There’s so much excitement about big data and the potential for it in work comp that it would be all to easy to forget some basics.

The law of small numbers chief among them.

Work comp accounts for a tad more than one percent of US medical spend – $30 billion in comp medical vs $2.8 trillion in total medical spend in 2012.

Many docs treat just a couple of work comp claims a year, and those who do handle a lot of WC claims see a wide range of injuries: knees, ankles, backs, shoulders, hands damaged by cuts, sprains strains and severe trauma.  When looking to compare providers – or procedures for that matter – researchers need enough data points to develop a statistically-valid sample set.  In most cases, no single provider has enough claims to enable clear-cut evaluation.  And, if they do, there aren’t any other providers in their service area with the necessary volume, making comparisons nigh-on-impossible.

The issue is statistical validity and statistical accuracy. Simply put, is the measurement procedure capable of measuring what it is supposed to measure. Without enough data, there just isn’t enough information to accurately assess performance.

That’s not to say researchers can’t do very meaningful and helpful analyses; the one just published on opioid prescribing by physicians dispensing docs to work comp claimants is a perfect example; the ongoing research by CWCI, WCRI, and NCCI provide plenty of additional examples.

The problem occurs when consultants, payers or managed care firms try to make definitive statements about individual providers based on inadequate data.  In my experience, provider rankings are often – but not always – based on little more than reimbursement or “savings” figures, and in no way account for “quality” measured by return to work, disability duration, cost-per-claim.  There isn’t enough data to case-mix adjust, not enough data to make comparisons, or really “rate” docs.

I would note that some payers, most often state funds, and some managed care firms, notably MedRisk (HSA consulting client) have a wealth of data and can (and do) make valid comparisons.

What does this mean for you?

Beware of rankings, ratings, and comparisons of individual providers.  Unless the underlying data is robust.

At the closing session at NCCI, Barry Lipton reported on their research on MSAs, research based in large part on 2200 files provided by Gould & Lamb. In 2011, CMS approved $1.1 billion of MSAs…

Key findings

• MSA dollars account for 40% of the average proposed settlement
• half of the MSA dollars are for prescription drugs
• “the differences between proposed and CMS-accepted MSA settlement are largely due to drugs”
• only 20% of submissions are for claimants <50
• 29% of submissions are for more than $300,000;
• but, most submissions are for much lower amounts;
• so 62% of costs are for the 29% that are for more than $200k.
• The highest initial submission approval rate was about t50% before 12/12; during 12/12, the approval rate zoomed up to 92%.  this happened to be the same month where processing vendors changed…
• median processing time has dropped dramatically to 41 days in Q4 2013
• Median MSA approved amounts have been very stable at around $42,000 over the last four years.
• recall CMS doesn’t cover off-label use of drugs, so any non-cancer claims with opioids are going to require full funding of future projected spend

The net – from my perspective, if you have an addicted claimant, CMS is going to want a lot of dollars set aside.

The geekiest part of NCCI is the research workshop that takes place after lunch on the second day – you know the people attending are committed if they are in Orlando on a Friday afternoon listening to economists…

Harry Shuford discussed the “mystery of declining claim frequency”, an oft-described trend that some believe will end at some point while others think it may continue ad infinitum.

While there has been a cynical pattern over the last 90 years, the overall decline has persisted since 1926 (using manufacturing claim rates, the only ones that go back that far).  That said, the decline steepened after 1990 and has continued to this day, and it is consisted across all states, industries, occupations, demographics, affected body parts…

Why?

Harry and his colleagues looked at a lot of factors to determine their correlation with injury rates and similar data points.  The correlation isn’t due to the decline in manufacturing in the US; the decline has happened globally and across all sectors of the economy, not just manufacturing.  

Harry then showed a graph of international work-related fatalities (across 120+ countries) which showed a similar decline trend, with a bit of leveling in the late eighties followed by a steeper decline till 2006  and an even more rapid drop after that.  On average, there’s been a 4.3% annual decline over the last 30 years.

The death rate decline also mirrored the increase in per-capita income, albeit at a lower rate (3.6%).

And it was, if not entirely consistent, at least similar across geographic regions.

Net – two drivers: time, perhaps driven by pressures to improve productivity; and as a country gets wealthier, there’s a decline in the injury rate.

The takeaway – the trend has been in place for 80 years (at least) and will very likely continue into the future.

What does this mean for you?

Frequency will continue to decline.

 

Held again in Boston, WCRI’s annual meeting is one of the better research-focused events on the annual calendar.

It always includes a timely topic as well as a deep dive into key issues explored by the geniuses at WCRI; we can also expect insights into the impact of the economy on comp, updates on WCRI’s Benchmarks analyses, and the always-terrific Alex Swedlow will educate on medical dispute resolution.

This year the focus is on the impact of Affordable Care on work comp, a subject near and dear.  I’m a bit surprised about the timing as PPACA has been the law of the land since 2010; payers who haven’t done their homework and prepared for tighter access, potentially more cost-shifting, revised hospital coding, and strengthened subro efforts by group payers are going to be far behind.

There was much that work comp payers could – and should – have been doing to prepare for PPACA over the last four years; those who have not are way behind. Fortunately, they can gain some insights from the several presentations on the subject at WCRI.

(Here’s a link to a nine-part series on Obamacare and WC; dozens of other posts are here.)

As always, I’ll be reporting from Boston, but won’t be nearly as prolific this time around…if you want to hear the good stuff, see you there.

The California Applicant Attorney’s Association says there’s a BIG problem with UR in California, and a recent analysis by CWCI is flawed and inaccurate.

I don’t see it.

Greg Jones’ piece in yesterday’s edition of WorkCompCentral digs deep into the issue; here’s the brevitzed version of the disagreement.

CWCI analysis indicates about 75% of ALL treatment requests are approved by the adjuster or surrogate referring 25% on to an elevated physician-based UR process.

Elevated UR denies or modifies 23% of the treatment requests for an overall denial/modification rate of about 6% (.25 x .23 = 6%). The denial rate falls even further to less than 5% for claims that go through the state’s new medical dispute resolution process, independent medical review (IMR).    CAAA’s consultant argues that the denial rate is higher, by:

• citing different studies from different years with different samples, thereby comparing apples to oranges;
• asserting that there’s a lot of variation in denial rates among payers, as if this was a bad thing or even meaningful (different employer types, different locations, different medical management strategies);
• claiming that CWCI’s analysis was in error because their study included medical-only and not just indemnity claims, as if a standard of care or UR should or could somehow be different for claims with lost time. Guidelines are guidelines; they apply to all injured workers and don’t vary by type of claim.

Moreover, the premise of CAAA’s argument – to the extent there is one – is fatally flawed.  

A denial rate of 5% is hardly a catastrophe – especially when one considers where California was before the 2004 reforms – known as the physician’s presumption of correctness. Treating docs decided what treatment was appropriate, based almost exclusively on their personal opinion, or for a relative few, how they could generate the most revenue.  Payers had few opportunities to challenge a treatment plan. Treatment costs exploded to over $12 billion a year.

Would anyone allow a vendor to completely determine what services they were going to provide at what cost to whom? 

Of course not. This is completely at odds with every other payer system’s medical management methodology/process, because it is wildly illogical.   Yet that appears to be the motivation behind CAAA’s “analysis”.

Today, about 5% of treatment requests are denied or modified, medical costs are half what they were and employer premiums are way down as well.

Let’s look a little deeper at the results.

CWCI found that 43% of elevated UR and a third of the IMR reviews were for drugs. About half of those RX reviews are for opioids and compound drugs. Most likely a relatively few docs have a disproportionate percentage of challenged treatments.

The key is what are they doing and why – if they are prescribing Oxys to patients not supported by medical evidence, that’s wrong.  And, the current IMR process has begun to fix that problem.

Yes there are workflow problems, problems that will be resolved. Yes the process needs to get a LOT more efficient and less expensive. Yes guidelines need to be constantly updated.

And No, we don’t need to go back to the days of never ending treatment at the whim of the treating doc and bankrupt insurers.