Compulsory bovine tuberculosis testing has been implemented since 1959 in NorthernIreland. Initial rapid progress in the eradication of the disease was followed by a situationwhere disease levels tended to fluctuate around a low level. This study explores recru-descence of bovine tuberculosis (bTB) in Northern Ireland herds by assessing risk factorsassociated with time from the six-month post-outbreak skin test until a further herd break-down. Bovine herds (n = 3377) were recruited in 2002 and 2003 and their survival analysedusing Kaplan–Meier survival estimates and a Cox proportional hazards model, with follow-up extending to August 2008. Exclusion criteria applied for study entry were bTB infection ina contiguous herd, changing of post restriction test to one of a higher risk status or chronicinfection. Chronic infection was defined as any situation where disclosure preceded thepost-outbreak test by two years or more. The application of these exclusion criteria meantthat herds recruited to the study were largely cleared of infection and not directly con-tiguous to other infected herds. Of the 3377 herds, 1402 (41.5%) suffered a further herdbreakdown before the end of follow-up. Median survival time was 582 days (interquartilerange = 336–1002 days). Breakdown severity (defined as the number of Single Intrader-mal Comparative Tuberculin Test (SICTT) reactors at disclosure test), local bTB prevalence,herd size and type were identified as significant risk factors (p < 0.05), as was the purchaseof higher numbers (n > 27.38 per year) of cattle. Consistent with other studies this workshows bTB confirmation to not be predictive of a future herd breakdown. This work showsbTB history as not being a risk factor for a future breakdown. This result could be reflectiveof the exclusion criteria used in the study, which may have selected for incidents wherehistorical status was of less importance.