Wednesday, July 19, 2017

What were attrition rates in surgical residency programs 25 years ago?

Last month I blogged about the 20% attrition rate of general surgery resident over the last 25 years, and a recent paper presented at a national meeting that found after following the general surgery resident class of 2007, 20% had dropped out for one reason or another.

A reader who calls himself Artiger commented on that piece asking, “Is there any data on resident attrition prior to 1992? Just curious if this has been a problem for more than the past 25 years.”

I responded that I wasn’t aware of any such studies but I would try to find out.

Most of the few papers written about attrition back in the day focused on one residency program or one medical school’s graduates.

Until the middle of the 1990s, many surgical residency programs were pyramidal—that is, they took more categorical first-year residents than they had chief residency positions. For example, when I began my training in 1971, my program had 12 first-year residents, decreasing to 8 in the second-year and only 4 chiefs.

Sunday, July 9, 2017

Parathyroids Anonymous

A One Act Play by Leo Gordon, MD

Dedicated to Parathyroid Surgeons
All proceeds from the production of this play go to Parathyroids Anonymous--An international organization dedicated to the well-being of those who perform parathyroid surgery

Scene: A sparsely furnished church basement. Rain is beating against the window panes. Folding chairs are arranged in a semi-circle. Participants are drinking from plastic cups. Some are in scrub suits. All appear tired.

Don: Hello. My name is Don and I’m a parathyroid surgeon.

All: Hello Don

Don: I will be your facilitator tonight. We have a new member so let me set the ground rules. All of us in this room are parathyroid surgeons. We maintain our anonymity as we discuss the mental and physical distress that parathyroid surgery engenders. There are no boundaries at our meetings. We speak openly and freely. Use your first name only. And of course, no patient names. Who wants to begin?

Miriam: (Nervously) Hello. My name is Miriam and I am a parathyroid surgeon.

All: Hello Miriam

Miriam: I’ve been here a few times but I’m a little bit nervous.

Don: Don’t be nervous Miriam. We all share the same problems.

Miriam: Well… last week. (Begins sobbing)

Don: Now Miriam, just relax. Please continue.

Miriam: I had a 56 year old woman referred to me with hypercalcemia. She had had elegant localizing studies at our hospital. Both studies – nuclear and sonographic - stated with metaphysical certainty that there was an adenoma of the right lower gland. Yet (sobbing) when we explored the area, there was no adenoma present. All we found was a normal sized parathyroid gland!

Al: (Hands Miriam a handkerchief)

Don: Go on, Miriam

Friday, July 7, 2017

The problem of “copy and paste” in electronic records

As opposed to text that is copied and pasted or imported from another part of the electronic record, the average amount of manually entered information in a progress note is

a. 18%
b. 29%
c. 43%
d. 55%
e. 70 %

A study of 23,630 internal medicine progress notes written by 460 different hospitalists, residents, and medical students found that a mean of only 18% of the text was created by hand with 46% copied and pasted from previous note or somewhere else and 36% imported from another part of the record such as a medication list.

The analysis, done at the University of California San Francisco*, was possible because the Epic electronic medical record used there can provide the provenance of every character entered in a progress note.

Medical students had the highest percentage of manually entered text and wrote longest notes—averaging 7053 characters, but even the shortest notes, by hospitalists, averaged 5006 characters. For reference, this post contains 1189 characters.

Manual entry comprised 11.8% of resident notes with 51.4% of the remaining information copied and pasted and 36.8% imported.

Think about it. For all groups, less than one-fifth of every progress note they wrote was original material. For resident notes, it was closer to 10%.

The authors cautioned that their study was limited to a single service at a single institution, but I suspect the results would be fairly similar in many if not most hospitals.

*Location of the study corrected on 7/7/17.

Tuesday, June 27, 2017

How to fix the problem of general surgery resident attrition

Over the last 25 years, about 20% of general surgery residents have failed to complete their five years of training. This compares unfavorably to other specialties such as orthopedics, obstetrics-gynecology, and medicine with attrition rates of < 1%, 4.5%, and 5%, respectively.

A paper presented at the American Surgical Association in April looked at the factors associated with attrition in one year’s resident class. In 2007, 1047 residents began their training and after 8 years of follow-up, 80% had become surgeons. How many non-finishers left programs by their own choice is not clear.

Some highlights of the research are as follows:

24% of women and 17% of men left general surgery training.

Size mattered because 23% of men and 25% of women left large programs compared to both sexes leaving smaller programs at a rate of just 11%.

Tuesday, June 20, 2017

Some general surgery residency graduates may not be competent to operate

A new study says 84% of general surgery residents in their last six months of training were rated as competent to perform the five most common general surgery core procedures—appendectomy, cholecystectomy, ventral hernia repair, groin hernia repair, and partial colectomy. However the percentage of those judged competent varied from a high of 96% for appendectomy to a low of 71% for partial colectomy.

When analyzing the other 127 core operations of general surgery, the investigators found that 26% of residents in their last six months of training were felt to not be competent to perform at least some of those procedures.

The study was presented at the annual meeting of the American Surgical Association in April 2017 and reported in ACS Surgery News.

Data were compiled from ratings of 522 residents by 437 faculty yielding 8526 different observations.

For all of the procedures rated, maximum resident autonomy was observed for 33% of cases, and the more complex the case, the less ready the residents were to perform it on their own.

The lead author of the study, Dr. Brian George of the University of Michigan, was asked whether the duration of surgery training should be increased. He answered, “The 20,000 hours of surgical residency should be enough to train a general surgeon to competence—it's up to us to figure out how.”

Thursday, June 15, 2017

Surgical residents have lots of problems, need more time off

A recent survey of surgical residents regarding their personal and professional well-being revealed that while most of them enjoyed going to work, they had many serious issues.

All 19 surgical residency programs in the New England region were invited to participate, and 10 did so. Of 363 trainees contacted, 166 (44.9%) responded to the survey with 54% of respondents saying they lacked time for basic health maintenance. For example, 56% did not have a primary care physician and were "not up to date with routine age-appropriate health maintenance such as a general physical examination, laboratory work, or a gynecologic examination."

I am not surprised that young men and women averaging 30 years of age or less have no primary care physician? I wonder what percentage of young people who are not surgical residents have one.

Should asymptomatic people in this age group or anyone in any age group have a general physical examination and lab work?

Thursday, June 8, 2017

More on artificial intelligence in medicine and surgery

Part 1

A survey published in the journal arXiv predicted with a 50% probability that high-level machine intelligence would equal human performance as a surgeon in approximately 35 years. See graph below. 
Click on the figure to enlarge it
We have already seen a machine beat the world’s best Go player. Although Go is a complicated game, it lends itself to mathematical analysis unlike what one might experience when doing a pancreatic resection.

A potential flaw in this study is that the surveyed individuals were all artificial intelligence researchers who predicted that machines would not be their equal for over 85 more years with the 75% likelihood of this occurring being over 200 years from now.

I suspect if surgeons were asked the same questions, we would say it would take over 85 years for machines to be able to operate as well as we can and 35 years until artificial intelligence researchers would be replaced by their creations.

[Thanks to @EricTopol for tweeting a link to the arXiv paper.]

Part 2

Similar to the question “who is responsible if a driverless car causes an accident?” is “when artificial intelligence botches your medical diagnosis, who’s to blame?” An article on Quartz discussed the topic.

[Digression: The article matter-of-factly states “Medical error is currently the third leading cause of death in the US… ” This is untrue. See this post of mine and this one from the rapid response pages of the BMJ.]

If artificial intelligence was simply being used as a tool by human physician, the doctor would be on the hook. However indications are that artificial intelligence may be more accurate than humans in diagnosing diseases and soon may be able to function independently.

If a machine makes a diagnostic error, are the designers of the software responsible? Is it the company that made the device? What about the entity owns the system? No one knows.

The Quartz piece did not address this. Who is responsible if a nonhuman surgeon makes a mistake during an operation?

I’m sorry I won’t be around 35 years to hear how this is settled.