Curbside Consult with Dr. Jayne 8/1/22

I’m always excited to read about technology that makes a difference at the point of care or improves the patient experience in a significant way. Two articles with solutions that fit the bill popped into my inbox this weekend.

The first discussed a smartphone app that can help identify skin cancer in older patients. Researchers from Stanford Medicine used an app called SkinIO after seeing that the COVID-19 pandemic had created barriers to patients accessing dermatology services. They noted that for older patients, a 37% decrease in visit volume correlated with a 23% decrease in skin cancer diagnoses.

The SkinIO app helps patients capture photos of concerning skin lesions and routes them to dermatologists. The app doesn’t try to make a diagnosis, but rather highlights lesions that meet certain criteria and lets the physicians know that additional review is advised. The study was small, occurring between November 2020 and July 2021. There were 27 patients enrolled, all of whom resided in senior living communities in the San Francisco area. Researchers visited the sites in person to collect skin images, using a dermatoscope to further evaluate high-risk lesions. The SkinIO app flagged 63% of the lesions as needing further attention, although researchers ruled out cancer in the majority of those submissions. However, three of the lesions were determined to be cancer, leading researchers to conclude that the app can be a helpful resource, although it’s not 100% accurate.

For the average patient who might be able to make a dermatology appointment, the app might not be as useful. But for a patient with mobility impairment or someone in a skilled nursing facility who might require additional services to support an in-person visit, it could be a useful adjunct. It could also be helpful for patients that have numerous moles and would benefit from additional evaluation beyond the traditional full-body skin exam they might receive once a year.

I checked out the SkinIO website to see how they were positioning the solution. I expected it to be marketed to dermatologists, and it is, with additional features such as automated mole-mapping and tracking of changes over time as key points. They’re also positioning it for health systems, perhaps to be integrated into primary care settings to allow greater referral of high-risk lesions to dermatologists within a given integrated delivery network. The angle I didn’t expect was marketing towards employers. At first glance, it might be a good employee wellness offering for industries where the conditions that contribute to skin cancer are present. Examples of workers who might benefit include those in the utility, landscaping, forestry, hospitality, entertainment, parks and recreation, and maritime industries. I’ve added the company to my watch list. It will be interesting to see how they do over the next couple of years.

The second article covered one of the industry’s buzzword topics: precision medicine. There are plenty of people excited about the topic. Being able to figure out the perfect therapy for a patient based on their genetic makeup or other measurable factors is exciting. Who wouldn’t want a treatment that was perfectly tailored for them? It sounds great, but actually making it a reality in our increasingly broken healthcare system is a challenge. For example, I recently implemented a new EHR that has some pretty exciting pharmacogenomic content. Unfortunately, the health system implementing it has not made the decision to store the results of pharmacogenomic testing as discrete data, rendering the content useless. The article talks about such patient-level data as the “fuel” that is needed to power the advanced analytics “engine” that drives precision medicine.

Once the organization makes the decision to incorporate discrete data, other factors need to be addressed, such as ensuring that clinicians know what changes to make based alerts that might appear, and that they trust the system to a level where they’re willing to take action. Clinicians will also need support using the tool, especially if incorporating it makes visits take longer due to the need for additional counseling and education for the patient. A physician who is already trying to cram a 30-minute visit into a 20-minute time slot is more likely to fall into old and familiar patterns rather than trying something new. If they need to research a proposed therapy or read deeper to understand what an alert is trying to tell them, then tailoring a treatment is likely to be a no-go.

The article notes this, and additional mentions that the complexity of our health care system is also a factor. Tailored treatments will also need to take into account patient factors such as symptoms, history, and preferences as well as societal factors such as access to care and impacts from social determinants of health. I’d go further to suggest that insurance coverage, treatment cost, and financial resources are likely to trump all other factors for the majority of patients in the US.

The article uses diabetes care as an example, where additional factors need to be added to precision medicine algorithms that are used for older adults — lived experience, support networks, current living situation, and more. For patients with cognitive impairment, limited social support, or other resource constraints, the situation is even more complex. In my experience, adding culture and personal / family values to the list of considerations is key.

I’ve seen patients flatly reject cost-effective treatments that are standard of care because they are perceived as being at odds with cultural practices or beliefs. I remember one family where the mother wouldn’t consent to a necessary surgery for her child because of a perception that the child would have to live with “mutilation.” Those are the elements that it will be challenging to add to an algorithm, because they’re not readily quantified and often only come to light in a crisis or through a longstanding physician-patient relationship, which is all to frequently the exception rather than the norm.

The authors propose that we need to go beyond the traditional thought of precision medicine as “the right treatment for the right patient at the right time” to incorporate the element of “each patient’s unique context.” I wholeheartedly agree and look forward to additional work as we continue to quantify those contextual factors in ways that allow us to take action at the point of care.

How is your organization approaching precision medicine? Leave a comment or email me.

Email Dr. Jayne.