Background Measured incidence rates of Sexually Transmitted Infections (STIs) are substantially higher among blacks compared to whites in the US. However, population-based data sources, such as routine STI surveillance systems, often lack information on race/ethnicity for the majority of cases. We compared black-white differences in gonorrhoea rates using both individual and neighbourhood-level measures of race/ethnicity.

Methods Publicly-available aggregate data on the number of New York City(NYC) adult gonorrhoea cases in 2009 by race/ethnicity and sex were obtained from the NYC Department of Health and Mental Hygiene’s online system for NYC’s 42 neighbourhoods. The proportion of black residents for each neighbourhood was obtained from census data.

Results The citywide gonorrhoea rate in 2009 was 81 per 100,000 (median for 42 neighbourhoods [median]: 29; intraquartile range across 42 neighbourhoods [IQR]: 2–105); 116 per 100,000 for males (median:44; IQR: 11–177); and 47 per 100,000 for females (median: 10; IQR: 0–67). Race/ethnicity data were missing for 49% of cases (median: 48%; IQR: 36%–61%). Using data on cases with complete information on race/ethnicity, gonorrhoea rates were 225 per 100,000 (median: 204; IQR: 108–321) among non-Hispanic blacks, compared with 33 per 100,000 (median: 6; IQR: 1–37) among non-Hispanic whites. The median black-white difference in gonorrhoea diagnosis rates was 104 per 100,000, and varied substantially across neighbourhoods (IQR: 23–223). Neighborhoods in the lowest quartile of the proportion of black residents (where 0.1%–0.4% of neighbourhood residents were black) had the lowest mean gonorrhoea rate 33 per 100,000, compared to 81, 162, and 224 in the 2^nd, 3^rd and 4^th quartiles (where 35%–76% of neighbourhood residents were black), respectively (p < 0.001).

Conclusions Analyses using individual-level and neighbourhood-level race/ethnicity measures both suggest substantial black-white differences within and across NYC neighbourhoods. More complete race/ethnicity information among persons diagnosed with gonorrhoea is critical to elucidate possible structural/neighbourhood determinants of black-white differences.