Table 2. Contrasts in honey-bee pathogen abundance by colony status.

	All CCD colonies vs. all non-CCD colonies				Weak vs. strong colonies in non-CCD apiaries
Target	ΔΔCT	SE	P-value	Fold change	ΔΔCT	SE	P-value	Fold change
ABPV	+2.23	0.96	0.02	4.57	+0.84	0.87	0.34	1.77
BQCV	+2.81	1.08	0.01	6.67	−1.70	1.59	0.29	0.32
DWV	+3.90	1.15	<0.01	14.26	−0.07	0.61	0.96	0.95
IAPV	+0.22	0.60	0.72	1.15	+1.64	0.98	0.11	2.83
KBV	+2.58	0.78	<0.01	5.49	+0.67	0.67	0.32	1.56
SBV	+0.28	0.64	0.66	1.27	+0.36	1.52	0.81	1.36
N. ceranae	+1.19	1.22	0.33	1.85	+1.70	1.59	0.29	2.41
N. apis	+3.94	1.07	<0.01	20.97	−0.03	0.14	0.81	0.98
Crithidia	+2.62	1.12	0.02	6.15	−0.79	1.95	0.69	0.58

The difference in mean ΔC_T values (normalized threshold cycle in qPCR reactions) was compared by ANOVA for two non-independent contrasts: all CCD colonies (n = 61) versus all non-CCD colonies (n = 63), and weak (n = 15) versus strong (n = 29) colonies in non-CCD apiaries. Weak colonies had six or fewer frames of bees and strong colonies had seven or more frames (see Materials and Methods). Non-CCD colonies include both sympatric and allopatric colonies, which were combined for increased statistical power. ΔΔ is the mean Δof the non-CCD population minus the mean Δvalue of the CCD population. Thus, positive numbers represent a decrease in mean threshold cycle and an increase in pathogen abundance. Fold change between categories is calculated as (1+ primer efficiency) ^ΔΔC _T. SE = standard error of population mean ΔC_T; P-value = probability of equal mean ΔC_T by ANOVA (i.e., that the true ΔΔC_T = 0); ABPV = acute bee paralysis virus; DWV = deformed wing virus; SBV = sacbrood virus; BQCV = black queen cell virus; IAPV = Israeli acute paralysis virus; KBV = Kashmir bee virus.