Risk Assessment measures in prediction of domestic / interpersonal violence:

Brief overview of some measures and issues

 

 

 

 

 

 

 

 

Report by 

Smita Vir Tyagi

 

Counterpoint, November 2003

 

 

 

 

 

 

 

 

 

Professional Education for Community Practitioners:

Technical Paper Series: User- Report 2003:01

 

 

 

 

 

 

 

 

 

 


 

 

 

 

 

 

 

 

 

 

 

 

 

 

With compliments from Counterpoint

 

 

 

 

 

 

 

 

 

 

 

 

 

 

                                    Technical Paper Series Editor : Smita Vir Tyagi, Ph.d (Candidate)

November 2003

The Technical Paper Series is a professional development service from Counterpoint for community based practitioners. It is an attempt to bridge the gap between research, academia and community based practice in the area of domestic violence. The series will appear from time to time and contact information for copies is given below. You are welcome to share these papers with your colleagues and photocopy the information. However we request that you acknowledge the source whenever you do so.

 

Contact information:

Counterpoint Counselling and Educational Co-operative Inc.

365 Bloor Street East, # 1005

Toronto, Ontario M4W 3L4

Phone (416) 920-0268

Fax: (416) 920-2387

 

 

 


 

 

 

Described in this snapshot paper are brief notes of risk measures in the field. As Dutton & Kropp (2000) point out diverse groups in North America [Police, Corrections and others ]have developed risk, lethality and prediction scales. Many are not easily accessible or do not have accessible published validity data. The scales reviewed here provide information on some popularly used as well as new risk measures that have been developed in response to changing needs. The paper also highlights some key issues in the use of risk assessment and risk prediction measures.

 

Important Caveats for use: Decisions to use one measure or the other should be made based on the following considerations

 

q       Context of use [Community, institution, type of setting, at intake or at program completion, used with victims or offenders]

 

q       Purpose [Exclusion from service, classification, evaluation of client, prediction of violence etc.]

 

q       Partner’s assessment in assessing risk. Recent research shows that victims assessment can be an accurate and important source of risk assessment and risk prediction (Weisz, Tolman & Saunders, 2000). Is such an assessment included in the risk measure?

 

q       Standardization and norms [Standardized on what population, sample size, diversity of sample, generalizability to Canadian population etc.] 

 

q       Reliability and validity [Psychometric properties as reported by scale developers]

 

q       Role of  ‘base rates’ in risk prediction [How often does violence occur in the population and it’s consideration in a risk instrument]. Other issues related to base rates are (to name one) difficulties in predicting homicide because it is rare.  

 

q       Information source for measure [From victim, from offender himself, from collateral sources, from file information. If information is from offender what about response distortions such as faking good or faking bad]

 

q       Theoretical framework of measure.

 

q       Content of measure [If measure incorporates previous history as well as current situation, how well does it address current ‘dynamic’ factors in the present, does the instrument address risk within a certain timeframe? etc.]

 

q       Ease of administration [Is it too long, can it be used on a phone intake etc.] 

 

q       Ease of interpretation [Is it too complicated or specialized for widespread use]

 

q       Training of staff [Does it require special training, qualifications]

 

q       Cost and availability [Is it copyrighted; Is the cost prohibitive; Will authors consent to share it for widespread use]  

 

 


 

 

 

SARA : Spousal Assault Risk Appraisal Guide (Kropp, Hart, Webster & Eaves, 1994; Kropp et al., 2000).

This scale uses empirically established risk indicators

-20 item instrument designed to screen for risk factors in males suspected of /being treated for family related assault. For use with adult males only 

-Purpose is not to provide relative or absolute risk measures using cut-off scores but to enhance professionals judgments about risk if person is a threat to family , spouse or other

-Scores based on information from multiple sources [victim, offender, addictions history, review of collateral records including police reports, criminal records, other psychological assessments where available]

-Well established psychometric properties

-Easy to score. Risk management flows from scores

-One critique is it’s variables which are derived from clinical judgments that rely on professionally trained staff and are subject to inter-rater disagreement. Also, it does not assess relationship status. 

-Copyrighted, sold by Multi Health System, Toronto, Complete kit $72 Cn.

 

DA: Danger Assessment  (Campbell, 1995).

One of the few instruments that uses victim as information source

-19  item paper and pencil checklist style instrument

-Uses information from the victim. Use applied to adult males only.

-Originally developed to assess risk of homicide among batterers but can be used to predict future domestic violence

-Psychometric properties have been published. Replication on larger samples promising (Goodman, Dutton & Benett, 2000 for other references see Danger Assessment website)

-Easy to use and interpret

-Copyright by author but available for widespread use with credit to author. See website for up to date information. www.son.jhmi.edu/research/CNR/homicide/DANGER.htm

 

 

PAPS: Partner Abuse Prognostic Scale (Murphy, Morrell, Elliott,& Neavans, 2003).

Scale obtained information from both offenders and partners in it’s development

-A prognostic index derived from 17 well established risk variables. For use with adult males only

-Found prediction of additive model (putting different risk factors together) was good

-Reported psychometric data are very promising

-Provides cut-off scores for risk

-As of Sept 2003 practice applications await further validation

-Authors can be contacted at University of Maryland

 

Kerry’s Femicide Scale (1998)

Scale based on information gathered from convicted killers and community men

-Scale is unique in that identifies characteristics of men who kill women with whom they’ve been in intimate relationship. For use with adult males only.

-Identifies a constellation of factors including psychological, emotional, physical abuse and attitudes to women.

-Based on an extensive study of perpetrators of femicide [killing of women]

-Preliminary psychometric data appears promising

-Author has indicated that scale would be available for widespread use 

-Author can be contacted via Correctional Services Canada, High Risk Family Violence Program, Kingston.

 

 

 


 

 

 

DOVE: Domestic Violence Evaluation (Ellis, D and associates, York University, Toronto).

New measure being developed on assessing, screening and predicting post-separation domestic violence. Information not available but awaiting imminent publication

 

PAS : Propensity for Abusiveness Scale (Dutton, 1995)

Scale based on information gathered from batterers and non violent community men

-Self-report scale developed by assessing items in a comprehensive assessment. Taps attachment style, anger response, trauma symptoms, parental treatment and self-concept stability. For use with adult males only.  

-Author suggests it maybe more useful for noncriminal populations and predicting emotional abuse and risk for physical abuse

-Preliminary psychometric data looks promising, still developmental

-Author should be contacted for more information @ Dept of Psychology, University of British Columbia 

 

 

Other Interpersonal violence scales

 

 

PCL –R : Psychopathy Checklist- Revised (Hare, 1991)

-20 item instrument is a structured interview and set of ratings based on the interview and corroboration from multiple sources. For use with adult males only 

-Multiple information sources include case history reviews, interviews with family members, criminal and psychiatric records.

-Uses information such as exploitation of others and chronically unstable lifestyle as predictors and has a few violence related items

-Well established psychometric properties, large number of studies supporting it’s use

-Robust predictor of violence including domestic violence although scale was not designed for domestic violence

-Easy to score. Risk management flows from scores

-One critique is that it was designed to predict general violence and criminal recidivism so it gathers little or no information on partner abuse. (See Dutton, 2000 for potential limitation as demonstrated in a recent study) 

-Copyrighted, Training for the instrument sold by Multi Health System, Toronto, Complete kit price $300 Cn.

 

VRS-E1: Violence Risk Scale Experimental Version 1 (Wong & Gordon, 1996)

 

-Uses 6 static (past history) items and 23 ‘dynamic (changing or situational) factors

-Provides a sum of scores. The higher the score, the higher the risk

-Comprehensive evaluation for risk for violent recidivism for incarcerated adult male offenders who will be released  into the community.

-Can identify risk areas which can be then targeted in treatment

-Can assess changes in risk levels as a result of treatment and used for evaluation purposes

-Promising psychometric properties

-Awaiting data on practical application (in general and also in relation to domestic violence)

-Authors can be contacted at Regional Psychiatric Centre, Saskatchewan  

 

 

 


 

 

 

VPS: Violence Prediction Scheme (Webster, Harris, Rice, Cormier & Quinsey, 1991)

and

VRAG: Violence Risk Appraisal Guide (Quinsey, Harris, Rice & Cormier, 1998)

-VPS was formed by combining the VRAG (12 item measure) with a 10 item clinical list called the ASSESS list

-Used for prediction of dangerousness and risk with adult males only

-For use with mentally disordered and serious offenders in prison populations

-Uses developmental, personality, violent and non violent history items. Includes PCL-R score.

-Predicts for 7 and 10 years the risk of violent (non-sexual) acts yielding probabilities

 

 

Other measures that have been used in risk assessment and program evaluation studies are listed below

 

ICB: Inventory of Controlling Behaviors (Meredith & Burns, 1990)

 

IBWB: Inventory of Beliefs About Wife Beating (Saunders, Lynch, Grayson & Linz,1987)

 

CTS 2: Conflict Tactics Scale 2 (Straus, Hamby, Boney-McCoy & Sugarman, (1996)

 

MAI: MultiDimensional Anger Inventory (Siegal, 1986)

 

HTWS: Hostility Toward Women Scale (Check, 1985)

 

Most of these are in the public domain and available for use by practitioners

 

Concluding remarks: Inspite of different measures in use there is a consensus on some key points 

-There are no short-cuts to assessing risk.

-Selection of instrument must be a systematic, thoughtful process.

-Available risk assessment / prediction instruments have important limitations

-Multiple sources of data are important (partner, probation officer, addictions worker etc.)

-Over reliance on any one method can be problematic

- Risk assessment should speak to a time frame (will offend in the next few days, in a few months, ever again)

-Risk assessment must present the conditions under which a batterer is likely to re-offend (example, under the influence of alcohol, on threat of separation etc.)

-Consultations with colleagues who work on the team improve chances of assessing risk

-Practitioners should be trained in risk assessments, be aware of the legal parameters of their

assessments and be prepared to develop and propose action plans for management of risk 

See Monahan and Steadman (1996) for an excellent review of issues in risk assessment for practitioners and Schopp (1996) for communication of risk assessments.

 

 


 

 

 

References

 

Campbell, J. C. (1995). Prediction of homicide of and by battered women. In J.C.

Campbell (Ed.). Assessing dangerousness: Violence by sexual offenders, batterers and child abusers (pp. 96-113).

Check, J.V.P. (1985). The Hostility Toward Women Scale. Unpublished doctoral dissertation. University of Manitoba.

Dutton, D. (1995). A scale for measuring the propensity for abusiveness. Journal of Family Violence, 10, 203-221.

Dutton, D.G., & Kropp, P.R. (2000).A review of domestic violence risk instruments. Trauma, Violence and Abuse,1, 171-182.

Goodman, A.L., Dutton, M., & Bennett, L. (2000). Predicting repeat abuse among arrested batterers: use of the danger Assessment Scale in the criminal justice system. Journal of Interpersonal Violence, 15 (1), 63-74.

Hare, R. (1991). Manual for the Hare Psychopathy Checklist-Revised. Toronto: Multi Health System

Kropp, P.R., Hart, S.D., Webster, C. D., & Eaves, D. (1994). Manual of the Spousal Assault Risk Assessment Guide (2nd ed). Vancouver, Canada: British Columbia Institute on Family Violence.

Kropp, P.R., Hart, S.D., Webster, C. D., & Eaves, D. (2000). The Spousal Assault Risk Assessment Guide (SARA) guide: Reliability and validity in adult male offenders. Law and Human Behaviour, 24 (1), 101-118.

Meredith, C., & Burns, N. (1990). The Index of Controlling Behaviours. Ottawa: ABT and Associates.

Monahan, J., & Steadman, J.H. (1996). Violent storms and violent people: How meteorology can inform risk communication in mental health law. American Psychologist, 51 (9), 931-938

 

Murphy,M.C., Morrell, M.T., Elliott, D.J., & Neavans,M.N.(2003). A prognostic Indicator Scale for the treatment of partner abuse perpetrators.  Journal of Interpersonal Violence, 18 (9), 1087-105.

Quinsey, L.V., Harris, T.G., Rice, E. M., & Cormier, A.C. (1998). Violent offenders: Appraising and managing risk. Washington: American Psychological Association.

Saunders, D.G., Lynch, A.B., Grayson M., & Linz, D. (1987). Inventory of Beliefs about Wife Beating: The construction and initial validation of a measure of beliefs and attitudes. Violence and Victims, 2, 39-55.

Schopp, F.R. (1996). Communicating risk assessments: Accuracy, efficacy and responsibility. American Psychologist, 51 (9), 939-944.

Siegal, J.M. (1986). Multidimensional Anger Inventory. Journal of Personality and Social Psychology, 51, 191-200.

Straus, M.A., Hamby, S.L., Boney-McCoy, S., & Sugarman, D.B (1996). The revised Conflict Tactics Scale (CTS-2). Journal of Family Issues, 7, 283-316. 

 

Webster, C.,  Harris, G., Rice, M., Cormier, C., & Quinsey, V. (1991). The violence prediction scheme: Assessing dangerousness in high risk men. Toronto: Centre of Criminology, University of Toronto

Weisz, A., Tolman, R.M., & Saunders, D. (2000). Assessing risk of severe domestic violence. Journal of Interpersonal Violence, 15 , 75-90.

Wong, S., & Gordon, A. (1996). Violence Risk Scale Experimental Version 1. Department of Psychiatric Research, Regional Psychiatric Centre, Saskatchewan, Canada: Solicitor General of Canada.