For patients, doctors, and app developers alike.
In the first season of House, M.D., the episode Histories begins with an unconscious middle-aged homeless woman being admitted with nothing more than a puzzling twitch to go on. While his team launches into listing all kinds of conditions that might explain the unconsciousness and the twitch, House makes a crucial yet basic observation.
“Who is she?”, he asks. “The only thing we know for sure about Jane Doe is that her name isn’t Jane Doe. Which means no medical history, allergies, medications, previous diagnoses, treatment, we have no baseline, no context for medical treatment”.
In medicine, figuring out the diagnosis is like solving a puzzle. Some are trickier than others, but they all require knowing enough about the puzzle in the first place. The symptoms the patient presents with start off this process of deduction. As the patient’s story unravels, the doctor rules in and rules out diagnoses that could explain what’s going on.
Arriving at the right diagnosis (or differential diagnoses) relies on having the right information. The patient history is by far the most crucial source of this information. It consists of the story behind the current problem; and, the patient’s medical background. The history helps in deducing a lot more than just the likelihood of potential diagnoses. It also gives an indication of what treatment escalation may be needed, or what treatment may not be safe. For example, a medication may work well in most patients, but giving it to someone who is allergic to it can be fatal.
Eliciting the story of the current presentation largely relies on the patient’s relatively short-term memory. The medical background, however, can be trickier to uncover.
While a lot of the patient’s past medical information is stored in great detail in clinical systems, that information is often siloed. The reason for this is that different providers often use different clinical systems and these systems struggle to talk to each other. Things get worse as patients change doctors, or move between primary and tertiary care. The result is that eliciting the background picture of health often requires going through stacks of paper notes, chasing down previous health records by fax or phone, or relying on the patient’s memory.
Unfortunately, the sicker the patient is, the more medical background there is to know, the more critical it becomes to know that background, and the less able the patient is to recall that background! An unconscious patient with no collateral history, like Jane Doe in House, M.D., can understandably become a diagnostic challenge.
Enter Apple Health Records
Earlier this year, Apple introduced an exciting initiative that tackles this problem head on.
The initiative, called Apple Health Records, is a new addition to the Apple Health app. And it does something quite remarkable. The patient data that has remained stuck in monolithic clinical systems for years can now be extracted and collated securely in a single app right on the patient’s iPhone. Giving your doctor an accurate and detailed insight into your medical background may become as easy as opening a single app.
For doctors, this could mean readily available information about the patient’s background that helps in formulating a management plan. Being able to see comorbidities, previous treatment escalation, response to treatment, and relevant imaging or lab results can quickly paint a picture of the patient’s overall state. This information forms, as Dr. House said, a baseline and context for medical treatment. In emergencies — like an unconscious patient with a curious twitch and no collateral history — having access to the background right away could save a life.
Apple has managed to communicate with disparate clinical systems by getting these systems to speak a ‘common language’. This language, or rather specification, is called FHIR. Essentially, FHIR provides a comprehensive healthcare ontology that lets computers send and receive medical information in a consistent way. As FHIR is an open standard, any software that chooses to speak in FHIR is then capable of understanding other FHIR-speaking apps. Apple has collaborated with close to 40 healthcare institutions so far. Once on the iPhone, health records are encrypted and protected by biometric security measures and a passcode.
Apple’s entry into health IT is a much needed shot in the arm for a sector that has traditionally lagged in consumer-focussed innovation. Apple has an ability to create user experiences that are simultaneously powerful and accessible. And it has a well-earned reputation for not misusing consumer data, as well as providing a brilliant platform for third-party developers. In fact, Apple opened up access to Health Records for it’s third-party developers just last week at WWDC 2018. Apple, it appears, is uniquely poised to change the game in a big way.
The scope of healthcare information stored on the iPhone is currently limited but I expect Apple to work on expanding this carefully. Once that happens, there are exciting possibilities for apps that provide patients with greater understanding and sense of control in dealing with their health.
Potential benefits of patient records on the iPhone
For most people, the smartphone is the most personal device they own. And having one’s healthcare record made available on this device has important implications.
A focus on user experience — like Apple’s — has largely been lacking in consumer-facing health apps. Given access to their health records through a user experience that is intuitive and inviting, patients may feel more compelled to interact with this information. In doing so, they may understand their health issues better, and feel a greater sense of control. The overall result is a patient that is more engaged in their own health journey.
Patient engagement is particularly important in managing chronic disease. Good self-management by patients can delay symptom onset and deterioration. Diabetes is a prime example of a chronic disease in which patient engagement can significantly influence the prognosis and quality of life.
Tracking disease progression through a well designed app — perhaps working together with the Apple Watch — could also provide healthcare professionals with the data to track treatment efficacy, and respond to deteriorations earlier. Take for instance, an app that lets patients with rheumatoid arthritis log an objective disease severity marker like the number of joints troubling them. This app, through Apple’s Health Records, could periodically make the affected joint count available to the rheumatologist’s clinical system. The data could show the rheumatologist how the patient is doing over time, and whether their current treatment plan is effective.
Integrating smartphones and clinical systems could also bring much needed ease and transparency to administrative tasks such as scheduling clinic appointments, and tracking the status of referrals.
Some of the uses cases Apple has suggested so far are medication tracking, nutrition planning, and medical research by integrating Health Records with Apple’s pre-existing ResearchKit.
There are a few tricky issues that I hope Apple and third-party developers keep in mind going forward.
Data sharing — As the recent Facebook and Cambridge Analytica crisis has shown, personal data can be used for nefarious purposes if it isn’t guarded well. It’s a safe bet that a breach of trust with respect to people’s health records would be even more unforgivable. Securing information through conventional app permissions may not be enough.
Information accuracy — Evidence-based medicine evolves at a rapid pace. Any medical information shown by apps would, therefore, need to be kept up-to-date.
Informed consent — Barring a few exceptions, nothing in medicine happens without the patient’s explicit and informed consent. In the context of interacting with an app, the notion of informed consent needs to be carefully thought about. In certain cases, a patient may not be able to give informed consent. In other cases, the information may belong to someone else, as in the case of potentially recording family medical history.
Keeping data in sync — Data stored in clinical systems and that reflected on the iPhone will need to be kept in sync. Getting this right will be critical in materializing the benefits of a unified patient record that resides with the patient.
Keeping things clear — Health information shown to a patient on their iPhone should serve to provide clarity and enable engagement. Haphazardly exposing large amounts of complex, jargon-filled information about their health may only end up making some patients anxious about their health. On the other hand, the significance of critical information should not be downplayed either.
No app for empathy
As exciting as innovation can be, it needs to remain an adjunct to providing care. Treating patients is, after all, a fundamentally human endeavour. Genuine empathy itself has therapeutic value, and no amount of technological progress should detract from this core component of the patient-doctor relationship.
Returning to the case of the homeless Jane Doe from House, M.D., the sad irony is that she would probably not be able to afford an iPhone. Or insurance that covers care from the prestigious hospitals Apple has so far partnered with. Innovation happens in steps, but I hope it reaches a stage that doesn’t leave out those who most need care, and yet face the greatest obstacles in accessing it.