20-25 minutes tops, and one lucky recipient will receive an iPad 2 as a token of our appreciation (make sure you include your contact info if you want a shot at the iPad).

My post earlier this week about the new opioid prescriber audit program recently announced by Texas’ Department of Insurance generated a rather interesting back-and-forth with TDI staff.
First, they were “disappointed” I didn’t post their entire response to the questions I emailed to TDI. There are a couple reasons for this:
1. TDI’s email response read like boilerplate, didn’t directly answer my five questions, and only indirectly answered three (I only figured this out after reading and re-reading the response several times; it was pretty convoluted).
2. I don’t post responses as a matter of course; if they are germane and responsive, I may well do so.
3. I received inconsistent guidance from TDI re the purpose of the audit – which is more accurately described as a peer review of 15 physicians’ prescribing patterns and comparison of those to guidelines. Nothing wrong with that audit, but as I noted in the original post, TDI’s reviewers will “have a good perspective on prescribing patterns for 15 docs, but…to what end?” The language in their email led me to believe the purpose was enforcement, but I was told that the purpose was actually “quality care for injured workers” (or words to that effect) – before I was told no, it was actually enforcement (in another communication). So, with no clear guidance, it was kinda tough to figure out the purpose of the audit.
Second, Amy Lee took me to task for neglecting to mention TDI had – in fact – published system-wide research on the types and usage of drugs in Texas. Lee was correct as I should have noted TDI has published results of two studies, one in 2011 and one here – (for some reason the embedded link doesn’t work) www.tdi.texas.gov/reports/wcreg/documents/Pharmaceutical_Prese1.ppt.
There was a lot more to the back-and-forth, but there’s no point in recounting the “he-said, she-saids”.
Instead, I’d suggest there are a couple take-aways.
1. I’ll continue to work to be more careful in giving credit where credit is due.
2. I never got an answer to my questions; why limit this to 15 docs?; and why not look at docs who prescribed meds for claimants with other prescribers? I’m still wondering, as are many others.
As I noted Monday, “payers have been required to report all manner of information to Texas for several years, with rather draconian penalties for failure to report. With this wealth of data, gathered at great expense and at no small cost to employers and their payers and vendors, it should be relatively simple to provide in-depth information on prescribing patterns around the entire Lone Star State. These data could be case-mix adjusted as well, something that isn’t mentioned in the TDI announcement on the current project.”
Now THAT would be helpful, and provide policymakers and other stakeholders with a wealth of information. While TDI did publish some research, in relation to controlled substances it was limited to system-wide script types, counts, costs, and number of claimants using (or more accurately billing for claimants dispensed) those drugs. There’s so much more that could be gleaned from the data; regional- or area-specific differences, scripts by type of injury, duration of care, case-mix adjusted comparison of claimants prescribed and not prescribed controlled substances, scatter plots or line plots of physician prescribing volumes by any number of variables…you get the picture.
Which leads me to point 3.
3. This all could have been avoided if TDI had responded to my query with direct answers. If that wasn’t possible, they could have called and said, hey, thanks for the query, but for reasons A, B, and C we can’t tell you that. Trying to get media – even we lowly bloggers – to play “find the answer in the vaguely-worded boilerplate” may sound like fun but in the end this hurts a lot more than helps.
You end up with a confused and frustrated writer, where you could have had another media outlet describing your yeoman efforts to improve things in your comp system.
FWIW, my original query, followed by TDI’s response is below.

Continue reading Texas responds…

NCCI released their latest research on the use of opioids in workers comp, and the news is chilling indeed. And there’s one outstanding puzzling conclusion.
Here are a few of the key take-aways.
– Per-claim narcotic costs have increased (almost doubling from 2001 – 2009)
– Far more claimants are getting narcotics early on in the claim now than in the past
– One percent of claimants taking narcotics use 40% of all narcotics
– Initial narcotic use (how much and how potent) is a good predictor of future use
One of the more (initially) surprising findings was somewhat counter-intuitive; narcotics’ share of total WC pharmacy spend is relatively stable. How can this be? If narcotic costs are going up and usage is going up and more claimants are getting narcotics now than in the past, how is this possible?
I will speculate this is directly related to the explosive growth in physician dispensing of repackaged drugs, and the much higher cost-per-pill of these drugs. Simply put, repackaged drugs are adding about a billion dollars a year (and growing) to employers’ work comp costs. The impact of the growth in physician dispensing far outstrips that of narctoic usage.
My educated guess is physician dispensed drugs will account for a third to forty percent of all drug costs in work comp this year.
Back to the study and the implications thereof.
This is yet another wake-up call to legislators, regulators, payers, PBMs, employers, physicians and claimants and claimant advocates. There is far, far too much usage of opioids in workers comp. Most of these drugs were developed solely for treating breakthrough cancer pain, a diagnosis all but non-existent in work comp. These drugs lead to addiction and dependency while having limited impact on pain while doing next to nothing to improve functionality.
So, why are we/you allowing this to happen?

States are starting to study opioid prescribing patterns – and well they should.
Last week, WorkCompWire arrived with an update on Texas’ plans to assess opioid prescribing patterns for work comp claimants in the Lone Star State. Actually, this isn’t an assessment of prescribing patterns, but rather a very limited review/audit of the top 15 physician opioid prescribers – with some major exemptions.
The opioid audit plan [opens pdf] covers claims with dates of accident in 2010 where the initial opioid prescription was less than 10 days from the date of injury; there’s more than a 30 day total supply of opioids for the injured employee; and the physician audited was the health care provider prescribing opioids to the injured employee.
I asked TDI (via email) why the audit was limited to 15 physicians, when and if the names of the top 15 prescribers will be published, why they aren’t looking at multiple prescibers, and what they will do with the results. The initial response was boilerplate and did not address those questions.
If there’s more follow up (I asked again) I’ll let you know.
The very limited nature of the audit is puzzling; payers have been required to report all manner of information to Texas for several years, with rather draconian penalties for failure to report. With this wealth of data, gathered at great expense and at no small cost to employers and their payers and vendors, it should be relatively simple to provide in-depth information on prescribing patterns around the entire Lone Star State. These data could be case-mix adjusted as well, something that isn’t mentioned in the TDI announcement on the current project. It is entirely possible the Medical Quality Review Panel will do that, but the memo from TDI says the Panel will “assess the medical necessity and appropriateness of prescribing opioids incases selected for this Plan-Based Audit by using their professional expertise and knowledge…”
In fact, case-mix adjusting may be irrelevant, as it doesn’t appear this is a statistical analysis, but rather a kind of peer review.
I’m a bit puzzled as to the intent and outcome of this effort.
While it is admirable to evaluate opioid prescribing, it’s unclear what the reviewers or regulators or enforcement authorities or employers will do differently after the audit.
They’ll have a good perspective on prescribing patterns for 15 docs, but…to what end?
Perhaps this is intended to be a warning shot across the bows of prescribing physicians, letting them know that high prescribers may come under scrutiny at some point. That’s purely conjecture on my part, as I’m not sure I understand the utility of the audit as currently described.
California, for all their ills, has done a creditable job studying and reporting on the physicians’ opioid prescribing patterns. The multiple reports, discussions, and publications resulting from CWCI’s research have led to a much better understanding of the dimension of the issue in California, one that may well have been instrumental in the state fund’s decision to incorporate prescribing practices in network contracting requirements.
I look forward to more dialogue with TDI, and will keep you posted.

Greg Jones’ recent articles in WorkCompCentral highlight the inherent problem in the debate in California around UR; there’s little data on which to base any assessment, much less draw conclusions. Moreover, the data that is available is not consistent; there are no common definitions or consensus around what constitutes a “denial”, an initial review, a secondary review, an appeal, a reconsideration,
Here are the key take-aways from Greg’s May 21 piece:
“Jerry Azevedo, a spokesman for the Workers’ Compensation Action Network, said
between the legislative proposal and reform talk, utilization review is an important topic. It would be beneficial to the entire community to have quantifiable data to understand the real frequency at which requests for medical treatment are being denied, he said.without statistics on utilization review it’s really only possible for stakeholders to base their debate on anecdotal evidence[emphasis added], which is not the best course of action.
“In terms of operating a system and in terms of what we want regulators to do, it needs to be based on statistics and what the numbers really are,” Azevedo said.”
The result of this lack of understanding is, at the least, mass confusion. At the worst, legislators and regulators are called on to make decisions affecting workers comp based on anecdotal information, press releases, histrionic statements, and data carefully selected to represent the perspective of the presenting party.
That’s no way to run a system.
Let’s start with a basic question – what, if anything, are the benefits of UR?
Well, the last credible study I could find was published in December 2007 by Alex Swedlow and John Ireland of CWCI. The study, entitled Analysis of California Workers’ Compensation Reforms; Part 1: Medical Utilization & Reimbursement Outcomes Accident Years 2002 – 2006 Claims Experience”, analyzed the impact of California’s work comp reforms.
Here’s the key data:
“In five of the six treatment categories, the average amount paid per claim for that type of treatment during the first 2 years following the injury declined – the only exception being surgery. Among surgery claims, the average amount paid for surgery services at the two-year valuation point was about 5 percent higher in the post-reform period than in the pre-reform period.”
Okay, I know, this was based on data that is now a bit old, and we need more current information and analysis. But CWCI’s research clearly and convincingly demonstrates that the reforms did have a significant, positive impact. Utilization declined across most medical treatment categories, and since then costs have declined dramatically. Really dramatically.
CWCI will be publishing updated research on this issue later this summer.
It is abundantly clear – from Jones’ article and the confusion surrounding the costs, benefits, and outcomes of UR, that we can’t make any decisions based on the information available today. It’s not sketchy, it’s almost non-existent. And, what we do have is contradictory and unhelpful as basic data field definitions are inconsistent.
As luck would have it, my firm (Health Strategy Associates, LLC) has just begun our first annual Survey on Utilization Review in Workers Compensation. We are surveying C level execs as well as desk-level folks (claims adjusters, claims execs) for their opinions concerning and results of UR.
The On-Line Survey should take 20-25 minutes tops, and one lucky recipient will receive an iPad 2 as a token of our appreciation (make sure you include your contact info if you want a shot at the iPad).
We will be publishing the results of the Survey in June, and hope it provides additional insight into the utility of UR across the industry.