Appraising therapeutic study (Part 2)

Now that we’ve assessed the study’s validity and methodology, the final step is to interpret the results.

This helps us answer the key question:

What do the results mean?

Understanding study results is crucial for determining the magnitude and significance of an intervention’s effect. Let’s break down the essential concepts used in clinical trials:

Key Measures in Clinical Trials

Results are often expressed using various statistical measures, such as:

• Relative Risk (RR) – Compares the probability of an outcome between two groups.

• Relative Risk Reduction (RRR) – The percentage reduction in risk due to the intervention.

• Absolute Risk Reduction (ARR) – The actual difference in risk between groups.

• Number Needed to Treat (NNT) – The number of patients who need to be treated to prevent one additional outcome.

• Number Needed to Harm (NNH) – The number of patients who would need to be treated for one to experience harm.

• Number Needed to Screen (NNS) – The number of people who must be screened to detect one case of disease.

Example Calculation

Imagine a study where 20% of participants in the placebo group develop a disease, compared to 15% in the treatment group:

Absolute Risk Reduction (ARR)

[ ARR = text{Risk in Placebo Group} – text{Risk in Treatment Group} ] [ = 20% – 15% = 5% ]

This means the treatment reduces the disease risk by 5%.

Relative Risk (RR)

[ RR = frac{text{Risk in Treatment Group}}{text{Risk in Placebo Group}} ] [ = frac{15%}{20%} = 0.75 ]

A Relative Risk of 0.75 means the intervention lowers the risk compared to the placebo.

Relative Risk Reduction (RRR)

[ RRR = 1 – RR ] [ = 1 – 0.75 = 0.25 text{ or } 25% ]

This means the disease risk is reduced by 25% in the treatment group compared to the placebo group.

Number Needed to Treat (NNT)

[ NNT = frac{1}{ARR} ] [ = frac{1}{0.05} = 20 ]

This means 20 patients need to be treated to prevent one additional case of the disease.

Precision of Results: Confidence Intervals (CIs)

To determine if the results are statistically significant, we use Confidence Intervals (CIs). A 95% CI provides a range within which the true effect likely falls 95% of the time.

Key Considerations for CIs:

• If the CI includes 1 for ratios like RR or Odds Ratio (OR) → the result is not statistically significant.

Example:

• RR = 0.75, 95% CI (0.60 to 0.90) → Significant (Intervention reduces risk).

• RR = 0.90, 95% CI (0.90 to 1.10) → Not significant (No clear effect).

Handling Dropouts & Noncompliance

Not all participants complete the study as planned. To address this, researchers use different analysis methods:

✅ Intention-to-Treat (ITT) Analysis

• Includes all participants in their original groups, even if they dropped out or didn’t follow the protocol.

• Preserves randomization, ensuring fair comparisons.

• Reflects real-world clinical practice, where patients may not strictly follow treatments.

❌ Per-Protocol Analysis

• Includes only participants who followed the study protocol.

• Can introduce bias by excluding dropouts, leading to overestimation of the treatment effect.

Key Principle: Once a participant is assigned to a group, they should always be analysed in that group, regardless of compliance.

Key Takeaways:

• Assess effect size using ARR, RRR, and NNT.

• Check statistical significance with Confidence Intervals (CIs).

• Look for ITT analysis to ensure real-world applicability.

By carefully interpreting study results, we can make evidence-based decisions and determine the relevance and reliability of research findings.

What’s Next? In the next session, we’ll explore how to critically appraise diagnostic studies. Stay tuned!