“Ask not what disease the person has, but rather what person the disease has” . — Sir William Osler
A patient’s context, including social determinants of health, shapes health experiences, outcomes, and costs in profound ways. We know from research, and our shared experience in the clinic, that a patient’s context is inextricably linked with their health: Context, — including environment, social and economic factors, and health behavior, — is estimated to determine 80% of a person’s modifiable health outcomes.
When patients have unmet needs or have competing priorities they are less likely to have control of her chronic medical conditions, are more likely to miss scheduled outpatient appointments, and more likely to have preventable visits to the emergency department or to be hospitalized.
Context, — including environment, social and economic factors, and health behavior, — is estimated to determine 80% of a person’s modifiable health outcomes.
As we move into the era of precision medicine with the inclusion of -omic data (including genetics, epigenetics, microbiome, and proteomics), we have the opportunity to further augment our clinical data and act upon it in a way that is tailored to each patient. At MCW, the Genomic Sciences and Precision Medicine Center is developing research and clinical programs to drive these changes. However the promise of this transformation to open a new era in Medicine and healthcare, it’s important to understand that context is a more significant larger driver than genetics, underpinning the adage that zip code is more predictive than the genetic code for poor health outcomes.
PatientWisdom Kickstarted Our “n of 1” Journey
Today, we are at a juncture where evidence-based medicine is meeting with personalized — n of 1 — medicine. Data and information are the common threads between the two approaches, which are anything but mutually exclusive. We have to have data about people, their context, their needs, values, and preferences, and be able to combine it with clinical data, and -omic data where available, to guide decisions.
Our experience with PatientWisdom over the last three years significantly advanced our ability to understand from patients, at scale, what was most important to them
Over the last three years, we partnered with PatientWisdom to integrate patient contextual data in our clinical practice. [n.b. Inception is an investor in PatientWisdom] This experience significantly advanced our organization’s ability to understand from patients, and at scale, what was most important to them, what their preferences were, and what barriers they perceived when seeking care. In order to evaluate the clinical effectiveness of the program, we followed a rigorous approach based on implementation science. We studied thoroughly the clinician and patient perspectives and the clinical outcomes of the program. Following these programmatic evaluations, we have demonstrated that patients who used the PatientWisdom tool were more likely to report visits going extremely well.
One of our most significant findings was that having the PatientWisdom profile significantly helped our broader care teams, such as our pharmacists performing disease management, or our medical assistants or nurse practitioners who may not have the deep understanding of who a person is as much as a primary care doctor who has known the patient for a long time.
If you want your MAs to know the patients really well, …then PatientWisdom helps a lot. I don’t think of it as a hindrance as we want to try to move our clinic towards a true patient-centered medical home that actually walks the walk and doesn’t just talk the talk.” — F&MCW Physician
PatientWisdom enabled us to begin to collect data incredibly relevant to the care of the patient. Interestingly enough, in several occasions patients are more willing to share sensitive information through a computer rather than in a face-to-face encounter. We heard this concept repeated in our qualitative work with our patients and physicians, yielding important implications for collection of social determinant of health data.
A New Opportunity to Harness Data For Personalized Care
Rather than using the term ‘precision medicine,’ which is more apt when discussing using genetic or molecular markers to guide therapy in targeted ways, I will use the term personalized medicine or personalized care. Personalized care takes into account clinical factors, contextual factors, personal factors (what is most important, what are the goals of the person), -omic data, and physiologic/sensor/wearable data (when helpful) to guide care decisions.
Evidence-based medicine has certain limitations. High quality evidence is only available for a subset of the decisions that we make in the clinic, and oftentimes our patients’ circumstances or presentation is not an exact replica of those in a clinical trial. This leads us to extrapolate from what data we do have.
Shared decision-making is increasingly a process that we do in every patient encounter. Even what used to be simple decisions require significant discussion (screening mammograms at 40 or 50, to get or not get the PSA and how often, to name a few).
Newer data tools may also help us understand what possible outcomes may be over time for similar patients. Using classification and machine learning techniques, we can identify patients with similar characteristics, adding a new dimension to evidence that shifts more to people most similar to our patient in our clinic rather than extrapolating from a population of study participants.
To advance all of this, we need high quality data, which is why contextual data, such as we began to gather with PatientWisdom, is so incredibly important to capture well.
Although we are still at the verge of this transformation, we shall constantly ensure the balance between data privacy and data utility for care. We will need to both collect these data in scalable and usable ways, while also enabling patients to remain in control over how their data are used. This is a critical consideration that current healthcare organizations are not fully equipped to manage and operate.
Developing personalized experiences built atop evidence-based medicine and quality and value based programs will enable care to be more responsive to individual patient needs. Our clinicians already are accustomed to tailoring diagnosis and care plans according to personal circumstances. We are building upon this to enable our teams, our processes, and our analytics to further support this ‘n of 1’ care.
Our next steps are to …
- Collect patient contextual data continuously, and develop computing tools to help us guide patients and clinicians toward effective actions
- Use patient-reported outcome measures at scale to track patients’ conditions and response to treatment
- Deeply integrate information and data that patients provide with the workflow and care operations
- Develop ways to more easily and efficiently add genomic data, such as starting with pharmacogenomics, to guide care where measurable value can be provided
- Continue to roll out tools that provide cost estimates to support decisions and high-value care/prescribing
- Integrate ‘efferent’ actions following what we learn about social determinants
- Create a digital experience for patients that is personalized to their needs, goals, and circumstances
We have plans in the works for each of these areas, and I’ll be sharing more about them over the coming months as they go live.