Why I quit medicine


I’m a doctor from the UK. Here’s how it works: I have patients in hospital. They are sick. It’s my job to make sure they survive and then to try to make them better.


However, I don’t have the tools which make my job that much easier. Considering I can talk to my smartphone and tell it to send a message to my dad or remind me to water the plants when I get home - why is it so difficult for me to get access to vital information about the people I look after?


Here’s an example: at the beginning of my shift, my patients are handed over to me by the team who have looked after them for the last 12 hours. Logical, yet this process is surprisingly painful. At the moment, I have to write things on a scrap of paper and carry that paper around. Each time I find out something new about one of my patients, I jot it down. When I then hand those patients over at the end of my day to another doctor, they write on - wait for it - a new piece of paper.


Some people use the ‘Notes’ application on their phone to jot this stuff down. Slight problem with that - you’re not supposed to put patient-identifiable data on your personal phone. It’s not very secure, is it? But then again, neither is carrying a crumpled scrap of paper with the same crucial information. It’s a daily occurrence that people accidentally leave that paper somewhere, which is easily done - you need two hands to take blood from someone or listen to their chest. The next thing you know, that scrap of paper is gone.

I recently talked to a senior doctor who proudly declared that their department had implemented electronic ‘handover’. He told me it was something that the hospital’s IT department had built in-house. Apparently it’s working really well and people are happy with it. Wow! This I had to see. I went there, and as a hacker, “wow” was my exact reaction. Can you guess what it was? Yep - a Word document.

A Word document installed on 4 hospital computers with a list of all the patients on the ward and some crucial information about their care. It would be nice if I didn’t have to call the person on the other side of the ward to close her window, so that I can edit my patient’s information on my computer - but perhaps that is too much to ask. With the current system, it is difficult to find out who wrote that this patient’s lungs sounded a bit wheezy, or that their family wants to discuss something. Having something which means people are accountable for what they write is crucial.

This process is so emblematic of how patient data is shared in healthcare. This is why I quit a great job - the one I spent years training to do. People need to find out about how broken things are in healthcare, and as one of the few doctors who can code, I believe I can help fix this. This is 2012. It’s high time that looking after patients was made a little bit easier.

After saying goodbye to my life as a doctor, I’ve built a simple, web-based prototype - which plenty of doctors are eager to use. At the moment I’m in Silicon Valley to try and take things to the next level. If you’re a talented hacker, who thinks health is important or wants to build something that solves a real problem - get in touch.

26 responses
Absolutely amazing! Glad someone is taking on an industry with such out-of-date processes.

Curious: What language/framework are you using?

I have spent so much time in hospitals with loved ones, and wondered why they can't get their act together. I understand it's a tough a stressful job -- but I'm glad someone with your real world experience is going to tackle it!
I don't get the problem with the paper relay system. Humans have figured out pretty well how to secure physical objects. As soon as you store patient data on a machine connected to a network, you're sacrificing their privacy.
@earlcochran Many thanks! I've been using Ruby on Rails atm :)
As nsaphra said, computer networks have their share of problems, that said any group of programmers can setup a multi user 'quite secure' system for a lowcost (close to peanuts for hospital budget). Plus you know the job and the mindset.

I'm just worried about legality, standards and certifications.. I guess any healthcare place has to conform its computer system to some kind of standard that requires costly fees. But considering what the other hospital uses .. I guess it's not a real obligation.

Good luck

I think you really have the right idea here with this. I hope that your software works out! The great thing is that you were a Doctor, someone familiar with the industry and who understands what Doctors want/need. Not Joe Programmer who has no idea what a Doctor could want/need in a program such as yours. You all have enough to worry about, so if there is something to make it just 10% easier, I would say it's worth it. Good luck to you!
It sounds like noone looked at the user experience of doctors before. Great that you've built something simple.

Possible alternatives:
- the open source GNUmed project at http://wiki.gnumed.de/bin/view/Gnumed
- an internal wiki like foswiki.org

Any chance you will open source it?
I come from a family of MDs who report similar complaints. Good luck with your project!

A few suggestions:

1) Assign a smart strip to each admission then configure all diagnostic and treatment equipment to report their results/status to a central server keyed to a specific admission on a real time basis. EG, Nurse A takes blood pressure of patient Z? Nurse activates Android tablet, scans strip, employes BP gauge configured to relay data to tablet which in turn relays data to central server. No input errors and who did what when is transparent.

2) Get your own radio frequency. To punch through concrete you would want frequencies between around 400MHz and 2GHz (personally, never liked 2.4GHz since that's the same as my microwave). Then encrypt the signals.

3) Physicians get sloppy with small input devices. Like it or not all physicians would have to lug around a light notebook, suitably WiFi-ed and encrypted (which of course would remain in the hospital). Trick here would be to keep all personal information in the 'cloud' and not on the device.

However, don't be surprised if at some point a patient wakes up and the following dialogue occurs: Where am I? In the village. What do you want? Information. Whose side are you on? That would be telling, we want information. Well you won't get it! By hook or by crook we will! Who are you? Number 2, you are number 6. I am not a number, I am a free man! Ha Ha Ha... [from the opening to the ancient 'Prisoner' series, replayed sometimes on PBS during pledge week].

If you are really serious, then the most effective approach would be to enable remote telemetry (biosensors, patient entered data, etc) that can be connected to the private physician. No data would transit to you, rather you would act like the agent who facilitates a skype call. Customer calls his health care provider (call goes to you, you sanitize his database and let him know what needs fixing, when all ok you transfer data connection to provider). Just like Apple, you take 30% commission-provider gets the rest.

OpenMRS is helping people in Kenya, but we cannot use it here because it's not HIPPAA approved (And it was invented here!)

Best of luck to you in this endeavor--when you're ready to integrate your app with a similarly forward-thinking EMR, let us know. www.practicefusion.com

(and we could help in your search for health-conscious engineers too)

The project is very interesting !

Germany has a bigger government project that starts to take care of the problem. The idea behind it was to centralize the patient data and give doctors access to this data. But it is a very political project and moves only very slowly and with to many features removed, changed and minimzed.

I have seen very funny hybrid solutions where nurses write their notes on paper and doctors have paper and a tablet pc (windows with a pen) and try to get things together.

I would start with the smallest step and get the notes in digital form on a tablet for the doctor and the nurses. A lot of other things are already digital and can be easily transfert (CT, Sono, ...). Direct linking between the devices could be very difficult IMHO.

Hi. I'm a doctor too in London and I really understand your point about the handover system. My partner's a hacker too, and I sort of understand that side of the technology world.

Yes we too have the "electronic handover system/just a word doc" which I stay clear of because quite frankly I don't see the benefit of it over the original way the acute take list is handed over.

I do agree though that hand over is crucial, but I don't think I can see how an electronic 'app' is going to make things easier.

Patient's who are handed over are discussed with you when you are coming on shift so the relevant history is already going to be verbally communicated to you. For me, I only write the jobs that I need doing for them on my 'list/scrap piece of paper' so I can cross them off when I'm done. What would the app provide? Paperless/electronic jobs list? How does the doctor get on to the system. Is it realistic that each on call doctor will have the time to get on to this system on the computer, log in and check off his/her list. Would the doctor not be rushing to do his next job instead after having written in the patient notes and signed and dated it and ready to get on with his jobs or see someone sick. Maybe a hand held device with a jobs list could be more efficient, but you will need to have all the doctors doing the same thing if you want no paper lists being used and lying around which is difficult to achieve. Furthermore, I don't know which hospital you have worked at, but for me the computers I've come across are simply old and slow. Frustrating!

I suppose my question really is what are you aiming to achieve from this new app/electronic hand over process. Is it to just provide an electronic form for jobs lists? Perhaps thats achievable. Or do you want to access patient history and investigations of people you are not familiar with when you are on call - thats tricky especially when we are still handwriting daily ward round entries/patient notes. This will need to be transitioned to electronic patient notes which I know is an aim for the healthcare system but, how much of the health budget is that going to cost?

Escherplex's first point reminds me of the computer system at A+E. They have a good way of gauging how far along each professional is doing for each patient. But that system is strictly for A+E.

I apologise to appear negative especially amongst all the positive comments above, but I haven't seen another doctor sharing his views. I'd be interested in your final plans, and I'm more than happy to share more opinions from my experience.

If you need help with databases (MySQL), I volunteer for my time for any medical institutions (or software that will end up getting used by medical institutions) http://www.jonathanlevin.co.uk/p/volunteering.html
Well, I live in Brasil, and almost every hospital (exept the public ones) have software to register patients conditions. I think that is a common problem, so why instead build your own, you could make contact with a brazilian company. The software take care of the entire hospital, it's like an ERP.
I am working for a company in Kentucky (USA), and we are currently working (with the state's Health Department) on EMR (Electronic Medical Records) software that addresses this precise issue.
Yes! That piece of paper! The problem is not limited to doctors -- nurses suffer from it, too. We call the piece of paper our "brain". More about this at http://www.nursesgetitdone.com and http://www.nursing.keller.com/whatshard.html
Sounds like an interesting problem - having worked in dev for healthcare before, I've seen some of the "solutions" that people end up using, and they really aren't suitable. Healthcare deserves better IT.
Hi Gautam,

I am another "refugee" who left clinical practice - I am truly sympathetic to many of your comments. I hope your project brings some useful order to the chaos of medical information sharing!


its really good source of information as after reading this post i come to know about this story.....
Online Safety Products
6 visitors upvoted this post.