Prospective multicenter randomized patient recruitment and sample collection to enable future measurements of sputum biomarkers of inflammation in an observational study of cystic fibrosis

Table 2 Bootstrapped Power Calculations^a

Row	Statistical Model	Outcome Variable	Biomarker, Concentration or Activity	Percent Power	α-Level	Estimate of Patients Required
1	Proportional Hazards	Time to first Pulmonary Exacerbation	HMGB-1	90	0.01	40
2	Proportional Hazards	Time to Lung Transplantation or Death	HMGB-1	90	0.01	30
3	Linear Regression	Acute FEV₁% Drop with onset of a Pulmonary Exacerbation	GMCSF	80	0.01	175
4	Linear Regression	Acute FEV₁% Drop with onset of a Pulmonary Exacerbation	GMCSF	85	0.05	125
5	Linear Regression	FEV₁% Drop over 2 Years	NE	80	0.05	32

^aBased on prior published results, we sampled patients and bootstrapped statistical models of predictor biomarkers for outcome variables, setting percent power and α-level in order to estimate the number of patients similar to prior patients needed to detect associations. See the study protocol for more details (Additional file 5)

ISSN: 1471-2288