Shopping then Sharing: Interoperability to reduce Patient Out-of-pocket Costs

Short version: Patients can save enormously by shopping for the cheapest scan; electronic transmittal of protocols, reports and images may alleviate some concerns.

Long version.

The lack of sharing infrastructure for medical images continues to fester.

In the United States, the cost of imaging, both acquisition and interpretation can be ludicrously expensive and vary enormously, depending on where the scan is done. The recent case of a pediatric MRI for which a Stanford hospital charged $16.5k for the scan alone (not including the anesthesia charges that brought the total to $25k) serves to highlight the issue, as does the recent move by Anthem to stop paying for such abuses.

One of the issues that the NPR discussion of the Anthem move raises is described by Leonard Lichtenfeld from the American Cancer Society:

'They have to go to a new outpatient facility, get the film, get it read and transmitted back to the cancer center," Lichtenfeld says. If, as often happens, the hospital and the imaging center's computer systems don't talk to each other, the patient may have to bring the results back to the doctor on a CD.'

I guess this gives new meaning to the term "out of network" from an infrastructure and technology (as opposed to insurance and billing) perspective.

So, all we need to do is solve the transmission problem, and everything will be hunky-dory and patients can more easily shop around for the scan with the cheapest out of pocket costs. Tedious, but essential, especially if you are "out of network" (from an an insurance perspective), as the unfortunate woman ripped off by Stanford apparently was.

Though personally, if I could save that much as a patient, a faxed or printed report and hand carrying a CD of images would not be the end of the world, so obviously image transmittal is not the only factor.

Some facilities already have the systems in place to do this (transmit images and reports), for handling outside referrals, undermining Lichtenfeld's assertion.

To be fair, there may be some value in a specialized interpretation, though that is no doubt difficult to quantify. And in some cases there may even be some value in a specialized acquisition, but is that ever worth a ten-fold difference in cost?

When visiting large academic institutions that tend to be on the high end of hospital charges, I observe that their equipment used for routine patient care (as opposed to research) is often out-dated compared to the latest shiny toy that an unaffiliated facility may have recently installed. It would be interesting to study this question (manufacturer model and version of scanner at different types of site), but I don't have any contemporary data. Not that there is necessarily anything wrong with using an older scanner. Certainly when I did a lot of multi-center clinical trial work, image quality was not an issue, comparing studies from community centers against those from academic hospitals, for routine oncology imaging. There is some evidence in the literature to the contrary though. It may be a less of a case what you have than how you use it.

The NPR article uses a study about reader variability in low back pain MRI interpretation to implicitly support the argument that more expensive scans might be better, though only going as far as asserting that imaging studies conducted by qualified providers may not yield comparable results. The study does indeed highlight variation, but does not provide evidence that more expensive acquisitions or interpretations are better. Nor is low back pain and MRI necessarily a great test case (see "worse than useless"). A better example might be oncology, and there is certainly some data to support reinterpretation of outside scans.

Specialized interpretation, if it truly adds value, or even remote protocolling to assure acquisition quality/relevance, can theoretically be addressed by teleradiology. Worst case reinterpretation of the shared scans by someone the referring physician knows/respects can be performed (and theoretically reimbursed if medically necessary, although this is challenging). Clearly greater collaboration between the site performing the scan and the person interpreting the scan, if they were financially and geographically separated, could provide an optimal solution.

But patients are deluded (or being misinformed) if they think that they will necessarily get better scans or even better interpretations by paying more or going somewhere fancy. It may or may not the case, and they should expect some supporting evidence (as well as some up front transparency in charges so they can make a cost/benefit assessment).

Speaking of radiologists, the ACR had its usual knee-jerk "save reimbursement at all costs" reaction to the Anthem move, and started a campaign against it. Since when did ACR become apologists for the hospital industry?

If one looks at the big picture, there is an interesting perverse incentive at work here. Prior to the Anthem move, it seems to have been in a hospital's financial interest to make it as difficult as possible for patients and physicians to share images beyond the enterprise. Anthem's action will reverse that, in that it eliminates the financial incentive to perform the imaging in the hospital, and to maintain the same quality of care, the hospital will now benefit in terms of efficiency by improving interoperability with the outside facilities where the patients will now have to go.

So everyone will win, except the ludicrously overpaid executives of both for-profit and non-profit hospitals who are ripping us all off.

Insurance companies believe, and patients need to accept, that scans are largely a commodity, at least in terms of acquisition, if occasionally not interpretation. The technology, standards and products exist to level the playing field of quality and cost (charges, anyway). It is only the perverse incentives that preclude deployment of a more distributed image and report sharing infrastructure, as well as greater control of the referring physicians over the manner in which scans are performed. So if payers can improve the situation, then good luck to them, even if their own motivation is strictly profit (or cost reduction) driven.

We may never have a "single payer" in the US, but we could theoretically have a "single imaging sharing network", but I dare say there isn't a snowball's chance in hell of that either. But it is possible on a local or regional level, so maybe the payers should be thinking about organizing/funding/implementing that, since some imaging facilities and hospitals seem to have trouble finding their own way out of a paper bag.

E.g., Anthem could build an imaging protocolling and sharing network, so that in addition to causing such a ruckus, they could provide some tools to circumvent some of the issues allegedly associated with it.

Then ACR could go back to focusing on improving quality and consistency and appropriateness, without trying to defend the indefensible.

David