Drugs and Drowning: The Toxicology of Adult Drownings in Baths and Hot Tubs in Australia, 2015–2024
Funding: This work was supported by National Drug and Alcohol Research Centre.
ABSTRACT
Introduction
Drowning is the third leading cause of unintentional injury death. We aimed to determine the toxicology and circumstances of all adult drowning deaths that occurred in bath/hot tubs in Australia over the period 2015–2024.
Methods
Retrospective study of all adult (≥ 15 years) drowning deaths in baths or hot tubs in Australia (1 January 2015–1 November 2024) retrieved from the National Coronial Information System (n = 195). In all cases the formal finding was based upon police, toxicology and forensic pathology reports.
Results
There were 195 adult drownings in baths or hot tubs. The mean age was 54.9 years (range 15–98) and 127 (65.1%) were female. Most fatal (171, 87.7%) incidents occurred in a bath. In 113 (57.9%) cases proximal substance use was noted in the coronial conclusions as contributory. The majority (108, 55.4%) were unintentional, with 73 (37.4%) deemed intentional. A psychotropic drug was detected in the blood of 152/179 (84.9%), most commonly hypnosedatives (77/179, 43.0%) and alcohol (75/179, 41.9%). Amongst alcohol positive cases the mean blood alcohol concentration was 0.176 g/100 mL (range 0.010–0.537). In 33 (16.9%) there appeared to have been a medical episode, such as a seizure or a cardiovascular event, that preceded drowning, and a slip or fall in 17 (8.7%).
Discussion and Conclusions
Substances were present in the majority of cases. The risk of drowning in such settings in the presence of drugs needs to be widely appreciated. Public campaigns that focus on the potential dangers of substance use in these settings would appear prudent.
1 Introduction
Drowning is the third leading cause of unintentional injury death worldwide, accounting for 9% of all injury-related deaths [1]. One potentially preventable form of drowning occurs in domestic baths or hot tubs [2-10]. Such cases may constitute a significant proportion of drowning deaths, with a recent US study reporting that 13% of all drownings occurred in domestic baths [2]. There are, however, few case series or reports on the risk factors or characteristics of adult deaths in such settings. One major risk factor that stand out from this scant literature is substance intoxication, which may cause unconsciousness and submersion leading to drowning. In particular, depressant drugs such as alcohol, opioids and hypnosedatives appear to be prominent [2-10]. In addition to direct effects on consciousness, intoxication may increase the risk of slips or falls (a prominent mechanism of injury in baths and hot tubs) and of a seizure or a cardiovascular event [10-12].
Reflecting the scant literature, there are few data on the demographics of adult cases of bath/hot tub drownings. Older people, aged over 60, appear at greater risk [2, 3, 5-9]. Cases of drug-related drownings in baths/hot tubs appear to be slightly younger, with a mean age in the 40s, reflecting drug-related deaths more broadly [5]. While women comprise the majority of such deaths [2, 3, 5-10], men were more common in a series of drug-related drownings [5]. Amongst drug-related drownings in the United Kingdom, mental health problems were documented in a third [5]. The role of intent in these incidents, however, remains unclear.
In recent work we reported on drug-related drownings in the United Kingdom in which opioids, hypnosedatives and alcohol were the most frequently detected substances [5]. That series was restricted to cases in which substances were present. To our knowledge, no national case series of all adult bath/hot tub drownings has been conducted in Australia or elsewhere. Several questions arise in characterising such cases. Firstly, what is the role of alcohol and other drugs? What are the circumstances and characteristics of cases? What is the role of intent? In order to address these questions, we conducted a retrospective study of all adult deaths due to drowning that occurred in a bath or hot tub in Australia across the 10-year period 2015–2024. Specifically, we aimed to determine the toxicology results and circumstances of all adult drowning deaths that occurred in bath/hot tubs in Australia over the period 2015–2024.
2 Methods
2.1 National Coronial Information System
The National Coronial Information System (NCIS) is a database of medicolegal death investigation records provided by the coroners' courts in each Australian and New Zealand jurisdiction, commencing in July 2000 for Australia (January 2001 for Queensland) and July 2007 for New Zealand. Only Australian cases were accessed in this study. A complete NCIS case file includes demographic information, a police narrative of circumstances, autopsy reports and toxicology reports (where these processes were conducted), and the coronial finding. Cause of death is ascertained by a forensic pathologist and documented on the autopsy and in the coroner's report. The forensic pathologist may report on: (i) the direct cause of death; (ii) the antecedent cause; and (iii) other significant conditions associated with the death. This advice is provided to the coroner, who makes a formal determination of the cause of death on the basis of the medical and other information. In all cases, the formal finding was based upon police, toxicology and forensic pathology reports.
2.2 Case Identification
A retrospective study of all adult (≥ 15 years old) drowning deaths in baths or hot tubs in Australia was conducted. All closed cases (i.e., in which the coronial investigation had been completed) that occurred between 1 January 2015 and 1 November 2024 that were coded in the NCIS Mechanism of Injury Coding fields as due to ‘Threat to breathing (Drowning/near drowning)’ were identified. The location of the fatal incident for each case was inspected, and cases in which the fatal incident occurred in a bath or hot tub were included in this study. Homicidal drownings were excluded, as the agency was external to the decedent. Ethical approval for the study was received from the Justice Human Research Ethics Committee, and the University of New South Wales Human Research Ethics Committee.
2.3 Measures
Data were retrieved from police narratives, autopsy reports, results of toxicology laboratory testing, and coronial findings. Manner of death, determined by the NCIS intent designation code, was classified as: (i) Unintentional; (ii) Intentional self-harm; or (iii) Undetermined intent. Information was collected on age, sex, mention of a history of alcohol or drug problems, whether substances were stated in the coronial conclusions as being contributory to death, and mention of a history of mental health problems. Information was also collected on the setting of the fatal incident (bath or hot tub), apparent medical episodes that preceded drowning, a documented acute injury that preceded drowning, apparent falls or slips that preceded drowning, whether there were other people nearby, and whether there was an ambulance or a doctor's intervention at the fatal incident prior to death.
Toxicological testing is centralised in each state and territory and was conducted according to local protocols. In cases of sudden death, such as those documented in this series, toxicology testing is routinely conducted. In cases of hospitalisation prior to death, antemortem blood samples taken on or near admission were reported, and drugs administered by medical staff were excluded. Blood toxicology testing was conducted and available for inspection for 179 cases. Data on blood samples were reported for those psychotropic drugs (i.e., substances that may alter perception, mood, consciousness, cognition or behaviour) that were detected: alcohol, antidepressants, antipsychotics, cannabis (Δ-9-THC), doxylamine, gabapentinoids, hypnosedatives, opioids, psychostimulants and volatile solvents. While these are quantitated, with the exception of alcohol, we report the proportions in which these substances were present. All samples were tested using a range of methodologies specific to that laboratory, including immunoassay, gas chromatography, high-performance liquid chromatography and liquid chromatography-quadrupole time-of-flight mass spectrometry for common drugs and selected therapeutic substances. While the time between sampling and testing was not known, all specimens were preserved and stored at 4°C prior to testing.
2.4 Statistical Analyses
Means, standard deviation (SD) and range were reported for age and blood alcohol concentrations as they were normally distributed. All analyses were conducted using IBM SPSS Statistics v. 27.0 [13].
3 Results
3.1 Case Characteristics
We identified 195 fatal drownings that occurred in a bath or hot tub. The mean age was 54.9 years (SD 18.3, range 15–98 years) and two-thirds were female (Table 1).
| N = 195 n (%) | |
|---|---|
| Age, years | |
| 15–20 | 6 (3.1) |
| 21–30 | 16 (8.2) |
| 31–40 | 20 (10.3) |
| 41–50 | 37 (19.0) |
| 51–60 | 40 (20.5) |
| 61–70 | 35 (17.9) |
| 71–80 | 22 (11.3) |
| > 80 | 19 (9.7) |
| Sex | |
| Female | 127 (65.1) |
| Male | 69 (34.9) |
| Documented history | |
| Substance use problems (alcohol or other drugs) | 54 (27.7) |
| Mental health problem | 101 (51.8) |
| Affective disorders | 96 (49.2) |
| Psychotic disorders | 6 (3.1) |
| Circumstances of death | |
| Location of fatal incident | |
| Bath | 171 (87.7) |
| Hot tub | 24 (13.3) |
| Intent | |
| Unintentional | 108 (55.4) |
| Intentional | 73 (37.4) |
| Undetermined | 14 (7.2) |
| Substances noted as contributory | 113 (57.9) |
| Apparent medical episode preceding drowning | 33 (16.9) |
| Injury documented | 24 (12.3) |
| Apparent fall/slip preceding drowning | 17 (8.7) |
| Another person nearby | 71 (36.4) |
| Ambulance and/or doctor attendance at fatal incident prior to death | 36 (18.5) |
The vast majority (171, 87.7%) occurred in a bath. Substance use problems were documented in a 54 (27.7%) and mental health problems in 101 (51.8%). In 113 (57.9%) proximal substance use was noted in the coronial conclusions as contributory to death. While the majority were unintentional (108, 55.4%), in 73 (37.4%) of cases, the fatal incident was deemed intentional, and intent was unable to be determined in a 14 (7.2%).
In 33 (16.9%) there appeared to have been a medical episode that preceded the drowning, most commonly a suspected seizure (20, 10.3%) or a cardiovascular event (12, 6.2%). An acute injury was noted in 24 (12.3%), and there appeared to have been a slip or fall that preceded the drowning in 17 (8.7%), all of which had evidence of traumatic injury. While other people were nearby in 71 (36.4%) cases, an ambulance and/or doctor attended prior to death occurred in only 36 (18.6%).
3.2 Results of Toxicology Laboratory Testing
Blood toxicology testing was conducted and available for inspection for 179 cases, of which 10 were antemortem blood samples (Table 2). In 16 cases, toxicological testing was conducted, but the reports were not available for inspection. A psychotropic drug was detected in the blood of 152 (84.9%), most commonly hypnosedatives (7, 43.0%) and alcohol (75, 41.9%). Alcohol was most commonly seen in conjunction with other substances. Overall, a depressant drug (hypnosedative, alcohol, opioid, gabapentinoid, gamma hydroxybutyrate) was present in 130 (72.6%) of cases, most frequently alcohol and opioids. Sizable proportions had antipsychotics and/or antidepressants present. In contrast, illicits such as cannabis and psychostimulants were uncommon.
| N = 179 n (%) | |
|---|---|
| Any psychotropic drug | 152 (84.9) |
| Alcohol only | 15 (8.4) |
| Other substance only | 77 (43.0) |
| Alcohol and other substance | 60 (33.5) |
| Hypnosedatives (benzodiazepines, z-class hypnotics) | 77 (43.0) |
| Alcohol | |
| 0.000 g/100 mL | 104 (58.1%) |
| 0.010–0.100 g/100 mL | 24 (13.4%) |
| 0.101–0.200 g/100 mL | 21 (11.7%) |
| 0.201–0.300 g/100 mL | 16 (8.9%) |
| > 0.300 g/100 mL | 14 (7.8%) |
| Antidepressants | 70 (39.1) |
| Opioids (buprenorphine, codeine, fentanyl, hydromorphone, methadone, morphine, oxycodone, pholcodine, tapentadol, tramadol) | 37 (20.7) |
| Antipsychotics | 29 (16.2) |
| Psychostimulants (cocaine, methamphetamine, MDMA, phentermine) | 19 (10.6) |
| Cannabis | 17 (9.5) |
| Gabapentinoids (gabapentin, pregabalin) | 9 (5.0) |
| Other (doxylamine, GHB, volatile solvents: 1, 1-difluoroethane, benzene toluene, xylenes) | 8 (4.5) |
- Abbreviations: GHB, gamma hydroxybutyrate; MDMA, 3,4-Methyl-enedioxy-methamphetamine.
Psychotropic drugs were present in 78/101 (77.2%) of unintentional deaths, 62/66 (93.9%) of intentional deaths and all 12 (100.0%) cases of unknown intent. Amongst alcohol positive cases the mean blood alcohol concentration was 0.176 g/100 mL (SD 0.123, range 0.010–0.537) or 0.044 mmol/L (SD 0.031, range 0.002–0.134). Of the 15 cases where a slip or fall preceded the drowning and toxicology results were available for inspection, 13 had a substance present in their blood.
4 Discussion
To our knowledge this study presents the first national data on the toxicological profiles of bath and hot tub drownings. Consistent with previous research [2-10], substance use, particularly depressants, was prominent and was formally deemed contributory to the drowning in more than half of cases. It was notable that a quarter of cases had documented histories of substance use problems, which is also consistent with previous research [5]. In three quarters of cases a depressant drug, whether prescribed or illicit, was present in the decedent's blood, drugs which may reduce the level of consciousness, most commonly alcohol and opioids. Alcohol was present in more than 40% of cases, and 30 cases had blood alcohol concentrations greater than 0.2 g/100 mL. In contrast, drugs such as THC were uncommon. More than a third of the deaths in this series were due to suicide. The role of suicide in such drownings has been reported in an earlier study [2]. In almost all intentional deaths substances were present in blood. A typical scenario was to intentionally take high doses of drugs and/or alcohol, fall unconscious and drown, often fully clothed. It was notable that in half of decedents there was a history of mental health problems, most commonly affective disorders, a finding consistent with previous research [5, 7].
What can be done to reduce this toll? The circumstances characterised here offer some clues. Substances were present in the majority of cases. While it is likely that most people are aware that intoxication is a risk factor in ocean, river and pool drownings, there needs to be awareness that there is risk in bathing intoxicated per se, regardless of the setting. Baths and hot tubs are not safe settings for an intoxicated person, particularly if the person is alone, with a real risk of falling unconscious or having a slip or fall that results in a loss of consciousness. Public safety awareness campaigns, that focus on the potential dangers of substance use in baths and hot tubs would appear prudent. The need for checking on the bather is illustrated by the high proportion of intentional drownings in this series, which could form part of such campaigns. Baths and hot tubs may also not be widely recognised as a setting for intentional drowning, and those living with at-risk individuals need to be made aware of the risk.
As in all studies, caveats must be borne in mind. The case series comprised only closed cases and in recent years there will be open cases in which the coronial process was still underway. Moreover, the diagnosis of drowning is a difficult exercise for forensic pathologists, with a combination of non-specific findings of drowning. Nevertheless, in all these cases it was the view of the forensic pathologist, and accepted by the coroner, that death was primarily the result of drowning. Details of clinical histories and circumstances surrounding death were restricted to those documented in case files and may thus be conservative estimates. Thus, we were unable to determine whether prescription drugs were prescribed to the decedent or, if so, for what clinical purpose. Similarly, we could not definitively determine whether events such as seizures, cardiac events, or slips and falls were specifically attributable to the presence of a substance, or the specific role of any particular substance in a drowning. Finally, in all studies of mortality suicidal intent is problematic. In this study, suicide intent was based upon the NCIS code for “Intentional self-harm”, which derives from the case circumstances and the conclusions of the coroner.
In summary, substances were present in the majority of these deaths. The risk of drowning in such settings in the presence of drugs needs to be widely appreciated.
Author Contributions
Professor Darke designed the study, conducted data collection, conducted statistical analyses and was the lead author in the write-up of the paper. Professor Farrell provided specialist medical comment and reviewed the manuscript. Professor Duflou specialist toxicological and forensic comment and reviewed the manuscript. Associate Professor Copeland provided specialist toxicological comment and reviewed the manuscript. Dr. Roberts provided specialist medical comment and reviewed the manuscript. All authors contributed to, and have approved, the final manuscript.
Acknowledgements
The authors acknowledge the Victorian Department of Justice and Community Safety as the source organisation for the data presented here, and the National Coronial Information System as the data source. We would like to thank the staff at the National Coronial Information System. Open access publishing facilitated by University of New South Wales, as part of the Wiley - University of New South Wales agreement via the Council of Australian University Librarians.
Conflicts of Interest
This work was funded by the National Drug & Alcohol Research Centre at the University of New South Wales. The National Drug & Alcohol Research Centre is supported by funding from the Australian Government Department of Health and Ageing. Michael Farrell has received untied educational grants from Seqirus, Mundipharma, and Indivior for post-marketing surveillance of pharmaceutical opioids. This organisation had no role in study design, analysis, and reporting, and funding support was for work unrelated to this project.
Open Research
Data Availability Statement
The data presented in this study are based upon confidential medicolegal documents from coronial reports. The authors are bound by ethical and legal agreements not to disclose data apart from those presented in the article.
