Gender issues in violence risk assessment and treatment in forensic psychiatry

Vivienne Vogel, Susanne de Haas, Vivienne dede Vogel, Jeantine Stam, Loes Eline Muller Eva de Hagenauw, Spa & Michiel&de Vries Robbé IAFMHS, Toronto, June 19, 2014

Ladykillers: Hurricanes with female names deadlier

Symposium outline Vivienne de Vogel, van der Hoeven Kliniek Introduction, multicenter study Susanne de Haas, van der Hoeven Kliniek Gender differences in motivations for offending Loes Hagenauw, FPK Assen Female arsonists Eline Muller, Pieter Baan Centrum Case example

Violence risk factors and incidents during treatment in female forensic psychiatric patients: Results from a Dutch multicenter study

Vivienne de Vogel Van der Hoeven Kliniek

Presentation outline • Female violence • Results multicenter study – Victimization – Psychopathology – Criminal characteristics – Treatment – Violence risk factors

Female violence • Female violence seems to be on the rise • Comparable prevalence rate men / women for: – Inpatient violence – Violence towards own children – Intimate partner violence • Intergenerational transfer

Nicholls et al., 2009; Serbin et al. 1998; De Vogel et al., 2012

Nature of violence by women

• Less visible: more domestic, less serious physical injuries • Different expression: more reactive and relational; less sexual and instrumental • Different motives: emotional, relational, jealousy

Fusco, 2011; Nicholls et al., 2009

Violence risk assessment in women • Significant differences men / women in the expression of violence, violence risk factors and manifestation of psychopathy • Most tools developed / validated in males • Questionable predictive validity tools

Are commonly used tools, like the HCR-20 or PCL-R well enough suited for use in women? Garcia-Mansilla et al., 2009; McKeown, 2010

Female Additional Manual (FAM)

• Additional guidelines to HCR-20 / HCR-20V3 for women: – Additional guidelines to several Historical factors (e.g., use of lower PCL-R cut-off score) – New items and additional final risk judgments • Preliminary results: promising reliability & predictive validity for self-destructive behavior & violence to others

De Vogel et al., 2012; De Vogel, & De Vries Robbé, 2013

FAM Gender-specific items Historical items • Prostitution • Parenting difficulties • Pregnancy at young age • Suicide attempt / selfharm • Victimization after childhood*

Clinical items • Covert / manipulative behavior • Low self-esteem

Risk management items • Problematic child care responsibility • Problematic intimate relationship

* This item is no longer needed with HCR-20V3

Multicenter study

Characteristics of women in forensic psychiatry

Acknowledgments • • • • • • • • • • • • •

Gerjonne Akkerman-Bouwsema (GGz Drenthe) Anouk Bohle (Van der Hoeven) Yvonne Bouman (Oldenkotte) Nienke Epskamp (Van der Hoeven) Susanne de Haas (Van der Hoeven) Loes Hagenauw (GGz Drenthe) Paul ter Horst (Woenselse poort) Marjolijn de Jong (Trajectum) Stéphanie Klein Tuente (Van der Hoeven) Marike Lancel (GGz Drenthe) Eva de Spa (Van der Hoeven) Jeantine Stam (Van der Hoeven) Nienke Verstegen (Van der Hoeven)

Multicenter study Aims

•

To gain more insight into criminal and psychiatric characteristics of female forensic psychiatric patients, especially characteristics that may function as risk or protective factors for violence.

•

Possible implications for psychodiagnostics, risk assessment and treatment in forensic psychiatric settings, but possibly also in general psychiatry or in the penitentiary system.

Multicenter study Method

• • • • •

Five Dutch forensic psychiatric settings Ongoing study N > 300 female forensic psychiatric patients N = 275 males matched on year of birth, admittance, judicial status Comprehensive questionnaire including several tools (a.o., PCL-R, Historical items HCR-20 / FAM and HKT-30) was coded based on file information by trained researchers

General characteristics N = 280 women

• Mean age upon admission 35.7 years • 84% born in the Netherlands • At the time of the index offense: – 40% had an intimate relationship – 53% had child(ren), but most of them were not capable of taking care of their children  82% of child(ren) not living with their mother  94% high score on FAM item Parenting difficulties

Criminal characteristics N = 280 women

• Majority had previous contacts with law enforcement: 72% – 20% without conviction • Mean age at first conviction: 23 years • Mean number of previous convictions: 4 • Mostly violent or property offenses

Criminal characteristics 275 women versus 275 men

90

Men: • Younger age first conviction • More sexual offenses, less arson

80

70

60

50

40

% Women % Men

30

20

10

0

Convictions

Contact without conviction

p < .01

Index offenses N = 280 women

Index offenses – Homicide – Attempted homicide – Arson – Violent offenses – Sexual offenses

25% 24% 29% 16% 4%

Victims of index offenses – (ex) Partner – Child(ren) – Relatives / friends – Treatment staff – Stranger

20% 16% 39% 9% 16%

Index offenses

275 women versus 275 men 30

25

20

% Women % Men

15

10

5

0

Homicide

Att.homicide

Arson

Violence

Sexual

All p < .001

Victims Index offenses 275 women versus 275 men

35

30

25

20

% Women % Men

15

10

5

0

(ex)partner

Child (own)

family/acq.

supervisor

stranger

p < .001

Victimization N = 280 women

High rates of victimization • 72% was victimized during childhood – Often by parent(s): 65% – Often a combination of sexual, physical and emotional abuse

• 54% was victimized during adulthood – Often by (ex) partner(s): 82% – Most often physical abuse

Victimization during childhood 275 women versus 275 men

70

60

50

40

% Women % Men

30

20

10

0

Emotional Physical

Sexual

All three

p < .001

Victimization during adulthood 275 women versus 275 men

50 45 40 35 30

% Women % Men

25 20 15 10 5 0

Emotional Physical

Sexual

All three

All p < .01

Psychopathology N = 269 women

High rates of comorbidity • 75% comorbid Axis I and II • High rates of substance use problems: 67% • Borderline personality disorder most prevalent: 59% + 21% traits • Narcissistic PD least prevalent: 3% • Psychopathy (PCL-R cut off of 23, FAM): 14%

Psychopathology

275 women versus 275 men 70

60

50

40

% Women % Men

30

20

10

0

Borderline

Antisocial

Narcissistic

All p < .001

Treatment

N = 280 women • Most had been in treatment before: 88% • High treatment dropout in history: 76% • Incidents during most recent treatment – – – – – – –

Violence Verbal violence / threats Manipulative behavior Self-destructive behavior Arson Victimization Other

34% 47% 55% 47% 8% 8% 68%

Incidents during treatment 170 women versus 170 men

60

50

40

% Women % Men

30

20

10

p < .01

0 Physical

Verbal

Covert

Self-destructive

Arson

Victimization

Violence risk factors HCR-20 / FAM 275 women versus 275 men

Men higher scores on: – Previous violence – Young age at first violent incident – Substance use problems – Psychopathy – Problematic behavior during childhood

Women higher scores on: – – – – –

Prostitution Parenting difficulties Pregnancy at young age Suicidality / self-harm Victimization after childhood

All p < .05

Predictive validity

Incidents during treatment (N = 280 women) • FAM / HCR-20 Historical subscale score modest predictor of physical violence, verbal violence / threats, arson, transfer to another ward due to problems (AUCs .67-.74) • PCL-R total score modest predictor of Manipulative behavior and verbal violence / threats (AUCs .65-.68) All p < .05

Predictive validity violent incidents 275 women versus 275 men • FAM / HCR-20 Historical subscale score: comparable predictive validity • Best predictor for violent incidents: – Women: Problematic behavior during childhood – Men: Young age at first violent incident

• PCL-R total score better predictor violent incidents for men than for women (AUC = .82 vs .68)

Subgroups • Psychopathy – Offenses: more ‘men like’, more often ‘bad’, less ‘sad’ – More treatment dropout and manipulative behavior • Borderline Personality Disorder – More severe victimization – More incidents during treatment, dropout • Intellectual disability – More prostitution – More stranger victims, less homicide All p < .05

Conclusions and implications • Overall, severely traumatized group with complex psychopathology, high comorbidity, many incidents during treatment • Significant differences between women / men and subgroups should lead to different or adapted treatment strategies

Implications • Gender-responsive treatment (e.g., more attention to trauma, parenting skills) • Clear policies (e.g., intimate relationships) • Staff: • Training, intervision, coaching • Support considering high burden BPD

• Collaboration general psychiatry

Future studies • Subgroups: e.g., offense type, diagnoses, intellectual disability • Effect on staff • Effect on children • Dynamic risk and protective factors • Predictive validity tools for women

More information: [email protected] www.violencebywomen.com

Gender differences in motivations for offending Susanne de Haas Van der Hoeven

After the crime has been committed • Why did he/she do this? What motivates people? • Helpful – Relapse prevention plan – Offense analysis – Decrease risk of recidivism?

• Object of research

Presentation outline • The refined taxonomy • Findings

– Reliability – Motivations

 Most common motivations  Index offences

• Conclusions and implications • Future research

Taxonomy of motivations inspired by Coid (1998) Mad

Psychotic, Compulsive urge to harm/kill

Bad

Expressive aggression, Power domination and control, Illicit gain, Excitement, Undercontrolled aggression

Sad

Cry for help/attention seeking, (Extended) suicide, Despair, Influenced by partner

Relational frustration

Revenge, Jealousy, Threatened/actual loss, Displaced aggression, Victim precipitation

Coping

Relief of tension/dysphoria, Hyperirritability

Sexual

Paraphilia, Sexual gratification, Sexual conflict

Reliability of the Clusters (n = 80) • Fleiss’ Kappa • Substantial: – Mad (.68)

• Moderate: – – – – –

Bad (.54) Sad (.55) Relational/social (.49) Coping (.49) Sexual (.56) All p ≤ .001

Most common motivations (n = 436) 35

30

25

20

Women % Men %

15

10

5

0 Mad

Bad

Sad

Relational/social

Coping

Sexual

Index offense and gender differences

• Index offenses: – – – – –

Homicide Attempted homicide Sex offense Arson Violent offense

Index offense: Homicide 35

30

25

20

Women % Men %

15

10

5

0 Mad

Bad

Sad

Relational/social

Coping

Sexual

p ≤ .05

Index offense: Attempted Homicide 40

35

30

25

Women % Men %

20

15

10

5

0 Mad

Bad

Sad

Relational/social

Coping

Sexual

p ≤ .05

Index offense: Sex offense 60

50

40

Women % Men %

30

20

10

0 Mad

Bad

Sad

Relational/social

Coping

Sexual

Index offense: Arson 50 45 40 35 30

Women % Men %

25 20 15 10 5 0 Mad

Bad

Sad

Relational/social

Coping

Sexual

p ≤ .01

Index offense: Violent offenses 50 45 40 35 30

Women % Men %

25 20 15 10 5 0 Mad

Bad

Sad

Relational/social

Coping

Sexual

p ≤ .05

Conclusions Significant gender differences motivations • In general: – Men more by Bad and Sexual – Women more by Sad and Relational/social

• Index offenses: – – – – –

Homicide: Mad (m), Sad (w) Attempted homicide: Bad (m) Sex offense: no significant differences Arson: Coping (m), Sad (w) Violent offense: Sad and Relational frustration (w)

Implications

• Increase effectiveness treatment • Decrease risk of recidivism

Suggestions future research • Refine taxonomy into structured assessment tool • Other influencing factors • Relationship Axis I and Axis II disorders

• Increase of effectiveness treatment and decrease risk of recidivism

Thank you for your attention! Thank you for your attention! • Questions?

• Contact: Susanne de Haas [email protected] / [email protected]

References Coid, J.W. (1998). Axis II disorders and motivation for serious criminal behavior. In A.E. Skodol (Ed.) , Psychopathology and violent crime (pp. 53-97). Washington, DC: American Psychiatric Press. Fusco, S.L., Perrault, R.T., Paiva, M.L., Cook, N.E., & Vincent, G. (2011). Probation officer perceptions of gender differences in youth offending and implications for practice in the field. Paper presented at the 4th International Congress on Psychology and Law, Miami, March 2011. Garcia-Mansilla, A., Rosenfeld, B. & Nicholls, T.L. (2009). Risk assessment: Are current methods applicable to women? International Journal of Forensic Mental Health, 8, 50-61. Klein Tuente, S., Vogel, V. de, & Stam, J. (2014). Exploring the criminal behavior of women with psychopathy: Results from a multicenter study into psychopathy and violent offending in female forensic psychiatric patients. Manuscript under review. McKeown, A. (2010). Female offenders: Assessment of risk in forensic settings. Aggression and Violent Behavior, 15, 422-429. Nicholls, T.L., Brink, J., Greaves, C., Lussier, P., & Verdun-Jones, S. (2009). Forensic psychiatric inpatients and aggression: An exploration of incidence, prevalence, severity, and interventions by gender. International Journal of Law and Psychiatry, 32, 23-30. Serbin, L.A., Cooperman, J.M., Peters, P.L., Lehoux, P.M., Stack, D.M., & Schwartzman, A.E. (1998). Intergenerational transfer of psychosocial risk in women with childhood histories of aggression, withdrawal, or aggression and withdrawal. Developmental Psychology, 34, 1246-1262. Vogel, V. de, & Vries Robbé, M. de (2013). Working with women. Towards a more gender-sensitive violence risk assessment. In C. Logan, & L. Johnstone (Eds.), Managing Clinical Risk: A guide to effective practice (pp. 224-241). London: Routledge Vogel, V. de, Vries Robbé, M. de, Kalmthout, W. van & Place, C. (2012). Female Additional Manual (FAM). Additional guidelines to the HCR-20 for the assessment of violent behavior by women. Utrecht, The Netherlands: Van der Hoeven Kliniek.