It sure isn’t ecstasy.  Yesterday, March 31, 2014 our US Senate passed a bill to delay the ICD-10 implementation  for another year.  It was supposed to go into effect on October 1, 2014.

For most people, ICD-10 is an unfamiliar term, but for many of us in the medical field this is a dark cloud that has been looming over us for years.   This is the new and improved coding system (60 thousand medical diagnosis codes so that WHO amongst others can have its stats on what’s happening).

I’ve been working on this for several months now, and it seems that many are seeing what is on the horizon.  October surpriseh/t Adrienne

Codes exist for being hurt at the opera (Y92253), walking into a lamppost (Y92253), walking into a second lamppost (W2202XD), getting sucked into a jet engine (V97.33XD), and being burned due to water skis on fire (V91.07XD).

Remember death panel overlords will be able to review these codes and know exactly what happened at what moment wherever to deny or approve what YOU may or may not believe is medically necessary.  (Forget what you or your physician are thinking.)  This is how decisions will be made.

From one of the comments.  Yes, I’ve heard these scenarios as well.

ICD-9 vs. ICD-10

In one example a CMS executive gave me, the new codes will impact patients and insurers in quite basic ways. Here’s the scenario she gave me. Under ICD-9: In April, Johnny falls off a neighbor’s tree, breaks his left arm, and goes to the hospital. Doctors fix him up and insurance pays. In May, Johnny falls off the same neighbor’s tree, breaks his right artm, goes to the hospital, doctors fix him up — but insurance balks, saying “We already paid for this.” That’s because ICD-9 doesn’t specify right or left limb. Under ICD-10, the doctor has to state s/he is working on the left or right arm. The insurance company sees it’s a different arm and paperwork goes through smoothly.
Perhaps? Of course, klutzy little Johnny who likes trespassing could ostensibly continue to break the same arm. But at least if he alternates arms it’ll be easier for his parents to get the bills paid — until they finally ground him or the neighbor gets a mean dog and/or a taller fence.

Social problems?  No worries, CPS can use those codes to take care of Johnny if his parents won’t fix him.  Just like this, and yes I am being sarcastic.

Z63.1: Problems in relationship with in-laws

According to a director of ICD-10 codes, this actually is a billable code that can be used to specify a diagnosis. Providing social context about trouble in the family could also be important for understanding the patient's source of stress or, in more serious cases, injury from domestic violence.</p><br /><br /><br /><br /><br /><br /><br /><br /><br /><br /><br /><br /><br /><br /><br /><br /><br /><br /><br /><br /> <p>(Image: All in the Family cast, Wikimedia Commons)

According to a director of ICD-10 codes, this actually is a billable code that can be used to specify a diagnosis. Providing social context about trouble in the family could also be important for understanding the patient’s source of stress or, in more serious cases, injury from domestic violence.

Or that bird bite.  Yes, I now am a parent to FOUR parrots so I will have to watch out for my fingers.

W61.01XA: Bitten by parrot, initial encounter

Ah, but let’s move on…how will this impact your doc?  I’ve listened to a few webinars that have recommended solo practices request a 6 month loan to cover operating expenses since this is a major overhaul of the healthcare system.  Cost to implement this is high considering that reimbursement for physicians has gone down, down, down.  Estimates range in the $85,000+ (for a solo practitioner) to several million dollars in cash reserves for different size practices implementing ICD-10.  Via October surprise.

The current version, the ICD-9, uses a 4- or 5-digit number to code for a particular disease, such as 540.9 for appendicitis.  The ICD-10 will have up to 7 alphanumeric characters to specify a condition, such as S52.521A for “torus fracture of lower end of right radius, initial encounter for closed fracture.”  And there are now over five times as many codes for doctors and hospitals to choose from.

But isn’t specificity better?  Sure it is.  Big data is the new frontier in medical research, making sense of the huge amount of generated health care data.  But can this go too far?

In an effort to push specificity to the limit, some ICD-10 codes have gotten silly.  Codes exist for being hurt at the opera (Y92253), walking into a lamppost (Y92253), walking into a second lamppost (W2202XD),

W22.02XD: Walked into lamp post, subsequent encounter

…getting sucked into a jet engine (V97.33XD), and being burned due to water skis on fire (V91.07XD).

Let’s get to the down and dirty.

First, medical practices and hospitals must know and have all of these 68 thousand codes readily available to add to the medical record in order to bill correctly and hope to be paid.  One more distraction for physicians, aside from all of the daily distractions of electronic records.  When physicians pay more attention to their computer screen or tablet than to the patient, guess who suffers.  This is the reason why texting and driving is illegal.

(Bold emphasis is mine.)

Second, electronic medical records (EMR) must be able to incorporate these codes into the exam or procedure report.  Are all EMR vendors up to speed on these codes?  Will their system upgrades work as advertised?  Or will they work as well as the website?  And if the codes don’t work, physicians and their practices don’t get paid.  Yet landlords, employees, and utility companies still want to be paid.

Third, will the insurance companies recognize each of these new 68 thousand codes, correctly match them to billed procedures, and promptly pay the providers?  If I treat a patient with macular degeneration with a monthly dose of a $2,000 drug, I now bill a single code, which insures that I will be paid.  Under the ICD-10, there will be 20 codes, specifying which eye(s) and severity, that allow payment.  Will every insurance company have each of these codes in its computers?  Will it recognize each code?  Remember that these are the same insurance companies that don’t even know who has actually paid their insurance premiums.

The American Medical Association announced this week that ICD-10 implementation will cost three times as much as originally estimated.  The  “costs of training, vendor and software upgrades, testing and payment disruption” could be $225,000 for a small medical practice and over $8 million for a large practice.  How do medical practices of marginal profitability absorb these costs?  With physician reimbursement rates set to grow at only half a percent per year over the next five years, far below the true rate of inflation of close to 10 percent, the financial writing is on the wall. This will accelerate the demise of private practice, already underway due to ObamaCare.  Come October 1, “The doctor is in” may be a phrase of historical interest only.

Electronic medical records have already slowed many practitioners in the real world although you will hear from many in the Ivory Tower how great they are.   So, let’s hear it from someone commenting who actually lives in the real world and has to do the actual work.

ICD-10 is “granular.”  It lets researchers, insurance companies, the feds, and anyone else who gets their eyes on the data do great things.  This data is valuable.  To get this data, I, a solo practitioner, must buy software, recode the diagnosis for every patient, code it again the next visit as the specificity has changed, and so on.

And what am I paid for providing this valuable data?  Why, nothing at all!  In fact, if the computer at the other end doesn’t like my code, I don’t get paid for my work.  And the payments here in primary care land are, shall we say, inadequate for the work we do, let alone for the data processing services we are required to provide for free in order to get our pay.


Gosh, he’s polite.  Here’s another reality check comment.

Sorry folks, but I’m one of hundreds of over 65 Family Docs in private practice who are planning to close our practices 9/30/14 if ICD10 isn’t delayed. It’s going to cost tens of thousands to convert and there will be zero return on that investment. Most of us are living hand to mouth paying our help now and the advice to “have six months revenue in reserve” or “open a line of credit” isn’t realistic. It makes perfect sense to use ICD10 at the hospital level and a much more limited set at the office. If you think there is any clinical benefit in letting an insurance carrier know if a kid has a left or right ear infection or if you hit your head walking into a lamppost you have major problems.

More mocking…my favorite

T71.161D: Asphyxiation due to hanging accidental, subsequent encounter.

This example turned up in the comments on the blog post I cited about the common misunderstanding of the meaning of “subsequent” in ICD-10. As a reader pointed out, in this case the formula has clearly been carried too far: Since by definition asphyxiation is fatal, “one would not have a subsequent encounter.”

How does this affect the patient?   This is supposed to be about the patient afterall.  The incentives to switch to EHR have now become much more difficult to navigate.  Time consumption is a big deal. Bottom line:  More time is spent at the computer than taking care of people.   The cost is high and excruciating in many ways.   Training for the these “improved” protocols is expensive.  Think staff training, new equipment, blah blah blah.

I say take a deep breath every morning!  Nom sent this one over because we don’t want Mcnorman to go bonkers.  Thanks for letting me rant.  BTW, ICD-10: Painful To Implement, Painful To Delay has more great mocking pics.

Great reading:  War on doctors.