John Blomster:聽Welcome to 红桃视频. I'm John Blomster. And today we're speaking with Dr. Geoffrey Baird, renowned pathologist and chair of Laboratory Medicine at UW Medicine in Seattle. Dr. Baird is a certified expert in anatomic pathology, clinical pathology and clinical chemistry. And he directs the clinical chemistry lab at Harborview Medical Center. And he is no stranger to the courtroom, as he has been called in regularly to share insights and expertise as part of some of the biggest cases over the years.
Moreover, he's a regular in 红桃视频 classrooms where he helps educate students on how to effectively analyze, interpret and wield forensic evidence on behalf of clients, which we know can make or break a case and we're speaking with him today for a primer on how to do just that. So, thanks for joining us today.
Geoffrey Baird:聽Thanks for having me.
闯叠:听In a very broad sense, what role do doctors in the medical lab play when it comes to being brought into a case that's coming to trial just in a very, very broad kind of picture.
骋叠:听So, there's two ways I think that doctors in my position get brought in. One is voluntarily and one is not voluntarily. So, if there is a mistake or something that has been done in the laboratory, or maybe not a mistake, maybe there's just a clarification, or something like a drunk driving case. So, I work at our county hospital, Harbor View, and many people who do drink and drive and then get into accidents would first come to Harbor View. And maybe their first sample that was taken of their blood would be used for clinical purposes and not for legal purposes. And so, we may get the first sort of I would say, the involuntary part of it would be we may be asked by under subpoena or something like that to testify or to provide documentation about what happened in the clinical lab. We do not use all of the legal methods for collecting specimens or testing them, because we're not producing evidence, we're trying to take care of human beings. And so, you wouldn't necessarily have chain of custody or do other types of tests, which aren't clinically useful, maybe legally useful, but not clinically useful. So, that would be the first time when we would be asked to testify about something that happened.
The second type of engagement that I've had has been through expert witnessing. So, when someone would call and ask, would we be able to review outside documentation, something that wasn't done internally, I do actually get a lot of requests to be an expert witness for things like drunk driving or drugged driving cases. But unfortunately, those cases oftentimes involve my own laboratory at Harbor View, because in King County, many of those cases they do end up in my hospital and then I'm unable to be an unbiased third-party expert for something that happened in my own laboratory. So, it's usually those cases: either expert witnessing or a subpoena or some sort of compelled testimony about what has happened in our laboratory.
闯叠:听One thing I was curious about is when you have to give testimony. Like when you had to give it, do that for the first time, was it nerve racking? Like what the process like as a doctor? You're brought in as an expert. Do you get nervous? Is it nerve racking? Or do you just come in and just zero in on what you need to do?
骋叠:听The times that I have been an expert witness and been deposed the first time that that happened to me I told Professor Bailey, that it was a bit like a seven-hour colonoscopy. It was just the most exhausting, painful thing where, you know, someone was trying to poke holes in everything that I said and question and distort things that I had said, asked about things that I had no idea why they were asking other than it seemed that they were trying to make me feel like I was at fault or to blame for something, or I was overstepping my bounds or if I was arrogant or if I was not actually knowledgeable in something too. And that has also, that feeling has continued I would say it, but been blunted as the subsequent times I've done depositions. But the opposing counsel have usually been very, very rough. And in talking with other attorneys, I have understood that is entirely par for the course. That's sort of what normally happens. So, I was worried, but I've gotten a lot of good advice, both from the counselors who retained me, also my wife's piece of advice, which was, say yes or no and don't say anything else. Which was very good advice. My wife didn't do a lot of litigation, but she had some very good advice for me too. So, I would say the first time I got deposed, it was very intimidating. And I felt very bad after the day, like physically exhausted, having just sat in the room for several hours, and then the subsequent times it's gotten less. But it is something that's very daunting and I think folks that are in my business aren't trained to do this, don't know what to expect for this and it helps a lot to have someone sort of helping you ahead of time to teach you what to expect.
闯叠:听One of the points that you talked about when you're working with students is it's not necessarily just about being able to talk about the facts, but how to interpret the information that鈥檚 coming in. Why is this a key distinction to make?
骋叠:听So, I think one of the things that I spend the most time harping on, something that in statistics is called Bayes theorem. But it's also an underpinning of how we do medical decision making. And honestly, it's one of the ways that human beings think. I could give you an example, which is, you know, if you're stuck behind someone in your car, and the person has their blinker on constantly, you know, at some point, you're going to think that they're probably not turning left. If this is on for, you know, 10 minutes, they鈥檝e forgotten it, right. And so, you would take all that information, and then when you finally have to make the decision to pass them, you're going to use your sense of reasoning to say, 鈥淲ell, are they going to hit me when I try to go around them? Or are they not?鈥 And you'd use that piece of information before, the fact that their blinker had been on for 10 minutes and you're going to tend to ignore that evidence, because you think it's a mistake. However, you know, if someone's been erratic or driving strangely, you would use that and think, well, you know, that actually might be unsafe, maybe I shouldn't go around it, that person's not blinking. And they鈥檙e swerving back and forth.
So, we use information from before we do tests or before we make decisions to help inform what we do. It's interesting because there's a conflict with the way that the law and evidence works there. So, what I've told the law students before is you have to understand what we would call in medicine, the pretest probability. So, we do this, say, clinically when someone has all of the risk factors for lung cancer. So, they鈥檙e a smoker and they're elderly, and they've been coughing up blood perhaps or something, and they have a nodule on their chest x-ray, we would think, geez, well, now when I go and do a biopsy, I'm expecting cancer. If it's not cancer, you'd think, did we miss the cancer? But if it was cancer, you'd think, 鈥淥kay, we got, we figured out what we knew.鈥 Likewise, if you had a child who was coughing, and you were to notice a little nodule in there and get a biopsy, you would be shocked to see lung cancer. And you would think, 鈥淲ell, is there a mistake? Did we miss something up? Is it a false positive?鈥 And so, the reason you think those things is because of what you thought before you did the test. So, a test is really not a yes or no answer. A test is something that modifies a pretest probability into a post-test probability. And so, if something is absolutely unlikely or impossible to be true, and then you get a test that says it's true, you should really think about that.
The example I use in the class often is what if you were to administer a pregnancy test to me. If it was positive you鈥檇 think, 鈥淗e's probably not positive, you know.鈥 You don't know. My physical appearance might be misleading, but you don't know for sure. But you'd think it's really, really unlikely. That's because before the test you were thinking, this person is phenotypically. They really appear to be male, probably not able to carry a pregnancy. And so, the pregnancy test, you go from maybe one in a million chance from being pregnant to if it's positive, you know, somewhat more likely, but still, you know, maybe 50 in a million or something like a very unlikely chance. Likewise, if there's a woman who has a big belly, and you can see the baby kicking inside and you do a pregnancy test, and it's negative, you would think, 鈥淲ell, that's something wrong with the test, right?鈥 So, patient鈥檚 pregnant, if they're obviously pregnant, they told you they're pregnant, they look pregnant, you do an ultrasound, they are pregnant, something went wrong with the test. So, you discount it that way.
The alternate answers there are even more interesting, which is if you get a pregnancy test on me and it's negative, or a pregnancy test on the person who's pregnant and positive, you knew that beforehand. Why did you do the stupid tests? So, on both ends, you tend to want to do tests at a time when you don't necessarily know the answer, but you definitely use the information you had before. Now the thing that's been difficult to relate from my time from the grand jury, and also in talking with other legal experts is the way that evidence works is you can't necessarily pull in all a bunch of unrelated evidence in a case. So, just because someone has 17 convictions for meth use, and a time for meth trafficking or meth possession, and now you have, you know, this patient was caught with a white powder, and they did a test on that white powder, and it was presumptively positive for methamphetamine, you're not likely to say, 鈥淲ell, that's a meth person, you know, a meth dealer, and they've been convicted, I'm going to use that to help me do that.鈥 Even though that is the logically and sort of clinically useful way of thinking. And that's actually the way that human beings think all the time with that driving example I gave.
So, that's why I think it's important to talk about the interpretation of test rather than there's simply the fact of the test of the answer of the test. Because even though you're not maybe allowed to use that approach in court, you need to think about that when you're weighing the evidence that you have in any particular case. You need to understand that, well, here were the things that were pretest issues that were going on. And so, I'm going to think about this positive or negative result in that context. So, if it really does sound like that was a false positive test, that person had a lot of good reasons not to have been doing that drug or have that result, then you would need to go back and really think about scrutinizing it. Maybe get an expert to say, 鈥淲hat are the possible reasons this could be false.鈥 Whereas if everything lines up and the person admits it, maybe you don't want to try to you know, it's going to be futile to do that.
闯叠:听Where do you see pain points, or the steepest learning curve when it comes to really being able to master, you know, the kind of concepts that you're going over when it comes to medical evidence?
骋叠:听I think that there is maybe, sort of, more global and more specific and some specific issues. The global one is that I think that there is a 鈥 illiteracy is a hard word to use 鈥 I would say, an under appreciation, perhaps of general sort of scientific, sort of, thought processes and basic sets of facts. Like, you know, how insane medical malpractice, you know, how medicine, physiology and chemistry actually work, you know that鈥檚 left, oftentimes as this black box and I get this impression that if there's a case that involves some sort of medical, scientific or biomedical scientific fact, at its core, generally, the idea is that there's going to be a fight, and that whoever gets the smartest or most expensive expert to, you know, bend the, you know, the fact or the interpretation, the fact wins, when in fact, you know, the actual core truth of it is something that, you know, everyone involved could appreciate, but that's usually the thinking is oftentimes defer to the expert, rather than, you know, perhaps, taken as the duty of the lawyers involved to really understand this.
And I've seen as a lay person, you know, there are attorneys out there who make it onto the news who have you know, made themselves names as being experts in DNA evidence and such like that. So, it is possible to do that, but I think a lot of times, the science is this thing that sort of, they draw a box around it, put it in a corner and that is something that, you know, it's another chess piece in this whole thing. Whereas for me, it's the actual underlying ground truth of what occurred. And maybe I'm naive to think that the underlying ground truth should be the thing that dictates what happens. It isn't always, I understand.
I think in specific cases, though, one of the things I've told medical students and residents too, is that medical education is a long process of disabusing yourself of believing all these things that seems so obviously true, but actually aren't. You know, like, it seems that you know, when it's cold outside, you know, and you have wet hair or somebody that you get a cold, right, and you shouldn't go outside when it's cold, you should bundle up. It turns out that, you know, flu viruses and cold viruses circulate in the winter for a variety of different reasons. So, it is cold when you get the flu, but it's not the reason you get the flu. And so you know, the germ theory of disease takes a while, you know, to understand how that actually operates. Now, there's a lot of other things like that, things that make sense. There's also a lot of misinformation on the internet, or I wouldn't even say misinformation but sort of half information. One of the common things you can do is you could Google, 鈥淲hat is the source of a false positive for my math test.鈥 And what you'll get is thousands and thousands of hits of people who have both reported it. There'll be scientific literature on it. There'll be drug enthusiast websites, things where people like doing drugs and post tips and tricks on how to evade things. There's an enormous amount of half information on that. And one of the idiosyncratic things about drug tests is there may in fact be a certain type of test that has a false positive or false negative, that's known, but it doesn't go to the second type of one. So, you really have to know exactly what happened, exactly what test. What form of the test? It was sold by a vendor, what vendor created that test? What generation of the assay was it? All those things to know exactly what that specific tests problems are.
There was a, I think a lotion, like a moisturizer that had caused some false positive THC test. So cannabinoid or marijuana tests. And I think this is reported on the East Coast somewhere. All of a sudden people are saying, 鈥淲ell, you know, I've been, you know, using moisturizer that clearly exonerates me from, you know, my THC test.鈥 And the answer is no, there was one specific type of screening test that had that problem. We don't actually use that screening test here in our system. So, it was completely irrelevant. But you have to understand that there's different ways of doing these things. And so I think that there's a lot of misinformation out there or half information, which sounds really great. But once you sort of cut through it, there's not much underneath it. So, I think disabusing folks of things that they believe to be true, because we've all thought that.
闯叠:听That's so interesting. So, it's not just the test result, the yes or no, but the concept, the parameters, and all of that is just as important and consequential. And so, when you're working with students and telling people this I mean, do you see any pushback in terms of well, then how are we, you know, if this test the parameters are too low, this one too high, like, how are we supposed to use any of this? You know, in terms of court of evidence, right?
骋叠:听The students do frequently ask a lot of questions in exactly the way that you're sort of framing it about what to do with this information, if it is then confusing. And I think that you can strike a happy medium and find where, you know, this result seems as plausible. Now with something, oh, maybe I'll reflect it back to the legal world here is I don't actually understand what words like probable cause mean, you know, in terms of using a lab test, when someone comes to Harbor View and crashed their car, and they need to find out if they have, you know, a medication on board that's causing a heart arrhythmia or something like that. I don't try to meet a standard of probable cause or a majority of the if a preponderance of the evidence or beyond reasonable doubt. I tell them, you know, that they have that medication on board or I tell them that it's negative, and we try to figure out what the actual truth is, and we act upon it.
聽It was very frustrating what I was a grand juror. I actually reached out to faculty in the law school here before I started when I was selected for grand jury. And I asked someone, could you please send me information about what probable cause means? And I read a lot about it. And what I have to say is, it's not entirely clear to me that probable cause has much to do with either probability or causality. It wasn't 51%, you know, chances because oftentimes the medicine we can actually give you those answers. If we can quantify, which we frequently in pretest probability, I can tell you how likely the test is to give a false positive or negative because we quantify that. We actually assign numbers to that. We can actually make risk calculators that say, 鈥淵eah, you are 51% likely to have had taken that drug鈥 or 鈥68% likely鈥 or 鈥渙nly 1% likely.鈥 We can actually assign those numbers, but then the legal system may not actually care about that number. It may be very important that probable doesn't mean 51%
闯叠:听I guess I'd like to kind of tie it all together with a, you know, an example of a case that you were, that you've been involved in that we talked about before. You know, one of the cases you talked about was a drugged driving case, which had to do with someone who was in question whether they had done methamphetamine or not. Could you tell us a little bit about that case? And how that played out your role and that trial, your experience?
骋叠:听Sure. So, as a head of a toxicology laboratory at the county hospital, I get a lot of calls from folks who want to know about drug testing and specifically drugged driving. And in one case that I remember that I've seen, the question was whether or not someone had been using methamphetamine during a car accident. And I believe it involved a company, you know, that there was maybe a corporate truck or something like that, that hit a passenger car or something like that. And someone wanted to know if there was a positive result. And when the person went for medical care and not into my own hospital, they had a routine urine drug screen, which is one of these more nonspecific tests that looks at drug classes and it was positive for methamphetamine or amphetamine, one of the two. And those are very closely related drugs. And so it's sometimes hard to tell them apart on the screen. And so the question here was whether or not this supported that the person had actually used methamphetamine when they were driving.
And the opinion that I gave, which is, I think a standard opinion that many, you know, those in the toxicology world would give is that, actually, methamphetamine screening tests are among the most unreliable tests that we do in the clinical laboratory, largely because there's a lot of drugs that we give that are structurally very similar. So, things like Sudafed, and over the counters and a large number of drugs of abuse, too. But there's a lot of things that have the similar molecular shape. And the screening tests we do all respond to the shape of the molecule, not necessarily some other property of it. And so if there's similarly shaped things they respond. So, methamphetamine and amphetamine, and maybe ephedrine you know, may on some test do react. Some tests may be more or less specific to some of those things too. Same thing with opiates. So, oftentimes, the screening test can't tell the difference between hydrocodone, which is Vicodin, or morphine or codeine, or dilaudid, which is hydromorphone. It just one of those ones is there or a mixture of them are there we can't really tell. And so what I said was, 鈥淣o, with a screening test, that isn't good evidence of it. It's more likely for sure.鈥 If you took 100 people who had positive screening tests and 100 people who have negative screening tests, there's going to be more methamphetamine users in the 100 people who had the positive test, but it's really a statistical enrichment. It's not definitive. What we then do clinically, if we need to make the distinction and then again, legally, you'd always have to make this distinction as we do a confirmatory test, which then measures methamphetamine itself by a different method. And this patient had not had that done, a patient who was also I guess, a defendant in this case, had not had that done. And so there wasn't enough evidence in that case to do that.
And so this is also a question where I had to, as I referred to earlier, try to figure out what was the actual screening test that was done. Sometimes, there are some versions of screening tests that are actually quite a bit better in determining that and some that are really awful. So, the type that you would maybe buy at the drugstore. You know, you pee on the little strip and you get a stripe or not, sort of looks like a pregnancy test. Those are pretty bad. Whereas if it was done in a big instrument in a big crime lab, generally you might think that that was a better test. And so I had some questions about that. But in general, a screening test, especially for methamphetamine should not be seen as evidence of use.
The other important thing that I try to teach to is that urine screening usually can only tell you if the patient has been exposed to the drug, meaning that they either ate it or got into them somehow, but it doesn't tell you if they were intoxicated or impaired. The reason for that is the level isn't very important. What's important for impairment is your brain level of something how much methamphetamine or cannabinoid or morphine or heroin or whatever got into your brain. The amount that's in your urine is sort of the average amount that you've metabolized over time, divided by the amount of water you drink. And so if you drink a lot of water, your level, it'll be pretty low. And if you don't drink much water, your level could be pretty high, and it could be wildly higher or wildly lower, and it doesn't really reflect what was going on. So, if you want to figure out what someone's brain level is, the analytically perfect thing, which would be totally unacceptable would be to take a piece of brain and measure it, but we can't do that.
闯叠:听Makes sense.
骋叠:听Exactly. But what we do is we measure blood levels. And if we measure in the blood the blood is close to the brain level in many cases. And so we usually have more evidence scientifically that the blood level tells you. That's why we do blood alcohol testing. And that's also right now in the state of Washington why we do blood THC testing, and so the blood level correlates better. I will mention, interesting about the THC level, the THC level is something that was also sort of picked out of a hat. So, the drugged driving level for THC is five nanograms per mil of whole blood. Whole blood isn't normally what you would test for THC in. You would test it in serum, which is a pedantic point. I think the reason it was done in whole blood is that's what we do alcohol testing, and then the cops will have that sample. But the five is actually neither very sensitive nor specific. So, there are people who use a lot of marijuana 鈥 you know, medical marijuana users, or just the types of people who use marijuana all day, you know, get up in the morning and use it till the end of the day 鈥 they can become somewhat tolerant and a level of five might be what they wake up with in the morning. And maybe they can drive at that. Maybe not, but maybe they could drive. Then for people who are more naive to it, five is quite a bit, but there are also active metabolites that they don't measure. So, you could have four of THC 鈥 four nanograms per mil of TCH 鈥 and then a little bit of an active metabolite and you would be impaired, but we would say that you were not above the level. So, you can both again falsely maybe accuse them of being impaired or miss someone who's impaired with that level. Again, as a matter of convenience, we have to say that there's a level that you're above which are positive and below which are negative. But that causes confusion and some errors that we know about.
闯叠:听It's been so fascinating to hear your perspective and how we analyze these tests. And I know that law students are going to take a lot away from this. And so again, we really appreciate you sharing so much because this is a really, really interesting topic. I guess, for law students who, you know, maybe want to be litigators, but maybe are a little bit more wary about how to handle medical evidence and just talking about these extremely complicated concepts, in general, what advice do you give, you know, to encourage people who may be a little bit unsure about how good they're going to be at handling this kind of evidence?
骋叠:听That's an interesting question. I'd have to say that one of the pieces of advice I got in my own profession, from actually from my father, I went into the My dad's profession, he was also a clinical pathologist, and someone that he trained from told him this, which was that 鈥淣o one gets any good at this until you're about 50.鈥 And so I'm not yet 50. So, I think I still have some time to get good at it. But I think that the point of that is that there's an enormous amount of lifelong learning that has to go into the law as well as medicine to. And I would think that if someone wanted to be a litigator in cases that were going to involve, say, medical malpractice or drug tests, you know, from, you know, liability cases, or I don't know whatever, it would do them well to actually participate actively in continuing education specifically targeted this and actually to have some sense of the actual means of what's going on.
The best attorneys that I've talked to are the ones for whom I could spend, you know, and I've done this before, too, as part of an expert witness thing, is spent simply a day with the legal team going over basic chemistry and biology on how these things work. And those who come to me, just like one a quick answer, just want me to write something down and say yes or no or want me to look through a bunch of paperwork to try to find, you know, the needle in the haystack error that got made. Those aren't the sophisticated ones. But the ones who are willing to sit down and actually say we need to be educated for hours about, you know, how does this type of blood testing work? How does this work? I think that's a great sign. And I think that it's inescapable to be good at this. I think you actually will need to get some domain knowledge. Now, you don't need to be an expert in this. You know, that's why that's why experts exist. And that's why you have the ability to hire them. But I think you can't treat it as in vacuous. So, I think the first step is to understand that you maybe aren't an expert in everything, but you need to get some level of expertise. I think in medical education oftentimes is the first thing you need to do to learn medicine is to learn what you don't know. And that's the first way we sort of trust people in medicine, is what we're really sure that someone knows what they don't know. Then we can have them you know, pick up the phone and answer a question because they'll always know 鈥淲ell, I don't know that. I'll go ask someone.鈥 So the first thing you need to know is what you don't know. Oftentimes we go as an attorney too. I think operating when you don't know that you don't know something, that's the most dangerous, because then you would go forward with a misconception. And so I think getting to the point where you, at the very least where you know, what you don't know about things or you know, for example, like, there's a bunch of different types of tests, I do know that the tests will act differently and have different sources of false positives. And false negatives. I need to get an expert to tell me which test was done here. And, is that one, is that test, the one that has this false positive or negative, but if you don't know to ask that question, then you'll never get there.
So, I think that the level that one needs to get to and as an attorney, is to get good at knowing when the expert is needed, and when not. The idea that an expert is needed for everything is a waste of money. Honestly, it's not helpful for me to go look through a bunch of pieces of paper, even though sometimes I've done that. You know, looking for the needle in the haystack or the typo in something that calls the whole thing into question. That's not usually likely in a crime Lab to have occurred. But never calling one too or not calling one when you should is another thing. So, I think that that's the sort of the fundamental level I think of education that someone should get to then get over that worry of not really knowing much about.
闯叠:听Dr. Geoffrey Baird is chair of Laboratory Medicine at UW Medicine in Seattle and a regular guest at the UW School of Law. He often serves as a guest lecturer and works with students to help them become more effective at wielding medical and scientific evidence in the courtroom in order to advocate more effectively for clients. You can learn more about his work on our podcast page at law.uw.edu. Check out the show notes for links to more information on all the topics we discussed today.
Thank you, Doctor Baird. This has been super fun. Thank you very much.
骋叠:听Thank you very much for having me.
