Sample size requirements for separating out the effects of combination treatments: Randomised controlled trials of combination therapy vs. standard treatment compared to factorial designs for patients with tuberculous meningitis
2011

Sample Size Requirements for Combination Treatments in Tuberculous Meningitis Trials

Sample size: 750 publication 10 minutes Evidence: moderate

Author Information

Author(s): Wolbers Marcel, Heemskerk Dorothee, Chau Tran Thi Hong, Yen Nguyen Thi Bich, Caws Maxine, Farrar Jeremy, Day Jeremy

Primary Institution: Hospital for Tropical Diseases, Wellcome Trust Major Overseas Programme and Oxford University Clinical Research Unit, Ho Chi Minh City, Vietnam

Hypothesis

Current anti-mycobacterial regimes are not potent enough and increasing the levels of effective anti-mycobacterial drugs in the cerebrospinal fluid will improve clinical outcome.

Conclusion

A pragmatic two-arm trial of combination therapy versus standard treatment is the preferred design for tuberculous meningitis due to the excessive sample size required for factorial designs.

Supporting Evidence

  • The study found that an 8-fold increase in sample size is needed for factorial designs to achieve adequate power.
  • Factorial designs may miss important interactions between treatments.
  • Combination trials are more feasible in diseases with limited patient pools.

Takeaway

This study shows that testing two drugs together for treating a serious illness needs a lot more patients than just testing one drug, making it harder to do.

Methodology

The study compared the power of simple randomized trials versus factorial designs using sample size calculations based on expected mortality rates.

Potential Biases

The reliance on assumptions for sample size calculations may introduce bias if the actual treatment effects differ.

Limitations

The study assumes no interactions between treatments, which may not hold true in practice.

Participant Demographics

Patients with tuberculous meningitis.

Statistical Information

P-Value

0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1745-6215-12-26

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication