September 1994

Suicide and Mortality Patterns in Anxiety Neurosis and Depressive Neurosis

Author Affiliations

From the Karolinska Institute, Stockholm, Sweden; the Department of Clinical Neuroscience and Family medicine, Section of Psychiatry, Huddinge (Sweden) University Hospital; and the Department of Psychiatry, Washington University School of Medicine, St Louis, Mo.

Arch Gen Psychiatry. 1994;51(9):708-712. doi:10.1001/archpsyc.1994.03950090040006

Background:  The diverging views on suicide risk in patients with morbid anxiety called for a sufficiently large study to estimate the suicide risk in patients with anxiety neurosis and depressive neurosis.

Methods:  The identities of all 9912 patients with anxiety neurosis and all 38 529 patients with depressive neurosis in the national Psychiatric Case Register in Sweden between 1973 and 1983, without any other psychiatric diagnoses, were matched with the national Cause-ofDeath Register. The observed causes of death among the 9910 patients who died in 1990 or earlier were compared with those expected in the general population.

Results:  There were 1481 determined and 265 undetermined suicides among the patients; ie, 18% of all deaths. The standardized mortality ratio of suicide before the age of 45 years among men and women with anxiety neurosis was 6.7 and 4.9, respectively; for depressive neurosis, 12.6 and 15.7, respectively. The suicide risk was much higher within 3 months of leaving the hospital. Standardized mortality ratios of death caused by ischemic heart disease and traumatic injury were marginally elevated among men in both diagnostic groups. Women in both categories were at increased risk for death caused by alcohol abuse and cirrhosis of the liver. Obstructive pulmonary disease was another notable cause of death, reflecting the aggravation of anxiety-depressive symptoms by airway obstruction or the effects of tobacco smoking.

Conclusions:  The risk of completed suicide among former inpatients with primary anxiety neurosis was higher than in previous, smaller studies and higher yet in patients with depressive neurosis. This hazard may hopefully be reduced by optimizing immediate and longterm treatment for the severely affected.