None. Repeat: NONE.

There is not a single major healthcare software company that genuinely cares about its end user experience, despite the fact that poor implementations have, amongst other things, been associated with increased patient death. The particular publication I’m thinking of is actually more reminiscent of a frustrated rant than a calm scientific explication[1].

Reasons why no one cares:

1. No one at the company really understands how to abstract healthcare workflows. The engineers are absorbed in the coding of functionality, the product managers have no idea how their stuff is being used to treat patients, the docs are too busy trying to use this crappy software to treat patients that they tend not to be available to give adequate feedback, and designers who are good are quickly snapped up by Silicon Valley companies who understand how ridiculously valuable they are.
2. There’s a huge barrier to moving away from a piece of healthcare software once it’s been bought and installed, because they’re all pricey and everyone has to learn a new set of workflows. Patient care almost invariably suffers during this adjustment phase, and the last thing anyone fighting in the trenches wants is to screw up a patient’s life.
3. Many morons in healthcare administration still have not figured out that “extreme customization available to fit your practice’s needs” is advertising BS for “we were too lazy to figure out this really important part of our software, but we don’t want to admit it, so we’re going to spin it like it’s a good thing and lay all the blame at your feet when the random crap you pick inevitably fails in some major way”.
4. The backend implementation for many large EMRs would have to change fairly drastically (read: be completely redone from scratch) to modernize it into something that coder-designers of recent generations would even want to touch. To give you some idea, EPIC, the largest EMR vendor, uses Cache[2]. This framework is based on MUMPS, which came out in the 1960s before humankind even knew what good programming methodologies were. Example for you programming types: there is no variable scoping in Cache. Let that sink in for a minute. Yes. All variables are globals. It is an unholy nightmare.

[1] http://www.pediatricsdigest.mobi…
[2] http://en.wikipedia.org/wiki/Int…

See Questions On Quora

This is a very open ended question, and the usefulness of the answers will vary for every individual. Nevertheless, on a base level, I feel some of the most common pitfalls are:

1. Trying to adopt/scale-down web and desktop designs: In an effort to preserve “branding” and minimize expenses, I often find app designers trying to maintain the appearance of their app on the web or desktop. Not only does this present an unfamiliar UX for the user, but it serves to do more harm than good for the “brand”.

2. Applying cross-platform design language: There’s nothing more annoying to the user than finding an Android app designed to be an exact duplicate of its iOS counterpart (I’m looking at you, Instagram). I imagine every platform user must feel the same way when they see a non-native appearing app. Avoid this at all costs. As Hilton Lipschitz has mentioned, pay attention to the specific interface guidelines of each platform and stay true to them as much as possible.

3. Slowing down the user with fluff: Mobile users generally multitask while using apps and would like the app to solve its purpose as quickly and efficiently as possible. Splashscreens and animations which take too long to complete (Like the Starbucks Android App) are one of the biggest pain points in using a mobile app and will have the user searching for a faster alternative.

4. Not providing feedback to user’s actions: “Did I tap that button?”, “Has the download started?”. Prevent your user from wondering about the consequences of his/her actions. Provide appropriate audio/visual/haptic feedbacks for actions and the resulting mechanisms in the app.

See Questions On Quora