Concepts I enjoy, vol. 2
Lost in mathematical multiverses
I enjoy the concept of multiple infinities.
I'm not saying I really understand all of that article (I read it three times and am still mostly like "huh?"), but I did learn a lot from it:
- Most importantly, that mathematicians just don't think like the rest of us.
- That while one can devise a series of rules for manipulating sets that "stabilizes the foundation of mathematics", those rules can create more questions than they answer.
- That sets of numbers exist whose similar property/properties can never be figured out. (What a puzzle)
- That some of math's current problems can be resolved by multiple infinities.
- That the phrase "mathematical multiverse" is one of the best in the English language.
A way to spur new ideas
This Creative Thinkering column is about using the mnemonic SCAMPER, first created by Bob Earle, to teach yourself how to brainstorm and innovate new ideas.
SCAMPER stands for:
S = Substitute
C = Combine
A = Adapt
M = Magnify and/or Modify
P = Put to other uses
E = Eliminate
R = Rearrange and/or Reverse
This is pretty cute (I love mnemonics), and helpful, and the author uses a number of strong historical examples for each part of the process. I think it's a notion that could come in handy in some way nearly every day, in a number of different settings.
Attention, please!
Check out this video, which asks you to count how many times a ball is passed between players. Can you keep up?
Apparently, that exercise fools about 50% of people. And now, check out this updated version (for those who already know the punchline to the first video):
And now read more about the experiment, and spend some time pondering all of the crazy things you probably miss every single day.
Laugh, Sing, and Eat Like a Pig
How did I miss this?! That'll teach me to take a vacation while a future best seller hits the marketplace!
When I first heard that Dave deBronkart (aka e-Patient Dave) was writing a book, several thoughts went through my mind. The first one being, "I wonder if he'll be kind enough to autograph mine for me?" Less selfish musings quickly followed. It's a story that needs telling. He'll definitely make it to Oprah. When it's a best seller--it couldn't have happened to a nicer guy. I'm sure hundreds, if not thousands of people have an e-Patient Dave story to tell.
I first encountered e-Patient Dave while scouting trends related to healthcare, specifically e-patients and social media. He kept popping up everywhere. After stalking him for a bit, I worked up the nerve to email him to see if he wanted to chat about common interests--I had visions of him coming to give a talk and enlightenening our organization about how patients were rocking the whole medical establishment. I had a great conversation with Dave, but somehow I botched the follow-up call with him and his physician Danny Sands.
My next encounter was last September where I heard Dave speak at Medicine 2.0 Conference. He gave the keynote address Gimme My Damn Data; I found him an engaging and moving speaker. He seemed quite approachable, so I worked up the nerve to re-introduce myself. In a quick minute, I suggested we might meet for dinner when he came to Philadelphia. To my great surprise he took me up on that invitation.
Obviously, the most memorable was being engaged in conversation for a few hours over dinner, live and in person. What a range of topics we explored! What a wonderfully humble and moving storyteller he was! What a generous a spirit he has! Just sitting at the same table with him scored me an invitation to the ePatient Connections conference the next day.
Well Dave deBronkart, I hope you can finally quit your day job. Much success with the book and in your future efforts on behalf of patients everywhere!
Whatever you do… don’t get sick in July.
Take two and call me in the morning! Wait, maybe take three? Or is it one... What's your diagnosis again?
I just finished reading the transcript of an interview on NPR with David Phillips, a sociologist at UCSD. He did a study on medication errors that spike in July... most likely because of new residents who aren't used to sleep deprivation and just plain aren't ready to practice on their own yet.
I guess it makes sense when you think about it... and we have thought about it recently during discussions about work hours restrictions that are placed on new residents.
I loved Dr. Phillips's first piece of advice - don't get sick in July. I'll keep that in mind!
What one can do when one discovers one is related to Lizzie Borden
this is your non-psychopathic brain not on drugs
Recently at work we’ve been exploring the potential for using information from brain imaging technologies to assess individual performance. For example, could we learn about how well a person multi-tasks by giving them some simple puzzles and scanning their brain? Anyway, this has been on my mind for the past few weeks, and then I heard this story on NPR this morning that really made me think.
A neuroscientist named James Fallon has been studying the brains of serial killers, and has noted that a specific part of a psychopath’s brain is dormant – the orbital cortex, which controls impulses and has been linked to rage and violence when dormant. Upon nudging from his own mother, he started poking around his ancestry to find the “cuckoos” that she suspected were there. Among many other less-well-known serial killers, he found that he was actually related to Lizzie Borden. You know the old tune, “Lizzie Borden took an axe, gave her mother forty whacks, when she saw what she had done, she gave her father forty-one”. (As an aside here, she shares a birthday with me – July 19th. I visited the Lizzie Borden Bed and Breakfast a couple of years ago… talk about creeeeeeepy! I also found a terrible metal band named after her. Whoa.)
Once you find Lizzie Borden in your family tree, the only thing to do is to find out whether your brain resembles that of a psychopathic murderer’s. And that’s exactly what James Fallon found. He had the same dormant orbital cortex in his PET scan. He also found what is referred to as the “warrior gene” in his DNA. According to his scientific findings, he should have been a murderer or at least full of rage. So why wasn’t he?
With a little digging, James found that almost all of the psychopath’s brains that were scanned in prison had experienced some sort of childhood abuse or violence. Lucky for him, James didn’t experience any kind of abuse as a child. This finding totally changed his mind about nature vs. nurture. He used to be almost 100% on the nature side of the argument, but looking at his own scientific data he had to conclude that nurture plays a bigger role than he originally thought.
This brings us to the following question: should we assume that everyone’s development is based on the same relative amount of nature and nurture? Or, in other words, might nurture play a bigger role in one person’s development, while nature is primary in another’s development? What might this mean for how the results of imaging technologies are interpreted and used?
Filter bubbles–what price personalization?
You know those recommendations you get from your favorite websites? A few come to mind--books from Borders, music from Pandora, just about anything from eBay? Sometimes they're spot on, sometimes not. I couldn't have been more surprised when I opened an email a few months ago, and the very first recommended item from Amazon.com was a walker. I didn't know whether to laugh or take offense. I'm no spring chicken, but seriously, a walker!! The thought did cross my mind...nah, there's no way Amazon could possibly know that bilateral knee replacements are common in my family.
Fast forward to a tweet by @susannahfox referencing a blog post by Ethan Zuckerman Eli Pariser on Filter Bubbles. Personally, I like filters. Filters get rid of the noise and allow me quicker access to what's relevant to me. When I search "Cocco's" at home around dinnertime, the first thing I see is the phone number to our local pizza shop. I never questioned who decided that's what I should see. I just took this convenience for granted. After all, what's wrong with personalization? One size doesn't fit all--a lesson learned in the 70s when someone got the bright idea that all women could be accommodated with the one-size-fits-all dress. It never entered my head that evil forces could be at work, and I might someday be all alone in my very personalized filter bubble. This may be okay if I could still have my pizza. But according to the article, it seems more likely that I would end up with the walker. Apparently, these filters are invisible. I don't choose the filter, it chooses me. I can't turn it off. This is good to know. At least now I can amuse myself by trying to trick the filters. I'll worry about this for a few weeks, then it will be something new. A while back it was personalized medicine. I figured there was the danger that no one would develop therapies tailored for me based on my genetics. If they did, I probably couldn't afford them. Eventually me and others like me would die out. Sadly now it's more likely that I'll waste away alone in my filter bubble with only a walker for comfort.
Can usability heuristics be used to evaluate patient experience?
I don't remember when I first learned about the concept of "usability", but I haven't been able to let go of it ever since. Usability is basically the ease with which something can be used to serve the intended purpose. For example, a usable website is one that gets you the information you want in an efficient, simple manner. A door knob is usable if it turns easily and opens the door. Systems and processes can also be examined with a usability lens - did the customer have to wait long at the register? Did the automated phone system eventually lead you to a person to address your question?
So given all that, I got to thinking about the usability of healthcare for patients. How might that be evaluated? Might frameworks for evaluating web/interface usability at least give us a starting place? So, using the 10 Heuristics for User Interface Design developed by usability expert Jakob Nielsen, I will attempt to draw surface-level parallels. (My comments are italicized.)
Visibility of system status - The system should always keep users informed about what is going on, through appropriate feedback within reasonable time. (How many times have you sat in an exam room with no idea when you would next see someone, or who they might be and what they might do to you? Hopefully it's not just me, or I'm going to get paranoid.)
Match between system and the real world - The system should speak the users' language, with words, phrases and concepts familiar to the user, rather than system-oriented terms. Follow real-world conventions, making information appear in a natural and logical order. (This has medicine written all over it. Ever tried to understand the notes your doc is typing in your medical record?)
User control and freedom - Users often choose system functions by mistake and will need a clearly marked "emergency exit" to leave the unwanted state without having to go through an extended dialogue. Support undo and redo. (This is an interesting one. An example of this might be if someone requested a physician's help with quitting smoking, and then changed their minds. It brings up the issue that usability from the patient's standpoint may not always be what's best for the patient.)
Consistency and standards - Users should not have to wonder whether different words, situations, or actions mean the same thing. Follow platform conventions. (Having trouble coming up with a unique example for this one - anyone else have ideas?)
Error prevention - Even better than good error messages is a careful design which prevents a problem from occurring in the first place. Either eliminate error-prone conditions or check for them and present users with a confirmation option before they commit to the action. (Imagine calling to make an doctor's appointment and asking them to confirm that all of the error-prone conditions have been eliminated - ha!)
Recognition rather than recall - Minimize the user's memory load by making objects, actions, and options visible. The user should not have to remember information from one part of the dialogue to another. Instructions for use of the system should be visible or easily retrievable whenever appropriate. (Physicians who turn the computer screen toward patients while they are typing, and refer to notes throughout the encounter, may be addressing this heuristic.)
Flexibility and efficiency of use - Accelerators -- unseen by the novice user -- may often speed up the interaction for the expert user such that the system can cater to both inexperienced and experienced users. Allow users to tailor frequent actions. (Could a patient-managed Personal Health Record be an accelerator?)
Aesthetic and minimalist design - Dialogues should not contain information which is irrelevant or rarely needed. Every extra unit of information in a dialogue competes with the relevant units of information and diminishes their relative visibility. (This is a complicated one! I'm thinking of patient literacy - what does this patient need to know? What does this patient want to know?)
Help users recognize, diagnose, and recover from errors - Error messages should be expressed in plain language (no codes), precisely indicate the problem, and constructively suggest a solution. (Error messages as diseases?)
Help and documentation - Even though it is better if the system can be used without documentation, it may be necessary to provide help and documentation. Any such information should be easy to search, focused on the user's task, list concrete steps to be carried out, and not be too large. (Patient education literature, for example.)
It seems as though some heuristics work better than others, but overall, I think there might be something to this with more exploration and unpacking of ideas. What do you think?
Evidence from Patients Like Me
What is evidence? I ask myself this question fairly often since I work in innovation. In trying to generate disruptive and transformative ideas, it is important to understand what has been "established", and why. The next question is: should it change? If so, how?
In Medicine, evidence often comes in the form of an article published in a peer-reviewed journal; hopefully the research demonstrates the principles of scientific rigor. In Law, evidence must conform to a number of rules and regulations in order to be admissible in court. In short: established disciplines try to ensure that decisions and future actions are based on good information, and thus create standards for how that information is captured and how it should be used.
This is a good thing. Not just anything should be considered "high quality" - if that were the case, we might see McDonald's publishing studies on the health merits of Happy Meals and McNuggets.
Unfortunately, though, the path to high quality information - or in other words, evidence - is resource-intensive and time-consuming, and requires expertise that only resides in a certain subset of the population. This rigorous process - as important as it is - no doubt excludes more than just "bad information". It can also exclude new types of information, which may be useful but which don't fit established standards for "evidence".
Case in point: A recent article in Technology Review discusses how the patient social-networking site PatientsLikeMe came to the conclusion that lithium did not slow the course of ALS. Eighteen months later, the journal Lancet Neurology published a study with the same conclusion. I think it is likely that many researchers did not take the conclusion seriously - or even know about it - until it was published in the Lancet.
This bears the question: what types of information (potential evidence) are being overlooked in patient social networking sites, and should Medicine be paying more attention? I could imagine, for instance, that a community of patients could identify chief complaints associated with diseases that are currently not in the published research literature. This could improve diagnosis. Patients could also identify previously unknown drug contraindications, just by describing their experiences with other patients.
Now, I'm not suggesting that we take information from patient communities and immediately consider it evidence, but shouldn't we consider its potential utility and find a way to apply it appropriately? I envision a physician-patient interaction going like so:
Mrs. Smith: Doctor, I was wondering if Fakeprofin might help my symptoms.
Doctor: Well, Mrs. Smith, Fakeprofin does address the symptoms you describe, but I want to make sure it is okay to take with your other medications. Let me see here....(type type type)...Hmmm...There are no clinical trial findings available yet. However, it looks like patients similar to you on PatientsLikeMe have had no issues and have had favorable experiences, so let's give it a cautious try. I want you to follow-up with me in 2 weeks or sooner if you experience any of these problems....
Okay, I'll stop rambling now - but hopefully you get the idea!
Will Foursquare incentivize life?
I joined Foursquare today. I am officially a newbie!
Why did it take me so long? Aside from a hectic schedule and generally being overwhelmed by information, I quite frankly didn't want folks to know my whereabouts at that level of detail. The gaming aspect alone was not enough to draw me - I could care less about earning points by checking in to cool placees, and becoming a mayor. For me, the tipping point was internally motivated; I thought, "Where might this go in the future?" and suddenly my brain imploded. I did a bit of research.
At its heart, Foursquare is basically a framework for incentivizing human behavior, using certain gaming principles. Users "check in" to locations and earn points and badges, and share information about locations with other Foursquare "friends". Right now, the incentives are basically popularity/social networking and competition. This is usual with these sorts of Web 2.0 start-ups, and for some people, that's enough of a draw. However, with additional financial incentives (e.g., check in here for 10% off this doohicky), Foursquare could change the world by better attracting a huge user population: consumers. It's already happening. For example, Starbucks created its own Foursquare badge in May to try to appeal to its more loyal customers, and of course Foursquare has plans to broaden its appeal (and financial incentives are a biggie.)
So in the immediate future, sure, folks might start using Foursquare to get a deal here and there - but what might come next? This is where my brain liquifies - not in a bad way. Imagine all the incentives and services that can be provided to humans that are geo-trackable, tagged with information (some profile-based, some near-real-time), plugged-in and engaged. This is an advertiser's dream. However, if the API remains open and the community can keep developing apps (i.e. purposes), the appeal will reach far beyond advertisers. Users themselves will create uses, demand functionalities, and dream up new and interesting ways of overlapping the virtual and physical worlds. You could do a voice search for the nearest person certified in CPR to help you with chest compressions (maybe they'll have a CPR badge?) You can find a Boy Scout to walk you across the street. You can find a random person nearby who also happens to be craving avocado icecream. You and everyone else in a given restaurant can complain together about the loud dude in the corner - God he's annoying, and look at all the bars he checked in to before coming here. Wait - is that the mayor?
I'm being silly, but hopefully you get the idea.