PubMed Central, : RMMJ Rambam Maimonides Medical Journal Rambam Health Care Campus 2012 July; 3(3): e0014. ISSN: 2076-9172Published online 2012 July 31. doi: 10.5041/RMMJ.10082.

Table 1.

Comparison between biased time-fixed data analysis and corrected time-dependent data analysis for the cohort study of inhaled corticosteroid (ICS) use and all-cause mortality in chronic obstructive pulmonary disease (COPD).²⁷

	ICS Users (n = 12,000)			ICS Non-Users (n = 12,000)			Crude Rate Ratio (95% CI)

	Deaths	Person-Years	Rate per 100 Person-Years	Deaths	Person-Years	Rate per 100 Person-Years

Biased Time-Fixed Analysis
Immortal and unexposed person-time	0	1,500^*		0	0
At-risk person-time	1,000	7,500		1,400	9,000
Total	1,000	9,000	11.1	1,400	9,000	15.6	0.71 (0.66–0.77)

Corrected Time-Dependent Analysis
Immortal and unexposed person-time	0	0		0	1,500^*
At-risk person-time	1,000	7,500		1,400	9,000
Total	1,000	7,500	13.3	1,400	10,500	13.3	1.00 (0.92–1.08)

^* Bold values represent the person-years of cohort follow-up misclassified as ICS exposed in the time-fixed analysis and properly reclassified as unexposed in the time-dependent analysis.

RMMJ Rambam Maimonides Medical Journal Rambam Health Care Campus 2012 July; 3(3): e0014. ISSN: 2076-9172

Published online 2012 July 31. doi: 10.5041/RMMJ.10082