Background The relationship between unemployment and suicide may be sensitive to demographic factors, national unemployment rates, and length of time without employment. This study investigated these factors in relation to suicide in Australia for the period 1985–2006, in an ecological study.

Methods The outcome variable was annual suicide rate by age group, sex and the eight states and territories over 22 years of observation (total observations=1760). The main predictor variable was the average duration of unemployment in the population, categorised into three time periods (<2 weeks, 2–4 weeks, >4 weeks). Poisson regression models were used to investigate the relationship between duration of unemployment and suicide over the years 1985–2006 in a series of cross-sectional analyses. Interaction analyses indicated significant differences during periods of declining or increasing labour market opportunity and by age group.

Results During periods of declining unemployment rates in the country, longer durations of unemployment were associated with higher male suicide rates. During periods of increasing unemployment in the country, longer unemployment duration was associated with lower male suicide rates. Effect modification was also apparent by age-group, with stronger associations between unemployment duration and male suicide evident in those aged 25–34 and 55–64, and weaker associations in those aged 15–24 and 44–54 years. Longer length of unemployment was not associated with an increase in female suicide rates.

Conclusions The labour market opportunities in Australia modified the effect of duration of unemployment on suicide, and the effect was more prominent in men and older age groups. This may reflect social norms and acceptability about unemployment, as well as life-stage influences associated with transitions into and out of the labour market.