Intimate partner homicide (IPH) is a global public health problem. One study conducted over 66 countries found that 13.5% of all homicides and 38.6% of female homicides were committed by an intimate partner. In South Africa, there were no published studies that examine alleged perpetrators of IPH that were referred for forensic psychiatric observation.
To describe the profile of accused persons referred for forensic psychiatric observation for a charge of murder or attempted murder of their intimate partners. Certain characteristics were further examined according to the psychiatric observation outcomes.
The study was conducted at Sterkfontein Hospital, a forensic psychiatric hospital in Gauteng, South Africa.
A retrospective record review of accused persons referred for forensic psychiatric observation for a charge of murder or attempted murder of their intimate partners was conducted. The period of the review was 19 years. The definition of intimate partners included current or former spouses and partners, same-sex partners and rejected suitors.
One hundred and sixty-three files, which included forensic psychiatric reports, were reviewed. The findings related to the profile of accused persons and offence characteristics indicated that: (1) history of violent behaviour is prevalent; (2) homicides mostly occur in private homes; (3) knives and firearms are most often used; (4) infidelity, separation and jealousy are common motives; (5) psychotic disorders, personality disorders and substance use disorders feature prominently. A total of 88% of the sample were found fit to stand trial and 82% were found criminally responsible. Factors significantly associated with being found fit to stand trial and criminally responsible following the forensic psychiatric observation were: male gender, having received a tertiary education, employment prior to the offence, earning a salary of more than R10 000, having no previous psychiatric or medical illness, a positive forensic history, previous intimate partner violence (IPV) perpetration, indicating a motive for the homicide, having no psychiatric illness at the time of the offence which would impact fitness to stand trial and criminal responsibility.
Factors significantly associated with being found not fit to stand trial and not criminally responsible following the forensic psychiatric observation were: female gender, having received a primary education, unemployment prior to the offence, having a previous psychiatric or medical illness, no forensic history, no previous IPV perpetration, not indicating a motive for the homicide, having a psychiatric illness at the time of the offence which would impact fitness to stand trial and criminal responsibility.
The characteristics highlighted in this study can contribute to the development of risk assessment tools which can be used to identify likely perpetrators of IPH. Other interventions, for example controlling access to knives and firearms, reducing substance abuse and improving mental health services, are also important in the prevention of IPH.
Intimate partner homicide (IPH), defined as ‘the intentional killing of one’s current or former partner’,
A global study conducted over 66 countries found that 13.5% of all homicides and 38.6% of female homicides were committed by an intimate partner.
Some literature on IPH considers the high prevalence of mental illness amongst perpetrators.
Significant gaps exist in the current literature with regard to the profile of accused persons and offence characteristics in cases of IPH referred for forensic psychiatric observation. Despite much research on the general population that commits IPH; few studies have addressed accused persons who are referred for forensic psychiatric observation.
Forensic psychiatric observations are conducted when the court has reason to believe accused persons may be suffering from a mental illness or intellectual disability which may be impacting on the individual’s fitness to stand trial (the ability to comprehend court proceedings) and/or criminal responsibility (the ability to appreciate the wrongfulness of an act and to act in accordance with such an appreciation).
Within the South African context, such accused persons are referred by the court to a psychiatric hospital for a forensic psychiatric observation in terms of sections 77, 78 and 79 of the
For those who are found not fit to stand trial and/or not criminally responsible, for a serious or major offence (for example murder, attempted murder, sexual assault or assault with intent to cause grievous bodily harm [assault GBH]), the accused is usually admitted to the forensic unit as a state patient in terms of section 42 of the
As there is no specific charge of IPH or IPV in SA, this research study focussed on murder and attempted murder of intimate partners in the context of forensic psycho-legal assessments in Johannesburg, SA. Attempted murder was included in this study as the authors wanted to investigate the profile of accused persons who had shown intent to commit murder of their intimate partners regardless of whether the act was successful or not. The authors are of the opinion that the profile of accused persons charged with IPH is more associated with the intent to commit murder rather than with the outcome of the act, and that this warranted the inclusion of attempted murder cases in this study. Persons accused of assault GBH against their intimate partner were not included in this study as the authors could not be confident that these accused persons had demonstrated intent to commit murder and thus the authors believed that their profile would be markedly different to those that were accused of murder or attempted murder of their intimate partners.
The primary aim of this study was to describe the socio-demographic, clinical and forensic profile of accused persons referred for forensic psychiatric observation, under the CPA, to Sterkfontein Hospital for a charge of murder or attempted murder of an intimate partner. The study also sought to examine offence characteristics and to describe the outcomes of the forensic psychiatric observation. The accused and offence characteristics, as well as diagnosis, were further examined according to the categories of psychiatric observation outcomes, that is, whether accused persons were found to be either fit to stand trial or not fit to stand trial and criminally responsible or not criminally responsible.
This was a retrospective record review of accused persons referred for forensic psychiatric observation to Sterkfontein Hospital with a charge of murder or attempted murder during a 19-year period, from 01 January 2000 to 31 December 2018.
The study was conducted at the forensic section of Sterkfontein Hospital, a tertiary psychiatric hospital providing both general and forensic psychiatric services, situated in Gauteng, SA.
The sample included records of all adult (age ≥ 18 years) males and females referred as observandi from 01 January 2000 to 31 December 2018 for the charges of murder or attempted murder of their intimate partners. The definition of intimate partners included current or former spouses and partners, same-sex partners and rejected suitors. The definition of IPH includes those homicides that were successful and those that were unsuccessful, that is, files included in this study were of those accused persons referred for forensic psychiatric observation with a charge of murder or attempted murder of their intimate partner.
Case files for IPH were identified via the forensic unit admission register, psychiatric reports and clinical records. Data collection took place between January 2019 and June 2019. Data were collected from psychiatric reports and clinical records. Data were captured using a data collection sheet which included socio-demographic, clinical, forensic, offence and psychiatric observation factors.
All statistical analyses were conducted using Python (Scipy.stats module;
The protocol was approved by the University of the Witwatersrand’s Human Research Ethics Committee (clearance certificate number: M180530). Permission to use Sterkfontein Hospital as the site for the research was granted by the hospital’s research committee. All data were collected by the primary investigator who was responsible for ensuring the anonymity, confidentiality and security of data obtained.
A total of 145 male and 18 female accused persons were included in this study (
Socio-demographic, clinical and forensic profile of accused persons.
Characteristics | % | |
---|---|---|
18 to 30 | 50 | 31 |
31 to 40 | 52 | 32 |
Over 40 | 61 | 37 |
Female | 18 | 11 |
Male | 145 | 89 |
Single | 73 | 45 |
Married | 59 | 36 |
Separated | 19 | 12 |
Divorced | 12 | 7 |
No | 79 | 48 |
Yes | 84 | 52 |
Mainstream | 156 | 96 |
Special | 3 | 2 |
No formal education | 4 | 2 |
No formal education | 4 | 2 |
Primary education | 20 | 12 |
Secondary education | 91 | 56 |
Tertiary education | 48 | 29 |
Employed | 91 | 56 |
Unemployed not on a disability grant | 65 | 40 |
Unemployed on a disability grant | 7 | 4 |
R0 – R5000 | 4 | 2 |
R5000 – R10 000 | 28 | 17 |
> R10 000 | 59 | 36 |
No | 130 | 80 |
Yes | 33 | 20 |
Antisocial personality disorder | 1 | 1 |
Schizophrenia | 10 | 6 |
Schizoaffective disorder | 1 | 1 |
Psychotic disorder due to a general medical condition | 2 | 1 |
Substance-induced psychotic disorder | 1 | 1 |
Major depressive disorder | 13 | 8 |
Mood disorder due to a general medical condition | 2 | 1 |
Bipolar I disorder | 2 | 1 |
Bipolar II disorder | 1 | 1 |
No | 106 | 65 |
Yes | 57 | 35 |
Previous head trauma | 29 | 18 |
Epilepsy | 21 | 13 |
Diabetes mellitus | 10 | 6 |
HIV | 6 | 4 |
No | 55 | 34 |
Yes | 108 | 66 |
Ethanol | 79 | 48 |
Cannabis | 46 | 28 |
Nicotine | 27 | 17 |
Stimulants | 11 | 7 |
No | 146 | 90 |
Yes | 17 | 10 |
Physical and/or sexual abuse | 10 | 6 |
Witnessed domestic violence | 5 | 3 |
Both of the above | 1 | 1 |
Other | 1 | 1 |
No | 103 | 63 |
Yes | 60 | 37 |
Murder | 4 | 2 |
Attempted murder | 5 | 3 |
Assault with intent to cause grievous bodily harm | 26 | 16 |
Domestic violence | 2 | 1 |
Protection order contravention | 10 | 6 |
Robbery | 20 | 12 |
Malicious damage to property | 1 | 1 |
No | 86 | 53 |
Yes | 77 | 47 |
Indicated – no | 34 | 21 |
Indicated – yes | 42 | 26 |
Not indicated | 87 | 53 |
No | 156 | 96 |
Yes | 7 | 4 |
DSM-IV-TR, Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision.
The vast majority of the victims were the current spouse or partner of the accused (
Offence characteristics.
Characteristics | % | |
---|---|---|
Murder | 124 | 76 |
Attempted murder | 39 | 24 |
Spouse | 75 | 46 |
Partner | 65 | 40 |
Ex-partner | 10 | 6 |
Ex-spouse | 5 | 3 |
Fiancé | 4 | 2 |
Rejected suitor | 4 | 2 |
Stabbing | 76 | 47 |
Gunshot | 43 | 26 |
Strangulation | 16 | 10 |
Blunt trauma | 13 | 8 |
Arson | 4 | 2 |
Poisoning | 1 | 1 |
Other | 6 | 4 |
Mixed | 4 | 2 |
Knife | 70 | 43 |
Firearm | 43 | 26 |
Bodily force | 14 | 9 |
Fire | 4 | 2 |
Poison | 1 | 1 |
Other | 27 | 17 |
Mixed | 4 | 2 |
Residence | 128 | 79 |
Street | 14 | 9 |
Other | 21 | 13 |
No | 112 | 69 |
Yes | 51 | 31 |
Indicated | 115 | 71 |
Not indicated | 48 | 29 |
Rage | 51 | 31 |
Infidelity | 36 | 22 |
Separation | 35 | 21 |
Jealousy | 21 | 13 |
Self-defence | 11 | 7 |
Possessiveness | 3 | 2 |
Financial benefit | 1 | 1 |
Retaliation | 1 | 1 |
Sixty-nine per cent of accused persons reported to not have used a substance at the time of the offence. Considerably more accused persons indicated a motive. Of those that indicated a motive, rage, infidelity and separation were the most common.
The Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision (DSM-IV-TR) classification system was used in this study. Half of the sample were found to have a psychiatric diagnosis (
Psychiatric observation outcomes.
Characteristics | % | |
---|---|---|
Not present | 82 | 50 |
Present | 81 | 50 |
Antisocial personality disorder | 20 | 12 |
Borderline personality disorder | 3 | 2 |
Narcissistic personality disorder | 1 | 1 |
Dependent personality disorder | 1 | 1 |
Substance use disorder | 29 | 18 |
Schizophrenia | 11 | 7 |
Schizoaffective disorder | 1 | 1 |
Psychotic disorder not otherwise specified | 2 | 1 |
Psychotic disorder due to a general medical condition | 9 | 6 |
Substance-induced psychotic disorder | 1 | 1 |
Major depressive disorder | 12 | 7 |
Mood disorder due to a general medical condition | 2 | 1 |
Mental retardation | 2 | 1 |
Dementia | 5 | 3 |
Fit | 143 | 88 |
Not fit | 20 | 12 |
Criminally responsible | 134 | 82 |
Not criminally responsible | 29 | 18 |
DSM-IV-TR, Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision.
Gender was found to be a significant predictor for fitness to stand trial and criminal responsibility (
Accused and offence characteristics and diagnosis according to the categories of psychiatric observation outcomes with
Characteristics | Fitness to stand trial |
Criminal responsibility |
|||
---|---|---|---|---|---|
Not fit (%) | Not responsible (%) | ||||
0.14 | 0.09 | ||||
18 to 30 | 50 | 12 | 14 | ||
31 to 40 | 52 | 6 | 12 | ||
Over 40 | 61 | 18 | 26 | ||
< 0.01 |
< 0.01 |
||||
Female | 18 | 44 | 56 | ||
Male | 145 | 8 | 13 | ||
0.75 | 0.53 | ||||
Single | 73 | 14 | 15 | ||
Married | 59 | 12 | 22 | ||
Separated | 19 | 5 | 11 | ||
Divorced | 12 | 17 | 25 | ||
0.02 |
0.01 |
||||
No formal education | 4 | 25 | 50 | ||
Primary education | 20 | 30 | 35 | ||
Secondary education | 91 | 12 | 19 | ||
Tertiary education | 48 | 4 | 6 | ||
< 0.01 |
0.02 |
||||
Employed | 91 | 4 | 11 | ||
Unemployed not on a disability grant | 65 | 20 | 25 | ||
Unemployed on a disability grant | 7 | 43 | 43 | ||
< 0.01 |
0.03 |
||||
R0 – R5000 | 4 | 0 | 25 | ||
R5000 – R10 000 | 28 | 14 | 18 | ||
> R10 000 | 59 | 0 | 7 | ||
< 0.01 |
0.02 |
||||
No | 130 | 8 | 14 | ||
Yes | 33 | 30 | 33 | ||
< 0.01 |
< 0.01 |
||||
Personality disorder | 1 | 0 | 0 | ||
Psychotic disorder | 14 | 64 | 71 | ||
Mood disorder | 18 | 6 | 6 | ||
0.01 |
< 0.01 |
||||
No | 106 | 7 | 10 | ||
Yes | 57 | 23 | 32 | ||
Previous head trauma | 29 | 24 | 0.07 | 31 | 0.07 |
Epilepsy | 21 | 24 | 0.17 | 38 | 0.02 |
HIV | 6 | 33 | 0.33 | 50 | 0.12 |
0.38 | 0.11 | ||||
No | 55 | 16 | 25 | ||
Yes | 108 | 10 | 14 | ||
0.75 | 0.73 | ||||
No | 146 | 12 | 18 | ||
Yes | 17 | 12 | 12 | ||
0.82 | 0.81 | ||||
Physical and/or sexual abuse | 10 | 20 | 20 | ||
Witnessed domestic violence | 5 | 0 | 0 | ||
Both of the above | 1 | 0 | 0 | ||
Other | 1 | 0 | 0 | ||
0.06 | 0.03 |
||||
No | 103 | 17 | 23 | ||
Yes | 60 | 5 | 8 | ||
0.06 | 0.19 | ||||
No | 86 | 17 | 22 | ||
Yes | 77 | 6 | 13 | ||
< 0.01 |
< 0.01 |
||||
Indicated – no | 34 | 32 | 47 | ||
Indicated – yes | 42 | 0 | 5 | ||
Not indicated | 87 | 10 | 13 | ||
0.05 | 0.21 | ||||
No | 156 | 11 | 17 | ||
Yes | 7 | 43 | 43 | ||
0.70 | 0.80 | ||||
No | 112 | 13 | 19 | ||
Yes | 51 | 10 | 16 | ||
0.02 |
< 0.01 |
||||
Indicated | 115 | 8 | 11 | ||
Not indicated | 48 | 23 | 33 | ||
< 0.01 |
< 0.01 |
||||
Not present | 82 | 0 | 1 | ||
Present | 81 | 25 | 35 | ||
Personality disorder | 25 | 0 | 0.09 | 4 | 0.09 |
Substance use disorder | 29 | 10 | 0.97 | 17 | 0.86 |
Psychotic disorder | 24 | 67 | < 0.01 |
83 | < 0.01 |
Mood disorder | 14 | 7 | 0.85 | 21 | 0.99 |
Mental retardation | 2 | 0 | 0.58 | 0 | 0.79 |
Dementia | 5 | 60 | 0.01 |
80 | < 0.01 |
, statistically significant.
Males were more likely to be found fit and criminally responsible whereas females were more likely to be found not fit and not criminally responsible.
The level of education completed also had a significant association with fitness to stand trial and criminal responsibility. Accused persons who achieved lower levels of education were more likely to be found not fit and not criminally responsible. Those who had obtained a tertiary education were more likely to be found fit and criminally responsible.
Employment status and salary were significantly associated with fitness to stand trial and criminal responsibility. Accused persons who were employed were more likely to be found fit and criminally responsible. Accused persons who earned more than R10 000 were more likely to be found fit and criminally responsible.
Previous psychiatric illness was found to be a significant predictor for fitness to stand trial and criminal responsibility. Accused persons with a previous psychiatric illness, particularly a psychotic disorder, were more likely to be found not fit and not criminally responsible compared to those without a previous psychiatric diagnosis who were more likely to be found fit and criminally responsible. Those diagnosed with a previous mood disorder were more likely to be found fit and criminally responsible.
Those with a medical illness were more likely to be found not fit and not criminally responsible compared to those without a medical illness who were more likely to be found fit and criminally responsible. Accused persons with a diagnosis of epilepsy were more likely to be found not criminally responsible.
Forensic history was significantly associated with criminal responsibility but not fitness to stand trial. Accused persons with a forensic history were more likely to be found criminally responsible compared to those with no forensic history, who were more likely to be found not criminally responsible.
A history of IPV perpetration and reporting a motive were significantly associated with fitness to stand trial and criminal responsibility. Accused persons with a history of committing previous IPV were more likely to be found fit and criminally responsible compared to those accused persons who were not previous IPV perpetrators, who were more likely to be found not fit and not criminally responsible. Accused persons who indicated a motive were more likely to be found fit and criminally responsible.
The presence of a psychiatric diagnosis at the time of the offence was significantly associated with fitness to stand trial and criminal responsibility. Having psychopathology in keeping with a psychiatric diagnosis, particularly a psychotic disorder or dementia, at the time of the offence, made it more likely for accused persons to be found not fit and not criminally responsible.
This study concurred with the findings in the literature that perpetrators of IPH are predominantly men.
Some studies indicate that approximately half of male perpetrators of IPH had not completed high school and the majority of female offenders have limited educational achievements.
This is contrary to our findings which showed that 14% of the sample had not completed secondary education.
A high percentage of accused persons (44%) were unemployed at the time of the offence, a finding consistent with those of most studies.
In a study that examined characteristics of IPH perpetrators, mental illness was rarely diagnosed before the incident,
Few studies have addressed the presence of medical illnesses amongst IPH perpetrators. Bourget and Gagné reported that 64% of women and 43% of men in their sample had chronic illnesses.
Literature suggests childhood trauma plays a significant role in the risk of IPH perpetration.
In this study, 37% of alleged perpetrators had a forensic history and 47% had a history of violent behaviour. These findings were similar to those in the literature which shows that approximately 25% – 50% of all male IPH perpetrators have been imprisoned for a previous brutal crime.
In SA, domestic violence is quoted as the main causal factor that results in IPHs.
[
This is further supported by the findings from two national studies conducted from 2003 to 2015 in North America.
Our study found a similar proportion of victims that were killed by their spouses (46%), in comparison to those that were killed by a non-marital partner (42%). This is contradictory to the literature which shows that IPH occurs more frequently within common-law relationships than those in marital relationships.
Alleged perpetrators killed their intimate partners most often in private residences, using a knife or firearm, which is supported by previous research.
Despite the high incidences of substance use in cases of non-fatal IPV, and a study indicating that a significant relationship exists between male perpetrators’ alcohol abuse and violence against intimate female partners,
Women rarely kill an intimate partner after the couple has separated but men are at greater risk of perpetrating IPH when separation of the relationship has occurred or is imminent.
Our study showed that 50% of the sample were found to have a psychiatric diagnosis at the time of the offence.
These results, however, must be interpreted with caution as PDs and SUDs were included in this category and these disorders alone do not impact fitness to stand trial and criminal responsibility. It is important to note, however, that PDs and SUDs may have an impact on IPV perpetration at large. Following forensic psychiatric observation (for IPH), studies indicate that psychotic disorders are most common, followed by mood disorders and anxiety disorders.
Our study found similar results: 15% had a psychotic disorder and 9% a mood disorder. Oram and colleagues, however, found contradictory results in that affective disorders were most common.
Multiple studies revealed that 14% – 20% of alleged IPH perpetrators, referred for forensic psychiatric observation, were deemed not criminally responsible because of the presence of a psychiatric disorder.
The study’s retrospective nature is a limitation, in that data may not always be complete and information gathered from others’ notes also has the potential to be inaccurate. Another limitation is that accused persons with PDs or SUDs were included amongst those individuals who were found to have ‘severe’ psychiatric illness whose fitness to stand trial and criminal responsibility were impacted. Additionally, these are alleged IPH perpetrators (who are still awaiting trial) so one has to be wary of drawing conclusions about actual convicted perpetrators. Assault GBH of an intimate partner was not included in the analyses, and the authors acknowledge the severe consequences of pervasive assault GBH over time which may be a precursor to attempted murder and murder. This warrants further study.
This is the first South African published study examining IPH within the context of forensic psychiatric observations. The primary aim of this study was to describe the socio-demographic, clinical and forensic profile of accused persons referred for forensic psychiatric observation, under the CPA, for a charge of IPH. The study also sought to examine offence characteristics and to describe the outcomes of the forensic psychiatric observation. The accused and offence characteristics, as well as diagnosis, were further examined according to the categories of psychiatric observation outcomes, that is, fitness to stand trial and criminal responsibility.
The main findings of this study were: (1) history of violent behaviour is prevalent; (2) homicides mostly occur in private homes; (3) knives and firearms are most often used; (4) infidelity, separation and jealousy are common motives; (5) psychotic disorders, PDs and SUDs feature prominently. Childhood trauma and previous IPV perpetration did not feature prominently. It is essential that an understanding of childhood trauma and IPV perpetration in the context of IPH is pursued, especially in developing countries, so that data can be compared globally. Furthermore, given the small number of alleged female perpetrators, we were not able to explore differences by gender. Future research is needed to examine gender differences amongst IPH perpetrators in SA.
From this study, the ‘typical’ fit and criminally responsible alleged perpetrator of IPH is a male who would have attained a tertiary education, would be employed prior to the offence and earning more than R10 000 a month. He would have no previous mental illness or medical illness but would have a positive forensic history. He would have a history of IPV perpetration and would indicate a motive for the homicide. Following the forensic psychiatric observation, he would not be found to have a psychiatric illness that impacted fitness to stand trial and criminal responsibility but may have a diagnosis of PD and/or SUD.
The ‘typical’ not fit and not criminally responsible alleged perpetrator of IPH is a female who would have attained only a primary education. She would be unemployed prior to the offence and have a history of previous mental illness. She may have a medical illness but would have no forensic history. She would not have a history of IPV perpetration and would not indicate a motive for the homicide. Following the forensic psychiatric observation, she would be found to have a psychiatric illness that impacted fitness to stand trial and criminal responsibility such as a psychotic disorder or dementia.
The study highlights multiple risk factors in those who are accused of IPH and are sent for forensic psychiatric observation. This information is valuable in that it can assist in preventing IPH, through the development of risk assessment tools which can be used for identification of likely perpetrators. Other interventions such as monitoring access to weapons such as knives and firearms, interventions to reduce alcohol and substance abuse and improving mental health services should all be addressed in order to reduce the incidence of IPH.
The authors acknowledge Devesh Naidu (Senior Quantitative Analyst) for his assistance with the statistical analysis.
The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.
S.N.V. was the principal author and U.S. and A.E. supervised the project. S.N.V., U.S. and A.E. collectively designed the study. S.N.V. collected and analysed the data and drafted the article. All authors critically revised the article and gave final approval of the version to be published.
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
The authors confirm that the data supporting the findings of this study are available within the article.
The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any affiliated agency of the authors.