Last month the Workers’ Compensation Research Institute (WCRI) released a study looking at how and if claim expense is affected depending on which party gets to select the doctor. The study, in excess of 100 pages, was entitled “The Effects of Provider Choice Policies on Workers’ Compensation Costs”. The authors were WCRI Public Policy Researcher Bogdan Savych and UC Irvine’s David Neumark.
WorkersCompensation.com reported in April that the report included 25 states, which were chosen based on availability of data. According to the report, “The 25 states chosen capture nearly 90 percent of benefits paid in all states where policies give control over provider choice predominantly to workers or employers. We had to assure that we could get as many states as possible, while at the same time making sure there were substantial records on workers’ compensation claims that we could use in the study.”
To a degree, the study results flew in the face of conventional wisdom by declaring that “we conclude that there is by no means an overwhelming case for concluding that overall medical costs per claim are higher when policies give workers more control over the choice of provider. However, there is fairly strong evidence that medical costs are higher for certain injury groups, including back and neck sprains, strains, and non-specific pain, or neurologic spine pain.”
It is difficult to argue with well-reasoned statistical analysis. However, when looking at this report, I cannot help but wonder if we are asking all the right questions, or even if in fact another related study should be undertaken by WCRI. This study focused purely on costs, which are easily quantifiable. A more difficult, yet useful, approach may be one that looks at outcomes based on similar criteria.
Despite the fact that the authors found average medical costs of employee choice states are more than 10% higher than that of employer choice states, they did not believe the variances were statistically significant. However, they also found that average indemnity costs in employee choice states were 20% to 24% higher than those claims based in employer directed states. I believe that may be a strong indicator of quality of outcome, and that is something that should be investigated further.
Note the last sentence of the report summary I mentioned prior. It says that, for states that allow employee selected care, “there is fairly strong evidence that medical costs are higher for certain injury groups, including back and neck sprains, strains, and non-specific pain, or neurologic spine pain.” Peter Rousmaniere, in a column that ran two weeks ago rightly picked up on that observation, noting, “With this finding, repeated often, the authors opened a door, although they do not step through it.”
It would be worthwhile crossing that threshold. Medical claim costs are certainly a significant percentage of expense for employers, and we should appreciate WCRI taking a look at the issue of physician selection. Still, medical costs alone do not reflect the true effectiveness of care or quality of outcome. Longer recovery periods, reflected by those extended indemnity periods, could be signaling the need for a more in depth look at this topic. Also, possibly missed in the focus on direct medical costs, are expenses from increased disability and lowered productivity that may (or may not) result as a consequence of who is selecting the physician.
My bias on the issue has been pretty clear. I am a proponent of “quality” employer directed care. I believe employers who find good occupational physicians and pay them well can see better outcomes and lower long term costs than those results achieved by employee selected doctors. But that is just my belief. I would like to see a study that can apply true statistical analysis to either prove or disprove my theory.
To do that, we need to be asking an entirely different set of questions.