Nursing home suicides – a psychological autopsy study.
Suominen K, Henriksson M, Isometsa E, Conwell Y, et al.;
Commented by , 28 Jan 2004
Aim and Objectives of the Study
To retrospectively analyse the sociodemographic and clinical features of all suicides among older adults in nursing homes in Finland over a 12 month period.
Methods
This study was part of the National Suicide Prevention Project in Finland; all suicides occurring in the year 1987-8 in Finland (n = 1397) were examined using the psychological autopsy method. The definition of suicide was based on the results of medicolegal examination required by Finnish law.
Data on suicides was collected from comprehensive interviews with relatives, attending health care personnel; and from police, social agency and medical records.
During the year of data collection, 13 residents of nursing homes committed suicide. 12 were over age 60 (mean age 76.1 +/- 5.7 years), and were included in this study. The remaining case was age 43 with a history of head trauma, and was excluded.
The data collected was: sociodemographics, suicide method, history of prior suicide method, history of prior suicide attempts (plus method used), time of placement in nursing home, lifetime psychiatric treatment, reported pain, psychiatric consultation and depressive symptoms in last 3 months, and medical/psychiatric diagnoses.
Attention was paid to assigning retrospective DSM-IV diagnoses with as much accuracy as possible. Two pairs of psychiatrists made independent diagnoses; all cases of diagnostic disagreement were re-analysed by a third psychiatrist. Inter-rater reliability for the whole sample (n = 1397) was good with high Kappa scores for most diagnoses.
Results
The 12 nursing home suicides represented 0.9 % of all elderly suicides. 9/12 were male (though 75 % of nursing home residents are female). 8/12 died by hanging, 3/12 by drowning, and 1/12 by shooting. 5/12 had moved into the nursing home in the last year of living. 4/12 had reported pain, all had co-morbid mental conditions diagnosed on Axis III of DSM.
1 or more psychiatric diagnoses were made for 12/12, 4/12 had previously attempted suicide. 9/12 were diagnosed with depression, alcohol abuse in 3/12. 8/12 had no lifetime history of psychiatric treatment; 0/3 cases of major depression were receiving antidepressants. 2/12 had had a psychiatric consultation in the previous 3 months.
Discussion
This simple study is of interest because it is the largest published series of nursing home suicides, a surprisingly under-researched area. The strength of the study is that it is based on a nationwide suicide population using explicit operationalised diagnostic criteria. The weaknesses, acknowledged by the authors, are the small numbers, the possibility of missed cases, and the limitations of assigning diagnoses retrospectively. Are these findings, now 16 years old, transferable outside Finland?
Despite these limitations, several themes can be drawn from the study. First and foremost, it is striking how many of the suicides had potentially treatable mental illness that was not being treated; this has obvious implications for suicide prevention.
Secondly, suicide is rare in older people and apparently very rare in nursing homes. Whilst many older people dread loss of independence and the potential indignity of being cared for in nursing homes, and may make suicidal threats, this does not seem to be translated into completed suicide. (Of course, passive suicide and parasuicide were not considered in this study).
Thirdly, it is noteworthy that 8/12 cases died by hanging and none by drug overdose; violent methods need to be considered in risk management of potentially suicidal people in nursing homes.
This is an interesting case series in an under-researched area, and again indicates that mental illness in older people is under-diagnosed and under-treated, occasionally with fatal consequences.