Sunday, March 27, 2016

Lingering Frustration And Bad Design

Mrs. Dalai loves HGTV, and so I end up watching it quite a bit. I think I've even become somewhat of a house connoisseur after all this time.

So let's look at a house, pictured above. Wouldn't you love to live in this house? It's a great house! 4 bedrooms, 3 bathrooms, gourmet kitchen with granite countertops, stainless steel appliances, the whole nine yards! The house contains every gadget you could possibly want.

But there's a problem. The house was designed by someone who lives in a cave and doesn't drive a car. It is part of, well, a rental community, owned by an absentee landlord who will never set foot in the neighborhood, living in a corner office suite in some skyscraper somewhere. Can you see what's wrong? Hopefully, you did before you moved in. Because once you're in, it's too late.

My metaphor might be a bit stretched, but I think it is clear nonetheless. It is all too easy to design something wrong when you don't know anything about those who will use your product, and when the end-users aren't even the targets of your sales pitch.

So it is with PACS and EMR's, and so I've been saying here and elsewhere for a very long time. But it seems that with the greater penetration (government-mandated or not) of EMR's has come a deeper understanding and acknowledgement of just how flawed they are from the user's standpoint, the only one that counts.

This graphic from Gomerblog nicely outlines the problem:

And I will shout about the problem as loudly as I can:


Moreover, there seems to be very little interest in correcting this.

But wait, the world is beginning to feel our pain, and attention is coming from a very unlikely source: the government! Read this article, "Frustrations linger around electronic health records and user-centered design" in Healthcare IT News, and be encouraged. At least some in relatively high places seems to get it:
In a provocative prime-time speech, meanwhile, Acting CMS Administrator Andy Slavitt threw down the gauntlet: "I'm certainly not bashful about what we need to do better, and I'm not going to be bashful here, even in the face of some very good reasons for optimism, about ways we need to take our game up across the board.”

The health IT industry has done very well in the years since the HITECH Act, said Slavitt. "But we're still at the stage where technology often hurts rather than helps physicians providing better care."
Wow. Someone from the government is here to help us!  He actually said out loud what I've been saying for years. These poorly-written, outlandishly expensive software extravaganzas can HURT patients. Yowza!

Mr. Slavitt, my new governmental hero, continues my, I mean his, rant:
To bolster his case, he rattled off a series of actual quotes from frustrated clinicians. One complained that in his EHR, "to order aspirin takes eight clicks; to order full-strength aspirin takes 16."

Slavitt said CMS is newly committed to taking a "user-centered approach to designing policy." He asked vendors to do the same, with a similar spirit of empathy: "Step back and look at what you don't think is working, and make it better."
This, from CMS? Someone pinch me...

Perhaps the additional attention comes from the fact that more and more physicians have become disgusted with the status quo:
That dissatisfaction is getting worse, not better. A study published this summer by the American Medical Association and the American College of Physicians found that physicians are more frustrated with EHRs than they were five years ago.

Forty-two percent of respondents said their EHR system’s ability to improve efficiency was "difficult or very difficult." Some 72 percent said the same about its ability to decrease workload.

We saw similar feedback in Healthcare IT News' first-ever EHR Satisfaction Survey this past fall. In addition to numerical scores, we also asked for anecdotal feedback from more than 400 people who took the poll. Opinions such as "not very intuitive," "cumbersome" and "too many clicks" cropped up over and over again.
So what's the problem? As usual, that can be stated with two little letters:  I and T. The article continues:
In his landmark book, The Design of Everyday Things, Don Norman, director of The Design Lab at University of California San Diego wrote:

"The reasons for the deficiencies in human-machine interaction are numerous. Some come from the limitations of today's technology. Some come from self-imposed restrictions by the designers, often to hold down cost. But most of the problems come from a complete lack of understanding of the design principles necessary for effective human-machine interaction. Why this deficiency? Because much of the design is done by engineers who are experts in technology but limited in their understanding of people."

Of course, in healthcare IT there are other challenges. EHR vendors would probably love to have all their products look as sleek and intuitive as the latest iOS release. But they also have to ensure they check all the boxes to comply with certification criteria from the Office of the National Coordinator – all 560 detailed pages of the 2015 Edition.

"I know some people inside big EMR companies who want to do excellent design, but in an organization that's owned by IT, it's difficult for even a design advocate to have their voice heard and affect the process," Amy Cueva, co-founder of the design-centric Health Experience Refactored conference, told Healthcare IT News in 2013.
I have been saying this in reference to PACS for much of my career, and it applies equally to the superset of EHR's. Let me emphasize these very important statements:

Why this deficiency? Because much of the design is done by engineers who are experts in technology but limited in their understanding of people.

(B)ut in an organization that's owned by IT, it's difficult for even a design advocate to have their voice heard and affect the process...

There is some light at the end of the tunnel, but...
That's changing, thankfully, as more and more efforts are being made industry-wide to make EHRs easier to use and perhaps a bit better-looking. One of those ONC certification criteria, after all, is that vendors employ a user-centered design process when developing their tools, and report the results of usability testing.

A study published this past November in the Journal of the American Medical Informatics Association took a look at UCD processes at 11 unnamed vendors, seeking to understand the challenges and opportunities for better design practices.

"Our analysis demonstrates a diverse range of vendors’ UCD practices that fall into 3 categories: well-developed UCD, basic UCD, and misconceptions of UCD," wrote researchers from MedStar Health's National Center for Human Factors in Healthcare, noting that the latter category might refer, say, to the mistaken belief that responding to end-users' requests and complaints qualifies as user-centered design..."
But we aren't there yet...
Dishearteningly, the researchers found some respondents still didn't see the business case for investing in UCD processes. It even found that some smaller EHR vendors didn't even have any usability experts on their staff.
And this is the bottom line. As with PACS, in most places, IT controls every facet of the EHR. Software wonks create these jumbled messes while insulated from the actual users. IT folks buy the stuff while equally removed from the wants and needs of those who have to use it. It needs to be reiterated...physicians use EHR's (and PACS) to take care of patients. What a concept. Healthcare IT is used for HEALTHCARE! This isn't some app for your iPhone that won't hurt anyone if it fails or if it isn't usable. This is life and death, people. That's not an exaggeration. It has to work and work well. And today, it doesn't. It gets between physician and patient, impairing instead of facilitating that sacred relationship.

We just aren't there, yet. But maybe someday...

No comments :