An introduction to randomised controlled trial statistics
Randomised controlled trials (RCTs) are a gold standard in medical research to test the efficacy and safety of new interventions. An RCT is a prospective, comparative, quantitative study/experiment performed under controlled conditions with random allocation of interventions to comparison groups. Interpretation of RCTs requires some basic knowledge of statistics to ensure that the findings are valid and reliable. In this blog post, we will discuss some of the basic statistics used when interpreting an RCT.
1. Sample size and power analysis
Sample size is a critical factor in the design of an RCT. The larger the sample size, the more reliable the results of the study. Power analysis is used to determine the optimal sample size for an RCT. Power analysis is a statistical method that calculates the probability of detecting a significant difference between the treatment and control groups. The power of an RCT should be at least 80% to ensure that the study can detect a clinically meaningful difference.
2. Baseline characteristics
Before randomization, baseline characteristics of participants should be similar in the treatment and control groups. This ensures that any differences observed between the groups are due to the treatment and not other factors. The baseline characteristics are usually reported as mean and standard deviation for continuous variables and as percentages for categorical variables.
3. Intention-to-treat analysis
Intention-to-treat (ITT) analysis is a statistical method that includes all randomized participants in the analysis, regardless of whether they received the treatment or not. ITT analysis is the most appropriate method to assess the efficacy of an intervention because it preserves the randomization and avoids bias. ITT analysis provides an estimate of the effect of the intervention on the population, not just those who completed the treatment.
4. Confidence intervals and p-values
Confidence intervals (CI) and p-values are statistical measures used to assess the significance of the results. CI is a range of values that is likely to contain the true effect size with a certain degree of confidence. The commonly used level of confidence is 95%. A CI that does not include zero indicates a statistically significant result.
P-value is a measure of the strength of evidence against the null hypothesis. The null hypothesis is the hypothesis that there is no difference between the treatment and control groups. A p-value of less than 0.05 is generally considered statistically significant, indicating that the null hypothesis can be rejected.
5. Effect size
Effect size is a measure of the magnitude of the difference between the treatment and control groups. It is usually expressed as a standardized mean difference or odds ratio. Effect size provides an estimate of the clinical significance of the intervention. A larger effect size indicates a more significant effect of the intervention.
6. Subgroup analysis
Subgroup analysis is a statistical method used to assess whether the effect of the intervention varies across different subgroups of participants. Subgroup analysis should be pre-specified in the protocol to avoid spurious findings due to multiple testing.
Conclusion
Many other factors, not just statistical terms outlined here, are also important when critically appraising results of a study. Study design, randomisation and blinding, bias and confounding as well as applicability to your setting and population all need to be considered.
There are many good resources to help with basic statistical interpretation including this BMJ Best Practice e-learning module. Practice makes progress so keep attending journal club and attempting to critically appraise important research papers in your field!
