In this month’s article, Mike Farahbakhshian decides to abandon the doom and gloom of previous months and focus on the brighter side of Innovation in Health IT. What innovations will make the world a better place in the coming years? As a hopeful spring approaches, Mike casts aside his natural pessimism and paints a hopeful picture of what wonders could happen. Time to read: 10 minutes. Suggested drink pairing: chilled prosecco Bellini.
Optimism! Maybe He Decides to Have Optimism!
It’s been a very interesting year to be alive. Between chaotic domestic politics, nerve-wracking international relations, worrisome market trends, and the ever-present specter of cybersecurity threats, it’s easy to take a grim, nihilistic tone. Others seem to agree: How else can we explain the smashing success of Nihilist Arby’s?
It would appear I’ve fallen into the same trap, and somewhere along the line, this column turned from “The Lighter Side of Health IT” to “Mike’s Apocalyptically Dystopian Hot Take of the Month.” Well, that’s not fair to you, the reader. For that, I apologize. Here, have a picture of an adorable sleeping puppy.
Clearly I need to write a happier column this month. Yet how?
Thankfully, whenever I have an existential crisis, I turn to the wisdom of Achewood, a quirky webcomic written by Chris Onstad in the early aughts. Surely there is something there that is analogous to the process of writing Meaningless Use. Right?
So, in the spirit of things, I have decided to have optimism. This month’s article will show all the great and wondrous things that evolutions in Health IT can do, with no harping on the downside.
Well, Except This
Okay, one exception, but I’ll get it out of the way quickly: Strava, are you !#$!#ing idiots? Your stupid heat map of biometric data from FitBits has allowed others to intuit sensitive information relating to U.S. military operations, including “black sites”. This isn’t just Strava’s fault. While it’s commendable that the Pentagon encouraged FitBit use to combat obesity, they should have used discretion in distributing them to personnel that were deployed in theater or performing sensitive operations. Unfortunately, it’s difficult to show that kind of discretion when top leaders are letting sensitive information into the IoT data cloud, along with the rank and file.
Incidents like these are why I turn into a Health IT version of Nihilist Arby’s. But since I have decided to have optimism, I’ll just leave these helpful words: If it’s the kind of information that you’d see in the classified MSAT system, don’t push it to a commercial cloud. Capiche?
Okay, with that out of the way, back to the optimism.
Let’s Get Optimistic About… Interoperability
Guys, it finally happened: The Office of the National Coordinator finally released the draft specifications for the Trusted Exchange Framework and Common Agreement (TEFCA). In addition to an updated Common Clinical Data Set – an evolution of the Meaningful Use data set that lists all the relevant data elements any health information exchange would need, the TEFCA provides a common set of governing principles and on-ramp path for health information networks (HINs) to connect to each other under the framework of a Recognized Coordinating Entity (RCE) and Qualified Health Information Networks (QHINs). An overview of the details can be seen in the TEFCA Draft Users Guide.
I am a bit of a standards geek, and for good reason; without a standard way for groups to interface with each other, nothing would get done. Imagine if railroads in different countries had different gauges and you had to use a crane to move cars from one width to another.
Okay, bad example. (And Russia is fixing its rail gauge anyway.) But imagine a world where different cities in the United States used different outlet prongs.
Like electricity, Internet data structures, and vehicles, standards are key to ensuring what works in New York will work in L.A. and everywhere in between. Given how critical Health information is – a set of standards and a governing body is key. TEFCA is the equivalent of our standards for 110 volt A/C, Interstate road widths, and HTML standards for Web pages.
I can’t stress enough how difficult it is to make a standard that everyone can be (mostly) satisfied with. The default tendency for standards developers is to become fractious and diverge wildly along a number of edge cases.
TEFCA is still in draft state, so there’s still a chance that things can go awry, but I choose to have optimism. This is because there are very smart people working on this, including By Light’s own John Moehrke, who is a notable standards developer and key to the evolving TEFCA and FHIR frameworks. His blog can be found here and it is significantly more informative than any of the drivel I write. He’s also a better optimist than I am, and I highly suggest you read this post on the bright future of health information exchanges in 2018 and beyond.
The bottom line is that TEFCA will present a common set of rules for the storage, securing, and transfer of Health Information that will allow provision of Healthcare to be truly portable. Once adopted nationwide, anyone can go from provider to provider with a fully automated system. Fraud, waste and abuse checks can be done anytime, anywhere, and not beholden to market-specific monitors. If adopted worldwide (remember: optimism), this paves the path to completely portable, interoperable Healthcare anywhere there is connectivity, from Tulsa to Tibet.
And that’s something to be optimistic about. But wait! There’s more!
Let’s Get Optimistic About… Combatting Opioid Abuse
The opioid epidemic is no joke. Drug overdose is the leading cause of accidental death in the U.S., with 52,404 lethal drug overdoses in 2015. Opioid addiction is driving this epidemic, with 20,101 overdose deaths related to prescription pain relievers, and 12,990 overdose deaths related to heroin in 2015.
It’s difficult to be optimistic about opioid abuse. The reasons are twofold:
- Opioids are incredibly habit-forming among humans. It’s like catnip for people, that can kill you.
- There is a very large industry that makes a lot of money off of #1.
In Big Pharma’s ideal world, we’d all be addicted to low doses of opioids that we couldn’t live without, and thus funnel a significant portion of our income to buying more opioids. Sounds pretty dystopian, right?
Thankfully, the opioid epidemic has reached such staggering proportions that I don’t think we will be lining up twice daily for a large smackaccino just yet. In fact, I am downright optimistic about the efforts to fight opioid addiction, for two reasons.
First, as the opioid epidemic worsens, the worst addicts will overdose and die. Dead people cannot abuse opioids, thus lowering demand.
Okay, okay, fine: in addition to lowering demand, as more people are lost to the opioid epidemic, we become more motivated to solve the problem as a society. Our righteous anger over losing so many of our people, young and old, to opioids will lead to greater communication, education and outreach. The CDC has put out many resources to aid in this, and our smartphone-enabled culture means that anyone has access to these resources. Sorry, action movie fans, but you’re not going to see the CDC come in biohazard suits wielding flamethrowers to stop a fictional zombie epidemic. Instead, you will see the CDC put out quality educational resources to stop a real opioid epidemic. We’re living in the action movie; the timelines are real, and the stakes have never been higher. Thanks to increased connectivity, we are going to win this fight… if we remain optimistic.
Second, and less cynically, we have significantly better monitoring of opioid consumption. This is thanks to Prescription Drug Monitoring Programs, or PDMPs. PDMPs, put simply, are databases that track prescriptions of controlled substances. A properly structured PDMP, especially one built with deep learning and big data analytics in mind, can cross-reference prescriptions against a patient’s diagnosis history, to determine whether the need for opioids is genuine or drug seeking. A really well designed one can also check for trends – over time, legitimately prescribed opioid use should taper off, not stay constant or increase. And, thanks to the analytics tools we have today, we can also check for inter-drug reactions.
Interestingly enough, until recently, every state in the union had a PDMP — except for Missouri. Multiple attempts to create one by legislation stalled until the governor finally signed an executive order to create a PDMP. I wonder why? Who benefits from the sale of opioids and has influence in Missouri politics? Clearly, it’s a mystery that will never be solved.
No matter, though, because PDMPs are here (even in Missouri!) and we need to ensure they can be truly useful. In order to be useful, a PDMP must have full reach back to a patient’s record, including diagnostics, vitals, prescription history, and genomics (where necessary). In short, a PDMP becomes more effective when it structures its data along the line of the Common Clinical Data Set used by TEFCA.
Next, a PDMP must be automated. One of the proposed Missouri PDMP plans that failed, failed for one reason: The proposed Bill did not set up a PDMP that functioned like others around the country. Namely, it did not give registered medical professionals direct access to a patient’s narcotic history.
Instead, the proposed Bill would have compelled doctors to send to the state health department the names of each patient they’re considering prescribing painkillers. The state PDMP would have automatically alerted the prescriber to any troubling patterns in that patient’s prescription history. Then, it would have been up to the prescriber to make a decision about whether to dole out the medication. While this is noble, we don’t have room for human error (or corruption) when it comes to combating such a deadly epidemic.
Finally, a PDMP needs to be portable. There are many use cases where prescriptions are filled across state lines. Military members, Veterans, folks using mail order pharmacies are some examples. This means that PDMPs need to be tied into health information exchanges, both at the local/state level and between localities. This once again fits into the TEFCA framework, where QHINs can communicate vital information to each other.
As a stretch goal, I think a PDMP – and a health record – should have hooks into a patient’s social media or financial records. As a privacy advocate, I’d prefer this to be an opt-in program, of course. Yet I feel that this has many benefits. Opioid addiction doesn’t happen in a vacuum. No one wakes up one day and says, “Wow, I sure would love to be dependent on Fentanyl for the rest of my life.” There are reasons people take painkillers – many life situations, such as stress, monetary issues, loneliness or an inability to connect emotionally and express oneself can magnify pain. You want to slow the opioid crisis? Use a PDMP. You want to stop it entirely? Use the PDMP in conjunction with finding out what’s making these poor souls feel they need these powerful drugs.
If PDMP architects and administrators can adhere to these core principles, then they can be used to proactively find trouble before trouble finds their constituents. Currently, we can use predictive analytics to root out fraud, waste and abuse. That’s great for the insurance industry, but for the rest of us, let’s use predictive analytics to save lives.
So there you have it. It’s been a grim, pessimistic 2017, but I didn’t want 2018 to be full of despair. So I threw us all a bone.
Let’s not confuse optimism with foolishness, however. There is a lot of work to be done and we must maintain constant (pharmaco-)vigilance. Optimism means we know it can be better, not that it will be. That part is up to us; that’s why it’s called willpower.
Yet, somehow, I have a feeling that we will be able to use the power of Health IT to solve these big problems. I don’t know why.
Call it optimism.
 Don’t worry, I won’t change the name of this column, mainly because there’s nothing I can name it that riffs off of “Common Clinical Data Set.”