Most cases involving allegations of adverse health effects involve an assessment of epidemiological studies and a detailed understanding of the underlying methods utilized. In a previous post, we discussed assessing and establishing causation in the courtroom.
Most cases involving allegations of adverse health effects involve an assessment of epidemiological studies and a detailed understanding of the underlying methods utilized. In a previous post, we discussed assessing and establishing causation in the courtroom. This process almost always involves epidemiological principles. In this post, we drill down on some of the broader concepts discussed in that post to discuss the general principles of epidemiological study as it applies to the courtroom.
Before digging in, let’s discuss what is in the forefront of most litigators minds: the Eyes Glaze Over (EGO) problem. Anyone who thinks epidemiological methods are hopelessly boring should watch Ben Goldacre’s TED Talks video in which he describes flawed epidemiological methods, which he refers to as “Battling Bad Science.”
Now let’s get started with a description of the different epidemiological study designs. To anchor the discussion, we’ll start with the gold standard, the Randomized Controlled Trial (RCT). This is the best way to evaluate whether an exposure (e.g., a drug, a chemical, or a supplement) is causing an adverse health effect. We refer to a RCT as an experimental study design because we are imposing an intervention on the study subjects and we are randomly assigning these subjects to groups.
(There are entire treatises written on RCT methods that you may wish to obtain and study: check out this one by a seminal figure in clinical trial design, Curtis Furberg and this one by my friend and colleague, Herb Weisberg, providing a critical assessment of what sometimes causes clinical trials to provide misleading answers.)
Although we would always like our answers generated from RCTs, this is not always feasible. The problem is that intervening is often too invasive. Cost and ethical concerns are the two most obvious and common constraints on performing interventional (i.e., experimental) studies.
The other methods available to study the effect of exposures on human health are all observational in nature (as opposed to experimental). In an observational study design, we don’t assign our subjects to groups, but rather we observe them after they have assigned themselves to groups.
Observational studies are subdivided into analytical studies and descriptive studies. The two most prominent types of analytical observational study designs are cohort studies and case control studies. The most prominent type of descriptive observational study designs is referred to as a cross sectional or prevalence survey. Analytical study designs are considered to be of a higher quality and to provide more reliable information than descriptive epidemiological study designs. We discussed the different levels and the quality of scientific evidence in a previous post.
Because they are the best observational study design (and the type most commonly referred to) I will focus below on describing the basic aspects of cohort and case control epidemiological study designs.
Cohort studies are modeled after (and resemble) RCTs on their surface. In a cohort epidemiological study, the investigator observes a group of individuals with the exposure of interest and compares those individuals to another group of individuals without the exposure of interest. Exposed subjects are then followed forward in time and compared with unexposed subjects for the rate of adverse health events. It is critical that the exposed cohort is selected in the exact same manner as the unexposed cohort and that the measures of disease outcome are made in a similar manner between the two groups.
At the end of a successful cohort study, the investigator calculates a Relative Risk and an associated confidence interval (CI), which he uses to evaluate the association between the exposure and the adverse event. If the rate of an adverse health outcome is higher in the exposed cohort compared to the unexposed cohort, we have some evidence to support an association between the exposure and the event. However, this is only a potential first step in a causation assessment and one must be sure that there is indeed a valid and reliable association (i.e., the role of bias, confounding, and chance must be ruled out).
Incidentally, cohort studies can be performed either retrospectively or prospectively. In either design, the cohort is followed forward in time to evaluate whether the individuals in the group had an adverse event outcome. However, in a retrospective cohort design, individuals are followed forward from a point in time in the past. In a prospective cohort design, individuals are followed forward in time from the present.
Case control studies turn the cohort study design on its head. Rather than following two groups (exposed and unexposed) forward in time, we look from disease onset back in time. Specifically, we look back at a group of individuals with a disease outcome of interest and compare them to a group of individuals without the disease outcome of interest. We then evaluate the rate of prior exposure in both groups.
At the end of a successfully performed case control study, the investigator comes out with an odds ratio (OR) and a confidence interval (CI), which he uses to evaluate the association between the exposure and the adverse event. . If the rate of prior exposure is higher in the cases (i.e., those with the disease) as compared with the controls (i.e., those without the disease), we have some evidence to support an association between the exposure and the event of interest. However, as with a cohort study, this is only a potential first step in a causation assessment and one must be sure that there is indeed a valid and reliable association (i.e., the role of bias, confounding, and chance must be ruled out).
As stated above, the way we ultimately measure the risk from an epidemiological study is generally referred to as a risk ratio. The risk ratio generated from RCTs or from cohort studies is usually referred to as a “relative risk” and the risk ratio generated from a case control study is usally referred to as an “odds ratio” (these metrics are both trying to tell us the same thing — is there an association — but they are calculated in slightly different ways).
This video provides a good general overview of the concept of an odds ratio and a relative risk in epidemiogical studies.
Whether you are defending a pharmaceutical product in a product liability case, a chemical manufacturer in a toxic tort or even a dietary supplement manufacturer in a consumer fraud class action, odds are you will have to grapple with epidemiological data. When doing so, it is important to identify and critique the study design and make sure that the expert testifying about the study is held accountable for the study’s methodology. If it is a defense expert using the study to make your case, it is important to ensure that the study is of a high quality and that any weaknesses in the study design are recognized (and that your expert is ready to address those points on cross). If it is an opposing expert relying on an epidemiological study to make the plaintiff case, you should dissect the study, including its methodology, and be prepared to confront the expert on any deficiencies noted.
They key with any clinical study is to evaluate sources of bias, confounding, or chance. Observational epidemiological studies are subject to all three aspects of error. No study is perfect and weaknesses can be found with any study design. These important topics will be the subject of upcoming posts.
Did you like this post? We are here to help! Schedule a consult with one of our experts.
Just fill out the form bellow and we will contact you with more information.
On Wednesday July 8, 2020, Dr. David Schwartz of Innovative Science Solutions presented at the IADC 2020 Virtual Annual Meeting on a panel titled The Use of Genetic Testing in the Courtroom. A complimentary copy of the panel presentation is now available for download. Read more
Dr. David Schwartz of Innovative Science Solutions will be presenting at the IADC 2020 Virtual Annual Meeting on a panel titled The Use of Genetic Testing in the Courtroom. Read more
Plaintiff experts having been asserting for decades that all mesotheliomas must be linked to some asbestos exposure. Indeed, this has led to the erroneous (but widespread) view that mesothelioma is a signature disease, only caused by asbestos exposure. Read more
The journey to scientific and commercial success is often complex and always critical, if you are looking for an expert partner to help steer you to confident solutions, contact us todayContact us