ED-treated nonfatal suicide attempts in ten US states, 1997-2004: Relationship between caseload and medical decision-making

Intro: While stabile regional differences in US rates of suicide have been well-described, little is known about the regional or temporal differences in nonfatal attempt rates, or the impact of case volume on post-attempt medical decision-making. This eight-year, 10-state analysis used annual counts of emergency department (ED) treated self-injuries obtained directly from state health agencies to quantify rates and characterize the relationship between caseload and dispositional patterns.

Method: The crude rates used for these analyses consisted of the number of annual, medically-identified intentional self harm events per 100,000 person years recorded in ten US states (Maine, Maryland, Massachusetts, Minnesota, Missouri, Nebraska, South Carolina, Tennessee, Utah and Wisconsin) between 1997 and 2004. Trend data were modeled with time series analysis, allowing for autocorrelation across years, and Pearson correlations were used to define other rate relationships.

Results: Overall, 319,060 self injuries occurring during 253,610,387 person years were available for analysis, for a combined ten-state, nonfatal medically-treated event rate of 126.99 events per 100,000 person years. This rate rose by an annual average of 3.66 /100,000 events per year (p<.001), and varied significantly by state. As ED rates rose, post-attempt hospitalization rates declined (r2=0.40), sinking to approximately 25% of all presentations in states experiencing more than 140 ED presentations /100,000 population.

Conclusions: ED-treated suicide attempt rates vary widely by region, but are rising in at least nine of the 10 states studied. Treatment of greater numbers of suicidal patients may be associated with greater willingness to assume risk during dispositional decision-making.