HEALTHCARE INDUSTRY

System in Hospital - A User Experience Case Study

The hospital recently purchased and installed monitoring hardware for each hospital bed
in the intensive care unit. This hardware continuously measures the six core vital signs for each patient and sends them to a central database.

There are nowhere near enough nurses to regularly measure every patient's vital signs. If a nurse misses a measurement, the patient might have a heart attack or stop breathing
without the medical team's awareness. This cost lives every year.

ROLE

UX Research & Visual Design

SCOPE

Hardware Application

TOOLS

Figma, Pen Paper

TEAM

Individual

The Problem Statement

Owing to the unforeseen circumstances due to the Global pandemic, there's a severe shortage of healthcare workers for them to be able to physically examine every patient, and continuously monitoring them is no easy task. A Lot of lives are being lost in failing to recognize and act swiftly on critical patients. The patient-nurse ratio at hospitals is alarmingly out of proportion. So I decided to design a central monitoring system that keeps track of the crucial vitals.

This is my take to design a solution to address this problem.

Understanding the problem

The way I saw this problem can be solved is by introducing a central monitoring system for every critical patient in their bed, so the nurse can monitor every patient's vitals constantly.
 
This can be achieved by connecting a physical device that detects vital signs and translating those values over the cloud to a database and displaying them in a meaningful way that the nurses can consume.

Then I had to quickly design the process flow to make the High fidelity user interface.

Research & Assumptions

Here are some Assumptions made before jumping into the UI Design

  • This System is mainly focused on Nurses so they can monitor patient's health conditions. Nurses are the Users who will be using this system.
  • There is now somehow an automated way of extracting vital signs from patients (maybe through attaching certain devices to the patients)
  • This design was concentrated on a single hospital rather than the full network of hospitals

The Research behind the process:

  • Research on how the current medical system is monitored, what data they display
  • Organizing the data accordingly so that no crucial values are left behind. This cost lives.
  • The mental model of nurses, how they are used to see this data ( this is crucial, we cannot design something they don't understand)
  • I had a call with my relative who is a nurse and took inputs from her on what type of data they measure and how they do it. Doing this gave me some interesting insights and findings.

Who are going to use this?

Since this system is going to be primarily targeted at nurses I spoke to some nurses and came up with a User Persona. This is a purely imaginary persona with inputs being legit and this is to be considered just for the sake of this assignment.

Existing Mental Models

Now that I have my process flow, defined my target users, have a rough idea of my process flow, now I had to do was to figure out when and where, and how this application would be used.

I quickly researched how the current working models for patient monitoring systems exist. Based on which I had to assume that nurses are mentally aware to see the monitoring systems as below

Initial Ideas

I then sketched out a few low-fidelity wireframes of the user interface. The objective of this task to come up with two or three solutions for me to begin with.

The main thing I had to keep in mind was to stay neutral to the ecosystem and not to go super fancy with my user interface.

Displaying the vital signs

Taking inspiration from the currently existing monitoring models I had to break down the real estate of each screen into 6 sections (Because we were given to measure 6 vital signs) and then appropriately adjust them accordingly.

The hierarchy is based on the type of data here's how I organized them :

- The signs that used graphs to visualize, I put them first and split them in a (Graph - Data ) Row
- The ones which don't use a graph, I arranged them next in the hierarchy.

Also, these were designed were iterated based on the feedback and dry user testing done on the wireframes, and keeping in mind the mental models of the nurses.

Here's what I came up with :

Testing the User Interface

To justify the direction of design I was going, I had quick user feedback with the relative nurse I mentioned earlier and showed her the design which I did (on the left) and the design I randomly pulled out from a dribbble shot( on the right ) and ask her which was easy for her to recognize the data/vital signs from

Without even thinking twice she said the first one because she felt she was familiar with that. and the second one sure looks beautiful but she was having a hard time understanding what was actually going on.

Of course, this is not the best way to measure the usability of the interface by putting it next to a dribbble shot, but this gave me an insight that not all beautifully looking interfaces are usable, at least in this case


I proceeded with my design. and choose not to redesign

These types of feedback are important as they tell how to make design decisions

UI States

Now that we have decided the basic layout of how a design is going to look like, now it's time to work on the user experience aspect

So I came up with the following design to make sure the nurses understand the condition if it is alarming or critical by choosing to highlight the respective vital signs with the high contrast colors

RED for critical

YELLOW for alarming

What problem did it solve?

The above system works fine for bed patient monitoring (in terms of not having to manually examine the users for vital signs) and these data are sent to a central monitoring system through which the nurses can keep a track of each and every patient

Below is the design I came up with for the monitoring system, which was categorized into Floors (assuming the hospital has different floors) and directly showing the monitoring screen of each patient on each floor while displaying the bed name and the patient name on the top

Interface in Action

Displaying every monitoring screen would be helpful to have a glance at each and every patient simultaneously without having to switch into another window or click to expand a screen and since this design is already familiar to nurses since they are mentally used to see the monitoring screens of the in-bed patient monitoring system(the design above) They will find it easy to navigate through.

Obstacles and Opportunities

  1. What Worked:

    • Feels native, and familiar and requires a little to no learning of the system
    • Familiar mental modal 
    • Each and every vital sign monitoring at a glance for every user
    • Simple UI with fewer complications and very little learning curve

    What Didn't:

    • Needs bigger screens to work (maybe 3 or 4 screens to monitor if the size of the hospital is big )
    • Cognitive overload and can be hard to navigate
    To solve the above problems, I have decided to go for another Modren UI interface approach

A Cleaner Approach

To tackle the obstacles in my first iteration, I decided to redo the design over a second iteration to test it over and see if we can overcome the imperfections.

Similar to the previous model, I decided to go with a touch of modern approach and get rid of the full screens of the monitoring dashboard that were previously visible and decided only to include the vital signs in the form of tables

They share a similar function visually, to the previous design in terms of notifying and alerting the nurses.

Clicking on the highlighted block, Pulls up the Native UI from monitoring device

The Outcome

  1. Information is now a lot cleaner and easier to navigate
  2. Getting rid of the full monitoring dashboard gives us a lot of screen real estate to accommodate more patients data.
  3. Included an option to search for patients quickly so the search can be done by either putting in the name of the bed number of the patient

Conclusion

Since most of this design process which we just went through is purely based on assumptions and a small amount of feedback from the users, the problem with this is we haven't tested the design with the real, potential users..yet, who will be using this system.
 
So to make sure if the design is production-ready, large-scale testing has to be done by involving the nurses and doing usability testing.
 
This could be easily achieved by giving them small tasks such as monitoring patients health and navigating to the home screen, maybe changing the current floor that is being monitored, etc
 
And upon keen observation of how the users are interacting with the design, we have to take notes and do further iterations.
 
Design is an iterative process.

 

And once we reach the point where the users are mindlessly navigating through the design also with a little bit of education on how the design works and selecting the best interface and practices while also ensuring the system is error-free and bug-free the design should be ready to be shipped.

Next Steps

Is the product is ready to go to market ?

Nope, not at the moment.

Healthcare is a really sensitive area and since we are dealing with the design that consists of patient vital sign data that is really crucial in deciding and making decisions based on them

Hence this is a huge responsibility from the user experience perspective and the development perspective to ensure there are absolutely zero to no errors that will occur in this design system and also to make sure that no information is left behind or un-monitored

What would I do different now ?

Identifying the stakeholders is vital decision-making of the design process show involving the stakeholder's area helps accelerate the process of production

Since this is a fictional project, I have not considered the accountability of the stakeholders for this one, but if I had taken that into account, I would be so sure to have made a lot better design decisions in the process.

The stakeholders, in this case, maybe doctors, investors, policymakers, providers. It is crucial to understand and map the stakeholders accordingly in order to ship the product
pencil