Issues of methods and interpretation in the National Cancer Institute formaldehyde cohort study

Table 2 Rate ratios and numbers of deaths for all causes combined by NCI update and peak FA exposure ^f

NCI update	Peak exposure (ppm)								p-value trend^d.	p-value trend^e.
	0		>0 - < 2.0		2.0 - < 4.0		≥ 4.0
	RR 95% CI	No. deaths	RR 95% CI	No. deaths	RR 95% CI	No. deaths	RR 95% CI	No. deaths
NCI 1994 update^a.	1.05	1,991	1.00 Baseline	2,554	1.21*	1,945	1.07*	1,996	0.013	0.014
NCI 1994 update (corrected)^b.	1.02 0.85-1.04	2,169	1.00 Baseline	3,201	0.96 0.91-1.02	2,012	0.86* 0.81-0.91	2,099	<0.001	<0.001
NCI 2004 update^c.	0.98 0.93-1.04	2,605	1.00 Baseline	4,996	0.95* 0.90-0.99	3,096	0.85* 0.81-0.89	3,254	<0.001	<0.001

^a.Hauptmann et al.[20].
^b.Beane-Freeman et al.[8].
^c.Beane-Freeman et al.[6].
^d.Two-sided likelihood ratio test (1df) of zero slope for continuous formaldehyde exposure among exposed person-years only.
^e.Two-sided likelihood ratio test (1df) of zero slope for continuous formaldehyde exposure among unexposed and exposed person-years.
^f.FA exposure metric calculated using 15-year time lag.
*95% CI does not include 1.00.

ISSN: 1745-6673