METHODS: Surveillance data were analyzed from a geographically representative probability sample of 3,790 gonorrhea cases reported in California between 2009-2011. Weighted generalized linear models were fit to model cumulative incidence ratios (CIR) for receipt of a non-CDC-recommended treatment regimen (as a binary outcome) with clinical setting the primary “risk” factor of interest. Potential confounders adjusted for in the analysis were: geography (operationalized by eight clusters geographically contiguous of counties) and whether or not the case was a man who has sex with men (MSM).
RESULTS: Overall, 16.7% (15.0, 18.3) of cases received a non-recommended treatment. The settings in which cases were significantly more likely to receive a non-recommended treatment (compared to STD clinics) were: military/VA facilities (CIR: 3.6; 95% CI: 2.0-6.7; p<0.001), emergency rooms/urgent care centers (2.4; 1.6-3.8; p<0.001), private physicians/HMOs (1.9; 1.4-2.7; p <0.001), and family planning facilities (1.7; 1.1-2.4; p=0.008). Additionally, cases treated in San Francisco were more likely to receive a non-recommended treatment (5.39; 3.4- 8.2; p<0.001) and, statewide, MSM cases were less-so (0.63; 0.47- 0.84; p=0.002). Region-specific models suggest some geographic variation in the effect of setting. Ceftriaxone 250 mg alone and azithromycin 1 g alone were the most commonly administered non-recommended treatments.
CONCLUSIONS: Barriers to adherence to the CDC treatment guidelines for GC appear to be greatest at military/VA facilities and emergency rooms/urgent care centers, and appear to also be experienced by private physicians/HMOs and family planning facilities. These patterns of risk may merit consideration in programming aimed at encouraging adherence to the treatment guidelines. Clarification of the importance of dual treatment with both ceftriaxone and azithromycin/doxycycline may help improve compliance.