Quasi-experiments are a lot of work, yet don’t have the same scientific powerto show cause and effect, as do randomized controlled trials (RCTs).An RCT would provide better support for any hypothesis that X causes Y. [As a quick review of what quasi-experimental versus RCT studies are, see “Of Mice & Cheese” and/or “Out of Control (Groups).”]
So why do quasi-experimental studies at all? Why not always do RCTs when we are testing cause and effect? Here are 3 reasons:
#1 Sometimes ETHICALLY the researcher canNOT randomly assign subjects to a control and an experimental group. If the researcher wants to compare health outcomes of smokers with non-smokers, the researcher cannot assign some people to smoke and others not to smoke! Why? Because we already know that smoking has significant harmful effects. (Of course, in a dictatorship, by using the police a researcher could assign them to smoke or not smoke, but I don’t think we wanna go there.)
#2 Sometimes PHYSICALLY the researcher canNOT randomly assign subjects to control & experimental groups. If the researcher wants to compare health outcomes of
individuals from different countries, it is physically impossible to assign country of origin.
#3 Sometimes FINANCIALLY the researcher canNOT afford to assign subjects randomly to control & experimental groups. It costs $ & time to get a list of subjects and then assign them to control & experimental groups using random numbers table or drawing names from a hat.
Thus, researchers sometimes are left with little alternative, but to do a quasi-experiment as the next best thing to an RCT, then discuss its limitations in research reports.
Critical Thinking: You read a research study in which a researcher recruits the 1st 100 patients on a surgical ward January-March quarter as a control group. Then the researcher recruits the 2nd 100 patients on that same surgical ward April-June for the experimental group. With the experimental group, the staff uses a new, standardized pain script for better pain communications. Then the pain communication outcomes of each group are compared statistically.
Is this a quasi-experiment or a randomized controlled trial (RCT)?
What factors (variables) might be the same among control & experimental groups in this study?
What factors (variables) might be different between control & experimental groups that might affect study outcomes?
How could you design an ethical & possible RCT that would overcome the problems with this study?
Why might you choose to do the study the same way that this researcher did?
In the last “Quasi-wha??” blogpost, I described 1 type of experimental design: Quasi-experimental. To review… In quasi-experimental designs, the researcher manipulates some variable, but either 1) doesn’t randomly assign subjects to a control and experimental group OR 2) doesn’t have a control group at all.
For example, the researcher may introduce pet therapy on unit #1 and avoid pet therapy on unit #2 and then afterwards compare the anxiety levels of patients on the 2 units. That study has a control group (unit #2), but because patients weren’t (& probably couldn’t be) randomly assigned to the units, this would be a quasi-experimental study. The control group in this pet therapy case is what researchers call a “non-equivalent control group.” Non-equivalent means the groups are different in ways that might affect study results! [Note: For review of what constitutes a true experimental study see first part of “Quasi-wha??” blogpost.]
Herein lies a weak link in the cause-and-effect chain. Quasi- designs are NOT as strong as true experimental designs because something other than our treatment (in this case pet therapy) may have created any difference in outcomes (e.g., anxiety levels). Why? Here’s your answer.
In an experimental study, randomly assigning subjects to a
control and a separate experimental group means that all the little, variable weirdities of all subjects are equally distributed to each group. Each group is the same mix of different types of people. This means we can assume that both groups are the exact same type of
people in regard to things that may influence study outcomes, such as attitudes, values, preferences, beliefs, anxiety level, psychology, physiology and so on.
In contrast, in the quasi-experimental pet therapy example above, there is probably something that caused a certain type of person to be on unit #1 and a different type to be on unit #2. Maybe it was their diagnosis, their doctor, their type of surgery, or other. Thus, we cannot assume that people in unit #1 and unit #2 groups are the same before pet therapy, and so any differences between them after pet therapy might have already existed.
So why do quasi-experimental studies at all?? There are great reasons! Stay tuned for next blogpost.
In a prior blog (Direct speaking about INdirect outcomes: HCAHPS as a measurement*), I argued that HCAHPS questions were indirect measures of outcomes. Indirect measures are weaker than direct measures because they are influenced by tons of variables that have nothing to do with the outcome of interest. But wait!! There’s more! HCAPS can sometimes be a DIRECT measure; it all depends on what you want to know.
(If you know this, then you are way ahead of many when it comes to measuring outcomes accurately!!)
If your research question is what do patients remember about hospitalization then HCAHPS is a DIRECT measure of what patients remember.
However if your research question is what did hospital staff actually do then HCHAPS is an INDIRECT* measure of what staff did.
Example:HCAHPS question #16 is, “Before giving you any new medicine, how often did hospital staff tell you what the medicine was for?” Whether the patient answers yes or no, the response tells us only how the patient remembers it.
Why is this important?
Because if you want to know whether or not RNs actually taughtinpatients about their medications, then for the most direct & accuratemeasure you will have to observe RNs .
However, if you want to know whether patients remember RNs teachingthem about discharge medications, then HCAHPS question #16 is one of the most direct & accurate measure of what they remember.
Is pain experience as diverse as our populations? This week I came across an interesting meta-analysis.
A meta-analysis (MA) is one of the strongest types of evidence there is. Some place it at the top; others, 2nd after evidence-based clinical practice guidelines. (For more on strength of evidence, click here.)
MA is not merely a review of literature, but is a statistical integration of studies on the same topic. MA that is based on integration of randomized controlled trials (RCTs) or experimental studies is the strongest type of MA. MA based on descriptive or non-experimental studies is a little less strong, because it just describes things as they seem to be; & it cannot show that one thing causes another.
Main findings?“AAs [African Americans], Asians, and Hispanics had higher pain sensitivity compared to NHWs [non-Hispanic Whites], particularly lower pain tolerance, higher pain ratings, and greater temporal summation of pain.” (https://www.ncbi.nlm.nih.gov/pubmed/27682208) (Temporal summation is the increase in subjective pain ratings as a pain stimulus is repeated again and again.)
Critical thinking:Given that this is a well-done meta-analysis and that the pain was created by researchers in each study, how should this change your practice? Or should it? How can you use the findings with your patients? Should each patient be treated as a completely unique individual? Or what are the pros & cons of using this MA to give us a starting point with groups of patients? [To dialogue about this, comment below.]
For more info?Request the full Kim et al. article via interlibrary loan from your med center or school library using reference above. It is available electronically pre-publication. Also check out my blog on strength of different types of evidence.
[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)?
If you want to know what someone thinks or feels, you ask them, right?
The same is true in research, but it is good to know the pros and cons of using the “self-report method” of collecting data in order to answer a research question. Most often self-report is done in ‘paper & pencil’ or SurveyMonkey form, but it can be done by interview.
Generally self-report is easy and inexpensive, and sometimes facilitates research that might otherwise be impossible. To answer well, respondents must be honest, have insight into themselves, and understand the questions. Self-report is an important tool in much behavioral research.
But, using self-report to answer a research question does have its limits. People may tend to answer in ways that make themselves look good (social desirability bias), agree with whatever is presented (social acquiescence bias), or answer in either extreme terms (extreme response set bias) or always pick the non-commital middle numbers. Another problem will occur if the reliability and validity of the self-report questionnaire is not established. (Reliability is consistency in measurement and validity is the accuracy of measuring what it purports to measure.) Additionally, self-reports typically provide only a)ordinal level data, such as on a 1-to-5 scale, b) nominal data, such as on a yes/no scale, or c) qualitative descriptions in words without categories or numbers. (Ordinal data=scores are in order with some numbers higher than others, and nominal data = categories. Statistical calculations are limited for both and not possible for qualitative data unless the researcher counts themes or words that recur.)
An example of a self-report measure that we regard as a gold standard for clinical and research data = 0-10 pain scale score. An example of a self-report measure that might be useful but less preferred is a self-assessment of knowledge (e.g., How strong on a 1-5 scale is your knowledge of arterial blood gas interpretation?) The use of it for knowledge can be okay as long as everyone understands that it is perceived level of knowledge.
Critical Thinking: What was the research question in this study? Malaria et al. (2016) Pain assessment in elderly with behavioral and psychological symptoms of dementia. Journal of Alzheimer’s Disease as posted on PubMed.govat http://www.ncbi.nlm.nih.gov/pubmed/26757042with link to full text. How did the authors use self-report to answer their research question? Do you see any of the above strengths & weaknesses in their use?
For more information: Be sure to check out Rob Hoskins blog: http://www.sciencebrainwaves.com/the-dangers-of-self-report/