Credible sources often disagree on technicalities. Sometimes this includes classification of research design. Some argue that there are only 2 categories of research design:
True experiments. True experiments have 3 elements: 1) randomization to groups, 2) a control group and an 3) intervention; and
Non-experiments. Non-experiments may have 1 to none of those 3 elements.
Fundamentally, I agree with the above. But what about designs that include an intervention and a control group, but Not randomization?
Those may be called quasi-experiments; the most often performed quasi-experiment is pre/post testing of a single group. The control group are subjects at baseline and the experimental group are the same subjects after they receive a treatment or intervention. That means the control group is a within-subjects control group (as opposed to between-group control). Quasi-experiments can be used to answer cause-and-effect hypothesis when an experiment may not be feasible or ethical.
One might even argue that a strength of pre/post, quasi-experiments is that we do Not have to Assume that control and experimental groups are equivalent–an assumption we would make about the subjects randomized (randomly assigned) to a control or experimental group. Instead the control and experimental are exactly equivalent because they are the same persons (barring maturation of subjects and similar threats to validity that are also true of experiments).
I think using the term quasi-experiments makes it clear that persons in the study receive an intervention. Adding “pre/post” means that the
researcher is using a single group as their own controls. I prefer to use the term non-experimental to mean a) descriptive studies (ones that just describe the situation) and b) correlation studies (ones without an intervention that look for whether one factor is related to another).
Last week’s blog focused on the strongest types of evidence that you might find when trying to solve a clinical problem. These are: #1 Systematic reviews, Meta-analyses, or Evidence-based clinical practice guidelines based on systematic review of RCTs; & #2 Randomized controlled trials. (For levels of evidence from strongest to weakest, see blog “I like my coffee (and my evidence) strong!”)
So after the two strongest levels of evidence what is the next strongest? #3 level is controlled trials without randomization. (Sometimes called quasi-experimental studies.)
Here’s an example of a controlled trial without randomization: I take two groups of mice and test two types of cheese to find out which one mice like best. I do NOT randomly assign the mice to groups. The experimental group #1 loved Swiss cheese, & the control group #2 refused to eat the cheddar. I assume confidently that mice LOVE Swiss cheese…
the participants are randomly assigned to groups, &
one group is a control group that gets a placebo or some inert treatment so that outcomes in that group can be compared to the group(s) that did get the treatment.
Non-experimental design in which the researcher doesn’t manipulate anything, but just observes & records what is going on. Some of these are descriptive, correlational, case, or cohort study designs for example.
One particularly interesting “experimental” design is one in which 1 or 2 of the experimental design ideal requirements as listed above are missing. These are called quasi-experimental designs.
In a quasi experimental design
The researcher manipulates some variable, but….
Either the participants are NOT randomly assigned to groups
&/OR there is no control group.
A quasi-experimental design is not as strong as a true experiment in showing that the manipulated variable X causes changes in the outcome variable Y. For example, a true experimental study with manipulation, randomization, and a control group would create much stronger evidence that hospital therapy dogs really reduced patient pain and anxiety. We would not be as confident in the results of a quasi-experimental design examining the exact same thing. In the next blog, we’ll examine why.
Critical thinking: Go to PubMed & use search terms “experiment AND nurse” (without the quotation marks). Open an interesting abstract and look for the 3 elements of a classic experimental design. Now look for “quasi experiment AND nurse” (without the quotation marks.) See what element is missing!
[After watching video: Note that the variable that is controlled by the researcher is call the Independent variable or Cause variable because it creates a change in something else. That something else that changes is the Dependent variable or Outcome variable.]
Based on the video, can you explain why true experiments are often called randomized controlled trial (RCT)?
Question: What is a randomized controlled trial (RCT)?And why should I care?
Answer: An RCT is one of the strongest types of studies in showing that a drug or a treatment actually improves a symptom or disease. If I have strep throat, I want to know what antibiotic works best in killing the bacteria, & RCTs are one of the best ways to find that answer.
In the simplest kind of RCT, subjects are randomly assigned to 2 groups. One group gets the treatment in which we are interested, & it is called the experimental group. The other group gets either no treatment or standard treatment, & it is called the control group.
Here’s an example from a study to determine whether chewing gum prevents postoperative ileus after laparotomy for benign gynecologic surgery: A total of 109 patients were randomly assigned to receive chewing gum (n=51) or routine postoperative care (n=58). Fewer participants assigned to receive chewing gum … experienced postoperative nausea (16 [31.4%] versus 29 [50.0%]; P=0.049) and postoperative ileus (0 vs. 5 [8.6%]; P=0.032).* There were no differences in the need for postoperative antiemetics, episodes of postoperative vomiting, readmissions, repeat surgeries, time to first hunger, time to toleration of clear liquids, time to regular diet, time to first flatus, or time to discharge. Conclusion? Postop gum chewing is safe & lowers the incidence of nausea and ileus! (Jernigan, Chen, & Sewell, 2014. Retrieve from PubMed abstract)
Do you see the elements of an RCT in above?
Let’s break it down.
Randomizedmeans that 109 subjects were randomly divided into 2 or more groups. In above case, 51 subjects ended up in a gum chewing group & 58 were assigned to a routine care, no gum group. Randomization increases the chance that the groups will be similar in characteristics such as age, gender, etc. This allows us to assume that different outcomes between groups are caused by gum-chewing, not by differences in group characteristics.
Controlled means that 1 of the groups is used as a control group. It is a comparison group, like the no-gum , standard care group above
Trial means that it was a study. The researchers were testing (trying) an intervention and measuring the outcomes to see if it worked. In this case the intervention was gum chewing and the measure outcomes were nausea and ileus.
Why should you care about RCTs? Because RCTs are strong evidence that an intervention works (or doesn’t) for your patients
Critical Thinking Exercise: Go to http://www.ncbi.nlm.nih.gov/pubmed In the blank box at the very top enter a few key words about the problem in which you are interested + RCT. For example: music pain + RCT. Then read 1 or more of the abstracts looking for random assignment (randomized), control group, and whether it was a study (trial). You’re on your way! -Dr.H
*Note: You may remember from other blogs that p<.05 means the difference between groups is probably cause by the intervention—in this case gum chewing.