Abstract
Most HIV behavioural surveillance programmes for gay, bisexual and other men who have sex with men (MSM) sample from location-based (offline) or web-based (online) populations, but few combine these two streams. MSM sampled online have been found to differ demographically and behaviourally from those sampled offline, meaning trends identified in one system may not hold for the other. The aim was to examine trends among MSM responding to supplementary repeat online behavioural surveillance surveys who had not participated in offline surveillance earlier that year in the same city, to see whether trends were parallel, converged or diverged. We recruited a total of 1613 MSM from an Internet dating site in Auckland, New Zealand in 2006, 2008 and 2011 using identical questionnaires and eligibility criteria to offline surveillance. Condom use was stable over time, HIV testing rates rose, the proportion reporting over 20 recent male partners declined, and anal intercourse rates increased, consistent with trends in offline surveillance conducted concomitantly and reported elsewhere. Variant trends included greater stability in condom use with casual partners among online-recruited MSM, and a rise in regular fuckbuddy partnering not identified among offline-recruited MSM. Among MSM recruited online, the frequency of checking Internet dating profiles increased between 2008 and 2011. In conclusion, supplementary web-based behavioural surveillance among MSM generally corroborates trends identified in offline surveillance. There are however some divergent trends, that would have been overlooked if only one form of surveillance had been conducted. As MSM populations increasingly shift their socialising patterns online and diversify, multiple forms of HIV behavioural monitoring may be required.