A retrospective ecological study was undertaken to identify social and envi
ronmental factors associated with increased incidence of meningococcal dise
ase in university halls of residence. A standardized questionnaire was sent
to UK universities and colleges of higher education outside London, for di
stribution to halls containing at least 50 students. Incidence rate ratios
of invasive meningococcal disease were obtained for a range of social and e
nvironmental variables. Multi-variable Poisson regression analysis identifi
ed 3 factors as having a strong association: a high proportion of first yea
r undergraduate residents (P = 0.0008), decreasing smokiness of the hall ba
r (P < 0.0001), and opening of hall bar before 1990 (P = 0.0001). The inver
se relationship between disease incidence and smokiness of bars was an unex
pected finding, and may be due to confounding factors. Universities should
continue to promote awareness of meningococcal disease, encourage vaccinati
on of first year students against serogroup C disease, and where appropriat
e, take measures to reduce overcrowding.