University of Texas made a great military hire and you can too


As active duty Navy and a UT alumnus, I’m thrilled with the decision by the University of Texas to (soon) appoint ADM McCraven as chancellor.  He’ll be a great leader at the university and will make UT alumni proud.  If you aren’t familiar with the Admiral or his work, UT did a great piece here.  There are a number of reasons I could point out as to why ADM McCraven is a great choice, but  I’d rather focus on how other employers can do the same.   So here are five tips for employers on how to hire superb senior leaders from the military.  Although it’s in my standard disclaimer, I must point out these are my opinions and not the government’s.

1.  Ask questions about their military assignments.  Specifically, you’ll want to know what their positions were, and how they showed leadership.  What you’re looking for here is variety and leadership progression.  If you read up on ADM McCraven, you’ll see he’s been a number of places and held all different levels of responsibility.  You should expect the same out of people you want to hold top leadership positions.  You’ll notice that ADM McCraven doesn’t have a lot of higher education experience.  He has a lot of experience.  Consider the same thing when looking at your candidates.  What you’ll want in a high level candidate is the ability to absorb information and make decisions.   You’ll want a candidate who can lead others.  These are the strongest characteristics of military leaders, so look for them.

2. Ask for information on their performance.  The military makes a lot of effort to document performance of its personnel.  For the Navy, this comes in the form of evaluations, or fitness reports.  Fitness reports do not have a lot to do with physical prowess, but more about the individual’s performance of their job.  Individual job applicants may have different opinions about sharing this information with a prospective employer, but it doesn’t hurt to invite them to share it with you.

I definitely would not hold it against a candidate who refuses to share their previous evaluations.  Sometimes these reports are not reflective of the candidate because their leadership didn’t invest the appropriate time.  Additionally, if your candidate is super cool, like ADM McCraven, there’s probably some stuff you’re not allowed to know.   If you do see the reports, focus on the narrative, which explains the level of leadership the candidate had at a particular job.  It could also provide some insight on projects they’ve worked on in the past.  If nothing else, those are great conversation starters for learning about your candidate.

3. Don’t be embarrassed that you don’t know the military lingo.  Chances are your candidate has made an effort to translate his/her resume into a civilian friendly style.  Unfortunately, some things don’t translate well.  I see this as the biggest barrier for military applying for high level jobs.  Take my linkedin profile for example.  Here’s a line from one of my positions. “Superior leadership led to hospital’s recognition as the top pharmacy in the Department of Defense with the 2008 Military Health System Excellence in Pharmacy Operations award.”  As a civilian this is difficult to quantify or understand.  I wouldn’t be surprised if a lot of my readers didn’t know military pharmacies existed. (In fact there are hundreds.) Similarly, there are individuals in the military who serve on “committees” that have greater leadership impact than a civilian board of directors.   When interpreting a resume or interviewing, try to nail down the impact, rather than the title.

4.  Respect why the person is leaving the military.  I’m undecided if it’s valuable to ask why a person is leaving the military.  Sometimes it will be clear (e.g. retirement), but sometimes it will put the candidate in a bad position.  Maybe they weren’t challenged by the military, or were frustrated by operating under a structured chain of command.  These answers might help a prospective employer in evaluating the candidate, but you also risk misinterpreting them.  This is particularly the case if you’ve never served on active duty personally.  If a candidate replies, “I was tired of the bureaucracy,” it could be interpreted as they don’t like systems or rules, when what they were implying, in fact, is that they didn’t like accounting for their presence at all times.

Bottom line: Be careful with this question if you ask it.  The military wasn’t just a job they left.  It was a complete life.  The candidate may have had “coworkers” who saved the candidates life.  They almost certainly didn’t make the decision hastily.  One final note on this: I only am referring to candidates who left the military under honorable conditions.  Other than honorable or dishonorable discharges require a different level of scrutiny.

5. Thank them for their service.  You may not hire them, but they have done a service for their country.  Most people will humbly accept the thanks, but it means more than you think.  If they were a decorated hero like ADM McCraven, or a technician that has never been to a war zone, they have given up some of their freedom for yours.

Bonus point: If a Medal of Honor hero applies for a job with you: hire them.  They deserve the job just for what they’ve done.  President Bush said in posthumous presentation of the Medal of Honor to LT Michael Murphy, “With this medal we acknowledge a debt that will not diminish with time and can never be repaid.” Remember you can’t do a favor for these heroes.  They’ve earned all they receive.



my idea to save 10,000 people

A few people have asked me about yesterday’s post.  What’s the idea?  I intentionally didn’t include it then for two reasons.  First, it wasn’t the point of the post.  Second, I was still personally considering how to bring it to market.  Last night, however, I considered something Elon Musk said in a recent interview with Stephen Colbert.  For those who don’t know, Elon is a founder of Paypal, SpaceX, Tesla Motors, and Solar City.  If you don’t know those companies, I suggest you look them up, and discover why I consider Elon one of the greatest visionaries of our generation.  During the conversation with Stephen, it was pointed out that Tesla Motors recently released their key patents for open use to anyone who wanted them.  When asked, “Why?” Elon’s response was priceless.  I’m paraphrasing, “It seems to be that we’re all on the same sinking ship, and we’re vigorously bailing out water.  Now if I have a design for a better bucket, isn’t it my responsibility to share it with everyone?”

So with that in mind.  Here’s the idea.

First the problem.  According to the CDC, “113 people die (daily) as a result of drug overdose, and another 6,748 are treated in emergency departments (ED) for the misuse or abuse of drugs.” It goes on to specifically to call out opioids (common pain killer Vicodin, Percocet, etc.) as the primary source of overdose deaths.  “Of the 22,810 deaths relating to pharmaceutical overdose in 2011, 16,917 (74%) involved opioid analgesics.”  Keep in mind these numbers don’t include those who die from heroin overdose, which is simply a more potent form of the same mechanism.  Interestingly, there is a pretty straight forward diagnosis and treatment for opioid overdose.  The treatment, naloxone, has been around for many years as a single injection to reverse the effects of opioids.  In fact, a new device, Evzio, just earned fast-track FDA approval for treatment of opioid overdose.  It functions with an auto-injector that you can stab in someone’s thigh when you believe they have OD’d.  It’s a great product, and there’s a lot of discussion about making it available over the counter, so that more deaths can be prevented.  It does have two fatal flaws.  First, it requires someone to recognize that a drug overdose is taking place.  Second, it requires a person to administer it.  As you know, that’s often not the case.  Just look at high profile celebrity deaths of Heath Ledger and Phillip Seymour Hoffman.  They could have had 10 injectors in their possession.  Without someone to recognize overdose and administer it, the device is useless.

Now we’re getting close to my idea.  So, let’s talk about the first problem of Evzio. Diagnosis.  There are different factors that can be used for opioid overdose diagnosis, but I’m proposing these three as sufficient.  You’ll see why in a minute.

1.  Exposure to opioid.  Common sense, right?  The person suffering the OD is aware before, during, and after the event they have been exposed to an opioid substance.

2. A resting respiration rate less than 8 breaths per minute.  This is ultimately what kills in opioid overdose, and is the hallmark feature.

3. Impaired consciousness.  For my device, I’m using the Pasereo Opioid-Induced Sedation Scale (POSS).  There are others, but I chose this one and you’ll see why soon.  For information on the different scales you can look HERE.

So, how do we make diagnosis easy?  We can automate the whole thing.  Here’s how:

1. Exposure to opioid.  By applying the device and activating the accompanying phone app, the user has indicated they have been, or soon will be, exposed to opioid substances. That one is easy.

2. Resting rate.  The non-app part of the device affixes to the skin, (you’ll see why in a second) and contains piezolectric sensor,s which can utilize the lung sounds to determine respiratory rate.  This technology was validated by some researchers from Poland in 2008 and is available as a commercial product today.

3. Assessment on the POSS scale.  This is accomplished by increasing alert levels of a smart phone.  Level one is detected by response to a simple ping and vibration.  Level 2 is indicated by response to  a more consistent vibration and soft sound.  Level 3 is indicated by a response to a consistent vibration and a loud alarm.  Level 4 is indicated by no response.

There. Now we’ve detected a probable opioid overdose without a person’s intervention.  So how do we stop it?  Simple.  The device, as I mentioned, is affixed to the persons skin.  In addition to the piezoelectric sensors, it contains a self-administration set of naloxone.  Again, this is not new technology.  The best technology I can compare to is omnipod.


Here’s a photo of the device.  It is for automated insulin delivery.  It could simply be reconfigured for naloxone delivery.  The device inserts a catheter into the skin and delivers the appropriate dose, all without human intervention.  Configuring this to interact with the sensors and the app could present the solution to opioid overdose without a second person’s intervention.

There area  few more tweaks I would propose to make this the ideal solution, but I’ll stop there because the post is already running long.  I’d love to hear what you think.

P.S. I named the device ODefense.  If someone else makes it, I’d appreciate the courtesy.

i could save 10,000 lives but it’s too hard

I have an idea.  It’s a brilliant idea (in my not-so-humble opinion).  It’s for a new medical device.  Based on death data in the United States, if this device was used on 1/3 of the population currently affected by this affliction, it would save 10,000 lives ANNUALLY.  That breaks down to just over a life an hour.

Unfortunately, this idea probably won’t happen.   That’s due to three nasty letters: F.D.A.

If you remember, in a blog not long ago, I praised the FDA on their resistance to approving the drug Tredaptive.  Unfortunately, their processes can also be harmful, and I’ll show why.

The first line of the FDA’s mission statement reads, “FDA is responsible for protecting the public health by assuring the safety, efficacy and security of human and veterinary drugs, biological products, medical devices, our nation’s food supply, cosmetics, and products that emit radiation.”  Most of that sounds good to me, but I’m a little hung up on one part.  Protecting the public health by ensuring… efficacy.  Now I understand the need to measure and promote efficacy.  Otherwise we’d have people peddling all kinds of stuff that wasn’t efficacious, like vitamins, or herbal supplements (which Jim Gaffigan calls, “the horoscope of vitamins”), even prescription drugs, described as supplements (I’m looking at you GNC.).    Well, let’s ignore that for a second, because maybe those are bad examples.

Instead let’s look at how the FDA stepped in on medical apps for smart devices.  The full link is here (APP guidelines.).  It’s an interesting read for sure, but it says basically three things.

1. Most apps aren’t devices and the FDA doesn’t care about them.

2. Other apps are devices, but they can’t really do any harm, so the FDA doesn’t care about them either.

3. Some apps are legitimate, FDA-devices, because they talk to another FDA-device, and the FDA sure wants to run those through the ringer.

It’s an interesting approach on how to define things.  I like the idea of products being free-to-market if they can’t cause harm.  The idea I have is in line with that definition.  It’s not benign, in that it does provide a medical intervention.  The device itself, though, can do no harm.   If it fails, the worse possible outcome is no different than no intervention at all.  So why does it require FDA approval?  It requires approval because it’s not an app.  It’s a physical device that touches a person.  Seems like a little bit of a double standard you’re pulling here, FDA.  I should be clear, I want the device to be 100% effective.   What’s really upsetting is what it takes to prove that to the FDA.

According to the Medical Device Manufacturer’s Association, in 2010, it cost approximately 24 million dollars to get a device through FDA approval.  You’re thinking, “Cf course! It costs millions to build a new smartphone as well.”  If that’s your thought then you didn’t read the statement correctly. I didn’t say costs to get from concept to market.  I said costs to get through the FDA.  One company reported spending upwards of 75 million dollars to get a product through the FDA.   The same report said the total concept-to-market cost for a new device is an average of 31 million dollars.  So to recap: 7M to build and market; 24M to get approval.  That seems a little off.

I don’t have 31 million dollars.  I wouldn’t know how to get 31 million dollars.  I would have to get 7 million dollars, to build a product to convince someone to give me 24 million dollars, to get the product approved.  Seems like this might make it hard for me to get the original 7 million dollars.

At the end of the day, I’m stuck where Sergey Brin (Google Founder) said I would be. “… generally, health is just so heavily regulated, it’s just a painful business to be in. It’s not necessarily how I want to spend my time. Even though we have some health projects, we’ll be doing it to a certain extent. But I think the regulatory burden in the U.S. is so high, I think it would dissuade a lot of entrepreneurs.”

So I have a great idea, but it’s too hard, and that makes me sad.

making twitter a reliable news source

When the tragedy (if that’s even a big enough term) of MH17 occurred, I found out through twitter.  It was reported within seemingly minutes of occurrence.  I followed the trend term for Malaysian Airlines and then MH17 for up to the second details of the event.  It’s not the first time I’ve done it, either.  I used twitter similarly to follow much less somber subjects, such as World Cup soccer, or the Google i/o Developers conference.  What I didn’t do was turn on my television, or go to the advertising-laden web page of a major news source.  I didn’t go to those sources because they were behind what I was seeing on twitter.  Sheer logistics made them slower.

I’m a pretty quick typist.  I can type about 350 characters in a minute.  That means, with the right content, I can submit 2 and a half tweets per minute.  Theoretically, I could summarize most of the facts from the MH17 disaster in less than ten minutes of tweeting.  It takes Fox News ten minutes to get Shepherd Smith’s hair right.  They just can’t compete in a timely fashion.  Where twitter starts to miss it’s mark is in the reliability of the information.  Because just like the facts, I can tweet out a substantial amount of lies in 10 minutes as well.   Is there a way for twitter to give you a reliability index for the tweets you’re reading?  I think so, and here’s how.

Let’s start with a simple example: a news event that we know occurred, and know where it occurred.   We know the Major League Baseball All-Star game happened last week on Tuesday.  We know it happened in Minneapolis, and, if I took thirty seconds to Google it, I could probably give you the GPS coordinates of the field.  So let’s start there.  The best and fastest facts will probably come from someone who is tweeting from a close, geographical proximity to the game.  Easy enough: if they’re close to the event, the reliability goes up.  Now let’s say Jim, a fictional guy at the game, tweets “Jeter hits a home run! Way to go out on top! #allstargame”.  Since he’s at the game, there is some reliability to that tweet.  But what if Jim tweets, “Babe Ruth hits a home run! #allstargame”?  Jim still has the proximal reliability, but is factually wrong.  So we can increase or decrease his reliability, by the presence or absence of a corresponding tweet from an alternate source.  This works for major events that people may care about (Jeter’s home run), but isn’t really important for inconsequential items: “The hotdogs at the #allstargame are delicious”.  We strengthen the reliability by the degrees of separation between the people tweeting the information.  If Jim and his follower Tom both tweet about Jeter’s home run it is less reliable than if Jim tweets it, and Anne, whom he only has a fifth degree connection, also tweets it.  Since both Tom and Anne tweeted about Jeter’s home run, and they were both near the event, we can assign a pretty high level of certainty that it is accurate.  The best part is this can all done by content matching and some fancy math.

Got it?  OK, let’s move on.

Now a substantially more complicated event.  A car crash on the DC beltway at rush hour that has traffic backed up for miles.  Bill tweets, “Guess I’m not getting home early #wreck #dctraffic.”  This tweet has no reliability at this point.  It’s something Bill put out, but he could have been commenting that he thinks DC traffic, in general, is a wreck.  Here’s how it gets reliability AND applicability (a two for one folks).  Natalie also tweets out “Accident on the beltway has traffic backed up for miles! #stayaway.”  Assuming they are in the same proximity, they have increased the reliability of the event.  Similar to the Jim and Anne example, the less they know each other the better.   More collaborating tweets is better and increases the applicability.

There are a number of different hurdles to overcome for this to work.  Word associations is probably the largest.  I can envision a path for machine learning on this, but this post is long enough.  I’ll leave it there and hope someone else has some ideas.  You’re welcome for the million dollar idea, twitter.


a cheaper way to cure hep c?

A lot of attention has been given recently to the new drug Sovaldi.  In case you don’t remember, it’s the really expensive drug used to treat Hepatitis C (hep c).  Let’s get this out of the way up  front: the drug is revolutionary.  It has phenomenal cure rates, compared to the previous standard of care.   The real reason that it made news, though, was because of its cost.  The quoted figure is $84,000 for a course of treatment, and yes, you can buy a modest house, or a really sweet car for that much.  Of course, anyone who has spent time with pharmacy benefit managers (drug insurance companies) understands that NO ONE is paying $84,000 for a course of treatment.  I’ll give the world a lesson in how drug pricing works at another point in time.  For now, just trust me: it will cost less.

According to a recent study by Express Scripts Inc (that I found via twitter), the state governments could be on the hook for up to $55 Billion of hep c treatment.  One fact I found particularly striking was, “(California) is expected to spend $6.6 billion if it treats all 93,000 of its Medicaid enrollees and prisoners who are estimated to have chronic hepatitis C.”  If you want to see the math, you can check it out here  Expensive for sure, but this is a disease state that has significant medical and economic consequences for its sufferers.  Sovaldi is a good drug, and I believe, if used appropriately and en mass, we could theoretically begin to stamp out Hepatitis C.  With the cure rates for a viral disease so high, it’s not improbable to think we could eliminate the disease from the United States.  Think about that: eradication of a disease we have struggled with since it was identified over 25 years ago.

To accomplish that we have to think of this agent, not as a treatment for the sick, but as a vaccine.   Now technically, this isn’t a vaccine.  Not even close.  It doesn’t stop someone from getting hep c, and it only works if you’re already infected.  It does effectively reduce the number of infected patients in the population.  By lowering the number of infected, the risk of contraction is lower.  As you drive both the number of infected down, and the number of newly infected down, you reach a point (ideally zero) where the disease is eradicated.

The first question you may have is, “Why don’t we just treat the symptomatic?”  Fair enough, we do that for lots of stuff.  The problem with that strategy here is that the population of newly infected will continue to persist.  This means the burden of treating the disease will stick with us for much longer if we conducted an all out assault.  By driving down the infected population, the number of people we treat over the next 20 years would be significantly lower.  So my first point towards making hep c cheaper to fight is Use Brute Force.  I’m talking Polio vaccination campaign force.  Drive the infected population down quickly, so we end up treating fewer people.  We don’t eradicate diseases often, so this is not a step to be taken lightly.

If you agreed with that somewhat contentious, point here’s your next question. “How do we pay for it?”  You thought the first part was tough to swallow? Then hold on.  Eminent Domain. I’ll let that soak in for a second.

Now, before people get bent out of shape, I’m not all about the government taking over everything in our lives.  I like my property and I want to protect the rights of citizens and companies, alike.  Additionally, I’m not a lawyer.  This may not even be legal, but I think it’s a radical idea worth considering.  California, the state in which Gilead (the maker of Sovaldi) resides, is on the hook (theoretically) for $6.6B worth of treatment for Hepatitis C.   What if instead of spending the money on treatment, they spent more (probably considerably more) on procuring the ownership of the drug?  Under eminent domain, they could, in my opinion, claim the drug as property in public interest and pay a fair market value.   Today the market cap on Gilead is about $137B, so it would probably be a substantial amount.  By accomplishing this we could begin spreading the treatment across the world in hopes of eradication.  Or we could wait until it goes generic… at a cost of about 350,000 people per year.

As a final note, there are even more effective medications coming to market soon and their cost is unknown.  The same logic applies.  Treat people quickly to drive the population of infected down.  How we do it is irrelevant.  This is a public health situation and not a personal health situation.  Let’s treat it that way.

Is our current system for medical literature in need of an overhaul?


I took time today to read the recently published results of the HPS-2 Thrive study. For those who don’t know, this is the study receiving a lot of media attention because it questions the benefit of Niacin in reducing bad outcomes (death for example) in patients who are at high risk (bad heart) and already taking a statin (#1 cholesterol medication class). After reading the article and doing a little digging on the history of the study drug Cordaptive (which I had never heard of), I came up with some interesting observations worth considering.

1. This is not new news. Sure it was published in the New England Journal of Medicine yesterday, but Merck (the manufacturer of the drug) decided to pull it’s FDA application over a year ago because of results in this study saying there were adverse effects (bad things) due to the medication. At that point the study was ended, which means no new results, which means the results were there over a year ago. Why did it take this long to get a full study on if Niacin works? You can be certain, though, that your doctor will start incorporating this “breaking news” into your health treatment today. Which leads me to the conclusion that maybe our current system of disseminating drug knowledge is not the fastest or most efficient. Granted, I understand the need for peer review, and I do want quality information. (Tweeting interim study results is no way to manage patients.) The questions still stands: Are we emphasizing the literature process a little too much? Especially given that we aren’t getting it right all the time anyway. For examples of failure see any law firm billboard in a metropolitan area.

2. The FDA got this right. The EU approved this drug and it was on the market (Tredaptive). The FDA did not. I’m not saying the FDA system is perfect, or even good, but it got this one right. I will be the first to criticize their slow pace on getting some items to market, but I should also give them credit for nailing this one. Good job, FDA. Don’t let it go to your head.

3. The system for post-approval research is abysmal. Niacin is not new. It’s a vitamin for goodness sake. We’ve used it for decades under various names. How did we discover that it might not be effective? Merck came up with a clever way to market an old product by adding something to reduce the side effects. Only because this combination required new drug approval and marketing did we conduct this study and find these results. Had Merck not been trying to skim easy money by repackaging an old drug, we probably wouldn’t have found this result. Good job Merck. Don’t let it go to your head… and sorry about the stock drop.

4. It’s possible that Niacin is still useful. WHAT!?! I just told you the study was discontinued early due to side effect, and you already know it didn’t prove to help people in the study.  Remember, this was a 4 year study. It was supposed to run a lot longer to be able to detect a difference. They stopped it because of the adverse effects. If the researchers let the study continue, it could have shown that Niacin actually prevented bad outcomes. Did they do the right thing by stopping the study? I say yes, the harm outweighed the good at the time they stopped it, but the point stands. The results don’t mean Niacin wouldn’t have eventually been proven helpful.

Now over the next few days and weeks, experts will talk a lot about outcomes. “Why are we measuring HDL?” and “Does HDL actually affect death rates?” In my opinion, we’re missing the point. Forget about how to dissect these results. Look at the glaring systemic issues above, and think about how we can reboot what is done in medicine. Getting these kinds of results is important. How do we get them more quickly, and more consistently, into the hands of people educated to make decisions based on the information?