Friday, August 24, 2012

ONC Whimps Out On Image Download in Meaningful Use Stage 2 Final Rule at GE's Request

Summary: ONC drops Stage 2 requirement to "download and transmit" DICOM images to third-parties, blurring the distinction between viewing and transmission; GE's negative comments at least partly if not wholly responsible.

Long Version.

You may recall that when the proposed rules for Meaningful Use Stage 2 were promulgated, many of us were ecstatic to see that a lot of attention had been given to imaging; there were some very promising features that I reviewed here.

Yesterday the final rules were released (you can find them at this page). I was excited and took a quick look at the (474 page) ONC Final Rule, did a search for "images", etc., only to be immediately disappointed to see that they had DROPPED THE REQUIREMENT for images to be "downloaded and transmitted to a third party".

Aaargh !

Here is a preliminary look at ONC's discussion of the comments and their response, which begin on page 76 of the final rule, and which I quote in its entirety here:

"Download and Transmission of Images
Comments. Commenters generally supported the principle of providing patients with access to images, however, only a few commenters outright supported our proposal. One commenter that supported our proposal suggested that images also be included in the “view” part of the certification criterion and stated that diagnostic quality is unnecessary for patient viewing. They encouraged us not to suggest a standard for image viewing by patients. Another commenter asked if we intended for images to be available for viewing in a basic distribution viewer or if small images embedded in the report or images viewed without tools in a browser would meet the certification criterion’s intent. They suggested that we require a basic distribution viewer to be part of the “view” portion of the certification criterion. One commenter stated that if we did not specify DICOM as a requirement for certification, that we should at least make available the option for EHR technology to be certified to the standard for the purposes of image downloads.

Several commenters strongly opposed or requested that we remove the capability and proposed standard. These commenters stated that including images for download and transmission by a patient would be a challenging requirement. They also contended that this capability exceeded the requirements in CMS’s proposed rule. Additionally, these commenters stated that images are typically stored in a system separate from EHR technology (i.e., a PACS system) and that this requirement would add significant complexity and burden to the certification criterion. They followed this comment by stating that the industry norm is for CDs with pertinent images to be given to a patient with an image reader that allows for viewing. A similar point was made by other commenters who stated that requiring DICOM for the transmission would force the recipient of the images to have a DICOM compliant viewer and to import the images into that viewer before they could be viewed. Many commenters noted that an image’s average file size would present significant storage and cost challenges for online downloading and transmitting. The JPEG file format was recommended as a potential solution since patients did not necessarily need diagnostic quality images.

Response. In consideration of the comments received and the complexity and potential burden identified by commenters, we have decided to remove the requirement for images to be available for download and transmission to a third party. We believe further industry dialogue needs to occur with respect to images and our policy goal of enabling patients to have ready, online access to their images. We expect to include this topic on the HITSC’s agenda for the next edition of EHR certification criteria we would adopt through rulemaking and intend to propose a requirement for online image access in a future edition of this certification criterion."

Images are big ? Hard ? Need a viewer? That's news ? Some whiny EHR vendors aren't ready for them yet? Boohoo.

Further industry dialog is required? What a cop out.

JPEG is sufficient? Patients don't need diagnostic quality images? Who are these clueless assholes?

Lost in the analysis of the comments seem to be two fundamentally important factors:
  1. The whole point of the "download and transmit to a third party" requirement is the "third party"! If that third party is a neurosurgeon or a radiotherapist or a pulmonologist or a rheumatologist or any one of a multitude of specialists with imaging expertise, your damn right they need diagnostic quality images.
  2. The intent of the incentives is not necessarily to apply to one single monolithic system that a recipient of those incentives might purchase; rather they are to motivate providers to better serve patients, whether it requires the use of one piece of technology, or two, or twenty; whether the images live in a traditional "EHR", or separately in a "PACS", should be irrelevant; what these systems are called and who sells them is irrelevant; PACS can contain functions of "Certified EHR Technology" too!
Not to the mention the fact that anyone who has ever windowed a CT image in their lives knows that JPEG as an interchange format for such images is completely inadequate.

The paternalistic attitude towards what the "patient" needs is equally depressing. Why doesn't the patient deserve a diagnostic quality set of images too? Frankly, any patient who can use a web browser can also use a DICOM viewing app quite handily, thank you very much. Many patients these days seem to be far more computer literate than the healthcare providers who are looking after them.

The comment that "the industry norm is for CDs with pertinent images to be given to a patient" is quite telling, in that it displays ignorance of the fact that the AMA safety panel made it quite clear many years ago (in a statement developed jointly by the professional societies and the imaging industry vendors) that a "complete set of diagnostic quality images" was the standard of care that needed to be met, not just "pertinent images" (begging the question of the definition of "pertinent").

The "norm" in the US right now is indeed the use of CDs, and fortunately these are now well standardized, but I thought the whole point of the Meaningful Use exercise was to move to doing things on the network, to make things better, not just ratify the existing, unsatisfactory, status quo. Sadly, the opportunity has been well and truly squandered this time around :(

There seems to have arisen tremendous confusion with respect to the distinction between the "viewing" use case, and the "download and transmit" use case, and the comments as described in the rule making, and the responses, seem to have further blurred the distinction to the point of making the assumption that if one can view the images (in a proprietary manner if desired), there is no need for "download and transmit". Clearly someone has forgotten about advanced visualization, surgical or radiotherapy planning, comparison with priors whilst reporting and the myriad of other scenarios that invalid the simple viewing requirement, but there you have it.

One very depressing element of the response is the statement that "only a few commenters outright supported our proposal". Most folks probably assume that when a proposed rule is promulgated, that the merits of the proposed rule are apparent from the outset, and that comment is primarily being sought on what is "wrong" with it, as opposed to nice cheery "good job, keep it up" supportive comments, but that is not necessarily the case. In the absence of strong supporting comments, anything that garners significant opposition might get dropped, as appears to have happened in this case.

Unfortunately, in retrospect, the ACR's comments did not specifically mention the "download and transmit" requirement, in positive or negative manner, though the RSNA's comments (to which I contributed a little) explicitly did. Surprisingly, or so I thought at the time, MITA (NEMA) did not submit comments.

As it turns out, if one goes searching to find out just who the naysayers are, amongst them turn out to have been none other than our friends at GE Healthcare; here is the text of their comment, much of which you can then see reflected in the ONC's summary and response. I wonder if GE has a product that (currently) does not support "such capabilities via a portal", or that is not integrated with their PACS in this respect? I guess they haven't bought LifeImage yet.

With GE in opposition, it is not much wonder that MITA made no comment at all on the MU Stage 2 rules, apparently because the vendor members "could not reach consensus"; at least that implies that some vendors must have been supportive, or MITA's comments would have perhaps been negative in this respect. I remain disappointed that MITA senior leadership could not have acted more strategically for the long term benefit of the entire industry, but they are, when all is said and done, a trade organization and beholden to their members.

All that said and done, at least ONC didn't drop the (separate) image viewing requirement, which still stands, albeit with links and without single sign on; that is what GE required in their comments.

By the way, if anyone knows how one can go about getting the entire set of comments that were submitted, what I believe the government calls the "docket", please let me know; I would have though this would be easy, and one would not need to make an FOI request, but I haven't been able to find them yet. It would be interesting to review them all to see how many other comments were negative, in the GE vein, or whether ONC just choose to prioritize GE's negative comments over RSNA's positive ones.


PS. There were also the following additional comments also mentioned:

"Comments. We received the following additional comments that did not fall within the general scope of the comments summarized above. One commenter proposed that a secure hyperlink to the image, supplied by the radiologist and conveyed via the Direct Project standard, become the method of making DICOM images and radiology reports available to patients and ordering providers. A commenter suggested that for image download a patient should be able to identify the location of a study to be referred to another provider as acceptable for the certification criterion. Last, a separate commenter asked that we specify for the “download and transmit” requirements, the IHE Portable Data for Imaging (PDI) profile.

Response. We appreciate commenters’ feedback. Given our decision to remove the requirement for image downloading and transmission to a third party, we will take this feedback into consideration for our future work with the HITSC as well as our next rulemaking."

PPS. If you are interested in an analysis of the "rest" of the final rule-making, the usual suspects have started their usual through analyses; here are Keith Boone's "What Changed" and John Halamka's "Rules Released" blog entries.


David Clunie said...

Ah ha !

Here is the link to the docket with all the comments:



JHT said...

Argh is right. What happened to the importance of image content? Agfa provides it with no download.

David Clunie said...

The docket link I supplied previously was just for the ONC comements.

The CMS comments, some of which also address imaging, can be found here: