Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Doctors ask engineers to spend more time in the hospital before building apps

cnbc.com

Post reply on HN
Print view View original

31–40 of 216 posts

Re: Doctors ask engineers to spend more time in the hospital before building apps

#31

Post by cordite »

oomkiller wrote: post #27

Unfortunately the problem is much greater than engineers not understanding doctors and other clinical staff, in my experience. For startups that want to sell to health systems and similar-sized/larger entities (really this is the minimum size that can work for most startups, practice sales usually have more friction than value), you unfortunately have to focus on the buyer, which is very rarely someone who is "in the…

Epic is often compared to Salesforce, which is to say even if there is a better localized app for a specialty (of which there are many in healthcare), the next questions are: How do integrate this into the other apps, how does it get into the record, how can people in other specialties receive information downstream to act appropriately from medical side to billing side. Then there's the last non-app part, what is the cost structure, what are the hardware requirements, who's going to watch it when it goes down, what disaster recovery strategies are available, what downtime-protocols should be followed when it goes down and up, who can I call when there's a problem I need fixed now, and finally who out there is already using it and demonstrated success with it?

I used to work at Epic and have seen in the field the requirements the immediate people need as I listed above. Billing driven documentation is an accurate way to label it. The software is implemented to maximize revenue for the hospitals and organizations, without reliable targeted information coverage agencies won't pay for what was supposedly done. There are whole teams in health care and applications from suites like Epic for such teams just for refining billing. A physician knows what they are ordering for a patient, a temp worker downstream refining billing data doesn't, therefore prioritizing accurate data from the physician will result in better likelihood of obtaining claims downstream. That however competes with the immediate need of the patient.

Re: Doctors ask engineers to spend more time in the hospital before building apps

#32

Post by chapium »

oomkiller wrote: post #27

Unfortunately the problem is much greater than engineers not understanding doctors and other clinical staff, in my experience. For startups that want to sell to health systems and similar-sized/larger entities (really this is the minimum size that can work for most startups, practice sales usually have more friction than value), you unfortunately have to focus on the buyer, which is very rarely someone who is "in the…

Cerner and Epic are billing driven because that is priority 1 for hospitals.

Re: Doctors ask engineers to spend more time in the hospital before building apps

#33

Post by chapium »

pragone wrote: post #16

I’m not at all surprised. As a semi retired software engineer and currently 4th year Med student, I still maintain a dream of building a usable EHR that puts the clinical, patient-focused side of things first. Currently implementations allow for increased billing recovery, but at a cost to both doctors and patients. I just don’t know how you’d start in competing with something like Epic or Cerner, from a business sid…

Dont beat them, join em. Build a solution to a problem and integrate it into one of their html5 modules. Both systems offer breadth, not depth when it comes to solving problems.

Re: Doctors ask engineers to spend more time in the hospital before building apps

#34

Post by sys_64738 »

Software developers are incredibly poor at gathering requirements. The best choice for gathering those requirements is to have a management consultant who does nothing but study business process and document the current clinical process. Once that's documented then you can discuss what needs to change from that document to extract the core requirements. Having a software person shadow a medical professional will only lead to the medical professional telling the software person what they think is needed.

In other words, decouple system analysis from system design.

Re: Doctors ask engineers to spend more time in the hospital before building apps

#37

Post by bm98 »

oomkiller wrote: post #27

Unfortunately the problem is much greater than engineers not understanding doctors and other clinical staff, in my experience. For startups that want to sell to health systems and similar-sized/larger entities (really this is the minimum size that can work for most startups, practice sales usually have more friction than value), you unfortunately have to focus on the buyer, which is very rarely someone who is "in the…

Standards like SMART-on-FHIR [1] and CDS Hooks [2] have the potential to allow innovation developed outside of Epic and Cerner inside of those products. Both of those vendors even have "App Stores" [3] [4]. So far, though, there aren't a lot of apps in these stores, and none of my doctors (all of whom work for large academic medical centers and use Epic) have access to any third-party apps - so I do wonder whether the vendors (or their customers) may be putting up roadblocks that are slowing adoption.

[1] http://docs.smarthealthit.org/ [2] https://cds-hooks.org/ [3] https://apporchard.epic.com/ [4] https://code.cerner.com/apps

Re: Doctors ask engineers to spend more time in the hospital before building apps

#38

Post by Trisell »

It also doesn’t help that to break into this market you have to overcome years of customization and millions upon millions of dollars in sunk costs that management isn’t likely to throw out for something that “works better for clinicians.” Not to mention the cost of retraining entire hospitals worth of staff to use the new system.

Re: Doctors ask engineers to spend more time in the hospital before building apps

#39

Post by gesman »

I managing projects that specifically integrating Epic data with dozens of clinical healthcare applications, HR data, medications access data as well as provider behavior data to offer security, privacy and unified monitoring capabilities for privacy and compliance officers in healthcare.

It really does helps to “eat our own dog food” and spend time in hospitals.

For example I spent few days in large New York hospital studying operations of secure drug dispensing cabinets before building solution for medications security analytics.

Tons of interesting data with valuable insights on how narcotics are secured and accessed.

Re: Doctors ask engineers to spend more time in the hospital before building apps

#40

Post by protomyth »

sys_64738 wrote: post #34

Software developers are incredibly poor at gathering requirements. The best choice for gathering those requirements is to have a management consultant who does nothing but study business process and document the current clinical process. Once that's documented then you can discuss what needs to change from that document to extract the core requirements. Having a software person shadow a medical professional will only…

Software developers are incredibly poor at gathering requirements.

I've had the opposite experience. Good software developers are good at gathering and understanding requirements. The information loss from management consultants and software developers is pretty nasty. Frankly, if the the people coding the system are going off a requirements document that they didn't have a hand in then you might as well outsource the development. If the software developers don't become domain experts then you should be worried about the correctness of the software.

I often wonder if the general dislike of DSLs is caught up in this separation.

Post reply on HN