My Role
I lead the complete end to end product design lifecycle for this feature… right from research to analysis, identifying gaps to defining the MVP. Designs are currently being socialized amongst senior leadership to influence upcoming roadmap.
Unfortunately, I was not able to see through to execution for this project.
Background
Tempus has 3+ million de-identified patient molecular and clinical data
Problem
We came at an inflection point where we asked the question “How can we use Tempus's in house Molecular and Clinical Data to help clinicians give better care for patients with Neurological and Psychiatric conditions?”
How can we leverage this patient data to help Clinicians give better care for their patients?
Solution
After conducting surveys and interviewing with our users, the team decided on an initial MVP. We created a cloud based precision medicine platform called Tempus N+ that analyses these complex datasets to help clinicians create a successful treatment plan for their patients. The tool provides a ranked list of medications from best to worst based on the patient profile.
Tempus N+ is a cloud platform that analyzes data for personalized treatment plans, ranking medications based on a patient profile.
Solution Spotlight
⚡ Take me to Final Design⚡
Research
Understanding the user journey through user interviews and survey.
We wanted to use clinical and molecular dataset to help clinicians make a better treatment plan. But what are our physchiatrists' pain points? Where are the gaps?
To understand this, we first had to identify shortcomings in achieving a successful treatment plan. We conducted 5 user interviews with different User personas(Generalists, Experts, Nurse Practitioners, PCPs). We also used a platform called Sermo to gather quantitative research insights.
Research Insights
Drugs are selected through a process of trial and error
On an average a patient is experimented with 3 drugs before finding the right one.
Lack of managing patients' treatment expectation leads to non-adherence
Patients become frustrated if they have unrealistic expectations about the extent to which a treatment will alleviate their symptoms, causing them to stop taking medications prematurely.
Misdiagnosis is the main cause of treatment failure
Misdiagnosis can result in the prescription of medications or therapies that are not effective for the actual condition, potentially leading to unnecessary side effects or complications.
Ideation
Aligning on feature prioritization with stakeholders using Empathy mapping
My PM and I socialized the key research learnings with our stakeholders. Since we had a broad set of stakeholders this helped us to come onto the same board in understanding who are users are and their painpoints. I took this a step further and facilitated an empathy mapping exercise. I did this to identify key solutions we wanted to build in our MVP. The key solutions we ideated on were:
Ranked list of Optimal Treatment options
Time on Treatment of a specific drug
Risk prediction of other disorders
Connecting the the Genetic Report and Tempus PRO ecosystems for easy access
A simple data diagram for further alignment between stakeholders
The diagram below was shared with the company to share a high level functionality of the Real World Data tool. This shows what data points the algorithm will utilize to provide the results. And finally how this helps with push forward Tempus's mission
Final Design
Moving from Trial and Error to Precision care
A Funnel chart to give a glimpse about the quality of data
The official word for this is called Data Completness. The diagram provides information about the data being inputted to the algorithm to inform results. This is important to instill trust in our users. This can be collapsed to provide more screen space to the list below
Using data from 3+ million patients to define most commonly used medications
Here we show a list of top medications that were prescribed to those who match the patient profile. This helps physicians eliminate trial and error and surfaces drugs that are not commonly seen in their inventory.
Another interesting layer onto this is the ability to overlay it with the patients genetic test result.
Access all Tempus offerings in one place.
Surfacing high level insights from the patient’s genetic test results helps with identifying the right medication for the patient.
Similarly the psychiatrist can also assign assessments to the patients to follow their journey using another Tempus offering called Tempus PRO. Here we surface only the most needed assessments for this patient profile.
A second opinion while assigning a diagnosis
Because the mind is so complex, we discovered that it is infact really hard to pinpoint the right diagnosis. We would have to walk a legal tightrope when we get really prescriptive about the diagnosis. So I experimented with this option of using a public dataset to identify closely related diagnosis.
This could help our psychiatrists by opening them to the possibility of a different diagnosis. We entrusted our psychiatrists to do their homework from here.
Managing patient expectations using Time on Treatment
Showing how a drug performs over the time of its treatment helps patients adhere to the medication for longer. When you know when the drug starts to kick in or even have an idea about the aide effects beforehand helps the patient to manage their expectations. Here there are two main elements we could determine from the data we have:
Drug effectiveness
Side effects
This is designed so that the physician can turn the screen around to the patient to share exactly how they will perform on this drug.