DocNotes: How I designed the process of hospital prescription writing in 24 Hours (UX Case Study)
Back in June 2016, I joined the AngelHack 2016 in Kuala Lumpur with 337 attendees from all different backgrounds. Attendees list vary from Hardware, Software Developers to UX Designers and Entrepreneurs. All gathers under the same roof in Berjaya Times Square.
My team and I wanted to solve a real world problem through the availability of existing technology. The requirement of this hackathon is to design and build a working prototype that is related to pre-selected topics in 24 hours.
The topics were: O2O Commerce, Big Data Analytics and Smart living.
We had the choice to either select one topic or combine two topics together to solve a problem. I personally was interested in the smart living category but it was very challenging to come up with an original product idea, instead we started by brainstorming and talked about how hospitals are still behind incorporating creative solutions and technology into their process.
Defining the problem:
When we looked at hospitals we realised there are countless and sometimes undocumented incidents where nurses write down the wrong prescriptions given to them verbally by doctors, or the pharmacy assigns the wrong prescriptions due to poor writing skills.
DoNotes comes in to reduce human-error by introducing a solution that could potentially prevent such errors to occur in hospitals. The ultimate objective is to save more lives. Introducing a well designed solution that can help doctors and nurses be efficient with writing prescription and digitalise the process to replace the traditional pen and paper work is a difficult task, but not an impossible one.
The idea: How do we solve it?
The traditional process requires patients to consult with a doctor in duty. If the doctor requires the patient to stay overnight. The patient will receives a smart wearable and get assigned to a hospital ward. Afterwards, all prescriptions will be sent directly into their wearable devices through a push notification. Nurses check each patient and mark prescription as taken or report an emergency (In case of an emergency)
- Smart wearable: Android wear(patient)
- Datacentre: Used to sync info to cloud (Hospital)
- Cloud software: Dashboard (Doctors & nurses)
The design process:
First and foremost, I quickly crafted user journeys and map out the process in a simple yet complex diagrams. This step is an important one to insure clarity in the process and connect the missing pieces. I began by drawing down the three personas side by side. We have the patient, nurse and a doctor. Refer to graph below for details:
Secondly, I wanted to craft down the features set and acceptance criteria for our software that will ultimately be accessible by doctors. The initial value of the dashboard is to provide useful information about patients packed with some primary features.
But before I start on that I need to define the primary actions that doctors would be able to accomplish using the system.
I came up with a list acceptance criteria:
- Add patient information to the system
- Add prescriptions
- Switch between hospital wards
- View current number of patients in ward
- Easily find patients that are marked with an emergency status
- Filter patients information based on (patient name, admission date, schedule, emergency and discharging status)
- View the history of patients’ information
I started to visualise the dashboard with quick and dirty pencil and paper sketches (That I currently don’t have anymore) and so based on the requirements above, I created a well thought Low-fidelity prototype. Here it is:
I had a clear idea of the creative direction that I’m taking earlier in the process, which in fact helped me to speed things up while creating the High-Fidelity visual prototype.
Lastly, I had to complete the process of the patients that have a smart wearable. I decided to work on the high fidelity prototypes for the wristband (to save time) and explain the process verbally to our android developer.
The direction of the journey was very clear to me that I wanted to make reading and marking prescriptions as easy as possible by using the native android watch swiping effect. I also wanted to make use of the voice command within the wristband to report any form of an emergency.
Without further due, the UIs are down below:
As much as I tried to cover up all possible scenarios, there’s always a possibility that I may have skipped or ignored certain parts of process. In addition to the fact that I only had 24 hours to come up with a working prototype, the whole process was based on hypothesis and ultimately assumptions. Which raised up questions that were tough to answer later on through the process. Here are some:
Will we require to design a hardware for patients to wear?
The answer is YES. The hardware needs to be a simplified version of the android wear (maybe) since the device will only be used to receive and send prescriptions data.
How many datacenteres are needed in one hospital?
Each ward within the hospital will have its own datacenter that will be responsible to send data back and forth between the DocNotes system and patients devices.
Will these devices have any side-effects on patients?
They must not.
- Designing digital products for hospitals is hard but not impossible
- Patients data are sensitive, therefore the datacenter is an absolute necessity to be located within the hospital
- Wearables my not be suitable for all patients
It was both fun and challenging to work on (in my opinion) a complicated project without prior user research. Most importantly, I had lots of fun designing it and now that I’m finally publishing an article about it, makes it such an accomplishment and personal milestone. Although this product idea will not make it to production any time soon (maybe never), I hope it will inspire you to look into complex problems in life and try to solve them with your superpowers.
I would like to thank my team and the organisers of Angelhack for this opportunity. We didn’t win the first price, but 3rd place under the smart living category is not bad at all :)
Share your thoughts with me!