The value of sharing data: What healthcare can learn from oncology

Again this year I was thrilled to be invited by Philips to participate in their Future Health Index project. This is among the most visionary annual healthcare overviews in the world. The full report – a 100 page PDF – is available free here.

Each year they’ve also asked me to submit a post for the project’s blog. Here’s my submission this year, touching on why oncology is ahead of most specialties in this area: the field decided years ago to align for patient benefit!

Let’s all do everything we can to help healthcare achieve its potential! Sharing information is part of that.

---

The value of sharing data:
What healthcare can learn from oncology

Decade after decade, innovations change the future of care. Microbes, anesthesia, surgery, transfusions, public health, radiology, penicillin, genomics … each development produced a quantum shift in what clinicians can achieve and in patients’ lives.

I wrote an article for the Future Health Index last year, Could data make you live longer?, which lists six different ways I as a patient can be empowered by better data flow. Similarly, it turns out one of medicine’s next great frontiers is not biological but technological: the ability for clinicians to share relevant patient data with others.

In this innovation, oncology is ahead of other specialties: the 2016 Future Health Index report found that 71% of oncologists across 13 countries share patient data electronically, while only 63% of other clinicians do.

Why? What can we learn from this?

A big factor is that oncology as a profession decided nearly a decade ago that electronic data sharing is core to their work, and spoke up about what they need.

“Generate and transmit cancer treatment plans”

A 2014 literature review[1] in the journal Cancer Informatics said this:

[In 2008 a] thorough analysis[2] supervised by American Society of Clinical Oncology (ASCO) … showed that the main functional requirements expected for an EHR to be efficient include the ability to generate and transmit cancer treatment plans and clinical summaries.

It then goes on to specifically list what data elements are important to share, for oncologists to achieve their potential. I suspect this was a key moment in oncology stealing a march by letting system developers know what their users want.

Clinicians’ involvement in writing the specifications is too often missing from the design of health data systems, resulting in widespread physician dissatisfaction with poor electronic health record (EHR) design and usability. A 2014 article in the Journal of Biomedical Informatics[3] found:

Physicians are likely to intend to use EHR when it is considered easy to use in their practice. … Involving physicians in the design of EHR, notably through user analysis, appears a good strategy to ensure that the system is intuitive for them and thus, more easy to use.

We all can perform better when we’re informed better

To be clear, I’m not an oncologist – you might say I’m in “oncolog-ee,” the recipient of an early immunological treatment[4] that saved me ten years ago from Stage IV kidney cancer. My odds were dismal and I was saved by an extraordinary combination of knowledge from many sources, all of them brought to bear on my case fast.

Some of the knowledge was about treatment options. Some was unpublished information that helped me survive the sometimes-lethal side effects. Some was about bringing a strong mental attitude to one’s case, which is too often not mentioned, even though it’s been well known at least as far back as Norman Cousins’ 1979 book Anatomy of an Illness as Perceived by the Patient. (Cousins made himself laugh by watching Candid Camera reruns; I used DVDs of Bugs Bunny and Saturday Night Live’s first season.)

But an essential part of my salvation was that all my clinicians had access to every bit of my health data.  How else could they have performed to the top of their training, to save my life? Could you excel at your own job if relevant facts were missing?

Let’s unite around sharing, not hoarding

To make the most of this opportunity will require not just best practices in software development, but further culture change. Some aspects of our medical culture view the clinician’s data – or a researcher’s data – to be their private property, the fruit of their intellect. Indeed, in January 2016 a firestorm erupted when an editorial in the New England Journal of Medicine noted “concern among some front-line researchers that the system will be taken over by what some researchers have characterized as ‘research parasites.’”[5]

A culture change is emerging around one key question: for whose benefit, whose need, were they paid to create the data? When Vice President Joe Biden spoke that May at the annual Health Datapalooza conference, he shared how lack of data mobility interfered with his son Beau’s case and how frustrating it was. He compared it to other industries funded by government: “We don’t do that in DARPA, we don’t do it in NASA – we only do that in cancer!”

Within days of the “parasites” editorial the NEJM backtracked, saying they’d been referring to some researchers who “spoke pejoratively in describing data scientists who analyze the data of others.”[6] And listen for the cultural assumption hidden in this sentence: “Many were concerned that data sharing would require them to commit scarce resources with little direct benefit.”

Little direct benefit? In whose opinion? Certainly not the opinion of the people, like the Biden family, struggling with the disease, who want research to bear fruit as soon as possible. Yet I’ve heard anecdotally of physicians and executives who feel that others have no fundamental right to access the data they’ve generated in their own work.[7] And I’ve tweeted “Researchers extracting data from patients and not giving it back is analogous to our being livestock for others’ benefit.”

Lives are at stake. Let’s do the best we can.

The potential power of digital data is one of the biggest opportunities on healthcare’s horizons. Unprecedented possibilities lie ahead.

Oncology is ahead, thanks in part, I believe, to their having defined in 2008 what data can be usefully shared. Let’s hope all of healthcare can learn from this: When patient data is shared electronically, net knowledge throughout the health system improves, redundant tests can be avoided, and all stakeholders can operate more efficiently for patient benefit and better outcomes.

^{[1] Fasola, G., et al. Health information technology in oncology practice: a literature review. Cancer informatics 13 (2014): 131.}

^{[2] Cox JV. ASCO’s commitment to a better electronic health record – we need your help! J Oncol Pract. 2008;4:43–4.}

^{[3] Gagnon, Marie-Pierre, et al. “Electronic health record acceptance by physicians: testing an integrated theoretical model.” Journal of biomedical informatics 48 (2014): 17-27.}

^{[4] HDIL-2 (high-dose interleukin-2, approved by the FDA back in 1994)}

^{[5] Longo, Dan L., and Jeffrey M. Drazen. “Data sharing.” (2016): 276-277.}

^{[6] Drazen, Jeffrey M. “Data sharing and the journal.” (2016): e24.}

^{[7] See Yale’s Harlan Krumholz MD keynote at Connected Health, citing a health system executive who told him “Why would we want to make it easy for people to get their health data? … We want to keep the patients with us. So why wouldn’t we want to make it a little more difficult for them to leave?”}