Figure 2

Sampling-adjusted seroprevalence measured by different assays, stratified by lifetime number of sexual partners and age at enrollment. Legend: Seroprevalence vs. number of partners (Panels A and C) and age (Panels B and D) is shown for HPV16 (Panels A-B) and HPV18 (Panels C-D). Lines represent seroprevalence by GST-L1 at the laboratory suggested seropositivity cutoff (solid line), GST-L1 at alternate cutoffs (dashed line), VLP-ELISA (solid line with diamond markers), cLIA (dashed line with triangle markers), and SEAP-NA (solid line with circle markers, HPV16 only). Alternate GST-L1 cutoffs were calculated to maximize agreement with VLP-ELISA (see Statistical Methods). Odds of GST-L1 seropositivity at the laboratory cutoff increased with each year of age for HPV16 (OR 1.20, 95% CI 1.03-1.40) and HPV18 (OR 1.20, 95% CI 1.04-1.38). Similarly, odds increased with each sexual partner for HPV16 (OR 2.06, 95% CI 1.49-2.83) and HPV18 (OR 1.59, 95% CI 1.18-2.15). For HPV18, one observation with a large weight was excluded for figure only.