Humanizing a complex medical device
2006 was an innovative year for diabetes management. A unique innovation was introduced to the market that altered the way diabetics can manage their illness. These devices are known as Continuous Glucose Monitors (CGM). CGMs measure glucose levels in the interstitial fluid space below the surface of the skin. A small and flexible wire is attached to a disposable body-worn sensor patch, which transmits measurements in real-time to a phone-sized monitor carried by the wearer. By delivering a steady stream of glucose readings and, when necessary, alerts, CGM has enabled diabetes patients to correct their insulin dosing ratios and head off hyperglycemic or hypoglycemic events before they occur. Traditional finger-prick testing cannot achieve the same.
While I was working for the consultancy frontend we were commisioned by a client to conduct some research and concept development for a CGM device. I volunteered to be the guinea pig who would wear the device and track the progress of my sugars over time. I had some personal motivations and interests in choosing to do this — my father is a Type-2 diabetic, and there is a history of the disease ravaging other members of my family. I was keen to see if there is any sign of dysfunction in my own pancreas. So I submitted to having a centimetre-long titanium wire embedded into my body for as long as it would last. My hope was I would learn something about myself through doing this and it would help me when it came time to design for this particular problem space.
Having a sensor installed into your stomach is an odd experience. Two colleagues helped me conduct the procedure, the manufacturer provided a helpful YouTube video to guide the process. The video instructed us to select an appropriately smooth part of my stomach that would be a safe place to insert the wire without it being dislodged by the creasing of the skin. The inserter was then pressed against the chosen section of my anatomy and through a complex multi-step movement the sensor was inserted into my skin. There wasn’t really any pain, although I was very aware of the sensation that there was a piece of wire inside my body. I could feel it shifting slightly as I moved.
The inserter that we used to install the sensor was a complex looking device. In appearance, it was not dissimilar to a large plastic syringe, the design of the device did not immediately signal how it should be used, as such the instructional video was required to guide the process. I am doubtful that we would have been able to install the sensor without guidance. It seemed to me that the design of the device was un-empathic to the conditions of chronic illness. My sense is that if you live with a chronic illness, in your own mind you will already be separate from your healthy peers. It seems to me that the medical devices that allow us to manage these issues should be designed to minimise any feelings of isolation and difference. It is not enough to merely create a device that works, our ultimate goal should be to improve the entire experience for the patient, as much as possible.
Following on from this the device that accompanied the sensor had to be calibrated. This involved me having to finger prick to extract some blood and then inserting the sample into a glucometer. In order to ensure accuracy, this has to be done twice, every day for the duration that I was to wear the CGM. This is a procedure that all diabetics know too well. It is a daily fact of the modern diabetic’s life that they have to injure themselves, slightly, several times a day in order to monitor their sugar levels. The promise of CGMs is that one day this irritating inconvenience will be removed from the experience altogether.
My initial interest was in what it would be like to be aware of how much sugar is in my blood. At first, as you can imagine, it is fascinating to watch how the line moves up after you eat and then drops again a short while later. After the initial phase of curiosity though, a type of mild anxiety sets in. You find yourself pulling the receiver out of your pocket at odd times just to see what is going on as if by checking it some new insight into your current experience can be gleaned. All you see, however, is the dotted yellow line plodding along.
The readings the sensor gave were not always consistent. Sometimes it would stop communicating with the receiver and there would be periods of sensor silence where the dotted line would be terrifyingly absent from the LCD screen. Until it returned, inexplicably, sometime later. My initial assumption was that I was of course terribly ill and needed medical attention immediately. However, after consulting with my colleague Michelle, our resident medical device designer, she informed me that this can sometimes happen due to dehydration. Note to all consumers of excess amounts of coffee, this will dry out the interstitial fluid under your skin. Be aware.
There were also alerts set up on the device to let you know when you moved out of your normal range. On a few occasions over the first few days, the receiver piped up to inform me that my sugars were below 60mg/dl and on one particular occasion it told me I was below 50mg/dl. In this particular instance, I endured a bout of short but intense terror. In the world of diabetes, a reading this low would mean I was hypoglycemic — in serious trouble. I sat for some moments, prone in my chair waiting for a diabetic coma to descend upon my sugar-starved body. The moments passed and my consciousness remained as intact as usual. The only notable symptom of sugar deficiency was that I was slightly hungry. It seemed that my CGM was not being entirely honest with me. However, after the first few days of use, these unusually low alarms seemed to taper off. It seemed it was getting used to me just as I was getting used to it. In the midst of all this fear and anxiety brought on by excess knowledge about my blood sugars, I began to speculate that a dystopian end-point for all this wearable technology and increased self-awareness could be a society of terrified hypochondriacs. As the Greeks told us “Know Thyself”, but perhaps there should be sensible limits to the extent of that knowledge.
When thinking about the monitoring experience we envisioend four unique personas that might use the device.
We conceived of a parent persona with a diabetic child, the parents as the caretaker of the child would need access to the childs CGM readings to help in managing of their condition through diet and insulin dosing.
We concevied of the average daibetic user who would use the CGM to help manage their condition as a replacement for the usual glucometer approach.
And lastly we conceived of an athlete type persona who wanted to use the CGM to help them manage their performance as they competed.
The proposition from a product perpsective was that the client could build a single application and using simple skinning and feature flagging they could alter the application for each personas needs. There would be obvious cost savings to be gained by taking this approach. The primary investments would be in the product marketing and then specific feature development as the needs of each persona were worked out.
Taking a persona centric approach to the development of technology not only makes sense from the perspective of creating technology that users find easier and more desirable to use, it also makes sense from a cost saving and development perspective.
By thinking of our users first it allows us to make smart decisions in how we build technology.