Gold Standard of Evidence: The Randomized Controlled Trial (RCT)
The Interactive Autism Network (IAN) was created in order to bring parents and researchers together with the goal of accelerating and expanding high quality, autism-focused research.
Research comes in many forms, and is conducted in many ways. The careful evaluation of a single case --called a case study--can help us describe and make sense of a certain situation or condition. (Leo Kanner's detailed description of only eleven children formed the basis of his groundbreaking article on autism in 1943.) Statistical techniques are used to analyze census or survey data from thousands of people -- and our own IAN data will be similarly used to discover new information about children with ASDs. Analysis of in-depth interviews exploring the beliefs and experiences of even a few dozen individuals can provide crucial insight, revealing, for example, the stressors faced by families with a child with an ASD or the perspectives and concerns of individuals with an ASD. All these methods have worth and contribute to our understanding.
When it comes to treatments and therapies for Autism Spectrum Disorders (ASDs), however, the ideal is to test any such intervention using a Randomized Controlled Trial (RCT).
What is a Randomized Controlled Trial?
Of all the many ways research can be conducted, the gold standard level of proof where treatments and therapies are concerned is the Randomized Controlled Trial (RCT). What is this method of carrying out scientific research, and why is it so highly valued?
A Randomized Controlled Trial is an experiment or study conducted in such a way that as many sources of bias as possible are removed from the process. Basically, scientific errors of the past have taught us where we can go wrong, drawing false conclusions from our research. RCTs are designed to eliminate these major errors.
Two Groups for Comparison
In most RCTs, the goal is to determine if a specific therapy actually makes a positive difference to the people receiving it. To figure out if it does, researchers need to be able to compare. Did the children receiving a new Educational Program A do better than those who got the old standard Educational Program B? Did the people who took a new medication do better than those who received a placebo, that is, a sugar pill?
The "control group" is the group that doesn't get the new treatment. They are the benchmark, the standard you measure against. Researchers testing a new intervention, program, or drug hope to show that the people receiving the new intervention end up much better off than those who were in the control group.
It's not enough to have two groups, one receiving a treatment and one receiving a placebo (or a different treatment). You need two groups that are randomized. That means that the researcher does not control who gets put in which group. Each person should end up in whatever group by chance in order to eliminate bias. For example, let's say you are testing a new drug. You don't want all the healthiest people to be placed in the group that will be taking the new drug, and all the sickest people to be placed in the group that's getting a sugar pill. If you do that, your treatment is going to look more effective than it really is. It's going to look like it worked wonders, when all that really happened is that the healthiest people were compared with the sickest ones.
To the greatest extent possible, you want assignment to groups to be random --beyond the control of participants, researchers, or anybody else.
Make it "Double Blind"
"Why give anybody a sugar pill?" you may ask. "Why can't the people in the control group just get nothing?" The answer is: because you don't want psychological factors to influence the outcome. You don't want people who got the pill (which they know must be the exciting new treatment) to get better just because they're feeling so hopeful about it, while those who got nothing get depressed because they know they got nothing. This will make a hash of your results, because now you're not comparing a drug treatment with a non-treatment, you're comparing people who feel hopeful with people who feel hopeless. If your results are due to that, and not to the effectiveness of the treatment under evaluation, you're in trouble.
Researchers can also be biased, of course. If they know that such-and-such a person got the new treatment, and that so-and-so did not, will it influence their observations of how that person is doing? Probably. Researchers are human beings, too, and if they are hoping the new treatment they developed is going to be the most wonderful thing since penicillin, their hopes can creep into the process, warping their observations and their results.
A "double-blind" study makes sure that neither participants nor scientists know who got what. The goal is to remove our all-too-human despair, hope, and ambition from the equation.
Of course, this is the ideal, and not easy to achieve in more natural settings, such as classrooms. A teacher will know what program they are teaching, for example. Still, researchers will try to create the "blind" condition by making sure those who assess children's performance at the study's conclusion are independent evaluators who do not know which program a child received.
During the Trial
During the trial, an ongoing review is conducted to make sure that everyone is safe and to detect if there are dramatic differences in how the people in the different treatment groups are doing. Occasionally, especially where medications are concerned, researchers stop the study early because of negative results, or because of results so positive that it would seem unethical not to give the new treatment to everyone participating. Usually, the trial goes to completion and the results are evaluated, confirming that either the new treatment is useful or that it isn't.
Autism Spectrum Disorders: An Urgent Need for More RCTs
Research is desperately needed to address treatments and therapies for ASDs. It is our opinion that any intervention claiming to treat ASDs effectively should be considered for a Randomized Controlled Trial.
IAN, especially in its role as a registry matching families to research projects, will help make it possible for more autism treatments and therapies to be tested in Randomized Controlled Trials.
Families have struggled for far too long with far too little information on which to base important treatment decisions. It is time they had real answers.