Skip to playerSkip to main content
  • 9 months ago
Every day, we are bombarded by attention grabbing headlines that promise miracle cures to all of our ailments -- often backed up by a "scientific study." But what are these studies, and how do we know if they are reliable? David H. Schwartz dissects two types of studies that scientists use, illuminating why you should always approach the claims with a critical eye.

Lesson by David H. Schwartz, animation by Augenblick Studios.
Transcript
00:00Studies have shown that taking vitamins is good for your health
00:10and bad for your health.
00:12That newly discovered herb can improve your memory
00:15or destroy your liver.
00:16Headlines proclaim a promising new cancer treatment
00:19and never mention it again.
00:21On a daily basis,
00:22we are bombarded with attention-grabbing news
00:24backed up by scientific studies.
00:26But what are these studies?
00:27How are they performed?
00:28And how do we know whether they're reliable?
00:30When it comes to dietary or medical information,
00:33the first thing to remember is that while studies on animals
00:36or individual cells can point the way towards further research,
00:39the only way to know how something will affect humans
00:42is through a study involving human subjects.
00:44And when it comes to human studies,
00:46the scientific gold standard is the randomized clinical trial, or RCT.
00:51The key to RCTs is that the subjects are randomly assigned
00:54to their study groups.
00:55They are often blinded to make them more rigorous.
00:57This process attempts to ensure that the only difference between the groups
01:01is the one the researchers are attempting to study.
01:04For example, when testing a new headache medication,
01:06a large pool of people with headaches
01:08would be randomly divided into two groups,
01:10one receiving the medication and another receiving a placebo.
01:14With proper randomization,
01:15the only significant overall difference between the two groups
01:18will be whether or not they received the medication,
01:21rather than other differences that could affect results.
01:24Randomized clinical trials are incredible tools,
01:27and in fact, the U.S. Food and Drug Administration often requires at least two to be conducted
01:32before a new drug can be marketed.
01:34But the problem is that an RCT is not possible in many cases,
01:38either because it's not practical or would require too many volunteers.
01:42In such cases, scientists use an epidemiological study,
01:46which simply observes people going about their usual behavior,
01:49rather than randomly assigning active participants to control and variable groups.
01:53Let's say we wanted to study whether an herbal ingredient on the market causes nausea.
01:58Rather than deliberately giving people something that might make them nauseated,
02:01we would find those who already take the ingredient in their everyday lives.
02:05This group is called the cohort.
02:07We would also need a comparison group of people who do not have exposure to the ingredient,
02:11and we would then compare statistics.
02:14If the rate of nausea is higher in the herbal cohort,
02:16it suggests an association between the herbal supplement and nausea.
02:20Epidemiological studies are great tools to study the health effects of almost anything,
02:25without directly interfering in people's lives,
02:27or assigning them to potentially dangerous exposures.
02:30So why can't we rely on these studies to establish causal relationships
02:33between substances and their effects on health?
02:35The problem is that even the best conducted epidemiological studies have inherent flaws,
02:40precisely because the test subjects are not randomly assigned to their groups.
02:44For example, if the cohort in our herbal study consisted of people
02:48who took the supplement for health reasons,
02:50they may have already had higher rates of nausea than the other people in the sample.
02:55Or the cohort group could have been composed of people who shop at health food stores,
02:59and have different diets, or better access to health care.
03:02These factors that can affect results in addition to the factor being studied
03:05are known as confounding variables.
03:08These two major pitfalls, combined with more general dangers,
03:11such as conflicts of interest or selective use of data,
03:13can make the findings of any particular epidemiological studies suspect,
03:17and a good study must go out of its way to prove that its authors have taken steps
03:21to eliminate these types of errors.
03:23But even when this has been done, the very nature of epidemiological studies,
03:27which examine differences between pre-existing groups,
03:29rather than deliberately inducing changes within the same individuals,
03:32means that a single study can only demonstrate a correlation
03:36between a substance and a health outcome,
03:38rather than a true cause and effect relationship.
03:41At the end of the day, epidemiological studies have served
03:44as excellent guides to public health,
03:46alerting us to critical health hazards such as smoking, asbestos, lead, and many more.
03:51But these were demonstrated through multiple,
03:53well-conducted epidemiological studies, all pointing in the same direction.
03:57So the next time you see a headline about a new miracle cure,
04:00or the terrible danger posed by an everyday substance,
04:03try to learn more about the original study and the limitations
04:06inherent in any epidemiological study or clinical trial
04:09before jumping to conclusions.
Comments

Recommended