Research Methodology 3: RCTs and comparing means Flashcards
What is a Randomised controlled trial ?
a study in which participants are allocated randomly between an intervention
(e.g. treatment) and a control group (e.g. no treatment or standard treatment)
What is a Confounder?
A confounder is a variable that influences both the dependent variable and independent variable causing a spurious association.
What is bias - in terms of statistics?
‘a tendency of an estimate to deviate in one direction
from a true value’
What does randomisation ensure?
Confounders will be distributed evenly between the samples
What does equipoise provide?
ethical basis for medical research that involves
assigning patients to different treatment arms of a clinical trial.
What is clinical equipoise?
genuine uncertainty in the expert medical community over whether one treatment will be more beneficial
than another - not ethical if one is known to be better
What is selection bias and when can it occur?
Refers to representativeness of sample to wider population
can happen when participants are asked to volunteer for a study
Systematic differences between baseline characteristics of groups that are compared
What is performance bias?
Systematic differences between groups in the care that is provided, or in exposure to factors other than the interventions of interest
What is attrition bias?
Systematic differences between groups in withdrawals from a study e.g. one
treatment has more side effects than another
What is observer/Detection bias?
- Outcome measure does not adequately capture outcome of interest
- Systematic differences between groups in how outcomes are determined
What are 2 ways attrition bias can be overcome?
• Intention to treat analysis – analysed in treatment group they were
randomised to, whatever happens later.
• On-treatment analysis or per-protocol analysis – only analyse
patients who finish the treatment according to the study protocol.
What assumptions do parametric tests make?
normally distributed
Give 2 examples parametric tests
t-test
ANOVA
What are non-parametric tests and give examples?
Described as “Distribution-free tests” as they make fewer assumptions
The order (or "rank") of the values is used rather than the actual values themselves
Examples: Mann-Whitney tests
Kruskal Wallis
Spearman’s Rho
What can the Mann-Whitney U test (Two-sample Wilcoxon ranksum test) be used to compare?
non-normally
distributed continuous variables between groups
What can a t-test be used to compare and what are the 5 assumptions made?
continuous variables between groups/quantitative data set comparing the means of two groups if the following assumptions are met:
- Randomly sampled
- Independent observations
- normally distributed
- Variances for each group are equal
- large sample
What is categorical data?
(nominal; binary; dichotomous)
e.g. smoking,
What is a scale variable?
(continuous; interval)
numeric measurement e.g. height
What makes a good RCT?
• Internal validity – is the IV causing the DV in this study?
• External validity – to what extent can these findings be generalised to
other people, situations and times?
What is the confidence interval?
- Range within the true value lies
- The confidence interval describes the range of values with a given probability (e.g. 95%) that the true value of a variable is contained within that range.
What is the p value?
The p value is the probability that the difference observed could have occurred by chance if the groups compared were really alike.
What is blinding?
Depending on the level of blinding participants, staff or the study investigator is unaware of group allocations, Rx and hypothesis
What is Type 1 error?
There is no true difference but a difference is observed
False positive
When the null hypothesis is true but you reject it
E.g a man comes in but is told he is pregnant
What is a Type 2 error?
True negative predictive value
A true difference exists but is not observed (Type Two True)
E.g a pregnant women comes in but is told she is not pregnant
When would you use chi-squared test?
qualitative data set, categorical outcomes
When would you use Whitney U test?
quantitative data set where there in no normal distribution.
