These things do happen--two broken systems.

Local Doctor Hit with Nearly $2 Million Judgment in Lawsuit

[ILLUSTRATION OMITTED]

Mr. A.G., 72, was awarded $1,900,000 yesterday in Superior Court, for a year's delay in diagnosis of cancer of the colon. Mr. L. Eagle, G.'s attorney, stated that G. had originally had an incomplete colonoscopy and never had a follow-up barium enema to exclude cancer.

"Everybody knows that you have to have either a complete colonoscopy or barium enema to exclude the disease." Eagle said. 'The doctor simply forgot."

Doctor S.S. admitted in trial testimony that he forgot to order the test and felt very badly about what happened. He said such a mistake is rare and that he has been very careful since then."

What do you think?

1. Damn lawyers.

2. Patient got too much money. He was 72 and might have had metastatic disease anyway.

3. The doctor blew it. How can you forget to do a BE after incomplete colonscopy?

4. Why didn't the patient, family or staff remind the doctor?

5. Has anybody changed the system?

Preventing errors is not a matter of "being careful" as much as it is having systems to catch errors early when they may be reversible. Dr. S.S. was unfortunately "the final garnish in the lethal brew that [had] been cooking for some time." (1)

One newspaper was quick to blame the gastroenterologist, and another cited the case as an example of the Board of Medical Examiners not disciplining physicians properly. Medical errors, transportation accidents, power failures, lost baggage, dropped passes--if something bad happens, find and discipline the person responsible. It feels good, unless you are being blamed or you want something to change.

Asking questions

It's important to ask why the mistake happened, and keep asking why!

1. Why did this occur? Because the barium enema was not ordered. TRUE.

2. Why was the BE not ordered? Because the physician forgot. TRUE.

3. Why did the physician forget? Because people forget! Because that day he had to cover his partners, he had been up most of the night on call and he had some problems at home (Hurry, fatigue, pre-occupied.) TRUE.

4. Why was it up to the physician to remember? Because good physicians don't forget. FALSE. Our memories are fallible and limited. We can keep only so many items in our working memory at one time. To those who disagree, what if you were fatigued, in a hurry, pre-occupied with your dented car or your child who is failing algebra? Don't you think you could forget?

5. Why wasn't there a good system to schedule barium enemas after incomplete colonoscopy? Because physicians will remember they need to do it. FALSE. (What is the evidence that they will? Has anybody proven it?)

Every GI lab needs a system where every incomplete colonoscopy automatically triggers an order for a barium enema. Physicians, patients and family members usually will remember to follow through, but why depend on memory? Further, the file would not be closed until both the physician and the patient knew the result.

Work processes comprise systems and the system design may reduce or enhance errors. A good system is efficient, capable, makes it difficult to err and is easy to understand. Additionally, good systems: (2)

1. Assume that people are imperfect and are designed with the expectation that people will make errors rather than assuming flawless performance.

2. Explore the power of constraints to make commission of an error impossible or very difficult. A well-known constraint in cars with automatic transmissions is the inability to shift into reverse without having the foot on the brake. Computer entry systems that do not allow an order to be carried out in the face of a known allergy are another example. We need more of these in medicine.

3. Have few steps and limited reliance on memory.

4. Use standardized vocabulary, labeling and approaches.

5. Engineer human failings out of the system rather than depending upon more rules, threats, or sanctions.

Where's the gallbladder?

Here's a system breakdown that was more complex.

A 36-year-old woman presents to the emergency department one evening with right-sided abdominal pain. She has had past surgery but doesn't know what was done. At 9 p.m., an ultrasound examination read by the radiologist (from home) shows sludge in the gallbladder and stones. At 1 a.m., the patient's abdominal CT read by the same radiologist shows a ureteral stone--and status post cholecystecomy. The patient is admitted.

A full day later, the patient is taken to the OR for a lap chole. It's a quick case. In retrospect, what was identified on the ultrasound as a gallbladder with stones and sludge was likely fat in the porta hepatis with shadowing from the surgical clips.

This did happen--at a large hospital in the suburb of a major city.

What do you think? "I would never make that mistake. How could anybody be so stupid?" Have you never done anything that you--or others--determined was really dumb?

[ILLUSTRATION OMITTED]

Not checking the blood type doomed a transplant patient at Duke. Several years earlier, it happened at Arizona. This is basic stuff, rookie mistakes. Yet, we all make them.

Aviation, maritime and nuclear power disasters have occurred that, in retrospect, were remarkable in how exactly the wrong thing was done at the wrong time. Here is how the gallbladder incident happened and where the system could be improved:

* The patient ought to know what surgery she had, but we can't always depend on that. Still, a process for trying to ensure people carry the significant portions of their medical records with them would be helpful.

* The ultrasound form had boxes to check for common surgeries, such as cholecystecomy. Unfortunately, there was no box saying "None of the above," so it wasn't clear whether the patient had been asked about surgery, which a "None of the above" would have clarified. The technologist's "gallbladder" comment on the images was a hint that the study was equivocal, but there were no qualifiers in the report.

* The radiologist was not informed that the CT scan was a follow-up on the prior ultrasound. At night, especially reading from home, such information should be available, improving the quality of both readings as well as preventing embarrassment. Picture Archive Communications System (PACS) may alert radiologists to this issue, but technology has to be used to work.

* With contradictory results, evaluating the patient's abdomen for prior scars (a lap chole puncture scar is small but visible) would have been helpful. Simple, but less high-tech studies, such as plain films, have their use. Had they been used here, they would have showed surgical clips.

* Different imaging studies performed on the same patient the same day should be hung together, so that the interpreter has additional information.

* The actual imaging studies, not just the report, need to be viewed by the surgeon. If the surgeon had suspected there was a previous cholecystecomy, and the ultrasound was showing a remnant, the CT images would have dissuaded him of that suspicion.

* Contradictory results should be a red flag, prompting complete clarification well before surgery.

It is worth playing "Jeopardy" with this case, asking "Why?" questions starting with the final bullet, "Why did surgery occur?" Because contradictory results were not a red flag.

Unfortunately, this case was not discussed at QA, so the chance for several groups of people--ED physicians, radiologists, surgeons, technologists, and OR staff--to improve a clearly flawed system was lost. Additionally, not having an open investigation lost the chance to look at the influence of human factors, such as fatigue, interruptions, or pre-occupation with other issues.

Human factors, or how people react to each other and their environment, are inextricably linked to flawed processes.

References:

1. Reason J. Human Error. Cambridge University Press, 1990.

2. Bogner MS, Human Error in Medicine. Laurence Erlbaum Associates, Hillsdale, NJ: 1994.

By Michael S. Smith, MD, MS

Michael S. Smith, MD, MS, a statistician, wants to help people in the medical community use statistics to make better, faster and easier decisions. He is self-employed and may be reached at 520-410-7917 or [e-mail].