In 2000 I started an experiment in self-quantification with the goal to empower patients.
Empowered patients are the key to more effective care.
Our health care system delivers the most advanced treatments in the world, but it is relatively weak in Continuity of Care. For chronic patents this means additional cost, impaired quality of life, and lowered life expectancy.
We record an abundance of information (Big Data) for billing of medical services, but we have a hard time understanding how a patient is actually doing over the course of care and beyond. Here are some questions that are at the core of a patient’s experience:
– Am I experiencing symptom relief?
– Does my ability to function improve?
– Am I able to live a fulfilling life?
What gets measured gets improved.
The main thing we are measuring in healthcare today is the type and quantity of medical services provided. These measurements leave a considerable gap to what matters to the patient and at times can actually work against the patient’s best interest.
To tackle this problem I looked to offer patients a better way to tell their story and allow us to measure and improve what ultimately matters. As a start I decided to tackle the most prevalent question that doctors ask their patients: “How have you been?”
Patients find it difficult to recall what happened since the last visit.
1. Doctors have little time to listen – typically a patient gets interrupted within 20 seconds by their care provider. In our healthcare system, time for patient interaction is short.
2. Patients generally have a hard time remembering exactly how they felt over a longer time span. When patients record their pain levels daily they tell a very different story than when trying to recall at once what happened during the last three weeks. Our long-term memory for symptom severity is just not very accurate.
Lack of Structured Data
In addition to lack of time and recall bias, there is the challenge of recording a patient’s story in a measurable way. Structured Data can help us analyze and understand if the treatment is working and what to do next. These analyses can be done in two ways:
1. By pooling the data of many patients in a way that de-identifies a patient’s data and protects the patient’s identify.
2. Looking at one patient’s data alone. Since the patient’s story would be captured over weeks, months or even years, it will be possible to just analyze one patient’s data with a N=1 Design to better understand what works.
Enabling Visual Insight and Machine Learning
Once we solve these data capture challenges the door is open for data visualization and Machine Learning to empower both patients and doctors to ‘get better’.
to be continued…