Lecture 27: Randomized Controlled/ Clinical Trials Flashcards
What are some questions that are asked and can be answered with analytical studies?
-Should I take the time to get the xxx vaccine
-Will the xx vaccine keep me from getting sick
-Will the xxx vaccine help me get back to work faster by making me less sick
Recall the criteria we use to determine whether an observational or experimental study was better to use, if none of those apply which test is better to use? (harmful, expensive not practical etc)
-If none of those are issues with answering your hypothesis, then an experimental study is a great option bc they often have high evidence
What are experimental studies and what the 2 different kinds?
-Researcher intervenes and creates the groups (ex administering a placebo injection to the “control” group and a vit C injection to the “treated” group
-Highest on the “hierarchy of strength” of research evidence for single analytical studies (there are 2 higher but they involve multiple clinical trials) and that would be the blinded RCT’s
-can have 2 types of experimental
1. Lab-based experiment
2. Randomized controlled trial
What are laboratory experiments and advantages and disadvantages?
-done in a controlled lab setting (lab barn at dairy Guelph, as long as controlled environment not going to multiple farms)
-Researchers create a controllable environment to test hypotheses (there is a four on quantitative data (just numbers and compare stats)
-Provides the best control of manipulative conditions like tested variables and the environment
Advantages: reduce confounding, can be repeatable
Disadvantages: lack of external validity (can’t really relate to real world well bc not real world conditions lab conditions), unrealistic
What are randomized controlled trials (RCT) and what are the pros and cons?
-Also known as randomized clinical trials
- Researchers create the groups in a “real world” setting (ie no control over environment, want the groups to be as similar aside from the treatment of interest)
-The best form is a blind randomized controlled trial (usually researcher and individual are blind helps prevent the placebo effect)
Advantages:
-Useful for some therapeutic studies (antibiotics, anti0cancer drugs)
-Useful for preventative studies (vaccines, feed additives, herbicides)
Disadvantages:
-Ethics, cost
What is the randomized (blind) controlled trial process?
- Start with your hypothesis
- Identify your target and source population
- Randomly select your study population (REAL WORLD)
- Randomly assign your study pop to either the treatment (group A) or control (group B)
- Administer the appropriate treatments to each group
- Follow up to determine the outcome of each group in a blind manner (get better vs didn’t get better)
- Conduct statistical analyses to compare the groups (ie odds of recovery in each group)
- Interpret the findings and make conclusions
What are the randomized (blind) controlled trial considerations?
10 Important issues:
1. A clearly defined question
2. Is it ethical to intervene?
3. What is the reference (target) population?
4. What groups are needed?
5. How are the groups created?
6. Is there evidence of challenge?
7. What are the outcomes?
8. How big of a study?
9. Avoiding bias in the follow-up
10. The statistical analysis
For the randomized controlled trial considerations what is explain 1-5 in more detail?
- A clearly defined question
Negative
-does the new treatment work (is it better than nothing ie sugar pill)
Positive
-Is new treatment better than the existing? (1 group gets new other gets existing)
- Is it ethical to intervene?
-Does-response studies: start with low dose often come with side effects
-control group (risk of living vs dying not ethical to give nothing)
-Should the established treatments that have been with us for a long period of time be subjected to re-evaluation
-Informed consent (lots of ethics approval) - What is the reference (target) population?
-We want the study outcome to be similar to the study objective
-External validity and internal validity - What groups are needed?
-Control group is essential to test a hypothesis bc allows for comparison (use current controls not historical)
-Does NEW treatment work (negative control group)
-Is NEW treatment better than existing - How are the groups created?
-Randomization is very important, usually blinded
-Each individual should have an equal chance of being in either group, reduces unidentified bias, think about the best approach depending on the study population
-Treatment and control groups should be alike in every other way (to avoid confounding bias like age, socioeconomic)
-Can use stats to determine if there is a difference in non-studied factors b/w groups
-What is cross over design: a study design in the allocation of individuals to treatment groups
For the randomized controlled trial considerations what is explain 6-10 in more detail?
- Is there evidence of challenge?
-Big issue for vaccine clinical trials (correct pathogen, correct strain)
-Is the challenge effective - What are the outcomes?
-Clearly defined, objective is always better than subjective (facts vs opinions)
-Good outcomes are: practically, medically, biologically or economically important
-uses at least 2 outcome measures to verify findings - How big a study?
-How many per treatment group is enough
-Number of subjects in a trial increases with many things - Avoiding bias in the follow-up:
-“any effect at any stage of an investigation tending to produce results that depart from the TRUTH”
-ex’s seen in RCT
-differential admission bias (selection bias, are young animals as likely to come to referral vet hospitals as older animals?)
-Recall bias
-Misclassification bias (poor sn and/or sp of a test used in selecting patients or measuring the outcome)
-Follow-up period: BLINDING to reduce bias (blind, double blind, triple blind)
-Double blind (just subject and observer) is most common - The stat analysis
-Use right tools (t-test, chi-square, linear model etc)
-Interpret the results correctly (p-values, confidence intervals)
-Both stat and biological significance need to be taken into account (a little improvement even in life/death situation is better, could improves longitudinal life of a person)
