This paper presents a statistical model constructed using logistic reg
ression to identify those at high-risk of repeating parasuicide. The s
ubjects in the study are Cork city residents who exhibited parasuicida
l behaviour between 1 January and 30 June 1995. Repetition of the beha
viour within six months of the index episode distinguishes repeaters f
rom nonrepeaters. The model was designed so that it could be used by n
on-clinicians and hence does not require information relating to psych
iatric diagnosis or use of psychiatric services. The proportion of sub
jects correctly classified remained stable across a range of cut-point
probabilities (mean = 86%, range: 83.9-87.5%). Using a cut-point of 0
.2, 96% of repeaters and 81% of non-repeaters were correctly classifie
d. Using 0.45 led to the correct identification of 81% of repeaters an
d 90% of non-repeaters. If these high levels of sensitivity and specif
icity are maintained in validation tests on future cohorts in Cork cit
y then the model could form the basis of an intervention programme des
igned to prevent the repetition of parasuicide.