Why we roll our eyes at shows like 'CSI'
  Pathologists' reaction to 'Ready, set, action!'

December 2007
Feature Story

Wendy M. Gunther, MD
Stephen J. Cina, MD

Recently the mother of a young woman who had died of a drug overdose more than a year before called one of the authors (W. Gunther) to demand angrily why no test had been done to determine what food her daughter had eaten before she died. Toxicologists had tested the stomach contents, and the tests had shown a fatal overdose of ingested cocaine. But her mother didn't want to know that; she wanted to know if her daughter had ingested a hamburger and Coca-Cola before she died. Her theory: The drug dealer persuaded her daughter to unknowingly ingest the cocaine by offering it hidden inside a hamburger and Coke. When asked what test she thought we could do to determine whether a person had drunk a Coca-Cola before death, she went blank. The best we can do, she was told, is look at the contents of the gastric pouch to see if we recognize any food. There is no massive database, no selection of magical tests that will identify hamburger if the food is fully digested, and no Coca-Cola assay. The poor mother was nonplussed. "Well, on 'CSI' they have a test..." she said. She had to be informed that television isn't always reality.

Despite the welcome attention that "CSI," "Bones," and other similar shows have brought to forensic pathology, their large dose of distortions and flat-out mistakes has affected our dealings with the people who need us most—the families of the decedents. The "CSI effect" has led to family members' unrealistic expectations of medical examiners, suspicions where none should exist, and juries vulnerable to traps set by savvy lawyers willing to meld fact and fiction to benefit their clients.

What do characters like Grissom of "CSI" do on television that real forensic scientists can't do? There are errors of fact and errors of proportion. The latter are more subtle and damaging because they often appear credible to the untrained observer.

Perhaps the most distressing error of proportion is the implication that one man can do it all. All it takes is Grissom and perhaps Sara (recently his love interest—generally poor office behavior) to do the scene investigation, collect all the evidence, commandeer the lab, and perform all of the specialized tests on every piece of evidence themselves. This requires forensic training such as no investigator in real practice ever has. They wander in and out of the autopsy suite, inappropriately attired, to absorb a summary of the entire gross autopsy in a line or two, and occasionally take over the body without permission of the forensic pathologist to perform something as ridiculous as a stab wound casting. They commonly interview witnesses, which is the job of the detectives. They do their own DNA examinations, handle their own toxicology, do their own Internet research, and probe infinite numbers of convenient databases. All this while getting enough sleep and looking great! A parallel in blood banking would be if a single employee were an intake specialist, phlebotomist, local area network computer specialist, hematologist, quality control specialist, and the person who makes the phone calls to physicians to alert them to critical values.

In real life, forensics, just like any other area of pathology, can be handled only by a team. No one person does it all. To suggest that scene investigators have the time and expertise to handle not only their job duties but also everyone else's minimizes the importance of teamwork and mutual reliance that is the hallmark of real forensic practice.

A major sore spot for forensic pathologists is the false notion these shows give that forensic science can accurately determine the time of death. Time of death is one of the more difficult scientific assessments to make; many would call it more art than science. Every year brings new articles in the forensic journals about the most recent attempts to scientifically determine time of death. However, these experiments are conducted under carefully controlled lab conditions; no forensic technique has proved to be a reliable predictor of the time of death in the uncontrolled conditions of a crime scene. Estimating the postmortem interval is so heavily dependent on temperature and myriad other variables that the best one can do is to make an educated guess. Factors that can lead to gross error in estimating the time of death include the ambient temperature; the temperature of the body when the victim died; whether the victim was frail, thin, obese, or muscular; what the victim was wearing or not wearing; whether blankets were covering the victim; and subtleties such as the microclimate (for example, when a transient sunbeam enters the room and lingers over the body for two or three hours but is gone when the body is discovered). In the absence of a thermostat-controlled room with no windows, determining the time of death is fraught with peril for even the most experienced investigator.

A real forensic scientist who is asked to estimate time of death always responds with one essential question, which is never asked on TV: What time was the victim last seen alive, and what time was he found? If you don't know these simple parameters, you risk your "scientific" guess falling outside the boundaries of the known facts.

Does the time of death really matter? It often matters to families who have been taught all they know about crime scene investigation by television shows. Medical examiners routinely field phone calls: "Doctor, what time did my husband die?" The guilt-ridden widow can be told what time he was pronounced dead in the ER, but that's not what she wants to know. She heard him gurgling in his sleep at 1:30 in the morning and found him blue and unresponsive at 8:30 in the morning. She wants to know whether she could have called 911 when she got up to go to the bathroom at 3:30 and is, therefore, to blame. Depending on physical findings such as rigor mortis and lividity, taken in concert with the circumstances, the medical examiner can provide her with a window when death may have occurred rather than a specific moment. She is not happy when told that medical science can't give her a more accurate answer.

Determining the postmortem interval also matters to police agencies and lawyers who are trying, based on time of death, to exclude a suspect from or place a suspect at the scene of a homicide. A miscalculation or scientifically unfounded proclamation by the forensic pathologist can result in the arrest, and perhaps execution, of an innocent person or the release of a killer. When the case goes to trial, if the medical examiner cannot tell a jury that a badly decomposed victim died between 1:00 and 1:30 in the morning, 13 days before discovery of the body, his or her credibility and expertise may be called into question. The jurors may be dubious of the entire autopsy findings if the medical examiner can't provide them with a simple time of death—because "CSI" and other similar shows have told them they can.

A point often omitted is that decedents develop not only rigor mortis but also concurrent livor mortis over the early postmortem interval. The settling of blood to a supine decedent's back under the influence of gravity is one of the most obvious signs of death and may be an actual forensic clue in a case where a scene has been altered by movement of the body after death in an attempt to conceal a homicide. It's understandable that living actors can't exhibit livor mortis—though if they can have pale faces through makeup, they can surely have purple backs. If a TV show is going to go to great expense to make a realistic dummy so they can pretend to have it undergo an autopsy, why doesn't it have livor mortis? The reason dead people are pale is because the blood has settled to their backs.

A popular myth is that science will always find the answer. On "CSI" there is a database or a test for everything and all you have to do is look it up. It's true that AFIS, the automated fingerprint inventory system, is a wonderful method for getting good matches on full and clear thumbprints of people whose prints are already in the system and on file. It may not be as effective for partial fingerprints, and a match tends to show up within hours to days, not minutes. And it can't do anything to identify the fingerprint of a person who has never been fingerprinted before. But just because AFIS works in some cases in the real world doesn't mean there is an equally rapid and universal database for everything.

It is true that forensic toxicology laboratories can detect and quantify drugs and a vast array of substances and that the resolution of many cases depends on their results. However, they can't do it in 30 minutes. Quality control issues, staff and equipment limitations, accreditation guidelines, suboptimal samples, and a constant stream of new designer chemicals mean longer turnaround times. On "CSI," the toxicology results are back before the end of the show. In real life, depending on the state or county budget, they are back within two and 102 weeks. A frequent, uncomfortable phone call goes something like this: "Doctor, you performed the autopsy on my son this morning. What did the drugs show?" It is sad and difficult to have to explain to that family member that although a drug overdose is possible or even likely, neither the family nor the doctors will know for weeks or months. Sometimes the families object that it's not that way on "CSI."

A colleague who works in a DNA lab recently shared her pet peeve: When the DNA analysts on TV come back to report the results, they are wearing the same clothes they wore when they went in to do the tests. The need for meticulous sample preparation and quality control measures to avoid performing PCR on the wrong sequence and the time required for a DNA expert to interpret a multi-specimen case has resulted in a huge backlog of DNA tests facing qualified laboratories. The minimum wait for results is days, if not months. The real DNA analyst doesn't have any set of clothes she wears for months.

The DNA analysts with whom we are familiar are all highly reliable scientists. They aren't the types of people who mind tedious work, repeated with care, if that is the way to get verifiable results. Then why does Greg, the DNA analyst on "CSI," come across as such an undependable and untrustworthy fellow? Of all the people in the lab to choose for the character flaw of unreliability, the DNA analyst should have been the last.

The idea that every medical examiner's office has an unlimited budget is a damaging error of proportion. No office we know of has its own police helicopter in which to fly over any area that might be interesting to look at from above. No office has the budget to turn out all its results in hours. No office has the budget to do it all and do it instantly. No police force has the resources to field endless scene investigation operatives evaluating every possible scenario. In our world, budgets are real, and necessary funding is won from harried governors and county commissioners through skillful representation by chief medical examiners.

Another myth popularized by television suggests that every investigator has the leisure to take on a single case at a time, investigate it as he or she sees fit, and come to a successful conclusion. And they always clear their cases and have a clean, well-organized desk. Busy, overworked police personnel in real life never have the luxury of handling a single case at a time and many cases are never cleared. Multiple overlapping cases, insufficient police personnel, and endless sacrifice on the part of real police heroes are the true story. By comparison, the infinite periods of leisure and implied infinite funds of television sleuths are almost insulting.

Many of us are also troubled by the notion that ultimately there will always be an answer and every case will be solved. If the star of "CSI" or other such shows can't figure out a problem, it isn't because science isn't good enough to answer the question; it's because he hasn't figured out how to apply the science to answer it—yet. The fact is, everyone who works in this field accumulates a mental list of cases where, despite exhausting all options, we could not figure out how and why someone died. I never hear operatives on "CSI"-type shows say, "I don't know" or "I tried the analysis, and it didn't work." But that is how real science is. Real science is real because it reports negative results. Much of science and medicine is admitting when you don't know the answer or when the test doesn't answer the question. In reality, more homicide cases are solved by confession than by science.

How prevalent are errors of fact? Not very. The science behind "CSI," for one, is usually sound. But any time science conflicts with the needs of television, it's science that must give way. For instance, many of the actresses of "CSI" have long hair, and when they enter a scene with a dead body in it, they kneel over that dead body and use their hands to push back those flowing locks, which would scatter random hairs and DNA across the decedent and the crime scene. Of course, advisors to "CSI" know real investigators wear hairnets and Tyvex suits at a scene, but this is one place where theatrics conflict with reality.

Though hairnets may be too much to ask of television, surely it isn't too much to ask Grissom to follow basic rules of evidence. If you see an item on the floor in a crime scene, your job is to document it. You may draw it on a diagram, take measurements, photograph it, or do all three. Only then do you collect it. You do not kneel and pick up the evidence so you can examine it intelligently in front of your nose, then lay it back down into position and take your photograph. Who knows what errors you might have introduced by handling the evidence before you documented it? Worse still, Grissom and colleagues have a disturbing tendency to taste things they find at the scene.

Another health hazard and accreditation violation seen on television is inadequate lighting of the laboratory. There is no reason why the autopsy pathologists on these shows all work in the dark. They track gunshot wounds through the body by following a single bright light against dramatic shadows. The first thing a real autopsy pathologist does is to get the decedent into an autopsy suite under as many fluorescent lights as the budget permits. We like to see lethal injuries, not blindly feel our way through them in the dark.

Inaccurate depictions of reality are not limited to the autopsy suite. For example, the crime scene technicians in these shows often collect blood-soaked evidence in plastic bags. There's no quicker way to make a bloody garment turn into a bloody, moldy garment than to enclose it in a plastic bag. The only suitable bag for decedents' hands and for most evidence is a paper bag. But the audience can't see through paper, and that would diminish the dramatic effect.

An example of outright nonsense is the idea that you can pour casting material into a stab wound to determine the size and shape of the knife. Everyone who works with actual bodies knows that the viscera change position as the body goes from standing to sitting to supine. Further, the internal injuries from a stab wound inflicted on a twisting, arching person who is trying to avoid a knife differ greatly from those associated with a wound track imparted to a motionless decedent lying flat on his or her back. In most cases, there isn't going to be a perfectly straight channel from the skin surface to the deepest point of penetration. There's more likely a meandering track through viscera that are no longer in the same position they were in when the knife entered the body. Lungs collapse, organs shift, hemothoraces compress adjacent structures. Casting material poured into a stab wound would never follow a well-defined path that conforms to the shape of the weapon; it would run into the body cavities and organs in a path of least resistance. You'd end up with a cast of the pericardial sac, peritoneum, or pleural cavity. And even if the casting material did follow the knife wound, the dimensions of the wound path would not necessarily correspond to the knife's shape and size. A person who is stabbed finds it uncomfortable and will attempt to move while the weapon is inside. This creates a wound with features of both a cut and a stab. It is impossible to determine the width of a blade when this occurs. The length of a blade is also easily over- or underestimated. The force applied by the assailant has a lot to do with how far a knife penetrates the body. It's been said that a strong arm can send a three-inch penknife from the skin of the belly right down to the spinal column (often twice its length). A cast of a wound caused by a penknife inserted into a twisting person's belly may suggest a butcher knife based on the length and width of the hypothetical cast. But we can't let a little reality stand in the way of a good story, can we?

Real technology is also misused in fictional television. In one scene, the crime scene technicians decided to use Luminol to detect occult blood in broad daylight. Luminol fluoresces when it reacts with blood, which means one must use it in a dark room to see it light up. There's no point in using it during daylight. It would be like looking at a black-light poster in the sun.

Perhaps our favorite scene in a television show was when the medical examiner looked up from a body and said, straight-faced, "He died of fear." He didn't specify acute or chronic fear, and without this critical information, how would it be possible for Vital Statistics to accurately SNOMED code this death?

Forensic pathologists have become the public face of pathology because of these shows. Though pathology as a specialty has benefited from the attention, the many myths the shows propagate have affected the ability of medical examiners to communicate with families of decedents and have led to miscarriages of justice. And thousands of young people watching the forensic television shows may be shaping career plans based on the jobs the shows' characters have. We welcome and need sincere and enthusiastic people in our field, but we must consider what they will do when they find out they can't do it all, can't solve only one case at a time, can't determine the time of death, don't have a database for everything, don't have a helicopter, can't take charge of a whole investigation and tell the detectives and doctors what to do, and don't always solve the case to a satisfying conclusion in a brief time. For families and juries and for eager young investigators, the myths of such shows have much to answer for.