I feel like perhaps I'm missing something here and this was already covered somewhere, but I don't see it in the post, the book splash page, or anywhere in these comments so here goes.
This is sensationalism. Bad design did not kill anyone. Sure, I bet it didn't help anyone, but part of the nurses jobs is to be able to use this software, they presumably were not all brand new to the job with no oversight, so as someone mentioned in a comment on the article, it's negligence on part of the nurses that caused this, not poor ui design. They knew how to use this interface, and they knew how to treat patients, it was their fault 100%. Unfortunately these kinds of mistakes happen in medicine. I could make the same claim for almost any other reason and it would be just as valid as saying the ui design was at fault for this person's death. Like I could say "one of the nurses had got a haircut yesterday, and they were talking about it, got distracted and overlooked the patient's treatment. The haircut killed jenny!" Nope. The nurses did. Or maybe chuck testa.
Has anyone ever thought about why all hospitals have horrendous old programs for patient management with poor UIs? Or did we just see this, agree that it was bad design, then jump on the bandwagon and start talking about all the magical ways we can fix it? It sure seems like the latter to me. This is just like the airline ticket redesigns, its just design masturbation with no actual research or backing. A post tackling this the right way would have talked to hospital managers about what they pay for the software, why they selected it. Ask if they'd be willing to upgrade, and why or why not. The gist of it is that re-training hundreds of employees is extremely costly. On top of that, if the result of the re-training of hundreds of employees is more medical errors and deaths, that's way worse. And finally, they'd need to figure out a way to migrate all their existing data into whatever structure the new system expects it in, which is expensive and time-consuming. So if it's working and employees can be trained to use it, even if it has a shitty interface and is generally a pain, it's still better than the massive cost and risk of a wholesale system swap. Smaller risks in other industries suddenly turn into enormous risks in medicine because peoples lives are literally on the line.
With the negativity out of the way, I want to say that I agree that these interfaces are horrible, and I want to improve them as much as anyone else does. My girlfriend also works in medicine and it drives me crazy seeing the amount of time wasted due to lack of quality medical software. This concept has my support 100%. I just don't want to see it publicized through an unscientific sensationalist post or book, I feel like it would be more productive for someone to take it on seriously by really picking apart the issue and seeing if it's feasible to solve.
These are both good points and many of the commenters of the article I wrote had the same points. Thats actually the main reason for the book, the topic is a big one that can't be fairly explained or assessed in a blog post.
The article I wrote was mainly to share that story, and the suggestions at the end were my best guess at the time. The book on the other hand will be more thought out, backed by research, additional cases, as well as exploring emotional harm, exclusion, and injustice.
Let me try to quickly respond to your 2 points though:
Preventable Medical Error cause 200,000 to 400,000 deaths a year (source) this isn't sensationalism. Its easy to blame the nurses. But the fact of the matter is the interface was distracting when it should be catching these errors. Thats the tragedy of it. We have very strict laws about distracting drives and recall any that are distracting. 200-400k deaths a year is a very serious issue, one that design can help aid and be part of the solution and as users/citizens we must demand that these designs be better.
I've been talking to a lot of people in the healthcare tech industry about this and I'm getting their suggestions. I'll be breaking down what designs and everyone can do to be part of the solution. Youre right in saying that the solutions and the environment that spawned the current ones are complex. But there are already some amazing people tackling this issue who need our support, companies coming in, and laws being written.
Great, this makes me feel a lot better about what you're doing. Based on the fact that the blog post had no actual research or backing, and also the pitch on the splash page didn't mention any actual research, I just assumed the book wouldn't have any. Perhaps adding a note somewhere in the pitch to this end would help ease skeptics like myself?
I must admit though, your claim about interfaces being the cause of medical error is just ridiculous. You simply cannot claim this without actual experimental backing. You especially cannot relate it to your "200-400k deaths a year" number, because that's just slamming big numbers that you know UI improvements could not change the vast majority of -- it's statistical deception. Interfaces cannot catch medical errors in most cases, and you are just assuming that they are causing some of them without figuring out the actual number that they cause, then presenting big unrelated numbers and hoping people won't notice.
Don't get me wrong, as I said I think that having better interfaces would be helpful and it's something I'd absolutely love to see. But the scientist in me just will not let me tell you it's ok to make really audacious claims like this without any evidence whatsoever. This is a touchy issue, which is why you have got the support you have so far. Do it justice and approach the research behind this book with very honest and very precise scientific methods, or you will see some serious backlash.
It wont solve it entirely. There are layers of error prevention and generation within a hospital organization. Design and tech can help aid many of these areas. In the example of drug distribution: Tech can be checking for proper dosage, patient compatibility, remind the nurse on how to administer it.
I'll make sure to add a few notes about the approach, and research that will go into it!
PS Would you be interested in helping me vet the book?
Sure, I don't disagree at all, and I think there are a lot of opportunities for design and dev both to make improvements to medical administration. I'm excited to see how it happens, but it is a very very tough market to make progress in because of the inherently monumental risk in any changes in administration.
Would be happy to do so! Find my email on the internets or hit me on twitter @jescalan
I very much value your perspective and the weight you hold this topic. I've never done anything like this, so I need the help of people like you who know both design and how to build arguments.
Thank you so much! I'll contact you soon.
You are obviously a very balanced and patient person to face down such harsh and direct criticism of what you are working on and turn it into something productive. For that alone you deserve whatever help I can give.
It was fair criticism. I really want to do the topic justice, to the best of my ability, whatever it may be.
Thanks again, I sent you an invite on LinkedIn.
Up until a few months ago, I worked on an ambulatory healthcare app, so let me help Jonathan out.
First, when a user interface prevents error, it can definitely save lives and improve health outcomes. Humans are prone to error, especially strung-out, overworked nurses working 20-hour shifts. If you enter the dosage for a medication and get the decimal point off by this much, it means the patient is in trouble. And when a weary nurse, who just wants to get home and sleep, accidentally puts an extra zero in that dosage, the UI should stop and say, "Hey, are you sure you want to do this 1,000 ccs? Usually it's 100ccs."
Second, no product is bullet-proof. No product or interface unilaterally solves a problem. It's the nature of product design; there are some cases that simply don't fit around what the product is supposed to do. You just have to do your best. And products in healthcare can do a lot better.
I'm hoping that Jonathan can make the connection that tragic design is not human design; it's design that doesn't consider how people actually live and work.
Looks really good. An important topic that needs more focus.
Looking forward to this book! Such an important topic!
Thats my wife!
A topic that deserves much more attention. Loving the type on the cover as well.
It's Calendas Plus which is a bueaitfully designed and robust font. It's chalked full of smart ligatures, and stuff!
Reminds me a bit of this: http://www.amazon.com/Design-Real-World-Ecology-Social/dp/0897331532
Papanek was more focused on "traditional"/physical product design rather than digital products... probably bc digital ui design didn't really exist as it does today at the time he wrote it... but definitely his lessons on the responsibilities designers have to do no harm and indeed help society are still very applicable today.
Looking forward to the book and your take on this topic :)
Great news!!! I am so pumped for this!!!! :D
STOKED for this, Jonathan!
I love that Medium post -- I've referenced it several times when giving talks on design to non-designers. Can't wait to read the book!
Thanks Joran! Even though writing this book has been the toughest project I've ever taken on, I feel so strongly that this topic needs some attention.
As designers were in such a big position to solve really important problems for the world!
Will Tragic Design have a happy ending?
(no, not that kind of happy ending...)
The book is about why its a problem and how we can fix it. But the ending is really up to us! Will we take on the responsibility we have as designers to make the world a better place.