Why be Critical? Flashcards
What is critical appraisal
systematically evaluating the validity, results and usefulness of a study for our patients.
What is validity
the degree to which the results of a study represent the true values that occur in a population.
What are the 2 types of validity that can be considered
Internal and external validity
What is internal validity
Internal validity is high when the design of the study makes it highly likely that the results reflect very closely the reality and is not influenced by any other variables than the one we measure.
What are confounding variables
These are variables other than the one we are measuring that affect the study
What is external validity
External validity is high when the results of the study are generalizable to the population it addresses.
Name some factors that affect external validity
Participants - only one economic background?
Setting - a place where the resources aren’t widely available
Operators - If only specialists can do it well probs aint that externally valid
Outcomes - How was the sensitivity etc. tested, is this realistic
Duration - less time = less reliable
Name some ways to increase external validity
- Random selection of patients
- Large sample
- Choose appropriate setting, general dental practices or third world countries etc
What is bias
Systematic errors in the design or reporting of a study that means the results do not the reality or “truth”
What types of bias are there
- Selection
- Performance
- Attrition
- Reporting
What is selection bias
This is when there is a systematic difference in the basic characteristics of the groups being compared.
e.g. one group is more motivated than the other, or one has less caries risk inherently
What is performance bias
This happens when the different groups get treated differently from each other
e.g. one group being checked on more than the other.
What is attrition bias
Attrition is the reduction in the number of participants which often happens in longer studies, but if those that drop out of the study are systematically different from those that remain in the study then there is a risk of attrition bias
e.g. 20% of group have allergic reaction to toothpaste and dip. if we didn’t know this then the toothpaste results would look better than they actually are.
What is reporting bias
This is when some outcomes are systematically reported whilst others are not.
e.g. extraction technique comparisons, researchers measured pain and bleeding but only found difference in bleeding and only reported that, then that’s fooked
