There is an overwhelming amount of healthcare information available at our fingertips. Unfortunately, the ease with which we access information makes us more susceptible to misinformation which can have significant repercussions when dealing with something as complex as our health. It can be easy to fall prey to the misinformation machine. The beauty of science is that its very foundation is rooted in curiosity with a hefty dose of healthy skepticism thrown in. Before turning to Dr. Google, here are some quick tips to help you make the best decision about your health. Part one of this two part series will shed some light on what constitutes good research and in part two we will discuss what that means when you are buying healthcare products.
Research shows that…not all research is created equal
Most companies will be more than willing to provide you with this information if you ask, or may have it available on their websites. That being said, research jargon can be tricky, so here’s a quick breakdown of some of the types of research papers out there, and their pros and cons.
Types of Studies for Natural Health Products:
Animal studies: Animal studies consider the effect of an intervention (intervention refers to the treatment, nutrient, drug, or variable that a study is testing) on an animal system. These are often used as a preliminary basis for starting larger human clinical trials.
Case studies: Case studies describe the experience of one individual with an intervention. Since there are no controls (aka a ‘normal’ to compare results to), and the study is based on a single subject, these studies are not ideal for making a claim that can be applied to the general public.
Observational studies: Also known as cohort studies, observational studies observe two groups over a period of time. They observe a group that has been exposed to an intervention and one that has not. While the findings of these studies describe a correlation between an intervention and a given outcome, they cannot demonstrate a cause and effect relationship.
Randomized Control Trial (RCT): These studies are the most reliable. RCT’s are able to control for many variables so the outcomes are more directly related to the intervention. The best RCT’s are those with a large number of participants, a control group that has not seen the intervention, a clear procedure, and objective measures (we will discuss these in more depth below). The majority of drugs on the market have been subject to rigorous testing using a three phase clinical trial model of RCT’s. This means that a large number of participants are studied over a long period of time to determine the effects of a drug or treatment.
Systematic review: These studies combine the results of a number of RCT’s that examine a similar intervention or disease process. Thus, you are increasing the reliability of results by making sure lots of different labs were able to find the same results.
Meta-analysis: Think of a meta-analysis as a systematic review with extra math. These studies use statistical algorithms to verify claims and gain an overall consensus.
Methods and Materials:
If you want to go even further in your research, it’s important to look at the method of the study. The goal for any well designed study is always to be reliable and highly valid. That means consistently and accurately testing for what you want to be testing for.
Here are some good questions to ask when approaching these studies:
When was this study done?
Results of a trial conducted in 1983 may prove a company’s point, but the results may not be relevant today. A good rule to follow is to reference studies done five to ten years ago (unless it’s a well-known foundational study that is still regarded as valid). It is also prudent to ensure that the study was conducted for an appropriate length of time.
How many people were involved in the study, and what were their inclusion criteria?
A good study will have a high number of participants, because that means the results can be applied broadly. It is also important to consider if all the participants stayed in the study, or if some dropped out, and why they may have done so. Next, see who was involved. Did the study have a control group? The control could be people who did not take the intervention, who were healthy, or those who may have taken a placebo instead of the intervention. It is important to have some sort of comparison to form a baseline for the study. The ‘randomized’ part of an RCT refers to the fact that researchers randomly assign subgroups within a pool of appropriate subjects.
What was their measure, and does it actually answer the question being posed by researchers?
This is important because many times the markers used to measure an outcome may not be appropriate. An example of this is studies involving curcumin bioavailability. These studies wanted to answer how much of a certain type of curcumin extract (the active ingredient in turmeric) actually makes it into the blood stream. The markers tested in this study were the metabolized form of curcumin. The issue here is that it is not the metabolized form that exerts benefits - it is the free form curcumin. Therefore, the marker is not appropriate to answer the question. Of course this example requires some scientific knowledge, but this example reflects how vital it can be to question how the researchers tested their hypothesis. Researchers should also have a clear statement about the assumptions that were made when designing the study. If these are reasonable assumptions, corroborated with evidence, the results of the study are more reliable.
Interpreting results is just as important as how they were obtained. Important measures of the results are the “number needed to treat” values or NNT’s. This value is defined as the number of patients that require the intervention in order to prevent one bad outcome. So you want a low NNT. This means that the intervention is very effective at preventing a bad outcome. Conversely, “the number needed to harm,” or NNH, is used to quantify how much exposure to an intervention can harm an individual. So if the NNH is low that means very little exposure can make you sick.
The best science does not care who you work for or what the results should be - it is objective and logical. Unfortunately, business and science are often intertwined and this can skew results. As long as researchers and companies fully disclose their motivations, we can add the appropriate grain of salt to their findings. Most scientific papers have a section at the end of the paper where researchers can list their affiliations and disclose any potential conflicts of interest. Paying attention to who funded a research venture can also account for potential biases.
Stay tuned next week for Part 2 on questions YOU should be asking about natural health products.
Let us know in the comments below how you stay informed! What questions do you always ask companies to make sure you are getting what you pay for?
You may also be interested in: