Chemnick | Moen | Greenstreet

Medical Malpractice. It's All We Do. 206-443-8600

Hindsight Bias in Radiology "Miss" Cases: A Get-Out-Of-Jail-Free Card for Negligent Radiologists?

By Tyler Goldberg-Hoss

Worst Case Scenario in Jury Deliberations in a Radiologist “Miss” Case

Juror #1: “You know, now I can see the tumor on the film, but that’s because the plaintiff’s lawyer has that huge picture of it and has been pointing to it over and over through the trial. Heck, I’ve got a picture of it circled in red right here.”

Juror #2: “You’re right. At first I thought: ‘Well if I can see it, why didn’t the trained radiologist see it?’ But the more I think about it, the more I think I wouldn’t have been able to see it without it being pointed out.”

Juror #3: “I for one tend to agree with the radiologist experts who are saying you can’t ever put yourself back in the position of the defendant at the time he was looking at these pictures. How are we supposed to fault a radiologist if it’s possible to have bias looking at these pictures?”

Juror #1: “You know, it’s not like he missed a scalpel or something obvious.”

In radiology “miss” cases, this is one of many possible nightmares a plaintiff attorney can have. This discussion, or one like it, might occur during deliberations in the actual case or in a preliminary focus group. It centers on the idea that hindsight bias affects all of us – the attorneys, the experts, the jury, even the judge – in a case where a radiologist is alleged to have missed a finding on a film.

Hindsight Bias Defined

One scholarly paper defines hindsight bias generally as “the after-the-fact feeling that some outcome was very likely to happen, or was predictable, even though it was not predicted to happen beforehand.”1 More specifically, visual hindsight bias can be defined as having after-acquired information that changes the way you look at something, often allowing you to “see” something that previously could not be seen.

There are countless examples of hindsight bias in day-to-day life. For sports enthusiasts, it is the Monday Morning Quarterback effect, where we pick apart the decisions made by players and coaches from the previous weekend’s football games.

For political junkies, it is waking up on November 5 and saying, “I just knew Inslee was going to win.”

For radiologists in medical negligence actions, it is the concept that it is only now, after something bad has happened, that we can look back and “see,” with the benefit of hindsight, something abnormal on a film. And it is often their strongest, if not only, defense.

Why Hindsight Bias is a Problem

Hindsight bias is a problem because it can be used effectively to defeat a claim of radiologist negligence in a number of ways.

First, be prepared for every other word out of the defendant’s mouth, when discussing the films at issue, to be “hindsight,” “bias,” or “retrospective.”

In a recent case I had, it was clear the radiologist had been conditioned to use these words to talk about the image at issue. Example: “I think in retrospect there is a very subtle finding, but that’s with a lot of retrospective bias.”

Second, the defendant’s experts will also couch their opinions in such terms and will rely on the concept for why they themselves would have missed the diagnosis. The defense radiologist will say: “I comply with the standard of care, and I would have missed it.”2 Implicit in this is the idea that the jury is now going to have to not only find the defendant negligent, but also all of the defendant’s radiology experts as well.

Another advantage for the defendant if it is not an obvious “scalpel-left-in” case is that defense counsel will be able to retain local radiology experts. Plaintiff’s counsel will likely have to resort to finding out-of-community or out-of-state radiologists to point the finger at the defendant.

Third, every time the plaintiff emphasizes the miss – with large blown up pictures, with their experts circling the abnormality, with computer illustrations – is another way in which we are“biasing” the jury. This can lead to confusion among jurors, and allow them to give the defendant a pass, particularly if they are inclined to want to let a doctor off the hook.

Ways to Combat Hindsight Bias

First, move to exclude their “expert” on hindsight bias. In the last couple of radiology cases our firm has had, the defendant retained the services of Dr. Geoffrey Loftus, professor and Ph.D. at the University of Washington. Dr. Loftus has testified countless times for criminal defendants regarding the reliability of eyewitness identification and is now being used in radiologist “miss” cases for the purpose of explaining hindsight bias.

Within this realm of radiology cases, Dr. Loftus has sometimes been allowed to testify, and sometimes not. In the cases in which he has not been allowed to testify, the trial court has found that his opinions are: 1) within the general understanding of jurors; 2) any probative value of his testimony is outweighed by the prejudice to the plaintiff of over emphasizing defense themes; and 3) that what his testimony really is doing is offering an opinion about the credibility of plaintiff’s own radiology experts.

I deposed Dr. Loftus in a recent case. He seemed like a very nice man, and soon after deposing him I moved to strike his testimony for the above reasons. Defense counsel called me and told me they would not be calling him at trial in any event, and the impression I got was that he was not particularly helpful to their case. His problem seemed to be that he could not get out of professor mode and talk like a regular person.

But in the future, I will be even more wary of Dr. Loftus if he starts publishing research in this field. So far, he is a general expert on hindsight bias that he applies to radiology. I believe a court will be more likely to allow him to testify if he has research and data that is radiology-specific upon which he can rely.

Second, attempt to minimize the impact of the defense by showing that your experts took efforts to have less bias than theirs did. In our most recent case, both of the defense experts knew prior to reviewing the films that they were being contacted by the attorney for the defendant, and they went down to the attorney’s office to view the films at issue in the attorney’s conference room on the attorney’s laptop.

In contrast, with one of our experts, I attempted to minimize the bias. I contacted this radiologist using an anonymous email account. I explained that I was an attorney and I asked him if he would be willing to review three plain films and form impressions of them. He agreed. I sent him an envelope with two smaller envelopes and letter of instruction.

He was instructed to open the first envelope, in which was a CD with the three images on it. Then, after he reviewed the films and formed impressions of what he saw, he was instructed to open the second envelope, which included my contact information. In this way, hopefully, a jury would conclude that, of all the radiologists they heard at trial, he was the least biased.

Another way to eliminate bias could be to send an expert three sets of films, two of which are “normal” while the third is the set at issue in your case, although there would be confidentiality and privacy issues with respect to the people in the “normal” films. One way to do this is could be to use other films from the same patient that have nothing to do with the issues in your case (such as previous arm x-rays when the issue in your case is a tumor in the patient’s chest). Another way (I have not yet tried) could be to use a consulting radiologist to put together a set of normal studies for you to send to your testifying radiologist along with the films at issue, although this may raise the same confidentiality and/or privacy issues as above.

Finally, do not let the jury forget the defendant is an expert searcher. In preparing my case for trial, I contacted a plaintiff’s attorney in Portland, OR, who had tried a very similar case to mine last year. In his case, the defense attorney had discussed the “Where’s Waldo?” series of books with the jury in voirdire in an attempt to convey to them this notion of hindsight bias (when you’ve found Waldo the first time, it’s really easy to find him again and again).

In closing, the plaintiff attorney brought this back up. He told the jury that it had been a while since his kids were of an age that they were reading these books, so he went to the bookstore and picked one up. At first, he had difficulty finding Waldo in one of the pictures, but eventually he found him. Sure enough, when he turned back to that page, it was easier for him to spot Waldo again.

But then he realized how much better at spotting Waldo he would be if he went to school for years to learn how to spot Waldo, and had gone through a residency program and passed tests to become an expert Waldo spotter.

Of course, he was not able to go to “Where’s Waldo” school, but he thought maybe if he came up with a system for spotting Waldo, he would be better at it. So he held his arm across the top of the next page, and slowly lowered it while his eyes scanned just below his arm. With that search technique, he found Waldo every time.

This is a great example of how to refocus a jury back to the finding on the film and how, hindsight bias or no, the defendant was an expert searcher with years of training who should have seen what was there to be seen.


Hindsight bias exists, but it is not a “get-out-of-jail-free” card for radiologists either. Managing the defense attorney’s ability to utilize these arguments can be a real benefit to your client’s claim.

Recommended Reading

Dr. Loftus has written a few articles on hindsight bias, the most relevant of which is “The ‘Saw-It-All-Along’ Effect: Demonstrations of Visual Hindsight Bias” (2004).For a prolific author on radiologist malpractice, including hindsight bias, look no further than Dr. Leonard Berlin. Dr. Berlin has been writing on this topic for over 30 years. A couple of articles to start with include: “Outcome Bias” (American Journal of Radiology 2004); and “Radiologic Errors and Malpractice: A Blurry Distinction” (American Journal of Radiology 2007).


  1. Harley, E.M., Carlsen, K.A., & Loftus, G.R. (2004). The “Saw-it-all-along” effect: Demonstrations of visual hindsight bias. Journal of Experimental Psychology: Learning, Memory, & Cognition, 30, 960-968.
  2. One can certainly attempt to limit this type of testimony as confusing to the jury since the standard of care is not defined by the practice of one radiologist. See Ketchum v. Overlake Hosp. Med. Ctr., 60 Wn. App. 406, 804 P.2d 1283 (1991).

Tyler Goldberg-Hoss is a WSAJ EAGLE member and second vice president of the New Members Committee. He is an associate in the Seattle firm of Chemnick | Moen | Greenstreet, which limits its practice to medical negligence claims. This article was first published in Trial News, April 2013.

Chemnick | Moen | Greenstreet
115 NE 100th St #220, Seattle, WA 98125 US
Phone: 206-443-8600
Fax: 206-443-6904