Monday, May 25, 2009

Gray Papers and Health Care IT Cost Savings

If we want to see where the money in health care is going, look to the stock market. Cerner, one of the big players in health information technology, has seen its stock soar 50% so far this year, much of it from the federal stimulus package. This is part of the grand plan to save costs in health care, after all: why ask questions? If we just pour a few more billion dollars to health care information technology, we're sure to see lower costs, right?

So this got me wondering, where did this "cost-saving" mantra come from?

Much of it comes from the backing of a single 2005 Rand Corporation research “study” that concluded "if most hospitals and doctors' offices adopted [health-care IT], the potential savings for both inpatient and outpatient care could average over $77 billion per year."

So I had to look this study up. Who is the RAND Corporation? How was the study done? What measures did they use? Is this conclusion of $77 billion on cost savings realistic?

The RAND Corporation mission is "to help improve policy and decisionmaking through research and analysis." In other words, it is a political organization. Importantly, it's so-called "research" is funded from federal, state, and local government agencies provide the largest share of the funding; however, RAND also conducts projects for foundations, foreign governments, and private-sector firms. Contributions from individuals, charitable foundations, and private firms, as well as earnings from RAND's endowment, offer a steadily growing pool of funds that allow RAND to address problems not yet on the policy agenda. As they say, they're a "think tank." They THINK they are doing research, I guess.

Here's the "highlights." Big "studies", you see, need highlights. (Other more politically-minded individuals would call them "talking points," but I digress.) If one reads these highlights, we note that there is no "Limitations" section, a standard section of any reasonable research study that mentions reasons why a study's conclusions might not be valid. Nope. These conclusions are absolutes.

And I'm still looking for the funding "disclosures." Can anybody tell me who funded this particular "study?"

But a little digging and we find these "methods:"
The RAND team drew upon data from a number of sources, including surveys, publications, interviews, and an expert-panel review. The team also analyzed the costs and benefits of information technology in other industries, paying special attention to the factors that enable such technology to succeed. The team then prepared mathematical models to estimate the costs and benefits of HIT implementation in healthcare.
Wow. Talk about a mish mash of data! I really have no worries now! Forget bias. Forget objectivity. Just throw everything in to a pot and, voilĂ , $77 billions of savings!

But it doesn't stop there. It seems much of the basis of this data came from "gray literature" (their words, not mine) defined as "the body of reports and studies produced by local government agencies, private organizations, and educational facilities that have not been reviewed and published in journals or other standard research publications." Most of us would call these "white papers" produced by industry, but when their blended with one's own politically- or industry-supported research and cost savings are extrapolated to make a political point, the papers become "gray."

Recently, there has been a flood of "investigations" in to conflicts of interest between researchers, their academic institutions, and the medical device and pharmaceutical industries by Congress. But I find these supposedly "independent non-profit organizations," funded by industry members of the insurance, information technology and pharmaceutical industry including Unitedhealth, Wellcare, Aetna, Blue Cross of California, Genentech, Amgen, and Intel (to name just a few) more insidious and covert. They promulgate messages that are not peer-reviewed, but peer-pressed into the media, government, and create policy decisions that we assume (naively) are unbiased and truthful.




Keith Sarpolis said...


Couldn't agree with you more! Having looked for references and studies that allude to the cost savings and improved quality that EMRs are to provide us, I have had a hard time finding much in the stanard literature. And most EMR companies have put language in their user contracts that forbid the end users from doing much of the research that could prove or disprove these benefits which should heighten ones supicion right away as to whether the claims are true.

More often than not, someone has performed one of these think tank studies and used the data to tout and market the benefit of their product which is neither peer reviewed or disectable.

All we should be asking before we drop all this cash on EMR technology is what we as clinicians should be asking every time we prescribe a medication or treatment; does the science back up the claims of efficacy and are we fully aware of the downsides. EMRs need to be treated like any other device or treatment in medicine and subject to careful study before we release them in droves upon our patients. I am not at all certain that we are not being oversold as to the great healing benefits of this elixer called EMR to our health care system. Sounds like snake oil to me!

DrWes said...


I would not go so far as to call this "snake oil," but I DO think that in an era where our current government debt obligations (including Medicare, Medicaid, SCHIPS, and Social Security) exceed (by some estimates) $65 TRILLION dollars, we'd better be damn sure we're not making the cost problem worse. For instance, are shorter hospital stays and shunting patients to "long-term care facilities" cost-saving to the system as a whole, or merely a means to covertly bury the costs so they don't show up on the government's hospital cost balance sheet?

I DO think there are efficiencies to be gained with IT by better communication, less re-testing, ease of completing ridiculous government-mandated documentation requirements, etc., but all of these savings are offset by the remarkable ease of ordering more tests electronically, especially in our system that refuses real tort reform and therefore continues to fuel more and more "defensive medicine" test-ordering. Further, the costs of these systems are growing and growing as they become more sophisticated, and they still do not communicate between each other.

NanaRCR said...

As one of those old engineers who has watched the Internet and information management become a piled up mess through the years, I agree wholeheartedly with your assessment, "Baloney!"

The same people who made a shambles of the Internet are in charge of our health records.

Does anyone remember the term GIGO (Garbage In, Garbage Out)?

EMR without proper assessment, design, planning and careful implementation is the road to nowhere.

It's too bad that EMR does not have to submit to the FDA approval process!

Anonymous said...

cerner outpatient product is terrible. support is terrible. stay away at all costs. i do mean all costs.