Comparisons, Conjecture, and Confusion Over Data Used by Testifying Experts

Sheet of Numbers

This has been an exciting year for data both big and small.  We’ve seen frequent news stories on data released by the government, secretly collected by the government, and stolen from private businesses.  On our blog this year, we’ve written about healthcare claims data from Medicare, highlighted reports of bicycle collision data in DC and data about judges calendars in Virginia, and discussed some of the ways in which collected data is changing litigation.  I think we might soon need a data set to support all statements we make that are based on experience, probability, or the “average” circumstances.

In Perkins v. Hansen, the DC Court of Appeals decided a closely related question: is an expert witness required to use relevant data?  The basic facts of the case are that in May of 2007, Margie Perkins needed a liver transplant.  Although Ms. Perkins was showing signs of needing a liver in early May, it wasn’t until late May that someone did something about it.  Unfortunately, Ms. Perkins passed away, still in need of a liver transplant on June 6, 2007.  So Ms. Perkins’ family hired an expert to say that it was more likely than not that if Ms. Perkins’ condition was realized in early May, instead of late May, she would have received a liver transplant prior to her death in early June.  The defense objected to this opinion by arguing that the expert had not reviewed the transplant list data for May 2007.  The trial judge agreed with the defense: no data, no expert.

In reversing the decision of the trial court, Perkins says that an expert’s failure to rely on relevant data goes to the weight of the expert’s opinion, not its admissibility.  The overall experience that an expert brings to the courtroom is sufficient to provide a reliable basis for an opinion.  The expert hired to testify for Ms. Perkins may be cross-examined on his failure to use the transplant list data, but it is a “red herring in assessing the admissibility of [his] testimony.”  So the Court of Appeals reversed and remanded for a new trial.

One reason that Perkins v. Hansen caught my eye is that the decision seems to arrive at the opposite conclusion as the Virginia Supreme Court did in Funkhouser v. Ford Motor Company that I posted about earlier this year.  In Funkhouser, the court ruled that when an expert’s opinion involves similar accidents or occurrences, those other accidents or occurrences must be substantially similar to the one at issue in the case.  At first, it may appear that this holding covers different ground than Perkins (indeed, some bloggers have referred to Funkhouser as only applying in products liability failure to warn cases, which I think is too narrow a reading of the opinion).  But in Funkhouser, the opinion states that it is the court’s “‘responsibility, when proper objection is made, to determine whether the factors required to be included in formulating the opinion were actually utilized. . . . If all the factors are not utilized, the court should exclude the opinion evidence.’”  Therefore, if the Perkins case had come up for review in the Virginia Supreme Court, it seems likely to me that the outcome would have been different.  Under the substantial similarity test, I would argue that the expert’s opinion that a liver would have become available in a certain time period is an opinion that requires the transplant list data from that time period (or a similar one).  And the penalty in Virginia would be that the expert’s opinion is inadmissible, not merely “valuable fodder for cross-examination.”

Another interesting wrinkle is whether the data must be actually available, or just plausibly in existence.  Footnote 11 of Perkins highlights that the parties only speculate that the liver transplant data exists at all.  It’s not as if the defense attorney held up a printout of the available livers during May of 2007.  In fact, neither side actually knew whether such data is collected and available.  The reference to it came up in the deposition of the expert, where the expert stated that UNOS (the non-profit that manages organ transplants in the US) “ha[s] statistics on everything.”  This was obviously an off-the-cuff remark, since that expert doctor hadn’t actually reviewed the statistics himself, he was merely speculating that such data existed and was in some way retrievable.  So does a positive response to a deposition question about data require the expert to go find that data and use it in Virginia?  I think a strong argument can be made that the answer is yes.  But in DC, Perkins says the answer is no.

The proliferation of technology that is able to gather, record, and retrieve information has made collecting and owning data an end in itself.  While the legal system is slow to adopt change, it is going to be forced to make decisions like Perkins and Funkhouser to deal with issues of how data is used in litigation.  An important question here is “who decides if the data is used?”  I suspect the judicial system will go through some growing pains before reaching a long term decision on this matter.

See our disclaimer.