5. Optimal information size calculations: fixed‐dose combinations versus single‐drug formulations (Comparison 1).

Outcomes	Assumed risk	Clinically important reduction		Optimal sample size 1,2
	Single‐drug formulations	Absolute	Relative
Treatment failure	2.2 %	0.5%	25%	6092
Relapse	2.3 %	0.5%	25%	4718
Death3	0.9 %	0.1%	4.5%	737,340
Sputum/culture conversion at end of treatment	88.7%	0.5%	0.6%	95,044
Serious adverse events	1.5 %	0.1%	6.7%	12,356
Adverse events leading to discontinuation of therapy	4.1 %	0.5%	24.4%	325,024

¹We based all calculations are based on: 1‐sided tests, with a ratio of 1:1, power of 0.9, and confidence level of 0.05.
 ²We performed all calculations using: http://www.sealedenvelope.com/power/binary‐noninferior/.
 ³ If there is truly no difference between the standard and experimental treatment, then 737,340 participants are required to be 90% sure that the upper limit of a 1‐sided 95% confidence interval (CI) (or equivalently a 90% 2‐sided CI) will exclude a difference in favour of the standard group of more than 0.1%.