Table 4.

Overall calibration and discrimination of the full and simple MSM HIV acquisition prediction models applied to Public Health—Seattle & King County repeat STD clinic testers (2001–2008) and Project Explore control arm participants (1999–2003).

Model and Sample	Expected Annual Incidence, %	Observed Annual Incidence, %	Expected-Observed Incidence Ratio (95% CI)*	1-Year AUC (95% CI)	4-Year AUC (95% CI)
Full Model
PHSKC STD Clinic	2.58	2.57	1.01 (0.96, 1.06)	0.72 (0.57, 0.79)	0.71 (0.62, 0.75)
Explore Control Arm	2.35	2.32	1.01 (0.97, 1.06)	0.68 (0.62, 0.75)	0.67 (0.62, 0.71)
Simple Model
PHSKC STD Clinic	2.62	2.57	1.02 (0.98, 1.07)	0.71 (0.57, 0.77)	0.69 (0.60. 0.74)
Explore Control Arm	2.34	2.32	1.01 (0.97, 1.05)	0.67 (0.60, 0.75)	0.66 (0.61, 0.71)

CI, confidence interval; AUC, area under the curve; PHSKC, Public Health—Seattle & King County.

^*Expected incidence divided by observed incidence. The observed annual incidence is the actual rate observed in the Public Health—Seattle & King County repeat testers and the control arm of Project Explore. The expected rate is the average of the risk predicted by the full and simple MSM HIV acquisition prediction models.