Wednesday, May 29, 2013

ED MD wants residency hours capped at 40 per week

Blogging at his site "Adventures in Emergency Medicine," Dr. Sam Ko says resident work hours should be limited to 40 per week. Via Twitter, I warned him that I would rebut his assertion.

Without any data or references except a tangential one, he bases his opinion on four premises.

1. Residents will be happier and nicer to patients because they will be less stressed. There is no proof that this is so. In fact, a recent paper in JAMA Surgery says about one-third of interns who work a maximum of 16 hours per day "demonstrated weekly symptoms of emotional exhaustion (28%) or depersonalization (28%) or reported that their personal-professional balance was either “very poor” or “not great” (32%)." And "at the end of their intern year, 44% [of interns] said they did not believe that the work hours limits led to reduced fatigue." This is not a very resounding confirmation of the theory that reduced work hours leads to happier or better rested residents.

2. "But we did it so you have to do it to." Under this heading, Dr. Ko says, "We are busier than they were 20-30 years ago. Before they probably got more sleep and had less patients in the hospital."

With the exceptions of more paperwork and the burden of the electronic medical record, I'm not so sure residents are busier today, but if they are, what's making them busier is REDUCED WORK HOURS. This recent paper from JAMA Internal Medicine concluded the following: "Compared with a 2003-compliant model, two 2011 duty hour regulation–compliant models were associated with increased sleep duration during the on-call period and with deteriorations in educational opportunities, continuity of patient care, and perceived quality of care." [Emphasis in bold added]

The supposition that there were fewer patients in the hospital 30 years ago is incorrect. When I was a resident over 30 years ago, cholecystectomy patients stayed in the hospital for 4 to 6 days. Even herniorrhaphies stayed 1 or 2 nights. Day surgery was in its infancy. Patients could be admitted for workups which are now done as outpatients. These people all needed H&Ps, had to be rounded on daily and notes had to be written. We had to draw routine and stat bloodwork and start IVs ourselves, we often transported patients to radiology and the OR. I could go on.

Dr. Ko is right about one thing. We did get more sleep when we were on call because we weren't cross-covering many patients that we didn't know very well. The abomination known as "night float" did not exist.

3. Residents won't get enough training. Dr. Ko dismisses this objection by pointing out that menial tasks should be delegated to others. But who are those others, and how will they be funded? In addition to the bolded portion of the sentence at the end of the paragraph above, here's another paper (of many such papers) documenting that many residents are already being poorly trained. And Dr. Ko wants to cut hours by half.

4. Less depression, anxiety and alcohol/drug abuse. He cites a statistic that 300-400 physicians commit suicide very year. That may be true, but there is no proof that decreasing work hours will alleviate that problem. Most papers on the subject seem to indicate that suicide is a problem of physicians who have completed training and are in practice. Did I mention that there are no work hours limits for doctors who are in practice?

Being a doctor is a stressful job. Sleep, or its lack, is not the only factor causing stress. Limiting resident training to 40 hours per week would be a catastrophe for residents, their education and most of all, their patients.

Tuesday, May 28, 2013

ICD-10 codes and politics

Senator Rand Paul has been getting some ink about a recent speech during which he mocked the Affordable Care Act for mandating the use of ICD-10 codes, some of which are pretty silly.

Theses codes are what hospitals and doctors must use when submitting bills to third-party payers.

Please understand that no one has gotten more mileage from making fun of the new ICD-10 codes than I have.

You may recall my posts on the codes for drowning due to falling from burning water skis, contact with (amorous) dolphins and getting sucked into a jet engine.

And I have to agree that the expansion in the number of codes from 18,000 in ICD-9 to over 150,000 in ICD-10 may be burdensome to most doctors.

Senator Paul is a physician, and he should know better. He is either clueless or disingenuous for blaming the ICD-10 code muddle on Obama.

As reported way back in January of 2009 by the Wall Street Journal no less, the ICD-10 codes were to be implemented by the Centers for Medicare and Medicaid Services, but CMS decided to delay doing so after protests by all sorts of medical people and organizations who said that they did not have enough time to comply.

In addition, these codes were not even developed by CMS. ICD stands for the International Statistical Classification of Diseases and Related Health Problems and the revised codes were formulated after many years of discussion by the World Health Organization (WHO).

So whether you love the new codes or hate them, they were going to be put into place regardless of the status of the ACA.

And regarding making fun of them, Mr. Paul is late to the party. My posts about the absurdity of some of the codes were on line in the fall of 2011.

Wednesday, May 22, 2013

Is it really best to take out a gallbladder in the daytime?

Under the headline "Best to take out gallbladder in daytime," MedPage Today reports on a study that says people who have laparoscopic cholecystectomies at night have more complications.

The work was presented at Digestive Disease Week in Orlando.

Ordinarily, I would not critique a paper that I had not read completely but I have to make an exception in this case.

There are some serious issues with both the research and the reporting. If the MedPage article is not read carefully, patients may receive inappropriate or delayed care.

According to the article, the paper comprised 549 patients who were mostly female (84%) with 65% having surgery in the daytime (defined as 7 a.m. to 7 p.m.), and 62% had surgery that was not elective—that is, urgent or emergent.

Those operated on at night had a longer median hospital length of stay, 3 days vs. 1 day and were more likely to have had non-elective surgery, p < 0.001 for both.

The article also says the nighttime patients "were more likely to have a discharge diagnosis." I'm only guessing, but I think they may have meant to say "a discharge diagnosis of acute cholecystitis."

"Bile leaks, bile duct injuries, retained stones, pneumonia, and readmission occurred at rates that did not differ significantly," says the report. The only complication that differed significantly was that of superficial wound infection, which occurred in 5% of the night and 2% of the day patients, p = 0.04.

Multivariate analysis showed that nighttime surgery increased the odds of complications by just over 3 times but with a wide confidence interval of 1.01-10.7 and a barely significant p value of 0.05.

So, what's the problem?

At the very end of the nearly 500-word article, we find that elective patients were excluded from the multivariate analysis with no explanation why. It could be that when the elective patients were included, there was no difference in outcomes.

The first part of the last sentence is even more revealing: "The authors also did not have data on postoperative length of stay and severity of gallbladder disease."

Perhaps some of the length of stay of 3 days for the nighttime patients was due to waiting for an available operating room, workup for possible common duct stones or stabilization of lab values.

But in my opinion, the factor that makes the entire study invalid is not knowing the severity of the gallbladder disease. A patient with a more severely inflamed gallbladder is obviously more likely to have a complication.

There is also no mention of co-morbidities like diabetes or heart disease which may have been more prevalent in the nighttime group.

I don't understand how this study ever saw the light of day, why it was selected as a featured paper by MedPage or why the misleading headline was used.

If you are a patient with a sick gallbladder, many recent studies have shown that you should have it removed as soon as possible—less time in the hospital, less cost better outcomes.

If your surgeon can do it at 8 p.m., please go ahead with the surgery. Don't wait until the next day.

Tuesday, May 21, 2013

Law schools start firms in order to employ their graduates

The New York Times reports that law schools in the US, smarting from the collapse of the job market for lawyers, are establishing law firms so they can hire their graduates and give them something to do.

Regarding such a scheme in Arizona, the article says, "Over the next few years, 30 graduates will work under seasoned lawyers and be paid for a wide range of services provided at relatively low cost. The school-based firms will be something like teaching hospitals for law school graduates."

Several schools have bought in to the idea to solve two "seemingly contradictory problems: heavily indebted law graduates with no clients and a vast number of Americans unable to afford a lawyer."

For a minute there, I thought the legal profession was actually going to break precedent and do some real charity work. [Please don't tell me lawyers do pro bono work. Very few do, and when they do, it is a paltry amount of time. A judge had the audacity to recommend that new lawyers do 50 hours of pro bono work before being allowed to take the bar exam in New York. That's about 6 days of work per year. A law blog called it "indentured servitude."]

But the article mentioned that the legal services provided will not be free. in fact, the plan for the Arizona project "is to charge $125 an hour in an area where the going hourly rate is $250." That's not exactly analogous to the way a teaching hospital works.

By what criterion is $125 per hour "relatively low cost"? Oh, I think I've got it. That's relatively low cost for legal fees. Of course, lawyers know how make up for low hourly fees. According to another New York Times piece, it called padding the bill or "churning." One expert commented that "churning, while not endemic, is an insidious problem in the legal profession."

A lawyer for a firm that is being sued for overbilling of hours allegedly worked said in an email to a colleague, “Now Vince has random people working full time on random research projects in standard ‘churn that bill, baby!’ mode,” Mr. Thomson wrote. “That bill shall know no limits.”

So the idea of law schools setting up graduates in business reminds me of a joke.

Back in the late 19th century, a lawyer moved to one of the many new towns springing up in the West. He was the first and only lawyer. For a year, he had nothing to do and nearly starved to death.

Then another lawyer came to town.

Monday, May 20, 2013

Two new posts are on line

On Physician's Weekly, I discuss, but don't really answer because I don't know for sure, the question "Is the impending physician shortage real?" To read it, click here.

On General Surgery News, I explore the impact of reduced work hours on residents. This post can be found here.

Saturday, May 18, 2013

Who Should Determine “Medical Necessity”?

The following is a guest post.

These days, there is a lot of talk about expanding scopes of practice for the group of folks who used to be called “physician extenders” and then “midlevel providers” and more recently “non-physician providers,” many of whom are now getting degrees with the title “doctor” incorporated.  While it seems to vary, these folks may include nurses, physician assistants (one day to be called “physician associates” perhaps), pharmacists, and more.  Lots of forums are discussing whether folks who are not doctors should be calling themselves “doctor” or whether they should be expanding their scope of practice via legislative rather than educational means. But a recent personal anecdote has made me wonder about a slightly different question: should non-physicians be able to certify “medical necessity”?

Two family members are covered under the same insurance policy.

He has had a history of knee injury, related surgery, and subsequent successful rehabilitation. He recently ran a marathon, and now has new knee pain that has persisted after 8 weeks of conservative therapy at home. After a long wait for an appointment slot, he finally sees an orthopedic surgeon with knee expertise. The doctor recommends an MRI to evaluate the nature of his ongoing pain. 

She has chronic back pain that seems to be acting up, and sees a chiropractor mostly out of convenience, since the office is in the strip-mall near her home and he can see her anytime as a walk-in. After just two sessions with the chiropractor, he suggests an MRI, since she isn’t responding as well as he had expected to the adjustments.

The insurance company immediately approves the chiropractor’s MRI, but denies the orthopedic surgeon’s.  The request was appealed, and again denied, on the grounds that it was not medically necessary. The insurance company issued a requirement that the physician first document the patient’s participation in a physical therapist’s prescribed self-care routine for at least 6 more weeks (recall that the patient has essentially already done this for 8 weeks, having had formal PT for the knee in the past, and familiar with the appropriate self-care, but this did not satisfy the payer). Only after jumping through this hoop may the doctor meet the standard of “medical necessity” to obtain the MRI.

Lest I be misunderstood, I do not object to the idea that the chiropractor can order imaging tests. A chiropractor is an autonomous professional after all, and is a doctor of chiropractic, just as a dentist is a doctor of dental surgery, and an optometrist is a doctor of optometry (but that doesn’t qualify him to do eye surgery– a debate for another time.).

In the modern landscape of myriad healthcare providers and payers, if a non-physician can certify “medical necessity,” surely a specialist physician should be able to as well.

Marjorie Stiegler is an anesthesiologist. She blogs about patient safety and decision-making errors at

Wednesday, May 15, 2013

Read the whole paper not just the abstract

Here is another installment in my series of posts about why you should read the entire paper and not just the abstract. (See others here, here and here.)

A paper in the February 2013 issue of the Journal of the American College of Surgeons describes 15 cases of median arcuate ligament syndrome treated with laparoscopic surgery.

Median arcuate ligament syndrome (MALS) is somewhat controversial. It is said to be due to impingement of the median arcuate ligament (a portion of the diaphragmatic crura) on the celiac artery causing a narrowing and decreased perfusion of the stomach. Symptoms are abdominal pain after eating, nausea and weight loss. It is often diagnosed in patients who have been worked up for many other suspected problems without finding anything.

The paper notes that 10% to 60% of people without symptoms have narrowing of the celiac artery.

The abstract reports resolution of the pain for 14 of the 15 patients who had the surgery as well as a significant mean decrease in celiac velocity indicating resolution of the narrowed area postoperatively.

It also mentions that one patient required conversion to open surgery but doesn't say why.

On reading the whole paper, one learns that the conversion to open occurred in the only case that was done with robotic assistance.

The authors state that the 2 mm injury to the aorta was the result of the robotic instrument being too large and "the absence of haptic feedback," which is robot-speak for "you can't feel anything."

That is one drawback of the robot. With robotic instruments the sense of touch is simply not present. Although the fingertips used in old-fashioned open surgery are much more sensitive than instruments used in standard laparoscopic surgery, those instruments do enable the surgeon to at least feel some variations in tissues

The aortic tear led to two liters of blood loss and an operative time of just under 8 hours.

The abstract says all but one patient had complete resolution of pain, but the paper says the amount of decrease in the Doppler celiac velocity "did not correspond to the degree of symptom resolution."

And you can see that the differences in preop (red) and postop (green) velocities are pretty modest in 7 of the 10 patients who had them measured even though the mean difference was significant at a p of 0.005. In addition, the postop values all hover around 200 cm/sec, which, in the presence of symptoms, was the threshold for doing the operation.

In fairness, of the 13 patients who were interviewed, all said they were satisfied with the outcome of the surgery and would go through it again.

In some ways, MALS reminds me of internal mammary artery ligation, which was once touted as a cure for angina pectoris (chest pain of cardiac origin). Over 50 years ago, randomized trials which included a sham operation—incisions were made, but the arteries were not ligated—showed that ligating the arteries was no better than the sham operation for relieving pain.

It might be time for such a trial in MALS, only let's skip the robot for this one.

Thanks to Dr. Michael Burchett for alerting me to the MALS paper.