James Fallows

James Fallows is a national correspondent for The Atlantic and has written for the magazine since the late 1970s. He has reported extensively from outside the United States and once worked as President Carter's chief speechwriter. His latest book is China Airborne. More

James Fallows is based in Washington as a national correspondent for The Atlantic. He has worked for the magazine for nearly 30 years and in that time has also lived in Seattle, Berkeley, Austin, Tokyo, Kuala Lumpur, Shanghai, and Beijing. He was raised in Redlands, California, received his undergraduate degree in American history and literature from Harvard, and received a graduate degree in economics from Oxford as a Rhodes scholar. In addition to working for The Atlantic, he has spent two years as chief White House speechwriter for Jimmy Carter, two years as the editor of US News & World Report, and six months as a program designer at Microsoft. He is an instrument-rated private pilot. He is also now the chair in U.S. media at the U.S. Studies Centre at the University of Sydney, in Australia.

Fallows has been a finalist for the National Magazine Award five times and has won once; he has also won the American Book Award for nonfiction and a N.Y. Emmy award for the documentary series Doing Business in China. He was the founding chairman of the New America Foundation. His recent books Blind Into Baghdad (2006) and Postcards From Tomorrow Square (2009) are based on his writings for The Atlantic. His latest book is China Airborne. He is married to Deborah Fallows, author of the recent book Dreaming in Chinese. They have two married sons.

Fallows welcomes and frequently quotes from reader mail sent via the "Email" button below. Unless you specify otherwise, we consider any incoming mail available for possible quotation -- but not with the sender's real name unless you explicitly state that it may be used. If you are wondering why Fallows does not use a "Comments" field below his posts, please see previous explanations here and here.

James Fallows: Technology

  • The Electronic-Medical-Records Email of the Day, No. 1

    "Just as cars are not all the same, Electronic Medical Records vary greatly. A Mercedes, a Maserati and a Yugo are all cars, but you certainly wouldn't accuse someone of rejecting a used Yugo as being a Luddite and hating all cars. Similarly, you shouldn't generalize physicians who reject terrible programs as hating EMR."

    Background: In last month's issue (subscribe!) I had a brief Q&A with Dr. David Blumenthal, who had kicked off the Obama Administration's effort to encourage use of electronic medical records. Since then, the mail has kept gushing in, as reported in previous as reported in in our April issue, about why the shift has been so difficult and taken so long. Previous multi-message compendia are available in installments onetwothreefourfive, and six

    As an operational matter, I am going to start doling these out one or sometimes two at a time, on a every-day-or-two basis. They'll have headlines based on this one's, and I will try to figure out some standardized image or illustration as cues that these are part of a series. Generally I'll post these without comment; they're meant to be part of a cumulative conversation among medical professionals, technologists, and the rest of us who are merely patients and bill-payers.

    Let's start with two—one from a patient, one from a doctor.

    Patient (and tech veteran): I can't stand filling out these damned forms over and over again.

    I've been in the high tech industry since I graduate college in 1986, watching it grow from a specialized industry to the giant, interpenetrated octopus it is now. My wife also is in high tech, and indeed started out ... installing EMR systems in hospitals in the early 90s.  Just a couple of quick thoughts:

    First, if someone—ANYONE—can come up with a system that would prevent me from having to fill out THE SAME information over and over again just because I'm seeing a different doctor, I WILL TAKE IT. You get the same information requirements, but they're all on different forms, in different formats, from different doctors. But all the base information is exactly the same: Name, address, social security number, marital status, kids, insurance info, and so on. It's all the same. I'm seeing a doctor who was recommended by my GP; why in god's name am I filling out yet another form by hand. In 2014. When what most offices do is take my information and ... enter it into their databases by hand. How inefficient can you get? Hell, some doctors require you to put the exact same info *on multiple forms*. There has got to be a better way. [JF note: This is also my experience-as-patient, and I share the exasperation.]

    I've long thought what we need is a card that is programmable, the size of a credit or insurance card, that you swipe through a reader, punch in a security code, and it downloads the info to the new doctor's system. Why no one has implemented this I have no idea.

    Another note: I'm sure that a lot of the difficulty is incompatible systems, systems that don't play nice with various insurance companies, systems that don't interact with each well, and so on. This is not an inherent flaw of the technology—it would be no different if they were doing everything on paper, and then found, shit, we're using legal-sized, but the insurance requires 8.5 x 11! Or some other mundane problem with paper records. I don't know of any way around the problem other than mandated standards—"Everyone will use Oracle," or some such—and that's not going to happen. But the answer isn't to go backwards, or we'll end up with ink pots and quills.

    Finally, I have to believe that the second doctor whom you quote is forced to use three systems partly by insurance-company requirements. I have to believe that if we had single-payer, that would simplify the record-keeping and IT problem considerably.

    Doctor: A female doctor—as she notes, her gender is relevant to one of her points—says it's important to distinguish between good and bad systems.

    I am a 50+ yo hospitalist (yes, the dreaded hospitalist bogeyman) and have been one for 17+ years. A couple of points, if I may:

    1- there's a lot of talk about EMR as an entity without really addressing the quality of the EMR's. Just as cars are not all the same, EMR's vary greatly. A Mercedes, a Maserati and a Yugo are all cars, but you certainly wouldn't accuse someone of rejecting a used Yugo as being a Luddite and hating all cars. Similarly, you shouldn't generalize physicians who reject terrible programs as hating EMR.

    They just enacted an EMR/CPOE [CPOE=Computerized Physician Order Entry] at my hospital. The reason this particular program was selected was money, savings by choosing a cheap program and avoiding the federal penalty. It is so difficult to use and (as many other commenters noted) fills your noted with drek and making the useful information difficult to find.

    The program is so awful, in addition to parts of it being mouse driven, you need to use function keys and arrow keys to navigate. (Just hit F9, Dr. Smith...) When was the last time, in 2014, you were forced to learn a new program that required you to navigate that way? You can't search, you need to know the specific names for tests (CT chest rather than chest CT, dysphagia exam versus video swallow) and you need to click up to 30-40 times to get through something that previously required you to write 1 order. You can accidentally (and dangerously) erase the patient's entire plan of care with 2 clicks (one poor nurse spent 2 hours trying to recreate it) but you need click to confirm and verify multiple things that are clinically insignificant.

    I would love an elegant program that enhanced patient care, was safe and made my job easier. Love, love, love it. But instead, I am painted (per lots of your communicants) as a intransigent luddite who doesn't want to move forward. Nothing could be further from the truth.

    (By the way, that picture you posted on March 24, with Xrays accessed on the left, trending labs and graphs, looked great! All that info at your fingertips, integrated into the system. What program was that?) [JF note: it appears to have been an "artist's conception" image rather than a real program.]

    2-I am an Apple fan. I don't care what the computer has regarding the hardware, I just want it to work, be intuitive and be reliable. (Not unusual for a woman, regarding computers or cars.)  However, many of my colleagues are uber-geeks. Just being over 40 doesn't mean we can't handle the technology. We are just less patient of bad technology. I don't use the same phone I used in 1997, don't expect me to use an antiquated, poorly written program which was developed in 1997.

    3--Another topic, but: Hospitalists are seeing patients because the primary care physician [PCP] chose that option. There are trade-offs for any system and thehospitalist system is no different. We may not have the longstanding relationships with people and families but we replace that with relationships forged under very emotional and intense circumstances. As with any physician, experiences vary  greatly. You wouldn't slam all orthopedic surgeons because you had one bad experience or bad doctor, so you should not generalize one experience onto the whole specialty.

    Also, the actual number of times people would actually see their PCP is lower than perceived, usually because of call schedules (seeing your doctor's partners instead) and going to hospitals where your PCP does not have privileges. I addition, your PCP is generally only in the house early morning and after office hours. When families come by in the middle of the day, I am available to talk to them. When someone crashes midday, I can handle it because I am there.

    I got hugs from 2 patient families yesterday, one for spending the time to explain why the orthopedic surgeon was recommending an amputation ( he was at another hospital by the time the family got there) , another for transferring a patient after a terrible, prolonged, critical illness to rehab. Neither had PCP's on staff.

    Previous post                                                                       Next post

  • If Doctors Don't Like Electronic Medical Records, Should We Care?

    "Yes, there are problems in any technology implementation and there always will be. But fewer people die. Yes, it is important to connect with the patient. But fewer people die. Yes, the opportunity to pad billing is obscene. But fewer people die."

    How critics imagine the new record-keeping system. ( Wikimedia commons )

    Dr. David Blumenthal, who now is head of the Commonwealth Fund, has been a friend since we both were teenagers. It was a sign of his medical / tech / policy skills that the newly arrived Obama administration put him in charge of encouraging a shift toward use of electronic medical records. It is evidence of his admirably good-humored big-tent personality that David still takes my calls after the many rounds of back-and-forth we've posted here in response to his original Q&A in our April issue, about why the shift has been so difficult and taken so long.

    For those joining us late, you can check out installments onetwothree,  four, and fiveHerewith number six, on the particular question of how the non-expert public -- those of us who experience the medical system mainly as patients and bill-payers -- should assess the opinions of physicians, nurses, and other inside participants. Should we give them more weight, because of their first-hand expertise? Less weight, because of possible institutional bias or blind-spots? Both at once? See for yourself.

    First, the concerns of two physicians. One on the West Coast writes:

    I am a family practice physician in western Washington state. I have been practicing for 25 years.  Ten years ago I was excited about about the potential of electronic technology to improve patient care. Today I am profoundly disappointed.  

    I am currently working in three different EHRs (electronic health records). Two are OK, i.e. allow me to efficiently document a patient visit with clinically relevant data.  The other one is cumbersome beyond belief. It is a company with outstanding marketing capability that won over our administrators. It falls far short of meeting the needs of those of us trying to improve patient care.  Intrinsically it fails to produce a note useful for other doctors. To achieve that end, I use time-consuming work arounds. Sad I think. 

    I believe that primary care is valuable to patients but also has potential to limit costs.....

    I have included a reference to one of my favorite articles from the New England Journal of Medicine, including the first paragraph of the article:

    "It is a widely accepted myth that medicine requires complex, highly specialized information-technology (IT) systems. This myth continues to justify soaring IT costs, burdensome physician workloads, and stagnation in innovation — while doctors become increasingly bound to documentation and communication products that are functionally decades behind those they use in their 'civilian' life.

    And from a doctor in Kentucky:

    As a 50 y/o it infuriates me when I read that only physicians less than 40 are comfortable with EMR’s because they grew up with them. Well that’s crap. My first computer was a Commodore 64 which I learned to program. I am very familiar with computers and have 4 networked together in my home.

    That being said I would agree with Dr. Wait [from this post] in that EMR’s are not ready for primetime. If EMR’s were so great, no one would have to bribe and penalize us to use less. They generate a tidal wave of information. The important data gets lost in the overwhelming volume of mostly useless information. I used to dictate my notes and they would then scanned into the computer. The note was legible and concise. I could find it anywhere. Then the EMR came. It takes 20 minutes to do what used to take 30 seconds. I get a note that is less than useful. It is full of errors that I can’t correct. Information that others have entered that is clearly wrong that I can’t remove. I no longer try. The only important part now of my notes are the HPI and the plan. The rest is just garbage.

    To give you an example my EMR won’t let me enter a subtotal hysterectomy in the past surgical history. Even when I supply the correct CPT code the EMR calls this  a Total hysterectomy, which is not correct and can lead to errors in determining who needs a pap smear.

    So EMR remain not ready for primetime. I’m not sure why I can’t continue to dictate and allow the transcriptionist to fill in the EMR. It would work so much better.

    Now for a different view, from an informed non-expert. This reader, a physics professor at a university in the South, uses the distinctive phrase of the day to suggest that we apply a discount to complaints from today's practitioners:

    I've been reading the back and forth over electronic medical records. It seems the opposition comes, by and large, from doctors. Because why?

    Because problems. There's lots of smoke and mirrors about interconnectivity, about interacting with the computer instead of the patient, about sleazy increased billing but all of that is in service of a single point of view: let's never change until we can change to something perfect. In other words, the underlying point is "don't make me change the way I'm used to doing things."

    This all misses the main point. To me, what is overriding importance is the undeniable fact that ANY system that does NOT rely on the memory of the patient for long term medical history storage is NECESSARILY a better system no matter how badly it sucks. The VA has proved this over the last couple of decades as measured by the fact that fewer people die. Better information management beats clever doctoring every time.

    Yes, there are problems in any technology implementation and there always will be. But fewer people die. Yes, it is important to connect with the patient. But fewer people die. Yes, the opportunity to pad billing is obscene. But fewer people die. Any large scale IT rollout has problems. The question is do the benefits outweigh the time invested in ironing out those problems. Most of us would say yes because fewer people die. I wonder why physicians are so reluctant to say that? Didn't they swear an oath or something?

    I also wonder how many of these physicians, when directing their gimlet eye to another field such as public education, are equally skeptical of, say, massive online courses or teachers attending to the computer instead of their students, or teaching to the test? I somehow doubt it.

    I think when you are the person dealing with a system day after day, it is easy to let your detailed knowledge of its problems overwhelm the vaguer notion of its benefits. You don't have a direct experience of a patient who didn't die, but you do have a direct experience of a technical snafu. 

     Thanks to experts and non-experts for writing in, and to David Blumenthal for opening this view into a world that affects us all.

    Previous post

  • How to Check if a Site Is Safe From 'Heartbleed'

    If your site reads Safe, it makes sense to change your password. Even if it doesn't yet, a change still makes sense.

    This post follows one a few hours ago about the Heartbleed security failure, and for safety's sake it repeats information I have added to that post as an update.

    Point 1: If you would like to test to see whether a site is exposed to the loophole created (over the past two years) by the OpenSSL bug, you can go here and enter the URL you are concerned about. (This tip via Bruce Schneier.) As explained in the FAQ, the test sometimes delivers "false positives" for vulnerability  -- that is, it may report problems with a site that actually is OK, or that is in the middle of taking steps to protect itself. But the site's creator explains why "false negatives" -- OK signals when there actually is a problem -- should be very rare, and especially if you perform the test several times. Update Here is another good test site.

    Point 2: If a site tests through as Safe, then it makes sense to change your password there. And all of my email and financial sites are now saying Safe, so the changes I am making there will stick.

    But even if a site does not say Safe, the people I have asked say that it still makes sense to change -- even though you'll need to change again when the SSL for that site is fully repaired.

    Reasoning: If you change it now, it's possible that a still-active hacker will capture info today. But if you don't change it now, anything exploited in the past two years is vulnerable. Also, many sites that are not yet fully protected are on higher alert than they would have been before this news, so hackers may have a tougher time in the new environment than when this was an unknown-unknown.

    Point 3: The guy who created the test site, a young Italian cryptologist based in Milan, has a donation button on the site.

    UPDATE: Here is another industrial-strength test site. I tried the same domain on it, and the score you see here is way, way close to the top of those it has tried. And here is another test site.

    Previous post

  • The 5 Things To Do About the New Heartbleed Bug

    Should you take the latest security scare seriously? I do, and here are the steps I am taking.

    [Please see important UPDATE in a newer post, and repeated at the bottom of this post.] Most flaps about scary new Internet bugs are just typical scary Internet flaps. This latest one, the Heartbleed bug, I am taking seriously. Potentially it means that username/ password combos for the sites everyone considered secure have in fact been hacked and stolen.

    Update: Just this second, I see that Bruce Schneier has declared the bug "catastrophic." Consider yourself warned. Schneier adds:"On the scale of 1 to 10, this is an 11." He has no track record as an alarmist.

     You can read more about how it happened, and why it matters, at this helpful master site and the dozens of useful tech links it includes. Here is also an overview from TechCrunch. (Update: and here is one of several useful test facilities to let you check the status of sites you're concerned about.)

    Simplest way to understand the problem: one of the protocols that many sites use to protect their own security, in an implementation known as OpenSSL (for Secure Socket Layers), itself has a previously unknown bug. That bug, in place for the past two years, could in theory allow an attacker to harvest large amounts of name/password combos plus other info from sites believed to be perfectly safe. Because exploitation of the bug would have left no trace, no one (except a potential hacker) yet knows how many names have been taken, or from where.

    A patched OpenSSL version exists and is being deployed. Until then, what should you do? Here's a five-point checklist, followed by explanations.

    1. Change the passwords for the handful of sites that really matter to you. I'll explain how you can do this in a total of ten minutes or less. This probably isn't necessary, but just in case...
       
    2. Do not ever use the same password at two sites that matter to you. Ever. Heartbleed or not, this lowers the security level of any site with that password to the level of the sleaziest and least-secure site where you've ever used it. 
       
    3. Use a password manager, which can generate an unlimited set of unique, "difficult" passwords and remember them for you.
       
    4. Use "two-step" sign-in processes wherever they're available, starting with Gmail.
       
    5. Read what happened in our family three years ago, when one of our Gmail accounts was taken over by someone in Africa, if you would like a real-world demonstration of why you should take these warnings seriously. It's from an article called "Hacked."  

    That's the action plan. Now the details.


    What I am personally doing about Heartbleed, and why.

    -  I am changing my password for a handful of "important" sites. My finance-related sites: bank accounts, credit cards, mortgage-payment, investment accounts. The email accounts I actually use, three of them in total and all Gmail-based. Plus all social-media accounts. Even though on most of these accounts I am dormant rather than active, I'd rather not have someone take over the account and cause problems in that way.  (UPDATE: In response to questions, you would need to do this again once the OpenSSL patch has been distributed or the sites have in other ways confirmed their safety. Nonetheless it seems worth doing even now, even given the possibility that a site is still vulnerable and could have new info intercepted as you're changing it, because otherwise you're exposed to any info collected over the past two years.)

    - I am abiding by the watchword of never using the same password on two accounts that matter. Whoever is in charge of security at, say, HottestCheerleadersPlusCheapMedicineFromThailand.com (not an actual site I have visited) might not know how to protect against hacks, or might even dishonestly sell its user info to hackers. They could then blindly try the combos elsewhere.

    - I am making all this easy on myself by using a password manager. The one I have used and liked for several years is LastPass, which was also the top choice in this recent PC Mag review. You can read reviews of a wide range of alternatives here and here. The idea behind all of them is that they store a vast range of passwords you could not possibly remember yourself; they automatically fill them in for your sites; and they have a range of very tough security measures to protect this precious central vault. In well under 1 minute per site, I can have Last Pass generate a new, "difficult," never-before-used password for important sites -- let's say u!YKhtAs7xQA , though that's not a real one -- and set my systems up to use that automatically.

    For now I'm not getting into the conceptual question of whether one centralized password trove is theoretically more vulnerable than the "distributed" approach of trying to manage this all on your own. In reality, I'm convinced that it's better to use a password manager, and safer than the alternative of trying to keep track of a whole list of passwords on your own. (For instance, you can read Last Pass's explanation of how it does encryption right on each user's computer, not at the central site, so that even someone who got the main controls wouldn't know your passwords.) The only password I keep in my mind is a very long password for Last Pass itself. It's so long that it could never be cracked by brute force, much as no one will win Warren Buffett's billion-dollar bet on the NCAA tournament. But it's very easy for me to remember, because it's a long passage I can reel off by heart.

    -- I am using two-step sign-in processes for every system that allows them, and you should too. Gmail does this, and in fact pioneered this as a free feature for mass, non-commercial users. Last Pass also does so. How this works: In certain circumstances, logging in requires not simply your password but an extra, real-time code that is sent to or generated by your mobile phone or other device. What it means: For all practical purposes, someone cannot take over your account from afar. Since so many destructive scams and hacks are carried out remotely -- from Russia, China, West Africa, Israel, the Stans, you name it -- this is the easiest possible protection you can take against a very broad category of attack.

    Two-step systems can be mildly inconvenient, but a lot of that has been buffed away. For instance, you can set Gmail so that it doesn't need the second password as long as you are using your own computer or phone. For more details, see this and this

    More as the story develops. The point for now: none of us can do anything about larger architectural questions of security, surveillance, vulnerability, and so on for the Internet. But along the spectrum of what that architecture makes possible, we can make ourselves less rather than more vulnerable. These steps will help.


    Update: Via Bruce Schneier, it is very much worth checking out this test site, to see whether a site you deal with frequently has been repaired to avoid the SSL bug. For instance, here -- fortunately -- is what you would see for the Atlantic's site:

    In theory, changing a password on a not-yet-fixed site could create new vulnerability, if a hacker has just decided to start watching it today. In practice, most of the people I have checked with say it's worth doing, because otherwise you're exposed to anything captured within the past two years. Then, when a site becomes safe -- as shown above -- it certainly makes sense to change the password. For further explanation, see this follow-on post

    Previous post                                                               Next post

  • But Seriously Now, Why Do Doctors Still Make You Fill Out Forms on Clipboards?

    "Meditative practices emphasize returning to one’s breath. The clinical equivalent of this is to return to one’s patient. "

    Growth of "Hospitalists," a relatively new medical specialty discussed in the last note below. ( Society of Hospital Medicine )

    We'll get back to St. Marys, Georgia, later today. For now, let's dip back into the mailbag for the latest array of views -- most from doctors or other medical professionals, some from technologists, some from "ordinary" patients -- on the pluses and minuses of the shift to electronic medical records. For background: my original Q&A with Dr. David Blumenthal, who directed the electronic-records program at the start of the Obama administration. That article also has links to four previous rounds of discussion -- and, why not, here they are again. One, two, three, and four. Now, eight more ways of looking at electronic medical records.

    1) "Unremitting folly" and "lack of leadership," and apart from that it has some problems. A negative verdict:

    I am a recently retired family physician and was formerly a physicist. Fifty years ago I was programming a mainframe computer in Fortran and am currently using the Python language to pursue several interests. I have experience with 4 different EHRs. Though not a computer expert, I am neither a technophobe nor a Luddite. 

    My purpose in writing to you is to draw your attention to the elephant in the room. In brief, the rollout of electronic health records (“EHRs”) in the United States is a story of unremitting folly, lack of leadership, opportunities wasted, and a stiff dose of medical academic hubris.

    Anyone involved with medicine or information technology (“IT”) has surely been aware for 3 decades or more that EHRs were coming, someday, somehow. The potential advantages were always clear enough. Broadly speaking, they were ready access to individual patient data at the point of care and aggregated patient data, “big data", to be mined somehow for new medical knowledge.

    Standards for medical records were developed, but were overly broad and insufficiently specific (see, for example, HL-7).  The Department of Defense and the Department of Veteran Affairs were interested in developing a systems-wide EHR, which probably discouraged any entrepreneur reluctant to develop a product only to see the government version become a national standard. 

    The lack of a clear standard is a major issue. EHRs, like computer operating systems, are a highly path-dependent technology.  The system you buy today will be yours to live with for the next 20 years, even if no system available today meets your needs.  A good example of this path dependence is the history of Unix-like versus Microsoft versus Apple operating systems.  Unfortunately, the EHR mandate ignores the lesson.

    We now see a technology not ready for deployment being imposed on hospitals and other health care systems. They can buy in with some help from the federal treasury or wait and be penalized for not being on line, an interesting new form of under funded federal mandate. Health care systems are scrambling to enlarge IT departments. Different vendors’ systems are largely not interoperable. This is more than a nuisance when patients self-refer between health care providers.

    For a physician seeing patients in clinic an EHR can be an astonishing impediment. We are rebuked, often deservedly, for being insufficiently engaged with our patients, yet now must spend more time in front of computer displays. (“Why can’t I find a nurse? They are in the patients’ rooms because the computer system is down.”)  

    The practice of medicine involves intensely personal encounters; indeed the patient-physician relationship is what makes being a primary care physician such a privilege.  The EHR does not accommodate narrative analysis of a patient encounter, also known as the personal touch. 

    Voice recognition requires time-consuming, highly distracting concurrent proofreading, quite unlike dictation transcribed by human intelligence.  

    Examination rooms are small (and are not going to grow), requiring that the physician’s back be toward the patient when addressing the computer. [JF note: several previous reader-messages have suggested solutions to this problem.] In the examination room the EHR is marginally effective and utterly inefficient. The human-machine interface is crude and by itself should have precluded widespread deployment of EHRs at this time. 

    EHRs have real potential for “encouraging” adherence to guidelines purported to improve “quality of care.” This is at best a mixed blessing. Many, probably most, guidelines are not solidly grounded in evidence or serve the self-interests of their authors. Until the guidelines industry is brought to heel, patients are at risk of negative benefit.  The diabetes-industrial complex is a good illustration of this. 

    The entire history of EHRs in the United States is worthy of a full-length book. An overdue technology, it is here to stay, as it should.  However, the fact remains that it was overpromised and recklessly deployed.  There are lessons to be learned, if and only if analyzed and reported by persons without a personal stake in the matter.

    2) "A patient's visit to the doctor is morphing into a billing session." From another practitioner:

    Maybe I’m late to the party here, but I thought I’d add a few additional perspectives regarding the matter of electronic medical record systems (EMRs).

    First, the good: A tremendous upside to EMRs is that they make the record so easily accessible. When I was a resident, I seemingly spent half of my time running around the hospital searching for patient charts and scans. Scans were the most maddening—the radiology file room was far from where my patients mostly were. Often, the file clerk wasn’t there. Other times, there were several teams ahead of me, and I’d waste 20 minutes standing there waiting for my turn. And then the scan may or may not even have been there—another team may have checked it out and taken it to their work room or the operating room.

    At my current institution (a large academic center) all of our scans are digital and can be viewed from any terminal in the complex and, via an encrypted connection, from any internet-connected computer anywhere. If one of my residents or a radiologist calls me regarding an important finding, I can be looking at the images and discussing the case in under a minute. I can show the images to colleagues, display them at a teaching conference, and use them to educate the patient and his family without worrying whether I’ll be able to get my hands on the films when I need them and without impeding anyone else’s access.

    Now the bad: Others have mentioned that EMRs make it easier to bill for higher levels of service. The larger issue is that, sadly, the patient’s visit to the doctor is morphing into a billing exercise with a clinical encounter appended to it. EMRs facilitate this process, but I think the causes lie upstream—with physicians, with the hospitals that increasingly employ us, and with our political choice to largely preserve a fee-for-service medical system.

    More recently, the billing imperative has been joined by the safety and quality imperatives. These are sorely needed, but they do sometimes distort medical practice and can even strain the doctor-patient relationship. Again, EMRs potentiate this but aren’t the cause. For example, one commonly used quality metric is a hospital or program’s ratio of observed to expected mortality. The numerator is straightforward, but arriving at the denominator requires prognostication based on the patients’ ages and the number, type, and severity of their various morbidities. Just as EMRs make it easier to document in such a way as to capture the highest possible charges, they also make it easier to document in such a way as to portray the highest possible severity of illness (and hence mortality risk). The hospital’s coders are constantly asking me to clarify various diagnoses that are unrelated to the patient’s presentation and that are often outside of my area of expertise. This diverts my attention away from direct patient care and instead toward the practice of massaging electronic medical records in order to optimize mortality ratios.

    For many physicians, the result of this pivot away from the individual patient and his clinical needs and toward the increasingly complex documentation of such is that medicine ceases to be an emotionally and intellectually fulfilling practice and becomes instead clerical work. We no longer spend a few extra minutes getting to know the patient and his family, perhaps learning something seemingly small but ultimately clinically important in the process. We instead spend unsatisfying time asking irrelevant questions (the review of systems) that allow us to check more boxes, bill a higher level of service, and make the patient appear as sick as possible.

    There’s a mental antidote to this pessimistic mindset, which is easier said than done given the cognitive loads under which we all labor—loads that are increased not only by the demands of using EMRs, but also by pagers, cell phones, various inboxes, etc. The antidote is to listen deeply and re-connect with the person in front of you. Meditative practices emphasize returning to one’s breath. The clinical equivalent of this is to return to one’s patient. A corollary to this is that my generation of medical educators, witnessing the end of the paper chart era while having many years of service ahead, must practice and teach the fundamentals of clinical medicine while helping trainees learn to marshal EMRs and other technologies appropriately.

    3) Comparison from France, and from Seattle

    Quote from one of your other readers: "[At] Group Health Cooperative in Puget Sound, electronic medical records were adopted decades ago, and are widely used and highly effective."

    Response: When I lost my insurance and the ability to stay with Group Health, I wanted to take my medical records. But they charged $45 to put them on a CD. Inexcusable even five years ago. They could just as easily have written a simple program to route records to a printer and handed me the stack of paper at nearly zero cost. Let alone providing the option to buy a USB stick for $5, with all records on it.... 

    Of (possible) interest: "The French way of cancer treatment", by Anya Schiffrin, from February 12, 2014.

    "In New York, my father, my mother and I would go to Sloan Kettering every Tuesday around 9:30 a.m. and wind up spending the entire day...feeling woozy, we'd get home by about 5:30 p.m.

    "[In Paris] A nurse would come to the house two days before my dad's treatment day to take his blood. When my dad appeared at the hospital, they were ready for him. The room was a little worn and there was often someone else in the next bed but, most important, there was no waiting. Total time at the Paris hospital each week: 90 minutes."...

    "When my dad needed to see specialists, for example...the specialists would all come to him. The team approach meant the nutritionist, oncologist, general practitioner and pharmacist spoke to each other and coordinated his care. As my dad said, 'It turns out there are solutions for the all the things we put up with in New York and accept as normal.' "

    Competition cannot provide these results, nor any market forces whatsoever. Regarding people as fellow humans can.

    4) And from Vietnam:

    [How it works there.] Go to the doctor. Begin the discussion at his/her desk. Your previous records have been reviewed in the data base. The doctor's hands rest on the desk.  She/he looks you in the eye and asks questions. Diagnosis made. Treatment recommended. If prescriptions are needed, they are input and transmitted electronically to the receptionist and the pharmacy. You make your co-pay pick up your drugs and depart.

    The efficiency is remarkable. I once had a CT scan at a gigantic clinic with a branch here and in California. The radiologist finished and said

    "Go get a cup of coffee and come back. I'll have your films in half an hour."

    EMR is a tool. A hammer is a tool. In the hands of persons with evil or avaricious intent, either one can do tremendous damage.

    5)  And Boston:

    About 4 years ago I changed health insurance plans and moved my business to a doctor who was a member of Partners Healthcare in the Boston area. I eventually discovered that the practice was connected to a medical records system that would allow any practitioner connected to that system to have immediate access to doctor’s notes, lab results, etc. related to my care. I could also email doctors, make appointments, obtain referrals, request subscriptions over the internet. I grew very comfortable with this. 

    Then my wife had a brain seizure and the EMTs took her to the nearest hospital. The hospital and the doctors who worked at that hospital were not connected to the EMR system we had been using. Problems ensued. 

    The hospital had no access to her history of care.

    I had to track down a doctor on a Sunday night and request complete information about my wife’s medications. The doctor had to send an email to my cell phone so that I could verbally communicate this critical information to the attending physician. 

    Drastic changes in medication were made with negative consequences. 

    I had not realized how much better care could be when you are using doctors who have access to an EMR system. But it is important that every doctor and hospital you use be connected to that system. 

    I will not consider using any medical service that is not connected to this EMR system in our area.

    6) And from a doctor's perspective in Boston:

    I am a surgeon who practiced in a solo private practice in a low income area in Massachusetts for 30 years. I bought an EHR in 2011 and participated in the incentive payment program from CMS and a subsequent audit in which the payment was recouped.

    As other physicians have pointed out , the EHR increased my workload by at least 20%.Dr Blumenthal and his team could have worked to make the VA EHR system, that the taxpayers paid to develop, available universally. Instead perhaps thousands of vendors were certified by the government . The price of these systems was always magically about the same: the $45000 in incentive payments that were promised by the CMS over 5 years.

    Once purchased, myriad other charges arose. The systems were clearly designed to maximize billing through justifying documentation modules. They also were set up to create reports to be forwarded to the government regarding "quality of practice." These mostly involved fairly crude measures like  bean counting how many patients had mammograms or colonoscopies.  With all this crammed in, the goal of creating  clear, informative documentation across a variety of specialties was bound to be lost .

    When these systems failed to serve particular practices or specialties well, , physicians were encouraged to develop their own templates and modifications. More time away from patient care and expense loomed.

    In Boston, there are three major hospital and physician practice systems based on the three medical schools: Tufts, Harvard, and BU. When a patient gets chest pain acutely , he will be taken by ambulance to the nearest facility.He may be transferred during his treatment to a different facility that may or may not be part of the hospital system where he was initially brought. His subsequent outpatient may again be not necessarily with physicians who work for the hospital system where he was treated. It is very likely that the various computer systems involved with the documentation of his care have no interconnectivity.

    At one of the many dinner meetings that we were invited to in 2010 and 2011 exhorting us to adopt the EHR , I queried an employee of the Mass ecolloborative, a federal  grant funded entity, about what priority CMS and the government were giving to the issue of interconnectivity. It seemed unlikely that the big, fiercely competitive  hospital systems and  the IT vendors would pursue this on their own . I specifically asked, when would an ER doctor seeing a patient at BU be able to see the records of the patient's previous care at Tufts or Harvard and she shook her head. So I ask if it would be in five years and she shook her head again . I tried ten years and she said "maybe" and then ,on prompting, said "they are talking about this."

    It seems: you are what you mandate, and the approach of Dr Blumenthal and his team, in my view, has  endorsed and augmented the free market model as regards IT and the large hospital chains and their internecine rivalries. The consequences to patients and independent practitioners are enfolding .

    So, what's a patient to do? In China, in the barefoot doctor days, they gave the paper charts to the patients  and let them carry them around.Not unlike in  the third world, many of my low income patients have smart phone access. In France, as TR Reid has reported, you can go to a doctor in their system and put your ID card through their reader and your updated EHR can be read off your chip. Patients need apps that can download and store these various differently configured EHRs. Like a lot of things regarding your health, when patients are empowered, things really can change.

    7) The technology has problems similar to the Pentagon's:

    1. Yes, some of the large health care systems such as Kaiser Permanente have deployed relatively effective electronic health record systems but what is seldom discussed are the huge cost-overruns associated with these deployments. 

    Health care IT procurement in the large delivery systems is similar to the problems that the Pentagon experiences when it buys weapons systems---the systems usually work, but the costs are often much higher than expected (therefore, the net benefits are lower than expected).  This problem is not unique to the health care sector---as you know, development and installation of enterprise software systems is notoriously complex and even some of the most IT savvy corporations and government agencies have experienced huge cost overruns and outright failures in this area. 

    Unfortunately, there is sort of a conspiracy of silence in the health care sector about cost overruns.  Both the software vendors and the executives who run these organizations are loathe to acknowledge this problem, instead they would rather focus on the benefits (which to be sure are real in many instances) and not talk about the costs---for example, Kaiser Permanente's staff has published 3 books touting the benefits of its electronic health records system, but none of the books discuss the costs or many of the daunting technical and organizational challenges they confronted in building their system.  

    2.  The interoperability problem in health care IT has two dimensions.  The first dimension (and the one that gets the most attention) is the lack of interoperability across health care organizations (as noted by the one of the physicians who commented on the VA's system).  The other dimension, which receives relatively little attention, is the lack of interoperability within organizations. 

    Most large health care delivery organizations decide to keep some of their legacy systems when they decide to implement a new EHR---for example, they may decide to keep their existing radiology and lab order systems, which means they have to spend alot of money creating middleware that can facilitate communication between the old systems and the new EHR.  The cost of developing the middleware is often huge because of the absense of industry standards---this is major reason why cost overruns in this space are so common.

    8) And to round things out, illustrating the complexity of working any change in today's health-care system, the complicating fact of that rapidly growing medical specialty, the "hospitalist":

    After years of only needing to see my doctor (the same one since 1977 until 2013) I've had an up close and personal experience with the new system that has required new doctors (a new medical condition and the retirement of my family physician). 

    What has that meant to me as a patient?  Like the doctor you quoted, when I see my new family physician (still the same practice that is the home of thirty some years of handwritten charts), she is looking at the computer instead of me.  She's also asking the same redundant questions over and over again.  There is a third party in the room--the computer--that is getting the major share of the attention. 

    On the other hand, I love having prescriptions entered immediately.  The scary part: I have caught a number of mistakes: which prescriptions I'm actually taking, what the dosages are, what diagnoses I've had in the distant past at another medical facility.  As they say: garbage in, garbage out. The only good thing is that people are mentioning the "garbage" and asking me if it is true because it is more obvious.

    But the computer is just one part of the problem.  Here's a much scarier thing.  An elderly man with Parkinson's is admitted for emergency surgery that has nothing to do with the Parkinson's.  He suffers from constipation--a common side effect of the disease.  He has a regular routine of over the counter medication to help with the problem.  His wife explains to the medical staff that this is what is prescribed by his regular physician. 

    But his care is now overseen by a hospitalist.  His wife is told that the constipation issue is being handled as usual.  It isn't.  After five days, he is extremely bloated and uncomfortable and nothing has been done.  His wife pleads for help for him in the form of an enema. Did I mention that she is the kind of person who doesn't like to be demanding? The hospitalist (who has almost never visited him and operates through the computer and the nursing staff) orders an x ray and then an enhanced x ray.  Meanwhile the patient gets more and more uncomfortable.

    Eventually, relief is prescribed in the form of--an enema.  A human  conversation in the form of a doctor to doctor discussion of the patient's prior conditions and accommodations would have made his recovery from the surgery so much more comfortable.  Instead the inevitable discomfort of the surgery was made worse by adding more discomfort.

    My conclusion: medicine human to human connection as well as technology. I want my doctors to use technology effectively, but I also want them to listen to me and connect with me as a patient rather than as a disease.  I am very fortunate to have found a new doctor who has this combination, but I worry for all those who aren't getting that kind of care.  I'm also convinced that a human connection with doctors and nurses and other medical people helps us trust our care better and helps us follow through with our treatments.  It's not just warm and fuzzy stuff; it's part of our healing.

    Thanks to all. This is about 5% of the mail that has arrived on the topic. Will keep looking through it. 

    Previous post                                                                          Next post

  • Electronic Medical Records: A Way to Jack up Billings, Put Patients in Control, or Both?

    "Digital records are also being aggressively used to maximize patient billings," and other imperfections on the route to a more sensible health care system.

    What we are leaving behind. ( Money and Medicine )

    Previously on this topic: my Atlantic Q&A with Dr. David Blumenthal, who supervised the Obama administration effort to move medical records into electronic form; and installments one, two, and three. Here is another round of reader responses.

    1) A new way to maximize billings. From Ronald Russell of Kenmore, Washington:

    As a member of Group Heath Cooperative in Puget Sound for over 20 years, I've seen first hand many positive aspects of computerizing patient records. Whomever you see, your records are instantly accessible- that's comforting when you land in the ER in the middle of the night.  Web access means you have access to some of those records yourself, and can communicate with your providers easily. 

    Unfortunately, I've also seen a negative aspect in how EMR's are being used that got only the slightest passing mention in this discussion, one that gets the incentives exactly backwards.  This is the reason I'm now a former GHC member. 

    Digital records are also being aggressively used to maximize patient billings. At GHC, it used to be the case that a standard office visit was a flat charge- most recently $80. Now, when your physician asks a question, responds to one of yours, and makes a note in your record, this becomes another billing code. The result is that a 15 minute office visit can easily run to several hundred dollars, perhaps just because you mention a concern or the physician asks another question.

    Every patient note entered in the digital record rings the cash register again in billing- and not in a way that anyone seems able to explain, or that physicians are aware of. The cynic in me wonders how long until they are compensated on commission, or get bonuses for entering more billing codes per visit. 

    The dollar amounts charged are often absurdly high, there is no accessible "price list" for consumers. My auto mechanic is legally required to explain his charges in advance, my health care provider never has to.

    Of course, for consumers with full coverage or copay-only plans, this would pass unnoticed, as just an accounting detial.  But for those of us with high deductibles that mean we essentially pay out of pocket for everything, this is a powerful disincentive  to discuss concerns with your doctor or interact beyond the minimum business at hand.  I don't believe this is good for patient care.  

    Fortunately, due to the ACA, I've been able to move to another insurance plan that mandates a flat charge for standard office visits, even before you meet the deductible. So perhaps that one small part of the market is working. Unfortunately, this brought up another problem with these records- ours are now locked up inside the Group Health system, and no longer accessible to me- at least, apparently, without paying for them. 

    These issues are not so much inherent problems with EMR's as they are symptoms of a broken health care system, in particular where the provider and the insurer are one and the same.  I put them out there to add to the discussion. 

    2) A way to get the doctor to look at you. In response to a previous complaint about doctors stare at their computers rather than their patients:

    [A previous reader says:] "No, at Kaiser, Northern California, they do not.  The computer is on a roll-around stand, and the doctor or nurse is facing me while using it."

    I'm in IT, and have worked at several hospitals where these stands are used, and the usual nomenclature is COW (computer on wheels, of course). I'm often reminded to be careful there are no women around when discussing the COWs in the room...

    3) Once again the VA is doing it right:

    I'm a 68 yr. old Vietnam vet (USMC) who is rated 90% 'Service Connected' disabled: hearing aids (I was in an artillery battery in Nam for 19 months); Type II Diabetes and Ischemic Heart Disease (Agent Orange exposure) and assorted other things.

    I can't praise the VA enough. Through HealtheVet I can re-order meds and have them mailed to me, same with hearing aid batteries. I can set up or cancel appointments or ask my Primary Care doc, or any of the physicians who treat me, questions and get an answer within 24 hours. I can go to ANY VA facility in the world (yes, there are VA clinics and hospitals outside the US) and they will have total access to my medical records.

    From the hell holes that VA hospitals were in the '70's, as depicted in the movie Born On The Fourth of July, they now are as good as it gets in the US. I give Bill Clinton props for the revamping he and his VA Secretary undertook that got the VA to where it is today. It may be struggling a little with the overwhelming influx from the Bush/Cheney/Rumsfeld fiascos but I have no complaints here in NYC. Semper Fi

    4) Promise from the patients' point of view:

    The use of EMR is obviously in early days, and flaws are easy to identify. My experience, however, shows the great promise of EMR to improve medical care and help patients make medical decisions.

    I have a mild case of MS, and go to [a major medical center] every year for a check-up. These include MRIs of my head every two years or so. I travel 200 miles to visit the clinic, so I want to get everything done on the same day. On MRI years, I’ll have the scan in the late morning at the imaging center that is allied with the clinic. When I see the nurse practitioner or physician’s assistant a few hours later, pictures of my brain are available in my electronic chart, and my medical professional looks at the scan with me, and explains what he or she sees. The reports of my last two scans are available to me right now on the MyChart website that [the center] makes available to its patients. And yes, my case is so boringly stable that there is no current need for me to see “the big man”.

    That’s cool, but the real power was revealed to me in 2011, when the radiologist who reviewed my scan discovered that I have a small benign brain tumor, a meningioma. I was referred to a neurosurgeon, also allied with Strong, where it was recommended that I should have my head screwed painfully into a frame so that a surgeon could aim what is essentially a killer death-ray at my brain.

    Except… when the medical professional and I looked together at the series of scans in my EMR dating back to 2002, there that pesky meningioma was, seemingly the same size as in the 2011 scan. This empowered me to turn down the surgery. I’d had that tumor for a decade or more with no ill effects. Prove to me that it’s growing, and I might consent to the surgery. A repeat scan a year later confirmed that the tumor is not growing.

    There’s one key element here, of course: all of the professionals involved are allied with the same large medical center, so communication between them is smooth and nearly instantaneous. My GP 200 miles away is not part of their system. Still, I’m very happy that I had access to a decent EMR system in this case, which helped me to make an informed decision.

    5) As long as the systems stay in touch:

    I have several chronic illnesses, and because of the specialized nature of them, I have a lot of doctors. I live in the San Francisco Bay area, and have specialists at Stanford, in San Francisco, and then my local team near home. Here’s the thing - all of these facilities have have state of the art EMR systems, but the systems don’t talk to each other.

    This means I spend a lot of my time bringing copies of records between various specialists, and communicating what one doctor said to another. As in telling my primary care  doctor “My rheumatologist is concerned that the medicine you want me to take will have an impact on my spine issue”.  Sometimes I get letters from one doctor that I have to carry to another. I keep my own records of my latest test results, culled from the various sources (included the online tools provided by some medical facilities), and bring them with me to appointments, so I can answer questions about when I last had a test, and what the results are.

    I also have a list of every medicine and treatments I have tried, and the outcomes, as new doctor often has a standard approach to the first thing they want to try. And they haven’t had time to go through all the records that were sent to them. Since the systems are designed around billing, they don’t have easy ways to extract care info, such “Show me the medicines to which this patient has had an adverse reaction”.

    Until there is a well integrated way for your doctors and their systems to communicate, coordination of care is going to be an issue. My career was in computer tech, and I know how hard it is to create interoperability standards.

    To me, yet another argument for single payer system, is that we could standardize on some basic data collection and exchange.

    6) From a Yank in Canada:

    I moved to British Columbia eight years ago from California.

    The first thing I had to get used to when I went to the doctor here was just... walking... out.  No stopping at the receptionist to deal with payment and/or insurance.  Just... walk... out.

    My clinic in California had started doing electronic records before I left, but I recall it as being kind of clunky.  Here, however it doesn't seem as clunky; it seems more integrated into the appointment.  Perhaps it is because I see young doctors (my clinic is a teaching clinic), but I think it's because the appointments are structured differently.  

    Here, the appointment starts with me sitting, fully clothed, in a chair, to the side of a desk.  The doctor sits at the desk with both me and the screen visible.  He or she asks me if anything has changed, and talk about why I came in.  Frequently, the doc will look something up on the web that is out of his/her area of expertise, and they are not shy about doing so.  (Usually not Wikipedia, something more like PubMed.)  If he/she needs to examine me, *then* I get given a gown.  

    By contrast, my recollection of appointments in the US is that they started with height/weight/blood pressure measurements by a nurse.  (This was true even when I was in my 20s and now seems like overkill.  Why did they need to take these measurements every time, when my measurements didn't budge for years at a time?)  Then the nurse would give me a gown, and I'd get undressed and sit on the exam table.  That meant that I would *start* the consultation sitting uncomfortably on the exam table.  (There often weren't even two chairs in the room; maybe there was a chair and a lower stool.)  In that configuration, it is not easy to position the computer so that the doc can see both the screen and the patient.

    Another thing that is different: I almost never fill out a form before my appointment at my regular clinic.  If I am going to a new practice (like an after-hours clinic), yes.  If I am getting some new and different procedure, yes. But they don't ask me to tell them who I am and where I live and what my insurance is and who my next of kin blah blah over and over again.  Occasionally they ask me verbally if anything has changed, and that's it.

    7) Allowing doctors to do more than just fill in the forms. From a librarian:

    One comment based on my experience, I appreciate the doctor who said the system would not let him record what he wanted to say. I think these systems should allow writing free-form notes, sketches, scanned items, etc. 

    I am a retired librarian and early in my career I worked on several of the early computer systems for recording the arrival of issues of magazines in libraries. I'm sure this is much less complex than medical records, but it is more complex than one might think. These early computer systems couldn't accommodate the creativity/inaccuracy of journal publishers and printers when there was an issue number 12 1/2, or, more often, the printer did not change the volume number in the new year until he discovered the mistake midway, so you have volume 14 for a year and a half but number 1-6 in one year were not the same as 1-6 in the next.

    In medicine, the doctor's free-form notes can express his knowledge of how complex things really are, in his best estimation at the time, or the questions he has (another issue, do you want this in a record that will be shared with the insurance company, and thus perhaps used out of context in litigation -- but if the electronic record is the only one you have, where else do you put the information?) 

    I'm all for having evidence-based guidance in medicine, but I want the doctor to be able to take all of this information and then see if I fit the profile the computer predicted. How is this going to happen if the information isn't even recorded?

    Previous post                                                                            Next post 

  • The Use and Misuse of Information Technology in Health Care: Several Doctors Reply

    One of them writes, "There is a very American tendency to look for technological fixes for significant problems.  In general, technological fixes only work in the context of appropriate institutional structures."

    Our new issue has an interview with Dr. David Blumenthal about why it has taken the medical system so long to adopt electronic record-keeping, and what it will mean when the switch occurs. (Blumenthal led the Obama administration's effort to encourage that change.) On Friday several technology experts and doctors weighed in with responses. Here are a few more.

    1) "Give us a cotton gin." Creed Wait, a family-practice doctor in Nebraska does not like the mandated shift to electronic records, at all. [I've added his name, as he sent permission to use it. Also he has moved from Texas to Nebraska.] For now, I am sharing his detailed complaint in full, rather than interspersing comments or "Yes, but" queries:

    The saying is, “Build a better mousetrap and the world will beat a path to your door. “

    The saying is not, “Build a different mousetrap, pay out nineteen billion dollars in incentives to use the mousetrap, mandate its use by law and punish those who fail to adopt it.  Then shove the world kicking and screaming against their will through your door.”

    So far, doctors have been paid $19B in incentives to buy EMRs [Electronic Medical Record systems].  No one had to incentivize the cotton gin.  It was simply a better product.

    The current EMR system is a mess because the current EMR systems in use by the majority of physicians were written in the Rube Goldberg School of Software Design and work poorly.  There is no ‘asymmetry of benefits’ as proposed by Dr. Blumenthal.  Unless, of course, what he means by this is that only the software companies are benefitting from these federal mandates.  Then, I would agree with him.  Yes, the benefits are asymmetrical.

    Build a better mousetrap and we will use it.  DVDs came out and they were better than VHS tapes.  Overnight the whole world invested in new electronics, we bought DVDs and we threw out our VHS tapes.  There was no need for $19B in incentives because DVDs were simply a better product.  Flat screen televisions came out and we stopped buying cathode ray tube televisions.  Why?  Because they were a better product.  Laws mandating the use of DVDs and flat screen TVs, bonuses for using them and punishments for failing to do so, were not needed.  The market chose the better products.

    Mandated EMR adoption requires carrots and sticks consisting of massive incentives and concomitant penalties because the products that are available work so poorly and are so severely user-unfriendly.

    Using the VA system, Kaiser, and Geisinger as examples of the successful use of EMRs is disingenuous.  These are massive systems with massive budgets and massive around-the-clock onsite IT departments.  The vast majority of physicians are not in these megalithic systems.

    Most of us are in much smaller practices.  We have IT departments but the salesmen and software engineers who sold us these magic beans are already down the road looking for the next unsuspecting rube and cannot be reached.  Our IT departments are swimming upstream trying to implement and maintain software that they do not understand while mandated changes to this software are being released before we can get the last update debugged and working.  The doctors are always screaming because the systems are down, we can’t work until the system is running and the IT guys have the harried and glazed look of caged prey.

    For the federal government to mandate the use of EMRs by every physician out there just because it works at the VA would be like telling the entire world, “OK, we made it to the moon.  Now it is your turn.  Any country that has not put a man on the moon within the next five years will be bombed.  Every country that complies with this mandate will get a check for $1B.  For those countries who fail to comply with this mandate, shelling will begin at 1:00AM, five years from today.”

    What the federal government can do with a bottomless supply of tax dollars cannot be used to reasonably mandate what happens in small offices constrained by budget limitations.

    One year ago in private practice I could see eighteen patients per day.  A transcriptionist typewrote my notes. These were typically three pages long, concise, complete and extremely useful.  Then our group bought an EMR.

    After one year I was seeing fourteen patients a day, my notes were twelve pages long, the vital signs alone required a half page and the notes bordered on being useless.

    My reimbursement per visit had increased, my face-to-face time with the patient was shorter, I was doing a poorer job, patients were less satisfied, and I was completely frustrated by trying to build each note out of dozens of pages of drop down menus.

    Before implementing an EMR I had approached each patient encounter with an attitude of, “What can we do today to improve your health, happiness and overall satisfaction with life?”  The patient and I would have a meaningful conversation about the pertinent issues.  Once an EMR was implemented, a subtle change began.  It was so gradual that at first I did not even recognize the poison.  But after a few months I realized that the visit had slowly evolved into, “Just a minute, we need to be sure that we have checked off every box on every screen and we need to be sure that a narrative of some sort has been entered into every required field.”  Then there were realizations like, “Oh, look.  If we add one more point to the Review of Systems then we can raise the billing code one notch.  Hold that thought while I click, ‘wears glasses’ under the ROS field!” 

    Well, time’s up!  The fields are all now completed and all goals have been met!  Next!

    The EMR had become the primary influence in the interview.  The dynamic had changed.  The patient and I were now both in the room to feed the hunger of the software.

    Don’t even get me started on CPOE (computer-based physician order entry systems).   Physicians used to write their orders and clerks would enter these data into the computer.  Under the new mandates, the physician is now a data entry clerk.  What’s next?  Is each hospital CEO going to be required to spend two hours a day manning the switchboard?  It is claimed that CPOE systems reduce errors.  In the real world, this is nonsense.  It is all in how one collects and reports the data.  Data collectors refuse to attribute errors to the CPOE system. Rather than blame the software, the physician is blamed for not understanding how to use the system correctly.  Just like with office-based EMRs we refuse to admit that the Emperor is naked.  I have seen physicians get past mandatory stop points during system entries, when the correct input was not an option, by inputting obviously erroneous answers so that they could keep working.  Then the physician would call the pharmacy and verbally correct the entry.  There are dozens of ‘back door’ fixes of this nature that allow physicians to keep working when a CPOE system locks up or cannot meet the needs of a unique patient.   In my own experience I estimate than a CPOE system adds 1-3 hours of work to each day.  

    I am not a typist, I never have been a typist and I never will be.  I can dictate a beautiful note.  A typist can then create an excellent document from that dictation.  There is no reasonable excuse for the government to mandate an end to this system when no one has a better product.  There is no reasonable excuse for the government to mandate that I will now be a typist.

    Now, on to the VA system.  They have a great and highly integrated information management system, with one glaring flaw.  As long as the patient stays completely within that system, it works.  As long as the patient never sees a physician outside of the VA, it works.  As long as the patient never gets a test, x-ray or is hospitalized outside of the VA, it works.  But the VA does not integrate well outside of their own system.  By the time non-VA physicians and hospitals can get records, reports or anything of value out of the VA, the patient is generally dead or cured.

    But this is exactly the problem that integrated EMR systems were touted to cure.  If the best EMR in the nation has not yet solved this one simple problem, why is the entire concept being shoved down our collective throats?

    So, you want to revolutionize data management in healthcare?  The starting point is a product that works.  Give us a DVD, a flat-screen television.

    Give us a cotton gin.

    2) The latest snake oil. More skepticism about the coming changes:

    I've been a clinical psychologist for many years, and it's long been clear to me that few patients are likely to benefit from the adoption of expensive, labor-intensive technologies, which do, indeed, make it easier for managers and other third parties to do their jobs.

    In recent years we've been seeing increasingly aggressive attempts by a variety of self-interested parties to insist that a certain change in how health care services are provided or paid for - a change which, coincidentally, would have salutary effect on their company's bottom line or their own career - is precisely the snake oil which is needed to "cure" the present system of its ills.

    Rather than making changes which would, in fact, be most likely to result in "the greatest good for the greatest number," what we've been seeing - and are almost certainly going to continue to see - are changes which reflect the outcomes of intense behind-the-scenes political maneuverings among "stakeholders," each of whom is trying to make sure that, when all is said and done, he will be among the "winners."  

    3) What the VA's experience shows, and doesn't. From a doctor in the upper Midwest: 

    I'm a practicing physician with significant experience in the VA system and at an institution that recently adopted a new EMR.  There are some significant qualifications about the potential of EMRs to improve care in the USA.

    1) The precedents of the VA system and systems like Kaiser are a bit misleading.  A very good criticism of EMRs generally being adopted in the USA is that they are fundamentally built on billing systems.  This is an inevitable consequence of the fact that the incentives to introduce EMRs are driven by reimbursement.  Systems designed to maximize patient information would be somewhat different. 

    The older, clunky but functional VA system is better in this respect than the modern EMR I use at my academic institution.  Dr. Blumenthal and his colleagues are inadvertantly partially responsible for this situation because the legislation incentivizing use of EMRs had to be built around reimbursement incentives-penalties.  In our fragmented system, I don't see that alternatives were available to Dr. Blumenthal and his colleagues but realism about the results is necessary.

    2) EMRs should, as they do in the VA system, reduce costs by reducing duplication of tests and services.  This occurs only, however, in the context of relatively large, integrated systems.  I routinely waste money by ordering tests that may well have been performed previously by other physicians because I don't have access to patients' medical records.  Big Data isn't big unless it can be aggregated and used broadly.  There may well be considerable consolidation among health care providers in the near future but any market or semi-market based system like ours is an obstacle to consolidation because it encourages inefficient winner take all behavior.

    3) An analogous point is that ostensibly data-driven changes in clinical practice will not emerge without someone or something to actually analyze the data and develop optimal care approaches.  We need something like the British National Institute for Health and Clinical Excellence (NICE), probably on a more ambitious scale, to push appropriate reforms in clinical practice.

    There is a very American tendency to look for technological fixes for significant problems.  In general, however, technological fixes only work in the context of appropriate institutional structures. 

    4) The good, the bad, and the worse. From a reader in Jerusalem:

    Good:  I'm a long-time software engineer and was recently talking to my hematologist about computers and medicine.  She was very grateful for the change.   She remembered patients coming in to the ER that the staff recognized, but they'd have a terrible time looking for the medical records.   Multiple staff searching through mounds of folders and not finding the right one.  Previous diagnosis helps a lot in figuring out what to do in an emergency.

    Bad: On the other hand, my GP spends more time facing the computer screen than his patient.  Prescriptions come out of the computer via his laser printer, which cranks away all day.  But all the Rx are in the system, anywhere I go.

    American docs used to leave the room in the middle of the visit to look things up, rather than turn away from the patient.  Bedside manner.  Uh, deskside manner.

    On the third hand:  Hacking. 

    I have new respect for the complexity and difficulty of this change. More to come.

    Previous post

  • On the Ramifications of High-Tech, Big-Data Medical Care

    As more of medicine comes into the info age, will we get better sooner? Or simply have a more detailed idea of why we're sick?

    As mentioned this morning, in our new issue I have an interview with Dr. David Blumenthal about the paradox of modernization in the American health care system. We all know that everything about medicine is becoming technologized, in ways good and bad. On the good, see previous interview with Eric Lander about the genomic-knowledge revolution. On the bad, see Jonathan Rauch on the industrialized process of dying. But we also know that nearly every visit to a medical facility begins with the tedium of filling out forms by hand.

    David Blumenthal was in charge of the Obama administration's effort to speed the adoption of electronic medical records, and in the interview he explains why that has been hard but will be worthwhile.

    Now, responses from readers in the tech and medical worlds. First, from David Handelsman, of a health-related data company in North Carolina:

    One of the things that Dr. Blumenthal didn’t include in his response was that the health care industry needs to continue to create a culture of evidence-based medicine, beyond the activities at those organizations that are further along the maturity curve regarding electronic health records and healthcare technology. 

    The reality is that much of healthcare is administered to patients based upon the practitioner’s experience (patients he or she has seen with similar conditions), the practitioner’s ability to accurately recall the appropriate care for the patient being seen and, where time has allowed, the opportunity to stay current on healthcare research and able to then apply that research correctly to the patient at hand.

    While I have the utmost respect for health care practitioners, there’s an awful lot of room here for what I’ll call non-optimized care.  The practitioner’s experience may be incomplete – he or she may not have seen a patient “like this patient”.  They’re ability to recall the best care options for “this type of patient” may be unreliable given the vast numbers of patients in their care.   They may be lagging behind in current research and recommendations because there isn’t enough time in the day, and if they are current, they may still have the same issue of recall regarding complex health decisions.

    Evidence-based medicine aims to provide optimal recommendations regarding patient care.   When electronic data is available, the patient’s current situation can be electronically compared to other similar patients AND their respective healthcare outcomes.  At that point, a recommendation for care for “this patient”, whose profile is aligned with “all of those similar patients”, can be made based upon the recorded outcomes.   Please note that this should be considered a recommendation – the responsibility for providing care ultimately falls on the practitioner, and not an algorithm, but that practitioner should always have the best information at hand to determine the best course of care.

    From a retired MD:

    There is a rub [among many, I suspect] as seen in our personal records from digitized offices.

    To respond to insurance company demands for documentation of visits, physicians can simply cut and paste the previous visit data onto the current visit, making such changes as are necessary. Every visit looks remarkably complete!

    The volume of material viewed makes finding anything new difficult.  The record becomes a document for the insurance company, barely useful for physician's own use or physician to physician communication.

    From the academy:

    I'm a PhD student in statistics working on prediction and causal inference using health data. I'd like to comment on a quote from your (great) interview with David Blumenthal about the promise of Electronic Medical Records: 

    Dr. Blumenthal says: "This will move us into a field that is taking shape right now, that of analytics. It will help us take these data and turn them into diagnostic information—into recommendations a physician can give a patient or that patients can get directly, online."

    He seemingly conflates 'diagnostic information' and '[treatment] recommendations'.  But they actually pose fundamentally different problems from a statistical perspective, and I think EMRs will play a much more transformative role in diagnosis than treatment. This is because diagnoses live in the realm of pure prediction, while treatment decisions live in the realm of causal inference. EMR data will be observational. Using observational data for pure prediction is completely valid, but using it for causal inference is only valid under strong assumptions of no unobserved confounding.

    A common fallacy of Big Data Hype is the assumption that gathering boatloads of observational data will enable us to solve problems that are fundamentally causal in nature. There will certainly be special situations where EMR data can reliably drive treatment decisions (and this will be a big deal!), but such cases will be the minority.

    By contrast, statistical algorithms should be able to almost always make excellent, reliable predictions about what conditions a new patient is likely to have or acquire given her own health history and the health histories of millions of other patients. These predictions, which are probabilistic diagnoses in themselves, can also guide decisions about which diagnostic tests to perform on which patients.

    From someone in the tech industry:

    This question in particular interested me:

    "JF: In the broadest sense, what difference will better information technology make in our lives and health?"

    And this part of the answer:

    "DB: This will move us into a field that is taking shape right now, that of analytics. It will help us take these data and turn them into diagnostic information—into recommendations a physician can give a patient or that patients can get directly, online."

    That’s where the future lies, and of course people want the benefit of it right now. Before, there was no market to make this sort of analytic product. Now that we have a growing electronic infrastructure for health information, there is a surge of traditional capitalist interest in turning that information into valuable knowledge, and selling it back to patients and doctors. That will happen. But it could never have happened until we got the data into digital form.

    To which I would add three things:

    1. Put simply, the short-term benefit is efficiency (easier/cheaper management of data for existing processes). The long term benefit is effectiveness (making people healthier)

    2. YCombinator (the #1 tech incubator, based in SV) yesterday put out their 'Request for Startups -- Breakthrough Technologies', which included a section on health, notably:

    "We’re especially interested in preventative healthcare, as this is probably the highest-leverage way to improve health.  Sensors and data are interesting in lots of different areas, but especially for healthcare."

    So certainly corroborating the "surge of traditional capitalist interest in turning that information into valuable knowledge." More here:
    http://blog.samaltman.com/new-rfs-breakthrough-technologies

    3. I share YC's view that preventative care holds huge potential returns. In general, I see "Health Care" being too focused on the "care" (e.g. I can get an appointment with an expert at John Hopkins within a week;  they have all the latest equipment;  the hospital is a nice building). And not focused enough on "health" (e.g. sixty year olds being in good enough shape to enjoy traveling and grand children; young people not being taken out of the work pool for a lack of basic medicine). Using data for preventative care could be a key component to redressing this - of course, it could also exacerbate it...

  • Why You Still Have to Fill Out All Those Paper Forms at a Doctor's Office ...

    ... and whether that might ever change.

    David Blumenthal MD

    In the new issue of the magazine (subscribe!) I have an interview with Dr. David Blumenthal. He is now head of the Commonwealth Fund, but during the first few years of the Obama administration he was in charge of moving America's medical-records system away from tedious paper-based filing to the digital age.

    I am biased, in that David Blumenthal and I have been friends since we were teenagers, but I think he does a very good job of explaining why it has taken so agonizingly long for medical records to catch up with the rest of our digitized life -- and what the payoff will be when the impending switch takes place.

    Please check it out. I've already received some retorts from other doctors (and insurance-company people), and will post in due course. 

  • Now This Is Team Spirit

    What the Wright Brothers have wrought. 

    I am taking no sides in The Big Game this weekend. I don't care.* 

    But I have to admire the combination of team spirit, precision flight planning, and disregard for practicality shown by the group at Boeing that produced this flight yesterday in solidarity with the Seahawks' "Twelfth Man" concept. Here is the radar record of the flight track, via Flight Aware.

     If you'd like to replicate the route, here are the waypoints, also courtesy of Flight Aware:

    SEA SEA146051 KS06G 4625N 12000W 4625N 11945W 4725N 11945W 4725N 12000W 4743N 12000W 4800N 11945W 4800N 11925W 4625N 11925W 4625N 11910W 4600N 11910W 4600N 11850W 4653N 11850W 4712N 11830W 4712N 11800W 4737N 11800W 4737N 11825W 4725N 11825W 4725N 11850W 4743N 11850W 4800N 11830W 4800N 11753W 4743N 11733W 4707N 11733W 4649N 11753W 4649N 11825W 4625N 11825W 4625N 11800W 4635N 11800W 4635N 11730W 4600N 11730W 4600N 11850W 4600N 11910W KS06G SUMMA SEA 

    It's up to you to find your own 747 to match** what Boeing flew.

    Update Here's the plane itself! Thanks to many readers in the Hawks diaspora who pointed me to stories about it (and this company photo).

    Update-update A reader who examined the Flight Aware charts adds this:

    Check the detailed flight data. They flew the 747 at 15,000 ft at 200 kts. In a way, that impresses me even more. Imagine flying your SR-22 at 1,500 ft and 80 kts for six hours straight.

    Yes, for an airliner this is quite low and slow -- comparable to early stages of an arrival/approach as an airliner is getting near an airport. For some other installment, what would be easier and harder about flying this way.


    * I grew up with the LA Rams: no más. My kids grew up with the DC NFL team: at this point, its continued flailing is not even interesting, the 15-year achievement of the league's worst ownership and management. So I decided that henceforth the community-owned Packers would be my team. For them, maybe next year.

    ** You'll probably also need to line up an RNP-style navigation system to plot out and follow this exact track, despite the powerful and variable jetstream winds blowing at those flight altitudes. I described the way some American-designed RNP systems were used for a different national-pride purpose, getting Chinese airliners into remote valley airports in Tibet, in China Airborne.  

  • Smaller-Town Startups: 'Stopping the Brain Drain' in South Carolina

    'People say, this is my ticket Out.' Then, they want to stay.

    "Code academy" room for The Iron Yard, inside the Next tech-accelerator building Greenville SC.

    Yesterday PCH International -- the company from Shenzhen, in southern China, that is run by my friend Liam Casey and whose exploits the Atlantic has chronicled from 2007 to 2012 -- announced yet another acquisition. It's of the e-commerce site ShopLocket, and the logic of the deal was an extension of what I've heard from Casey all along. The main function of his company (and others) has been to shorten the distance -- in time, money, effort -- between the idea for a new product, and the reality of that product in a customer's hands.

    Inside Marriage Special Report bug
    Reinvention and resilience across the nation
    Read more

    You can read all about it in the announcement, but here's the connection to our American Futures journey. ShopLocket is a service for the "maker revolution" -- the small startups, all around the U.S. and elsewhere, that are producing new things, and that are using the advantages of today's distributed commerce to help small, new companies do what only big companies could do before. That is, they can more quickly and easily: get startup capital; refine prototypes for their products; find suppliers and subcontractors; line up distributors and test markets; respond to shifts in demand; and all the rest. This is what Liam Casey was describing to me a little over a year ago, and it's what you can read about on the ShopLocket site here.

    Which brings us back to the Greenville-Greer-Spartanburg area of "upstate" South Carolina. A big question we have been asking is why high-value companies end up where they do, and how and where new companies get their start. In Burlington, Vermont, this involved asking what Dealer.com was doing there. In Redlands, California, why the big geographic-software company Esri (our partner in the project) had started and --  more interesting -- stayed in a place far from existing software centers.

    In Greenville, we spent a very interesting afternoon at the locally well-known firm The Iron Yard, which was housed in a new tech-incubator building that said "NEXT" out front.

    Next Innovation Center at night, photo by Brad Feld.

    As my wife mentioned before, and as we're sure to mention many times again, just about everything in the Greenville area reflects the fruits of the "public-private partnerships" that have rebuilt the downtown, attracted international manufacturing firms, created surprising new schools, and in other ways tried to reposition the town as a modern technology/culture/good-life center.

    [Before you ask, I've received a lot of mail about Greenville's troubled past in race relations and other barometers of inclusiveness. Greenville County, for instance, was the very last one in South Carolina to observe Martin Luther King's birthday as a holiday.  We talked the past, present, and future of the area's "openness" with lots of people there and will say more about it in upcoming dispatches.]

    The Next project is run by the Greenville Chamber of Commerce, with "public-private" guidance from local officials, businesses, developers, and so on. But it is located separately from the main Chamber of Commerce building and is designed to look and feel like a Boston/SF/Shenzhen-style startup center rather than some normal civic building. When we visited there last week, John Moore, a Chamber executive who runs the Next project, told us that it tried to run lean startup-style too. "We had the advantage of starting with a blank sheet of paper," he said.

    The Iron Yard, discussed below, part of the Next complex.

    "Because there were no existing entities to protect their turf, we were able to leapfrog," Moore said -- much as some developing countries jump entirely past the wired-telephone stage to create nationwide wireless networks. "We went from the idea for the center, to finishing the building and opening it, to having it full, all within three years. If we'd had to start with a university or an existing city facility and tried to change its model, it would have been a lot harder and slower." 

    The purpose of Next is to make it easier for new companies to start in the Greenville area. Moore pointed out that this "upstate" region of South Carolina had become famously effective in recruiting big, established firms to set up operations there: GE, BMW, Michelin, and on down a long list. "We've been so good an attracting other companies that we may not have done enough to develop our own," he said. Thus Next and related enterprises -- which connect startups with angel investors, provide physical space to get started, offer advice from mentors and startup veterans, and generally supply the sort of surrounding entrepreneurial information and advantage that can come automatically from being in startup centers from Boston to SF. 

    Some of the startups with offices in Greenville's Next building.

    Has it made any difference? Can it make any difference, I asked Moore, given the scale and distance handicaps of a smallish place like Greenville?

    "If you'd asked me five years ago, during the toughest times economically, I would have said, I hope so," Moore told us. "We had eight software companies in our program in 2006. They hadn't known each other. Now we have 134 companies, all new, in all kinds of industries, from manufacturing to genomics to game software." He said that he expected 200 local startups to be involved with Next soon. The main building has space only for 20 to 24; the rest are part of a network for advice, financing, and other services. Moore said the companies Next is looking for are ones "that can compete on a global scale but are based here."

    "They're now coming without recruiting. It's become a kind of flywheel. The momentum, the acceleration -- it all shows the potential. But of course I'm from the Chamber of Commerce, so you'd expect me to say that!"

    Indeed, but then he put it in more tangible terms, gesturing to an office across the hall: "A few years ago, people like Eric Dodds would never have stayed here."


    The Iron Yard's Eric Dodds, via Global Accelerator Networks.

    Eric Dodds, whom we met at the Next building, is a co-founder and the chief marketing officer of The Iron Yard, a multi-purpose software startup based in Greenville and with operations in Spartanburg, nearby Asheville, NC, and other southeastern locations. You can read more about its operations here.

    Dodds grew up in Greenville, and always dreamed of getting away. "Boulder, Portland -- that's where My People would be." Then, after going to Clemson and working for national branding companies, he came back and noticed that the place where he started out had changed. The Iron Yard's co-founder and CEO, Peter Barth, grew up in Florida and had worked in New York and the Midwest and was planning to work in Charlotte. He stopped for lunch in Greenville, walked through its famously renovated downtown, and decided this is where he wanted to stay.

    The Iron Yard's CEO, via Vimeo.

    Some other time I'll go into The Iron Yard's whole business model, which is a combination of "code academy," business incubators, kids' classes, and other features. The code academy charges $10,000 for a three-month session, and offers a full refund if graduates can't get an appropriate job. "We can guarantee an entry-level job, but entry level in this field might be $65,000 or $75,000," Barth told us. So far they have not had to give any refunds. (More here and here.)

    The point for today is an effect we heard about time and again. This was a change in this area's ability to attract and retain people like Barth or Dodds -- whom you might normally expect to find in Boulder, Portland, Boston, and who might have expected to find themselves there.

    "Greenville is great once you get here, but it can be hard to get people to come and take a look," Peter Barth said.  Eric Dodds added, "It's been really interesting rubbing shoulders with people in our classes who say: 'I’ve gotten here, I’m going into your program, and this is my ticket Out.' Then after a while they say, 'I’ve seen this culture, I think I’m going to stay around here.' It’s been very interesting in stopping the brain drain."

    More on this theme, the dispersion of opportunity, in coming installments. And before this coming week's State of the Union address, a reminder that the resilient capacity of America is more evident and encouraging city-by-city than it seems in national discussions.

    As a closing bonus, in case you were wondering what the Blue Ridge Parkway looks like from above during the Polar Vortex era, here is your answer.

     

  • Why Do Tech Companies End Up Where They Are?

    "You get some clusters, and some stand-alone firms far from anyone else.  But rarely anything in-between."

    Two days ago I mentioned that Redlands, California, posed a question similar to one we'd encountered in Burlington, Vermont. Namely: what were sizable but standalone Internet-based tech companies doing in these smallish towns? In Redlands's case, this meant Esri -- a leader in the mapping-software industry, and a partner in our "American Futures" project. In Burlington's, it was Dealer.com.

    Inside Marriage Special Report bug
    Reinvention and resilience across the nation
    Read more

    In the Seattle area or San Francisco or a dozen other big tech-cluster cities, it would seem commonplace to find such firms. But how had they gotten going far from a surrounding tech ecology that would naturally supply all of the supporting elements of a start-up culture, from potential employees to financial or marketing know-how to design studios?

    Several readers who themselves work in the tech industry write in with explanations. First, from someone who moved from another part of the world to work in one of today's dominant tech clusters, the SF Bay area:

    Yes, silicon Valley (SV) and its environs (SF) are a tech hub. So if you're dreaming up a new communication protocol (Twitter), you need to be surrounded by people who can share in your vision of the future -- and that means (for the most part) that you have a shared 'now', and that's why Twitter was only going to be built in a tech hub. If you're building a better search engine (Google) and Yahoo! is just dont the road, then you can strike up a deal to power their search (which funds you) and if you're surrounded by techies then you have your early adopter users on your doorstep. 

    Now, if you're building software for car dealerships then the #1 thing you need to grow is access to car dealerships. Lots of tech folks I know in SF dont even own cars -- cars are not top of mind. You'd have the same problem in NYC. On top of that, because a lot of city techies dont care about cars, they're blind to the opportunity for selling software to car dealerships. On top of all that, as good a business dealer.com is, I suspect the hardest part of building that business is selling to the dealers, not writing the software (that's still easy to mess up, of course, as proven elsewhere). And I doubt SF car dealers a significantly earlier adopters of tech than anywhere else in the country (I have tried selling tech to them, so I speak from experience :)

    Good software can be written anywhere that there are smart people. So my mental model is not to ask, "The puzzle, again, is why -- and why here?" it's to ask "Why couldn't it be built here?". High-tech spinouts from universities is one possible answer. A need for a cluster of early-tech-adopters (mainly B2C) provides another. 

    Inside Dealer.com, in Burlington; company photo.

    From another person in the California tech industry:

    Some thoughts on why you get some clusters and some stand-alone firms far from anyone else.  But rarely anything in-between.

    As you note, it is now possible to run a high tech company from anywhere, and have employees scattered all over the landscape.  For example, the small company I work for [which produces network analysis software] is "headquartered" (i.e the founder and CEO lives) an hour-plus south of Silicon Valley. I am a couple hours northeast of there. I get down there maybe once a year.  Our third employee is on the East Coast.  The company is over a decade old; later this month we will have our first-ever all hands face-to-face staff meeting. 

    A company can spring up anywhere, as your examples also prove.  But how do you get a cluster?  

    New start-ups/spin-offs frequently happen because someone wants to do something new and different, and can't persuade his management to go for it.  But for that to happen, he has to have been where he can talk casually with others in the same field.  And growing will generally require recruiting from what is essentially a single local pool of talent: the base company.  (Recruiting can also be done via networks built at professional conferences.  But it's harder than recruiting over lunch or at the park.)  That makes getting a second company going very difficult.

    As a result, you can get a stand-alone company anywhere.  It's unlikely, but it can happen.  However to get a cluster, you have to somehow get a second (and preferably a third) company in the same area, before additional companies can be started (relatively!) easily.  Essentially, that means repeating the original process for starting a stand-alone company.  Obviously that's do-able; stand-alones do get started.  But new stand-lones don't happen often -- which means the odds of it chancing to happen multiple times in the same area are really, really low.  And only when you get lucky 2-3 times in one area do you have the conditions which will allow a cluster to blossom.

    Esri cafeteria, in Redlands, via Armantrout Architects.

    And one more explanation:

    I have to assume your question "Why here?" regarding ESRI was rhetorical. It was pretty obvious to me why - because they could. They loved the town, figured others would too, and they succeeded there.

    This is really an interesting connundrum. You (and others) have accurately written about the importance of a support system (probably more properly "supply chain") as being essential to a region's large scale success, esp in modern manufacturing. It would be incredibly difficult (though not impossible) to open an electronics assembly plant in some random town in America, because it's so much easier, cheaper, and faster (by most measures) to do so in China.

    But as ESRI has shown, and Dealer.com as well, if the founder is dedicated to a place, is willing to forgo some, or even most of the easy money, and the location has an appeal that is understood by people other than the founder, then success can happen. It won't be as fast, or as profitable, but those aren't everybody's measure.

  • Landing at the Wrong Airport: Sigh

    America is covered with small airfields. Usually that's a plus. But not always.

    Many people have sent me links about the unfortunate news of a giant cargo airplane, a modified 747 known as a "Dreamlifter," landing at a smaller airport in Kansas rather than at its intended destination, McConnell Air Force Base. 

    How could this happen? Of course it shouldn't happen -- airports have distinct GPS locations and "airport identifiers," all but the smallest ones have instrument-approach frequencies that you dial in (even in good weather) to be sure you're lined up with the right runways, you're supposed to be constantly looking for landmarks, and so on. But if it were to happen, you could sort of understand in a case like this, where there are three airports with similar orientation lined up one after another on the Kansas plain. Via Google Earth, here is how it might look if, like this plane, you were headed in from the north:

    The airport at the top of this view is McConnell Air Force Base, where the plane thought it was headed. The one in the middle Beech airport, whose runway is laid out in the same direction. At the bottom is Jabara airport, also with a similar layout, where the plane actually touched down -- and, fortunately, managed to take off again the next day for the minutes-long flight to McConnell.

    For previous cases of adjoining, similar-looking airports leading pilots astray, see accounts from South Dakota and from Florida

    Main reason to mention the story, apart from the interesting videos in some of those older, linked posts? The air-traffic control tape of the mistaken landing and its aftermath truly is fascinating. We all know about the pilot sangfroid that captured in the beginning of The Right Stuff. "You may feel a tiny bit of bumpiness here," the Yeager/Tom Cruise-accent-aspirant pilot might say as the plane is tossed up and down through a thunderstorm. Bear that in mind as you hear flight crew and controllers sounding nonchalant about what they all recognize, but are not mentioning, as a big embarrassment.

    In case you were wondering, here is how the three airports in question look on an FAA "VFR Sectional" chart. Jabara (actual landing site) at the top, Beech in the middle, McConnell (intended destination) at the bottom.

    To keep things in perspective: no one was hurt, the plane managed to take off again, and the only thing damaged was pride. May all aviation misjudgments have effects this benign. 


    On the airport theme, something new has come to PEK, the Beijing International Airport where I sit at the moment, since my last visit here. There's free wifi: Great! But to get your code, you have to put your passport into a scanning machine that reads and records your coordinates before issuing you a wifi code. I guess there are still tricks for the NSA to learn.

  • Tech Tips for the Day: Gmail Fix, Doo.net

    If you really hate the new Gmail compose interface ...

    1) Gmail de-lobotomization. The Gmail design team is constantly coming up with fritterware "improvements" to the system's look and feel. In the face of most of the changes, all you can do is sigh and adapt. Or switch to a more formal email front-end, like Outlook, Thunderbird, Entourage, Apple Mail, and so on.

    But there does seem to be some recourse against the latest Gmail change, the shrinking and dumbing-down of the Compose window that became mandatory not long ago. You can follow the instructions shown at Ghacks, a German site run by Martin Brinkmann. The site tells you how to download and install a Chrome extension that restores the previous Gmail look. (The fix works only in Chrome.) Follow the instructions on this site exactly, since there are tricks involved for both downloading and then installing the extension. It's conceivable that the extension has some undesirable side effect, but in using it for a while I haven't run into any problems, and Brinkmann has a track record of sound advice. Take a look, and thanks to reader DS for the lead. 

    2) The unusual Doo.net. This is turning out to be German-software day. A new indexing and retrieval system called Doo has come out from a small startup company based in Bonn. It works on five major platforms -- the latest versions of Windows and Mac, plus iPhone, iPad, and Android -- and at the moment is free.  

    Doo is designed to parse, categorize, organize, and retrieve documents of all sorts in a more intelligent way than existing indexers. A minor example: it automatically comes up with a list of all duplicate files on your computer or your cloud storage areas, with the option to delete all the redundant ones. (Hundreds of them, in my case.) 

    Like other "interesting" programs in their nascent phase, I find this one both intriguing and somewhat puzzling. But I have installed it on my MacBook Air (which is now my standard workplace computer -- I connect it to a big display and nice keyboard when I'm at home) and on my iPad and Android phone as well. I find myself using it a little more each day. Give it a look too. 

  • The Odd Fallibilities of Flight Aware

    Technology, our friend and foe: Chapter 3,189.

    Inside Marriage Special Report bug
    Reinvention and resilience across the nation
    Read more

    At our American Futures special-report site, you'll find a new post explaining the difference between the flight-plan map you see above, and the radar-track image you see below. The comparison is an intriguing (to me) illustration of the partial-but-not-total accuracy of the popular Flight Aware flight-tracking site.

    In the other post I explain what caught my eye about the difference between these renderings, and why they matter. I've sent a note to Flight Aware (which is always careful to say it's not offering perfect reports) to better understand what is going on.

    Now, out to a factory. Soon, actual substance reports.

Video

Where Time Comes From

The clocks that coordinate your cellphone, GPS, and more

Video

Computer Vision Syndrome and You

Save your eyes. Take breaks.

Video

What Happens in 60 Seconds

Quantifying human activity around the world

Writers

Up
Down

From This Author