• Abstract and Figures
• Public Full-text

“Iron” Mike Webster was one of the greatest health outcome. It may also be admitted to prove centers to ever play professional football. Known the safety and efficacy of a product, to explain a for playing sleeveless in freezing temperatures defendant’s actions, or as the basis for an expert’s to intimidate his opponents, Webster spent 17 opinion. Given their general nature and inherent seasons in the NFL and helped lead the Pittsburg bias, epidemiological studies alone do not answer Steelers to four Super Bowl Championships in the questions of specific causation, i.e. whether an 1970’s. He was later inducted into the Pro Football agent caused a particular individual’s disease or Hall of Fame and named to the NFL’s all-time health outcome.7 team—a true football legend.1 CATEGORIES OF EPIDEMIOLOGIC STUDIES But the glory of football quickly faded after his There are two categories of epidemiologic retirement. Unable to keep a job and losing his studies: experimental and observational. In an marriage, Webster became homeless, depressed, experimental study, a researcher selects two indebted, drug-addicted, and plagued by chronic groups of individuals from a given population. pain that made sleep almost impossible. At one He then exposes one group to a suspected agent point Webster became so desperate for rest that while leaving the second group unexposed. he would shoot himself in the leg with a Taser until The researcher later evaluates both groups for he lost consciousness. In 2002, Mike Webster died development of the disease. of a heart attack at age 50.2 Experimental studies include randomized controlled What happened next would change professional trials and are often double-blinded, making them football forever. During Webster’s autopsy, forensic the “gold standard” of epidemiological evidence. pathologist and medical examiner Dr. Bennet However, experimental studies in humans are Omalu discovered that Webster suffered from ethically prohibited when an agent is known to be Chronic Traumatic Encephalopathy (CTE)—a brain potentially harmful. As a result, most epidemiologic disease never before seen in professional football studies are observational. players.3 In 2005, Omalu published his findings There are two main types of observational studies: in the medical journal Neurosurgery. CTE was cohort and case-control. Both types of studies catapulted into the national spotlight. But was have a comparison group and determine if there is football really the cause of Mike Webster’s rare an association between exposure to an agent and brain disease? The answer to that question is an a disease. If an association is present, the study epidemiological one. then measures the strength of that association. EPIDEMIOLOGY DEFINED In a cohort study, a researcher selects a group of “Epidemiology is the field of public health and individuals who have been exposed to the agent in medicine that studies the incidence, distribution, question and a second group who have not been and etiology of disease in human populations.”4 exposed. The researcher then follows the groups Epidemiology attempts to understand the cause and compares their development of the disease.8 and prevention of disease5 and is considered Behind randomized controlled trials, cohort studies among “the best evidence of causation in the mass are the strongest form of scientific evidence.9 torts context.”6 In contrast, case-control studies begin by Epidemiological evidence is often tendered as identifying a group of individuals who actually evidence of general causation, i.e. whether an have the disease and a second group who do not. agent is capable of causing a particular disease or The researcher then compares the groups’ past 11 12 exposures to the agent in question. If the agent positive association could result from random tells researchers that they can be 95% sure that not based on the random selection of patients. causes the disease, the researcher should find a error even if no association were in fact present.”19 the true relative risk is somewhere between Rather, researchers gain access to brains through higher proportion of past exposures among those For example, a p-value of 0.05—which is the most 2.14 and 3.36. The more narrow the confidence donations from families or by direction of the who have the disease.10 Case-control studies common significance level—means that there is a interval, the more precise the result. However, if the players themselves before their deaths. Because fall just below cohort studies in the hierarchy of 5% chance that the study will erroneously find an confidence interval includes 1.0, the result cannot the families of players who exhibited CTE-like scientific evidence.11 association where no true association exists. Thus, be “statistically significant.”20 symptoms during their lives are more likely to the researcher can be 95% sure that the observed donate a brain for research, the sample of brains Case reports, like the one regarding Mike Webster, 2. BIAS association is true. As long as the observed received is not representative. describe clinical events in a single patient and are The overarching objection to the existing body p-value for the study falls below the preselected among the weakest forms of scientific evidence.12 of CTE research is that it is based on inherently Even Dr. Ann McKee, director of Boston significance level, then the relative risk or odds Alone, case reports cannot establish a causal biased case reports. Bias refers to anything University’s CTE center and now the leading ratio can potentially be “statistically significant.” link between an agent and a disease primarily “that results in a systematic (nonrandom) error neuropathologist in the study of the disease, because there is no comparison group, and they A second way to assess random error is by using in a study result and thereby compromises its admits that “an autopsy series is terribly biased” are generally excluded at trial.13 Even Dr. Omalu a confidence interval. A confidence interval is validity.”21 There are two primary types of bias and by itself is unable to detect incidence and conceded that his case report of Mike Webster “by a range of values within which the true value is in epidemiological studies: selection bias and prevalence of the disease.23 In other words, case itself [could not] confirm a causal link between likely to fall. Suppose a study finds a relative risk information bias.22 reports cannot establish why or how frequently professional football and CTE.”14 (discussed below) of 2.66 with a 95% confidence the disease occurs in a given population. Selection bias results from the method by which interval of 2.14 to 3.36. The confidence interval PRESENCE OF ASSOCIATION study participants are chosen. CTE research is Information bias results from “inaccurate information However, Dr. Omalu’s diagnosis of CTE in Mike about either the disease or the exposure status of Webster was just the beginning. In November the study participants or a result of confounding.”24 2006, Omalu published a second case report15 after Because concussions in football are underreported finding CTE in the brain of former NFL player Terry and largely undocumented, it is difficult to Long, a 45-year-old former Pittsburg Steeler who reconstruct accurately a player’s medical history. As committed suicide in 2005 by drinking antifreeze.16 a result, researchers must often rely on interviews He later found evidence of CTE in the brains of with family members about the player’s exposures retired NFL players Justin Strzelczyk (age 36) which test memory and are speculative at best. and Andre Waters (age 44).17 Omalu’s research These interviews can also encourage “recall bias,” in appeared to show an association between playing which family members are more likely to report past professional football and CTE. But were these exposures once the disease has been confirmed. associations true, or merely the result of random error, bias, or confounding factors? ALONE, CASE REPORTS 1. RANDOM ERROR Epidemiologic studies are often based on CANNOT ESTABLISH A relatively small sample groups. As a result, a study CAUSAL LINK BETWEEN may erroneously find an association where one does not actually exist, or not find an association AN AGENT AND A DISEASE where one does exist, simply due to “chance” or PRIMARILY BECAUSE THERE “random error.”18 IS NO COMPARISON GROUP, One way to assess the potential for random error is by calculating a p-value. “A p-value AND THEY ARE GENERALLY represents the probability that an observed EXCLUDED AT TRIAL. 13 14 3. CONFOUNDING FACTORS by the incidence rate of disease in the unexposed association between the exposure and disease. Many jurisdictions will only admit epidemiologic A “confounding factor” is an “extra” factor in a group. A relative risk of 1.0 means that there is no A relative risk greater than 1.0 means there is a evidence if the relative risk is greater than 2.0—a study group which independently increases both association between an agent and a disease, and positive association between an agent and disease, level that permits an inference that the disease was the risk of disease and exposure.25 If not properly that the risk of contracting the disease is the same which could be causal. The higher the relative risk, more likely than not caused by the agent in question. accounted for, confounding factors can skew in both exposed and unexposed individuals. A the stronger the association. Others courts reject this reasoning and will admit the results of a study by producing an observed relative risk less than 1.0 means there is a negative epidemiologic studies with a relative of 2.0 or less as For example, suppose as a hypothetical that a association when no true association exists.

Load more

  4

Copy Link