Objectives: To critically review the scientific literature relating to the timing of deliberate self harm behaviour and completed suicide.
Method: A literature search of the Medline and CINAHL databases from 1970-2002 was performed, using deliberate self harm, overdose, self poisoning, suicide, parasuicide, and time, timing, day, week, month and season as key words. Relevant secondary references were retrieved and hand searching of important journals was done.
Results: The time of day of non-fatal self harm shows a marked diurnal variation, with an evening peak that is related to non-violent episodes, concomitant alcohol use, and a younger age. It is not conclusively linked to the degree of suicidal intent or particular psychiatric diagnoses. Completed suicides more commonly occur earlier in the day, at the beginning of the week and during springtime, but show no overall increase during many national events and holidays.
Conclusions: Circadian biological mechanisms involving the serotonin-melatonin axis, Cortisol secretion and sleep abnormalities appear to be implicated. Psychosocial explanations for these epidemiological findings include alcohol use, a sense of personal isolation and the ‘broken promise’ effect.