Workers’ comp – for hospitals, it’s where the money is

Two recent articles in Health Affairs highlight a growing issue for employers and taxpayers; some hospitals are increasingly looking to work comp as a profit maker.

Depending on the state, facility costs can account for anywhere from around 32 – 40% of total work comp medical expenses (different states classify locations-of-service differently).

Ge Bai and Gerard Anderson examined the fifty US hospitals with the highest charge-to-cost ratios and found their markups over Medicare-allowable costs were three times higher than the average hospital.

This is critical in work comp because state work comp regulations often base facility reimbursement on charges – despite NO evidence or requirement that those charges have any basis in reality.

Fully 20 of the fifty hospitals are in one state – Florida – that uses a percent-of-charges reimbursement methodology for hospital outpatient services (manual is here).

Bai and Anderson’s latest work provides a deeper dive into hospital profitability.  A few key quotes:

  • Hospitals with for-profit status, higher markups, system affiliation, or regional power, as well as those located in states with price regulation, tended to be more profitable than other hospitals.
  • Hospitals that treated a higher proportion of Medicare patients, had higher expenditures per adjusted discharge, were located in counties with a high proportion of uninsured patients, or were located in states with a dominant insurer or greater health maintenance organization (HMO) penetration had lower profitability than hospitals that did not have these characteristics.

The methodology used by Bai and Anderson is somewhat different from that used by other researchers in that it excluded income from non-patient care services. I infer that they did this to focus specifically on the actual care delivery cost and not factor in other revenues from services such as parking, gift shops, investment income, etc.

So, what are the implications?

  • Work comp is a soft target for facilities in many states
  • The percentage-of-charges methodology is a license to…profit
  • More profitable facilities have likely already figured out how to make the most revenue possible from every source – including workers comp
  • Less profitable hospitals are going to learn from their more profitable competitors

Correction – Are work comp medical costs really dropping?

Last week’s announcement at NCCI AIS that medical costs for lost time claims dropped for the first time ever was a shocker. Talks with experts and industry pros after Kathy Antonello’s talk led to much head scratching and wondering.

The likeliest contributor is…California.

My mistake – California is NOT an NCCI state.  I was under the mistaken impression that, while CA is not an NCCI state, CA does share data with NCCI and therefore was included in the data used for this research.

Today’s WorkCompCentral opened with the news that California’s work comp rates are dropping 5%, driven primarily by reduced medical costs.  In turn, that decrease was due to favorable medical development on older claims – which means those older claims are turning out to be less expensive than originally forecast.

As California accounts for more than 20% of ALL workers’ comp premium, it should not be a surprise that the reforms that have stripped out a lot of the egregious profiteering and waste (e.g. double billing for surgical implants, reduced reimbursement for certain procedures, reductions in costs for physician-dispensed drugs) have actually lowered cost for older claims.

What’s not apparent is the undoubted improvement in patients’ medical outcomes. By reducing incentives for too many surgeries and drugs, patients aren’t getting as much unnecessary care that prolongs disability and has dangerous side effects.

Notably, if Los Angeles was removed from the data, results would be a LOT better. That county has most of the worst physicians treating work comp patients.

What does this mean for you?

Don’t write work comp in LA County.

Opioids, spines, and dead people

Friend and colleague David Deitz, MD, PhD, was kind enough to provide his perspective on two seemingly-unconnected items in the current issue of the New England Journal of Medicine that are highly relevant for medical providers treating occupational injuries.  Here’s his view:

Deitz – The first is an editorial by Drs. Thomas Frieden and Debra Houry from the Centers for Disease Control (CDC) reviewing the new CDC opioid prescribing guideline. It’s a concise review of what led the CDC to develop the guideline, as well as a clear statement of what CDC hopes to achieve. The money quote is this one: “We know of no other medication routinely used for a non-fatal condition that kills patients so frequently.”

Included in the same issue is one of a regular series of Images in Clinical Medicine – this one entitled Resolution of Lumbar Disc Herniation without Surgery. You don’t need a medical education of any kind to interpret this one – the pair of MRI images beautifully demonstrates a large disc herniation which resolves over a 5-month period. Nothing surprising to students of low back pain, there is abundant literature demonstrating that the best care for the majority of patients with lumbar disc herniation is conservative – maintaining physical activity as much as possible while waiting for the natural resolution demonstrated again in this case.

While I don’t think the Journal editors intended the irony, it’s sobering to think about how many opioids have been prescribed to injured workers over the last 20 years for this condition, and its (often unnecessary) surgical consequences. One of the most common conditions in WC, and a routinely-prescribed medication with potentially fatal consequences. Hopefully, we’re starting to do better.

Paduda – In a related piece, Michael Van Korff ScD andGary Franklin MD MPH summarize the iatrogenic disaster driven by opioid over-prescribing.  Over the last fifteen years, almost 200,000 prescription opioid overdose deaths have occurred in the US, with most deaths from medically-prescribed opioids.

Doctors prescribed opioids that killed well over a hundred thousand people.

Here’s one


Today, about 10 million Americans are using doctor-prescribed opioids; somewhere between 10% – 40% may have prescription opioid use disorder – they may well be addicted.

Van Korff and Franklin note that 60% of overdose fatalities were prescribed dosages greater than a 50 mg morphine equivalent.

This despite evidence suggesting “neither high opioid dose nor dose escalation improves patient outcomes.”

The authors suggest three immediate steps we can take:

  1. Avoid ill-advised and unplanned initiation of COT (chronic opioid therapy). Don’t prescribe more than 10 pills initially, check the Prescription Drug Monitoring Program database, educate the patient.
  2. Regulators and legislators need to change policies and regulations to reflect what we KNOW about COT and its inherent dangers.
  3. Considerably enhance population surveillance of opioid prescribing and safety.  The FDA should expand its postmarketing surveillance program for long-acting opioids to patients using short-acting versions.

What should you do about this?

  1. Do NOT allow opioids for “herniated” disks.  (I know, easier said than done…)
  2. Require a pre-auth for ALL acting opioid scripts, and all increases in dosage above 50 mg MED.
  3. Wherever and whenever possible, ensure prescribing docs check PDMPs, educate patients, limit initial scripts, complete an opioid agreement.
  4. Educate patients – for those already on excessive dosages, have your nurses contact the patient to educate them on the potentially fatal risk inherent in long-term use of opioids.

Monday catch-up

Happy Monday! here’s a few items you may have missed.

King v CompPartners – the California case may have implications for UR, IMR, and the “exclusive remedy” foundation of worker’s comp.

Here’s a very brief summary (see url above for more detail).

  • The underlying issue – did CompPartners’ UR reviewer do the right thing? is not in question.  The treating doc’s request was appropriately rejected as it was inconsistent with California’s evidence-based treatment guidelines.
  • However, the patient allegedly suffered seizures due to sudden cessation of the medication, and contended that the UR physician had a “duty of care” to inform the patient of that risk and recommend a weaning process.
  • The plaintiff took the case outside the work comp judicial process to civil court, where he lost.  It then went to Appellate Court, where the ruling raised this “question”: could Utilization Review be considered medical treatment, and the reviewer a treating provider?
  • This is contrary to all work comp precedent; the case is now before the State Supreme Court, which has stayed the Appellate Court’s ruling pending a decision.

Implications – talking to those who know better than I, the Supreme Court will likely reject the Appellate Court’s validation of civil court as an appropriate venue for the case, thereby reaffirming the “exclusive remedy” inherent in workers’ comp.

One issue that strikes me about this case; as the medication in question was prescribed by a physician for a condition deemed not covered by workers’ comp, why did the patient not a) pay for the medication himself or more likely b) get his health insurer to cover the script?

This would have allowed the patient to continue taking the drug and avoid the health issues experienced by the patient allegedly due to suddenly stopping the medication.

If you are in ChicagoLand and/or looking into value-based networks, read this. Really interesting piece on how a big provider system thinks about narrow networks, contracting, and what it wants to get paid for high-end services.  And will “eat” on commodities, such as MRIs for $100.

Here’s a shocker – media is all over reports on how chocolate helps athletes – even if the underlying study is pretty much nonsense. A much more important study that determined a very common spinal procedure is fraught with danger and likely counter-productive – was all but ignored.

From HealthNewsReview:

“Provocative discography” is a diagnostic procedure that’s used up to 70,000 times a year in the United States at great cost to the health care system. It’s commonly performed on patients with so-called “degenerative disc disease” who are considering spinal fusion surgery — a $40 billion per year industry”

If you have to rely on MCM to hear about critically-important research, there’s something really wrong with the mass media.

Looking forward to NCCI next week; will be on a panel moderated by Peter Burton with Mark Walls and Bob Wilson discussing regulatory issues.

Hope to see you there.

ACA’s 20 million increase in insureds – implications for workers’ comp

That’s a bit of a misstatement; ACA alone is not responsible for increasing the number of insureds by some 20 million, but there’s no question it was the primary causal factor.

Be that as it may, let’s examine who the newly-covered are, what they do, and where they reside.  The insured population’s demographics may be of interest to workers’ comp payers.

As noted yesterday, the newly-insured population is poorer, more likely to be recent immigrants, and much more likely to be Hispanic than the rest of the country. For work comp, what may be of more interest is the jobs they hold and where they live.

First, the percentage of part-time workers insured rose by 5.8 points, while the full-time population’s coverage went up 2.8 points. Those concerned with so-called Monday-morning injuries, may see this as a plus for work comp as more working people have insurance to pay for non-occ injuries.

Next, what do these workers do?

Pretty much everything; of particular interest to the work comp community, several high-severity &/or high-frequency industries saw significant jumps in the percentage of workers with health insurance. (details below)

  • agriculture +5.4%
  • construction +4.7%
  • transportation/warehousing +4.0%
  • manufacturing +3.3%
  • natural resources + 3.9%

Why is this important?  A few reasons.

Insured people are healthier than the uninsured, so they will heal faster if they do get injured on the job.

Work comp payers won’t have to foot the bill for medical conditions non-occ-related for insured workers.  This isn’t the case for claimants who do not have health insurance; actually work comp payers technically don’t need to pay for non-occ conditions, but end up paying for those conditions if by so doing the claimant gets better faster.

Monday-morning injury frequency may be reduced (if it is a real problem and not just commonly-accepted wisdom).

(chart below from NYT article)

Screen Shot 2016-04-19 at 12.33.44 PM

I bring this to your attention, dear reader, because clients, friends, and all manner of industry folk are keenly interested in the “impact of ACA on work comp.”  Fact is, we don’t know what it will be, but we can prepare if we look closely at what’s happening and make some educated, experience-based guesses.

What does this mean for you (work comp payers)?

A long term and incremental plus…perhaps.

Why be a crook when you can be a dispensing doc?

WCRI’s latest report on physician dispensing confirms what we weary soldiers have known for years; the physician dispensing industry is way better at figuring out how to screw employers and taxpayers than workers’ comp payers and regulators are at stopping them.

We’ve tried eliminating the upcharge for repackaged drugs; they came up with custom-manufactured medications.


Here’s the summary from WCRI:

the mechanism involves the creation of an opportunity to assign a much higher AWP to these new-strength and new-formulation products. Consider cyclobenzaprine HCL (a muscle relaxant), for which the most common strengths are 5 milligrams and 10 milligrams. If a new strength of 7.5 milligrams comes to market and the original manufacturer of that new strength sets a new AWP, this AWP could be much higher than the AWPs set by the original manufacturers for the existing 5- and 10-milligram strengths. These new strengths and formulation, almost all dispensed by physicians, are labeled as drugs made by generic manufacturers, not repackagers, and therefore, are not subject to the new reimbursement rules targeting physician-dispensed repackaged drugs.

Shockingly, Florida and California, two states that have attempted to control doc-dispensed drug costs with a repackaged drug cost cap have seen these “new” drugs become the most popular versions of the drug – and the most costly, with an average price of $3.01 per pill compared to $0.38 for the “regular” formulations.

Why has this 7.5mg version become so popular?

Is it better than 5mg or 10mg versions?

Of course not.

Make no mistake, these dispensing docs – and the industry that supports them, are quite clear about the money.

Proof.  More proof. And even more proof.

WCRI used data from 2 years ago; if anything it’s way worse now.

The solution is both simple in concept and difficult in execution.  Enable employer direction to pharmacies, a situation that currently exists in NY and MN (and in some cases in CA as well).

Yes, limiting doc dispensing to the first few days helps – legislation in IN and PA has been quite helpful in limiting the shameless profiteering of corrupt docs. However, the dispensing industry is quite creative in coming up with ways to circumvent regulations; don’t be surprised if:

  • docs rent a corner of their office to a “pharmacy”, and/or
  • docs get ownership in a pharmacy down the hall, and/or
  • companies are setting up vending machine-like dispensaries in medical office buildings

In fact, these all – and likely other maneuvers – are already operating in many states.  As I noted a year ago, these bad actors “will find any loophole, whether in a states’ pharmacy licensing process, medical board regulation, work comp statute or scope of practice to find a way to continue screwing employers and taxpayers.”

Because that is precisely what they are doing.

What does this mean for you?

It is long past time to stop playing nice.

Opt Out and work comp – the definitive report

At the 2016 WCRI Conference, several hours were devoted to opt out – what’s happening in TX and OK, variations among and between proposals to expand opt out to other states, employer views and challenges and problems and opportunities and…

No stone was left unturned.  Now, some folks think this was way too much time spent on what is a pretty small issue.  I’d suggest the exhaustive and complete review was helpful and needed, providing attendees, reporters, and you, dear reader with a source for a 360 perspective.

Trey Gillespie opened the Opt Out session with a dispassionate, thorough, and compelling demolition of the idea itself and execution thereof.

There are four different types identified by Gillespie

Tx – WC is not mandatory – so companies “opt in” to work comp

OK – qualified employers must have a benefit plan that meets specific requirements

TN – a proposed hybrid of the TX and OK models

SC – pending legislation proposes both models

Really, opt out moves occupational injury coverage from work comp to an ERISA plan – a federally-regulated benefit plan.  Gillespie identified a number of differences between ERISA and work comp; the ones I captured are below (I may well have missed others).

  • ERISA – there is no statutory or contractual entitlement to benefits
  • eligibility is based on employment status at the time of benefit – not the time of injury
  • employer decides what – if any – injuries are covered, and which employees, if any, are eligible.

Opt-out coverage commonly excludes industrial diseases caused by asbestos and silica and similar substances, along with assaults and terrorism.  It’s also much harder to “file a claim” as the reporting requirements, conditions, and limitations are much stricter than under work comp statute.

This last is key; according to NCCI, less than 20% of LT injuries were reported on the date of injury.  As opt out plans typically require immediate reporting, there’s a reasonable question as to the impact of opt out on those workers who can’t or don’t report their claim “immediately”.

There are also quite a few restrictions around the kinds and types of medical care that is covered.  Definitions such as “medically necessary” are fungible and, more disturbingly, almost all of the OK approved plans incorporate language that allows the Claims Administrator to terminate or change a previously-agreed-upon treatment plan at any point.

All in all, this makes a mockery of employers’ responsibility to make employees injured or hurt on the job whole.

More tomorrow…

More from WCRI – What happened after “reform” to states’ work comp systems?

Rather than inundate your in-box with multiple posts last week, I decided to delay posting on some of the research for a few days.  Today, WCRI’s analysis of work comp reform’s impact.

I had to listen very quickly and type even faster while listening to WCRI’s Carol Telles discussion of the impact of reform efforts in four states…as a result I probably missed most of the really good stuff.

When IL changed their Fee Schedule in 2011, medical payments per claim dropped about 19% while overall prices paid for non-hospital services dropped 27%. What’s with the discrepancy?  Did utilization or intensity of services increase to partially offset the intended 30% decrease in the fee schedule?  I might’ve missed the answer…

North Carolina also tried to reduce facility costs by changing the fee schedule from one based on charges to one based on Medicare for hospitals and Ambulatory Surgical Centers – there is a progressive decrease for services each year from 2015 to 2017, resulting in inpatient at 160% of Medicare and outpatient and ASC at 200%, with no separate billing or mar-ups for surgical implants.

In our favorite state – California – the implementation of SB 863 led to slight but significant decreases in medical payments per claim after many years of continual steady inflation. This was driven by reduction of reimbursement for ASCs from 120% of medicare to 80% effective 1/1/13, the elimination of additional reimbursement for surgical implants, and as of 1/1/14, a gradual transition to a Medicare-based FS for non-hospital providers.

Not surprisingly, ASC payments per claim decreased dramatically, dropping 24% in 2013/2014.  Imaging was also hit hard with a 23% drop, while physical medicine payments increased 28%.  This isn’t surprising as it is consistent with CMS’ desire to increase reimbursement for cognitive services.

Interestingly, the shift to the Medicare RBRVS system  resulted in a change in billing practices; the “rise in billing more complex office visits…stopped…after RBRVS transition.”

Last up was Louisiana.  State-set medical treatment guidelines were introduced five years ago, and there have been slightly fewer visits per claim after that intro.  The biggest drop was in pain management injections followed by a 5.7% drop in physical medicine (PT, OT, chiro).

The net – reform can effectively reduce cost if effectively targeted and well-designed.

Physician dispensing, opiods and efforts to control same

The last session at WCRI focused on my least favorite topics – doc dispensed drugs and opioids. note findings are preliminary and subject to change.

From Dongchun Wang, doc dispensing.

My takeaways.

Price-focused controls don’t work to control physician dispensing.  Sure, they work over the short term, but the dispensing industry quickly circumvents those price controls by coming up with novel new drugs, increasing the volume, or finding higher base-cost drugs to dispense to their patients.

In fact, prices for doc dispensed drugs-actually INCREASE quarter by quarter post-implementation of price-based controls.

For those of us who’ve been stuck fighting a barely even battle against the profiteering crappy docs and their supporters who do this, this is NOT new news.

Perhaps the to-be-released study will energize payers and employers to finally ban doc dispensing, and/or drive adoption of pharmacy direction (this last is the only real solution), we’ve seen doc dispensing rise even in states that technically ban or severely restrict doc dispensing


Opioids.  Vennela Thumula PhD talked about opioids.  Double Argh.

Okay, the good news is the amount of opioids per claim has decreased somewhat over the last few years, with almost every state seeing a drop (except WI).

  • About 30% of patients that get opioids only get one script – which is fine.  Acute injury, quick treatment, all good…
  • but 70% of so get more than one – and therein lies the issue.
  • NY LA and PA have much higher opioid usage than the average, with NY and LA patients getting well over 3000 MEDs per claim. THIS IS INSANE.
  • the average worker in Louisiana got 7 scripts, due largely to the large percentage of workers who used opioids for more than six months.
  • A significant percentage of opioid-taking claimants were also dispensed benzodiazepines.  WTF are these people thinking?

Drug testing has increased over the last few years – which is fine, except that the top 5% of claimants in LA are getting tested 11 times for 12 substances per test – and the average test costs just under $1200.

This is almost certainly driven by physician-owned labs, which have proliferated over the last few years.  (full disclosure – Millennium Health is a consulting client).

What does this mean for you?

We have a very, very long way to go.

Health care delivery varies a LOT – and there’s your opportunity

So, medicine is a science right?

If it is, then the delivery of care should be consistent across the country for patients with identical conditions, right.

Absolutely not.

That’s the quick takeaway from a terrific panel this morning at WCRI; below is the detail.

I’ve long been intrigued by the huge variation in medical care delivery across geography – why medical care for identical conditions for the same type of patient varies greatly from place to place is pervasive, fascinating, and, more to the point, driver of low quality and high cost care.

Dr Jon Lurie of tjhe Dartmouth Institute for Health Policy is one of the nation’s leading experts on this issue.  I’ll get right to the big finding –

There’s tremendous variance in “preference and supply-sensitive medical care” across hospital regions, defined as medical care for procedures such as vertebroplasty, spine surgery, total joint arthroplasty, and, in reality, most musculoskeletal procedures.

The most gross example is vertebroplasty, which varies by a hundred-fold.

That’s right, if you live in one area, you may be 100 times more likely to get this procedure than in another area.

Frequency of the medical procedures done in work comp varies widely across the country, and even within states.  Discussing one type of procedure, authors of a study found; “orthopedic surgeons’ opinions or enthusiasm for the procedure was the dominant modifiable determinant of ara variation.”

In English, doctors’ opinions and enthusiasm – not science, evidence, or outcomes – greatly influences what procedures get done how often.

Shockingly, reimbursement also affects procedure usage.  Washington and California have very different approaches to spinal fusion due to regulatory influences, with WA regulating the procedure much more tightly.  As a result, in WA, costs are lower, outcomes much better, there are far fewer spinal fusions, and the surgeries that are performed are less complex.

Yep, costs are lower, outcomes are better – and, not coincidentally, patients are much better served due to WA’s widespread use of evidence-based medical guidelines.

Next up was WCRI’s Dr Oleysa Fomenko – who got everyone’s attention with the opening statement “why are injured workers in one state three times more likely to get surgery than workers in another state?”

Key takeaway – in general, the higher the rate of surgery in group health, higher the rate of surgery in WC.  So, a payer can look at Medicare data and get a fairly accurate picture of what they can expect to see among their work comp patients.

However (there’s always a however), states that pay really, really well for surgery for work comp patients have a lot more surgeries than one would expect.

Alas, the Land of Lincoln is, once again, our poster child for bad outcomes – the work comp surgery rate is 2.5 times higher than one would expect, due perhaps to the $11,000 higher reimbursement for the procedure in IL vs the other study states.

NCCI’s Barry Lipton led off the panel with a discussion of cost variation across six states, using a methodology that took out fee schedule variations. The takeaway – costs for initial care for knee injuries varied by 71% across the six states, with KY CO and IL well above the other three (MD IN MO).

For knee injuries, one of the differentiators is, not surprisingly, utilization – with MD IN and MO exhibiting low utilization.  Utilization of surgery and physical medicine [PM] are the primary drivers.  There are also differences between and among the high-cost states. KY has much higher surgical costs, with IL spending a lot more on PM.

The other differentiator is the cost associated with diagnoses; cost per diagnosis varied widely across the study states.

Across the three high cost states, surgical utilization accounts for 35% of the cost compared to 23% in the low cost states; in contrast diagnostic imaging accounts for 32% in low cost states and and 24% in high cost states (other cost areas are pretty similar).

That said, looking at elbows and knees, most of the interstate variation is due to surgical and PM utilization AND how specific conditions get different treatment in different states.

For those patient and nerdy enough to make it this far, give yourself a new pocket protector as a reward.

What does this mean for you?

Medicine is a lot art and varies widely, and therein lies the problem – and for smart payers, the opportunity.