The American Journal of Forensic Medicine and Pathology
22(2):134–138, 2001.
©2001 Lippincott Williams & Wilkins, Inc., Philadelphia
Characteristic Features of Suicidal Drownings A 20-Year Study
Roger W. Byard, M.D., Gillian Houldsworth, Ross A. James, F.R.C.P.A., and John D. Gilbert, F.R.C.P.A.
Although suicides by drowning account for a relatively low number of deaths in South Australia, cases may present diagnostic difficulties. This study was undertaken to analyze specific features of drowning suicides to determine whether particular characteristics differed depending on the age and sex of the victim or the location of the drowning.
A retrospective study of cases of drowning suicide was undertaken at the Forensic Science Centre in Adelaide, South Australia for the period April 1980 to March 2000. A total of 123 cases were found, with 76 males (age, 16–88 years; average, 50.5 years; standard deviation [SD], 20.1 years) and 47 females (age, 34–88 years; average, 60.6 years; SD, 13.9 years). There were 66 fresh water drownings and 57 saltwater drownings. Female victims were significantly older than male victims for both fresh water and saltwater drownings (P .05 and P .01, respectively). Deaths in young women were rare. No temporal trend in drowning suicides could be shown, with annual numbers varying from 0 to 12 cases (average, 6.15; median, 7). Women preferentially chose the ocean or bath to drown themselves in, whereas males chose rivers, ditches, and lakes. Swimming pools were rarely used for suicide in this population; alcohol use was not usual; and there was often a significant history of mental illness. Key Words: Drowning suicide—Mental illness.
MATERIALS AND METHODS The files of the Forensic Science Centre in Adelaide, South Australia were examined for all cases listed as suicide by drowning during the 20-year period from April 1980 to March 2000. The Forensic Science Centre provides autopsy services to the State Coroner for the state of South Australia, Australia, which has a population of approximately 1.5 million persons. Cases had been designated as suicide only after full police investigations and forensic autopsies were undertaken. Equivocal cases were excluded from the study, and cases in which death may have been caused by an accident, natural disease, or homicide were also not included. Details of the year of death; age and sex of the victim; medical and psychiatric history; drug or alcohol use; location of the drowning; and any additional props or suicide techniques used were extracted from the records and tabulated. Cases were divided into fresh water and saltwater drownings; fresh water drownings were subdivided further into bath, waterway (i.e., river, ditch, or lake), swimming pool, or miscellaneous groups. Saltwater drownings were subdivided into sea or beach; off jetties; inside cars; or miscellaneous groups. An “unknown” category was used when information was lacking. Statistical analyses were performed using the 2 and Student’s t test.
Manuscript received June 1, 2000; accepted August 2, 2000. From the Forensic Science Centre, Adelaide, Australia. Address correspondence to Roger W. Byard, Forensic Science Centre, 21 Divett Place, Adelaide 5000, Australia; email:
[email protected].
134
CHARACTERISTIC FEATURES OF SUICIDAL DROWNINGS RESULTS A total of 123 cases were found in which death was attributed to suicidal drowning. There were 76 males and 47 females, with an age range for men of 16 to 88 years (average, 50.5 years) and, for women, of 34 to 88 years (average, 60.6 years). The total number of fresh water drownings was 66 (male:female [M:F], 1.9:1; male age range, 16–88 years; average, 51.4 years; standard deviation [SD], 20.4 years; female age range, 34–83 years; average,
135
60.8 years; SD, 13.8) with 57 saltwater drownings (M:F, 1.4:1; male age range, 17–84 years; average, 49.3 years; SD, 19.8; female age range, 39–88 years; average, 60.4 years; SD, 14.2; see Fig. 1). No temporal trend in drowning suicides could be shown, with annual numbers varying from 0 to 12 cases (average, 6.15; median, 7) (Fig. 2). There was no significant difference in numbers of cases per month in each of the categories throughout the year.
FIG. 1. Average age of victims of drowning suicides for males and females in fresh water (FWD) and saltwater (SWD) taken from 123 cases of drowning suicide in South Australia between 1980 and 2000.
Am J Forensic Med Pathol, Vol. 22, No. 2, June 2001
136
R. W. BYARD ET AL.
FIG. 2. Numbers of drowning suicides in South Australia per year from 1980 to 2000 (dashed lines, incomplete calendar years).
Female victims were significantly older than male victims for both fresh water and saltwater drownings (P .05 and P .01, respectively). The numbers of drowning suicide deaths per decade of life for both males and females are shown in Figure 3. Table 1 provides a summary of the numbers of victims found at each of the different sites. Sites chosen for drowning suicides varied, with significantly more women than men choosing to commit suicide in the bath (P .02) and more men than women drowning themselves in waterways (P .02). Although there was a trend in saltwater drownings for men to favor jumping off jetties or using cars, with women either found floating in the sea or washed up on beaches, this did not reach statistical significance.
FIG. 3. Numbers of drowning suicides per decade of life for males and females in South Australia between 1980 and 2000.
Am J Forensic Med Pathol, Vol. 22, No. 2, June 2001
Routine toxicologic screening was performed on each case for opiates, amphetamines, alcohol, cannabis, tricyclic antidepressants, and benzodiazepines using urine enzyme-multiplied immunoassay test screens followed by gas chromatography and high-pressure liquid chromatography. The percentage of victims using drugs or alcohol is summarized in Table 2. There was a significantly increased number of individuals who tested positive for drugs or alcohol among those who drowned in the bath (P .01) compared with other subgroups. In a small number of cases, additional steps were taken to ensure that death would occur. These consisted of incising the wrists in six cases (bath, 4; unknown, 2); incising the throat in two cases; and shooting in two cases. Although potentially lethal
CHARACTERISTIC FEATURES OF SUICIDAL DROWNINGS TABLE 1. Summary of the location of drowning and sex of 123 victims of drowning suicides in South Australia between 1980 and 2000 Male (no.)
Female (no.)
Total (no.)
7 21 3 3 9 43
11 8 1 0 3 23
18 29 4 3 12 66
10 6 4 13 33
15 3 1 5 24
25 9 5 18 57
FWD Bath Waterway Swimming pool Other Unknown Total SWD Sea/beach Off jetty Car Unknown Total
137
TABLE 3. Numbers of cases of known mental or physical illness in 123 cases of drowning suicide
FWD Bath Waterway Swimming pool Other Unknown Total SWD Sea/beach Off jetty In car Unknown Total
Mental illness
Physical illness
Total
13 13 1 2 2 31
0 1 1 0 4 6
13 14 2 2 6 37
11 8 0 3 22
1 1 0 2 4
12 9 0 5 26
FWD, fresh water drowning; SWD, saltwater drowning.
FWD, fresh water drowning; SWD, saltwater drowning.
injuries were incurred from these additional activities, the primary cause of death was considered to be drowning. In nine cases, weights were used to help keep the victim’s head below water. These consisted of a load of bricks in three cases and diving weights, a rock, a toolbox, and a dumbbell in one case each. In two cases, the type of weight was not specified. A number of victims were known to be suffering from mental or physical illness at the time of their drowning. These results are summarized in Table 3. An extremely high percentage of individuals who drowned by jumping off jetties had psychiatric illnesses. These consisted of schizophrenia (3), depression (1), Huntington’s chorea (1), and unspecified mental illness (4). Physical illnesses consisted of terminal cancer, recent myocardial infarction, recent cerebrovascular accidents, and severe arthritis. Suicide notes were found in only 18 cases (14.6%).
TABLE 2. Number of cases where alcohol or prescription and illicit drugs were detected at autopsy after drowning suicide Drugs Alcohol Illicit Prescription Unknown Total FWD Bath Waterway Swimming pool Other Unknown Total SWD Sea/beach Off jetty In car Unknown Total
8 8 0 0 2 18
0 0 0 0 0 0
10 4 1 0 1 16
2 0 0 0 0 2
13 9 1 0 3
4 1 1 1 7
0 1 0 0 1
2 0 0 2 4
0 0 0 0 0
5 1 1 3
FWD, fresh water drowning; SWD, saltwater drowning.
The results of diatom examinations were not considered useful. DISCUSSION Although suicidal drowning appears to be a rare event in South Australia, accounting for only 3.6% of suicides autopsied at the Forensic Science Centre during the period 1985 to 1997, the incidence varies among communities. For example, drownings accounted for 8.9% of all suicides in a study from Newfoundland, Canada, being the fourth leading cause of suicidal death (1). Other studies have shown rates of between 2.86% and 7.9% (2–4). It has been proposed that part of the reason for this variation in rate is accessibility to water, with higher drowning suicide rates in areas adjacent to the sea or large inland lakes (5). Access to other methods of suicide certainly has been suggested to explain differences in the preferred method of selftermination (6). Part of the problem in assessing the rate of suicidal drowning is in the evaluation of individual cases (3,7,8). The diagnosis of drowning itself may be difficult as there are no specific diagnostic features at autopsy; the Gettler chloride test (which compares chloride levels in blood from the right and left sides of the heart) is not considered reliable; other electrolyte assessments are not helpful; and diatom examination is useful only if the results are positive (9–12). In addition, it also is possible that studies such as ours report a falsely low rate because a certain number of drownings that have been classified as accidental are actually disguised suicides (4). Homicidal suicides also may not be correctly classified if police investigations do not reveal suspicious circumstances (13). It may also be difficult on occasion to determine whether underlying natural diseases played a role in the terminal episode (7). Am J Forensic Med Pathol, Vol. 22, No. 2, June 2001
138
R. W. BYARD ET AL.
In the present study, there were significantly more males than females who committed suicide by drowning. This also differs between communities. A Danish study has demonstrated more female drowning suicides than male (14), contrasting with a Finnish study in which there were more males (5). There was also a marked difference in location chosen for drowning, with a strong male preference for rivers, ditches, or lakes and a female preference for the sea or baths. There is also variation in incidence of drowning suicide with age, with no drownings reported in a study of suicide under the age of 17 years in South Australia (6) compared with 27.5% of all suicides in the age group 50 years and older in a Newfoundland study (1). Although Auer demonstrated a predominance of males between 21 and 30 years of age (5), the average age in the current study was much higher, at 50.6 years for males and 60.6 years for females. One of the striking findings in the present study was the absence of female cases under 30 years of age, contrasting with relatively high numbers of males between 20 and 39 years of age (Fig. 3). Given their rarity, cases of apparent suicidal drowning in young women therefore should undergo very careful investigation before accidents or homicides are excluded. Suicide by drowning has been shown to be increased among schizophrenics, many of whom have been on leave or have absconded from the hospital (15). Consistent details of hospitalization were not available in the Forensic Science Centre files on the present cases; however, psychiatric illness was noted in quite a high percentage of victims (23%). This approached 89% of the cases of those individuals who drowned after jumping off jetties into the sea. Although it has been stated that individuals who commit suicide by drowning often are intoxicated (16), this has not been the experience in other studies, with alcohol only being detected in 9.1% to 41.4% of cases (1–3). Our results were intermediate, with 20.3% of cases in which victims tested positive for alcohol. Although the use of illicit and prescription drugs in our cases was very low, there was a significantly higher rate of use of prescription medications among those who committed suicide in the bath, as has been shown in other studies of bath suicides (17). In summary, this study has shown that although suicide by drowning in South Australia occurs at
Am J Forensic Med Pathol, Vol. 22, No. 2, June 2001
all ages, it tends to be a method favored by older individuals and not by young women. Women prefer to drown themselves in the ocean or bath, whereas males choose rivers, ditches, and lakes. Swimming pools are rarely used for suicide in this population. Alcohol use is also not usual, and there is often a significant history of mental illness. The use of weights was uncommon, and back-up techniques, such as wrist cutting, were also rare. Suicide notes were not found in the majority of cases. Acknowledgments: The authors thank the South Australian State Coroner, Wayne Chivell, for permission to publish details of these cases.
REFERENCES 1. Avis SP. Suicidal drowning. J Forensic Sci 1993;38: 1422–6. 2. Davis LG. Suicidal drowning in South Florida. J Forensic Sci 1999;44:902–5. 3. Copeland AR. Suicide by drowning. Am J Forensic Med Pathol 1987;8:18–22. 4. Nowers MP. Suicide by drowning in the bath. Med Sci Law 1999;39:349–53. 5. Auer A. Suicide by drowning in Uusimaa province in Southern Finland. Med Sci Law 1990;30:175–9. 6. Lester D. The suicide rate by drowning and the presence of oceans. Percept Mot Skills 1989;69:338. 7. Byard RW, Markopoulos D, Prasad D, et al. Early adolescent suicide: a comparative study. J Clin Forensic Med 2000;7:6–9. 8. Smith NM, Byard RW, Bourne AJ. Death during immersion in childhood. Am J Forensic Med Pathol 1991;12: 219–21. 9. Geertinger P, Voigt J. Death in the bath: a survey of bathtub deaths in Copenhagen, Denmark, and Gothenburg, Sweden, from 1961 to 1969. J Forensic Med 1970;17: 136–47. 10. Gettler AO. A method for determination of death by drowning. JAMA 1921;77:1650–2. 11. Knight B. Immersion deaths. Forensic pathology, 2nd ed. London: Arnold, 1997:391–406. 12. DeNicola LK, Falk JL, Swanson ME, et al. Submersion injuries in children and adults. Crit Care Clin 1997;13: 477–502. 13. Modell JH, Davis JH. Electrolyte changes in human drowning victims. Anesthesiology 1969;30:414–20. 14. Byard RW, Lipsett J. Drowning deaths in toddlers and preambulatory children in South Australia. Am J Forensic Med Pathol 1999;20:328–32. 15. Kringsholm B, Filskov A, Kock A. Autopsied cases of drowning in Denmark 1987–1989. Forensic Sci Int 1991; 52:85–92. 16. Shah A, Ganesvaran T. Suicide among psychiatric in-patients with schizophrenia in an Australian mental hospital. Med Sci Law 1999;39:251–9. 17. Giertsen JC. Drowning. In Tedeschi CG, Eckert WG, Tedeschi LG, eds. Forensic medicine, vol 3. Philadelphia: WB Saunders, 1977:1317–33.
