CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES THE RVD-BCN RISK ASSESSMENT GUIDE FOR VIOLENCE AGAINST WOMEN PERPETRATED BY THEIR PARTNER OR FORMER PARTNER Barcelona, 15 December 2011 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Institutional author: THE BARCELONA NETWORK TO FIGHT VIOLENCE AGAINST WOMEN Authors and originators of the idea: • Marta Álvarez Freijo Equip d’Atenció a les Dones (Women’s Services Team, Barcelona City Council) • Antonio Andrés-Pueyo Faculty of Psychology (University of Barcelona) • Mónica Augé Gomà Equip d’Atenció a les Dones (Women’s Services Team, Barcelona City Council) • Anna Choy Vilana Mossos d’Esquadra (Catalan Police Force, Ministry of the Interior) • Rocío Fernández Velasco Servei de Mesures Penals Alternatives (Alternative Penal Measures Service, Ministry of Justice) • Carme Fernández Rodríguez Casa d’Acollida (Shelter, Barcelona City Council) • Hubert Foulon Area Bàsica de Salut Besòs (Besòs Basic Health Division, Catalan Health Institute) • Sandra López Ferré Equip d’Assessorament Tècnic Penal de Barcelona (Criminal Technical Advisory Team of Barcelona, Ministry of Justice) • M. Teresa Martínez-Izquierdo Emergency Service (Hospital del Mar – Parc de Salut Mar) • Cristina Martínez Bueno Sexual and Reproductive Health Programme (Catalan Health Institute) • Margarida Saiz Lloret Direcció del Programa de Dona (Women’s Programme Directorate, Barcelona City Council) • Lídia Serratusell Salvadó Àrea de Reparació i Atenció a la Víctima (Victim Recovery and Support Division, Ministry of Justice) 2 The authors will be stated in any exploitation of the work authorised under the license. CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 3 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 4 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES CONTENTS PART I: THE ORIGIN OF THE RVD-BCN Risc Violència Dona [Risk of Violence Against Women] – Barcelona) (p. 7) PART II: THE RVD-BCN AND SHORT USER MANUAL (p. 17) PART III: THE RVD-BCN MANUAL (p 27) 5 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 6 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES PART I THE ORIGIN OF THE RVD-BCN (Risc Violència Dona [Risk of Violence Against Women] – Barcelona) 1 The Barcelona Network to Fight Violence Against Women 2 The RVD-BCN responds to the shared need of various services and institutions 3 The process of scientific validation 4 The keys to success 7 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 8 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 1 The Barcelona Network to Fight Violence Against Women It was founded in 2001, promoted by Barcelona City Council and the Barcelona Health Consortium, with the compelling mission to foster coordination and cooperation between the different institutions involved in the fight to eradicate violence against women in Barcelona, as a pre-requisite in order to offer quality care to wo- men experiencing situations of violence, as well as to their children, and taking the preventive dimension into account. The organisation seeks to give impetus to effective collaboration and cooperation between the different services and professionals that intervene in different sectors (social services, health, police, justice, the judicial sphere and education) in order to take more efficient action in situations of violence against women, to afford better quality care to those affected and to promote preventive strategies. The Barcelona Network to Fight Violence Against Women presents itself as a strategy to move towards achie- ving the following objectives: • To foster collaboration and cooperation between the various public services involved in order to afford comprehensive, effective and quality care to women that experience or have experienced situations of gender-based violence, as well as to their children. • To promote the prevention of violence through pro-active measures and proposals that put an end to the perpetuation of the causes and manifestations of violence. • To prevent secondary victimisation through coordination mechanisms between different services, and provide professionals with specialised training in gender-based violence that facilitates the improvement of the comprehensive intervention approach. • To advance and innovate by way of the analysis and reflection on projects and practices that are being developed in the domains of care services and prevention, to improve intervention approaches, to adapt the services to new scenarios and to put forward future courses of action. The network comprises managers and professionals from institutions and public services from all the areas involved in addressing gender-based violence in Barcelona: the health, social, police, education, justice and judicial sectors. The Government of Catalonia (regional government), Barcelona City Council (local government), Office of the Public Prosecutor (central government) and health, education and social services consortia of the city of Barce- lona are directly involved. 9 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES The network’s structure comprises: • The Technical Directorate, owned and led jointly by the Barcelona Health Consortium and Barcelona City Council. • The network’s Technical Coordination Committee in the area of the city, in which representatives from all the public institutions involved in the fight to eradicate gender-based violence take part. • The Technical Secretariat, coordinated by the Barcelona Health Consortium and Barcelona City Council. • Territorial committees made up of ten networks at district level. • Inter-institutional and multi-disciplinary sub-committees for specific issues. It is within this context that the RVD-BCN Guide came into being. The existence of the Barcelona Network to Fight Violence Against Women for the past ten years has made it possible to build a common discourse that has allowed us not only to agree on the guide presented herein, but also to obtain a qualitatively better result than any of the various assessment instruments employed by the participating services thus far. For this reason, institutional authorship corresponds to the Network as a whole, while the conception of ideas is shared by all those comprising the working group that compiled it. It can be said that without the Barcelona Network to Fight Violence Against Women, the RVD-BCN would not have been possible and, by the same measure, the existence of the RVD-BCN is an indicator of the success of the Network. Why? Because the existence of the Network has created the climate necessary to combine the services of various insti- tutional units in devising a new instrument of use to all of them. This climate has emerged from: • Mutual knowledge: over and above the formal powers and responsibilities of each service, belonging to the Network has given them an insight into the scope and limitations of the different services from the point of view of reality, and has facilitated not only the adaptation of the expectations and requirements of the respective services but also methodologies of analysis and combined work. • Mutual trust and respect: which has been earned over a decade of collaboration among the different services. • Possibility of joint growth and technical creation: through a broad spectrum of means: - The exchange of information between services and devices. - Thematic debates. - Case studies and the creation of a procedure and a common technical culture in the provision of care and coordination between services. - Shared ongoing training on specific topics of common interest. - Setting up working groups on various topics, such as the one drawn up by the RVD-BCN for instance - Organising annual conferences open to all the professionals of the entire body of participating services both in the city as well as in the ten districts. In conclusion, the Network has made it possible to achieve the necessary requirements for effective and produc- tive collaboration and cooperation which has facilitated the creation of a shared tool that is much better than any of those employed to date by the different participating services. It is for this reason that the guide incorpo- 10 rates the initials of the city of Barcelona in its name. CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 2 The RVD-BCN responds to the shared need of various services and institutions THE SITUATION PRIOR TO THE RVD-BCN The situation prior to the RVD-BCN is characterised by the following: • Each service or institution employs their own assessment tool, for internal use, which is not shared with the rest of the services offered by other institutions that are also involved in the intervention. This renders collaboration difficult since risk assessments for the same case may differ according to the instrument used. • Many of these instruments are subject to a high degree of professional subjectivity: they do not quantify the risk factors or establish a matrix for the risk assessment (low, moderate or high). Assessment hinges on the criterion and interpretation of the professional in question. • Not all the risk assessment instruments prior to the RVD-BCN take periodic reappraisals of the situation into consideration. • Not all the current instruments allow appropriate assessments to be undertaken in situations that are unique. • Most of the instruments currently used do include risk factors that depend on the woman’s vulnerability or her perception. • Most of the instruments used do not incorporate the function of warning of foreseeable circumstances that may heighten the risk in the future. WHY WAS THE RVD-BCN CREATED? Because a risk assessment tool is needed which: • Is shared among the different services that intervene in a given case, so that the same information yields the same assessment. • Eliminates professional subjectivity to assess risk, but at the same time allows this assessment to be adapted to the unique nature of specific cases. • Features reappraisal as one of its defining characteristics. • Incorporates the woman’s vulnerability factors and her perception. • Incorporates the function of warning of foreseeable circumstances that may heighten the level of risk in the future. 11 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES WHAT IS THE RVD-BCN? A tool that helps professionals that deal with women experiencing situations of gender-based violence to assess the risk of severe acts of violence perpetrated by their partner or former partner occurring in the short temporal delay. The guide, the explanation of its characteristics and its instructions for use are part of the second and third sections of this document. 12 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 3 The process of scientific validation of the RVD-BCN Guide (from February 2010 to June 2011) The RVD-BCN Guide was subject to a process of scientific validation that began in February 2010 and ended in June 2011. This validation process was led by the University of Barcelona’s Group of Advanced Studies on Violence (GEAV), and was financed equally by Barcelona City Council, the Barcelona Health Consortium and the Catalan Institute for Women (ICD). Professionals and services from all the sectors involved, i.e. justice, health, social services and police forces (Mossos d’Esquadra [Catalan Police Force] and Guàrdia Urbana [Barcelona Municipal Police Force]) have taken part therein. It should be mentioned that health services cannot take part in research in Catalonia without the prior approval of a clinical research ethics committee. In this case, on 1 April 2010, the Ethics and Clinical Research Committee of the IDIAP Jordi Gol i Gurina (Primary Care Research Institute) granted its final approval to the RVD-BCN pro- ject. The objectives of the validation process are as follows: • To verify the usefulness of the RVD-BCN. - To compare predictive validity. - To compare reliability. • Internal consistency • Inter-observer reliability - To compare the sensitivity and specificity for different forms of violence/severity. • To adjust the scale and cut-off points according to: - The severity of violence - The time interval - The population of victims • To compare the usefulness of the RVD-BCN in the different services that are users of the Network. In view of the results of the validation process, the working group that authored the guide came together once again to make pertinent decisions on improving the initial version, to obtain an instrument that adapts better to the objectives proposed, that is to say, to optimise its predictive power. The RVD-BCN Manual contained in the third part of this document features a summary of the main findings of the validation process. For a full description of this process and the results obtained, please see the report submitted by Dr. Antonio Andrés-Pueyo, director of the Advanced Studies Group on Violence (GEAV), to the two institutions that manage and coordinate the Network and the members of the Technical Monitoring Committee for the Process of Validation and Implementation of the RVD-BCN. 13 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 4 The keys to success 1. The RVD-BCN was conceived from a real need felt by professionals in different services and insti- tutions that cater for women experiencing situations of gender-based violence. It was precisely one sector of professionals that put forward the proposal of creating an inter-institutional working group to the Network’s plenary session, which based on the instruments used by each service, would build a new joint tool that would improve the predictive power of the risk of violent acts against women perpetrated by their partner or former partner. 2. Institutional backing both in drawing up the RVD-BCN, which took ten months, and the process of scientific validation proposed by the working group that authored the guide in presenting the findings to the Network’s Technical Committee. This backing is particularly visible on the following occasions: • Three institutions financed the validation process in equal measures: Barcelona City Council, the Barcelona Health Consortium and the Catalan Institute for Women. • A letter was sent, signed by the councillor for Women and Young People of Barcelona City Council, the general coordinator of the Health Corporation of Barcelona and the president of the Catalan Institute for Women, addressed to the upper echelons of the different institutions likely to take part in the validation process of the RVD-BCN, and within which their collaboration was requested. • Each institution designated the services and the professionals that would participate in the validation process. • Subsequently, a technical monitoring committee was appointed and each institution appointed their repre- sentatives. • Once this process was completed, the report outlining the findings of the scientific validation process was submitted to the highest authorities of the two institutions that coordinate and manage the Network and the members of the Monitoring Committee. 3. Scientific rigour was guaranteed at all times: • From the outset, the working group had the collaboration of an external expert, Dr. Antonio Andrés-Pueyo, director of the University of Barcelona’s Advanced Studies Group on Violence (GEAV). • Expert management of the RVD-BCN scientific validation process was guaranteed by way of the GEAV. 14 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 4. An institutional body was appointed that assumed the role of coordinating the entire process with the agreement of the other institutions involved. This function was undertaken by the Women’s Programme Directorate of Barcelona City Council. 5. The participation of all the institutions involved was ensured in the overall monitoring of the validation process and implementation of the RVD-BCN in the day-to-day practice of beneficiary services through the constitution of a technical committee with the capacity to introduce the necessary regulation measures throughout the process to facilitate the achievement of the proposed objectives. This Committee is made up of representatives from the following institutions: • Barcelona City Council • Ministry of Justice • Mossos d’Esquadra (Catalan Police Force) • Sexual and Reproductive Health Care Service (ASSIR) • Primary Health Care Division • Barcelona Health Consortium • Guàrdia Urbana (Barcelona Municipal Police Force) • Social Services Consortium of Barcelona 15 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES THE RVD-BCN (Risc Violència Dona [Risk of Violence Against Women] – Barcelona) and SHORT USER MANUAL 16 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES PART II THE RVD-BCN (Risc Violència Dona [Risk of Violence Against Women] – Barcelona) and SHORT USER MANUAL 17 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 18 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES THE RVD-BCN .Risk assessment guide for violence against women perpetrated by their partner File no. / Record no.: Date of assessment: / / Woman’s name and surname: Spanish or Foreign National Ientification no. / Passport no.: Date of birth: / / History of violent conduct by partner or former partner YES NO UNKNOWN 1 Assaults or physical and/or sexual violence towards the woman or former partners in the last eighteen months. 2 Assaults or violence against other persons whether or not family members (children or others). 3 Assaults on the woman during pregnancy. 4 Police/criminal record of violence against the partner/former partner (current partner or other partners in previous incidents). 5 The offender has breached the woman’s court protection measures. Threats and/or severe abuse of the woman YES NO UNKNOWN 6 The woman has received serious and credible threats to her physical integrity, and/or with the use of weapons. 7 The woman has suffered severe emotional and verbal abuse in the last six months. Aggravating circumstances YES NO UNKNOWN 8 The woman informs the alleged offender of her wish to separate or the separation occurred less than six months ago. 9 Increase in the frequency or severity of the violent incidents in the last six months. 10 The alleged offender abuses drugs and/or alcohol. 11 Diagnosis or history of severe mental disorder in the alleged offender. 12 Possession or easy access to weapons by the alleged offender. 13 Suicide attempts or ideation on the part of the alleged offender. 14 Extreme control of women’s conduct out of jealousy or similar feelings. Woman’s vulnerability factors YES NO UNKNOWN Woman’s social isolation and/or lack of personal resources, and/or the 15 woman’s justification of the violence inflicted by the alleged offender, and/ or the presence of the woman’s children that are minors and/or dependents. Woman’s appraisal of the risk situation YES NO UNKNOWN 16 The woman believes that the alleged offender is capable of killing her personally or having her killed by third parties. SUM OF POSITIVE RESPONSES / RISK ASSESSMENT TOTAL UNKNOWNS The RVD-BCN is useful if information LOW RISK (1 to 7 positive responses) is available on seven or more risk factors Otherwise it is recommended MODERATE RISK (8 to 9 positive responses) that final assessments not be made until the minimum number of the afo- HIGH RISK (10 to 16 positive responses) rementioned risk factors is reached. 19 The authors will be stated in any exploitation of the work authorised under the license. CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Other rellevant risk factors to assess the violence risk for this woman that are missing in the above list (list but do not score) She is pregnant when this assessment is being undertaken. There are indications that the partner or former partners intends to kill her. Other (specify): - - - RISK ASSESSMENT CONDUCTED BY THE PROFESSIONAL The risk assessment by the professional is made based on the aforementioned sections (score obtained, number of risk factors without information and the presence of other important risk factors for this assessment). LOW RISK MODERATE RISK HIGH RISK Observations MANAGEMENT PLAN ACCORDING TO THE CURRENT EXISTING RISK ASSESSMENT FORESEEABLE CIRCUMSTANCES THAT MAY HEIGHTEN THE LEVEL OF RISK IN THE FUTURE (ALERT) (For instance: the offender is released from prison, the offender is returning from their country of origin or other destinations, etc.) Professional conducting the assessment: Name and surname(s) or identification no.: Email address: Tel.: Fax: 20 Spanish or Foreign National identification No. / Passport No. Address: The authors will be stated in any exploitation of the work authorised under the license. CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES THE RVD-BCN SHORT USER MANUAL 1. Objective To afford professionals that deal with women experiencing situations of gender-based violence with a tool that helps them assess the risk of severe acts of violence perpetrated by their partner or former partner occurring in the short term. 2. Characteristics • It includes compared and valid risk factors to assess the probability and immediacy of the risk of acts of violence occurring. • It has an indicative but not evidential nature. It helps the professional assess the risk of acts of violence occurring. 3. Functions • To assess the risk. • To record what action is being undertaken according to the current existing risk assessment. • To provide warning of possible foreseeable circumstances that may heighten the level of risk in the future. 4. When is the use of the RVD-BCN necessary • When the professional is aware that the woman is experiencing a situation of violence or the woman herself states so. • In the event of changes in the woman’s circumstances that may in some way alter the risk of violence. • As a preventive measure every three months. 5. When is it the use of the RVD-BCN not necessary • When there is a current risk assessment and we do not have additional information. (For instance: the woman has 24-hour police protection by court order; when another service has already applied the guide in the last month and the circumstances have not changed since then, etc.). • The offender has been imprisoned without probation, except when: - There is a possibility of him of causing an assault by way of third parties (family members or others). - His release from prison (probation or definitive release) is due and the warning function of RVD-BCN must be activated. 6. Structure 6.1 Risk assessment • List of factors: a total of 16 risk factors were selected with three response options (“yes”, “no” and “unknown”). They are grouped under five categories: 1. History of violent conduct by partner or former partner 2. Serious threats and/or abuse of the woman 3. Aggravating circumstances 4. Woman’s vulnerability factorsr 5. Woman’s appraisal of the risk situation A description of each risk factor can be found in the final section of these instructions. 21 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES The sum of affirmative responses indicates a level of risk according to the range score defined by RVD-BCN norms: 1. Low risk (1 to 7 points) 2. Moderate risk (8 to 9 points) 3. High risk (10 to 16 points) • Number of risk factors on which information is held: this guide is useful if there is information on seven risk factors or more. Otherwise, it is recommended that definitive assessments not be made until the mini- mum number of the aforementioned risk factors has been reached. • Presence of other rellevant risk factors to assess the risk that the professional is missing in the above list: there are risk factors that, though important, only occur in some cases, which is why they have not been included in the list; however, they must be borne in mind when making the final assessment of the existing risk. For instance: - She is pregnant when this assessment is being undertaken. - There are indications that the partner or former partners intends to kill her. 6.2 Description of the action undertaken according to the risk Brief description of the intervention carried out which is directly linked to the risk assessment. 6.3 Foreseeable circumstances that may heighten the level of risk in the future (alert) For instance: the offender is released from prison (on probation or definitive release), the offender is returning from their country of origin or other destinations, etc. 7. Methodology • The professional dealing with the woman must fill out the form. • They must do so based on the information given to them by the woman herself, and if they should have access, also based on the information provided to them by other services and available documents (court rulings, etc.). • The questions do not have to be formulated as if a questionnaire were being administered. Often the wo- man provides useful information spontaneously, or it is the professional who poses questions to the women in a flexible manner and at the opportune time in the context of an interview, or, depending on the type of service in question and how the woman feels, over the course of several interviews. • It is not designed to be a questionnaire that the woman may fill out directly. 22 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 8. Description of risk factors History of violent conduct on the part of the partner or former partner 1. Assaults or physical and/or sexual violence towards the woman or former partners in the last eighteen months. Non-accidental action that during the last eighteen months has led to physical and/or sexual harm. Physical violence is understood to be any assault (punching, slapping, scratching, pushing, kicking, etc.) with the result or risk of causing physical injury or harm. Sexual violence covers any act of a sexual nature without the consent of the woman, inclu- ding exhibition, observation and the imposition of sexual relations by means of violence, intimidation, abuse of the offender’s position or emotional manipulation. For further information on this risk factor, please see the RVD-BCN Guide manual on page 48. 2. Assaults or violence against third parties whether or not family members (children or others) The alleged offender committed a physical/sexual assault or attempted assault on: • Members of the family (whether blood or political ties) and/or • Acquaintances (male or female friends, work colleagues or neighbours) and/or • Strangers. For further information on this risk factor, please see the RVD-BCN Guide manual on page 49. 3. Assaults on the woman during pregnancy Physical and/or sexual assault during the woman’s pregnancy by her partner/former partner. It also includes severe and persistent psychological abuse. For further information on this risk factor, please see the RVD-BCN Guide manual on page 50. 4. Police/criminal record of violence against the partner/former partner (current partner or other partners in pre- vious incidents). The criminal and/or police record of the alleged offender shows a history of physical, sexual or emotional abuse of a current or former intimate partner. For further information on this risk factor, please see the RVD-BCN Guide manual on page 50. 5. The offender has breached the woman’s court protection measures The offender has breached a protection measure. He has been subject to convictions or police investigations as a result of having breached a protection measure for his partner or former partner. For further information on this risk factor, please see the RVD-BCN Guide manual on page 51. 6. The woman has received serious and credible threats to her physical integrity, and/or with the use of weapons. The alleged offender threatens, intimidates or coerces the woman with or without the use of weapons. For further information on this risk factor, please see the RVD-BCN Guide manual on page 52. 7. The woman has suffered severe emotional and verbal abuse in the last six months. Non accidental action that has caused emotional harm in the last six months. Severe emotional and verbal abuse co- vers degrading, humiliating, insulting, threatening, criticising, belittling, ridiculing, vexing, demanding obedience and submission, verbal coercion or any other limitation of their freedom. For further information on this risk factor, please see the RVD-BCN Guide manual on page 53. 23 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Aggravating circumstances 8. The woman informs the alleged offender of her wish to separate or the separation occurred less than six months ago. At the time of the assessment and/or incident, the woman informed the alleged offender of her intention to end the relationship or they have been separated for less than six months. For further information on this risk factor, please see the RVD-BCN Guide manual on page 53. 9. Increase in the frequency or severity of the violent incidents in the last six months. Increase in the frequency and/or severity of the aggressive conduct towards the partner or former partner over the last six months. For further information on this risk factor, please see the RVD-BCN Guide manual on page 54. 10. The alleged offender abuses drugs and/or alcohol. The alleged offender is dependent or abuser of legal and/or illegal substances (cocaine, alcohol, hashish, heroin, psy- chotropic drugs, etc.). For further information on this risk factor, please see the RVD-BCN Guide manual on page 55. 11. Diagnosis or history of severe mental disorder in the alleged offender. Diagnosis or history of severe mental disorder in the alleged offender, considering as such, for instance: • Personality disorder with anger, impulsivity and emotional instability. • Schizofrenic disorder, major depression, bipolar disorder, paranoid disorder and similar disorders. For further information on this risk factor, please see the RVD-BCN Guide manual on page 56. 12. Possession or easy access to weapons by the alleged offender. The alleged offender has attempted suicide, has had or has thoughts about suicide. For further information on this risk factor, please see the RVD-BCN Guide manual on page 57. 13. Suicide attempts or ideation on the part of the alleged offender. The alleged offender has attempted suicide, has had or has thoughts about suicide. For further information on this risk factor, please see the RVD-BCN Guide manual on page 57. 14. Extrem control of the woman’s conduct out of jealousy or similar feelings. The alleged offender manifests controlling behaviour, stalking and restriction of the woman’s freedom motivated by jealousy or similar feelings (for instance, cultural beliefs or beliefs of another kind). For further information on this risk factor, please see the RVD-BCN Guide manual on page 58. 24 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Woman’s vulnerability factors 15. Woman’s social isolation and/or lack of personal resources, and/or the woman’s justification of the violence inflicted by the alleged offender, and/or the presence of the woman’s children that are minors and/or dependents. The woman does not possess or have personal, social, family, economic and/or work resources at hand, or they are very scarce. Isolation occurs in one or several of the aforementioned areas, and/or the woman justifies, minimises or denies that the alleged offender has inflicted a form of violence on their person (he is a good person but when he drinks he loses control and becomes violent; he does not want to do it but sometimes he gets nervous; the thing is that I do not do things well and he gets angry, he acts in accordance with his customs and beliefs, etc.)., and/or the woman has children that are minors and/or dependents in her care, which may or may not be common to the alleged aggessor. For further information on this risk factor, please see the RVD-BCN Guide manual on page 59. Woman’s appraisal of the risk situation 16. The woman believes that the alleged offender is capable of killing her personally or having her killed by third parties. The woman has a real perception of the high risk to which she is exposed and is convinced that the alleged offender is capable of killing her or fulfilling his death threats through third parties (family members, friends, hitmen, etc.). For further information on this risk factor, please see the RVD-BCN Guide manual on page 60. 9. Services that can use the RVD-BCN Guide The RVD-BCN is validated for use in the following areas/services: • Ministry of Justice (Crime Victim Support Office, Criminal Technical Advisory Team and others) • Coordinating Prosecution of Domestic Violence (Supreme Court of Justice) • Police forces (Mossos d’Esquadra and Guàrdia Urbana) • Health services (emergency, primary care services, hospitals) • Social services (general services, care division for gender-based violence and the remainder of the care divi- sions) • Catalan Institute for Women (ICD) For further information, please see the RVD-BCN Manual in the third part of this document. 25 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 26 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES PART III THE RVD-BCN MANUAL 1 Introduction 2 Specific aspects of intimate partner violence (IPV) 3 IPV prevention: violence risk assessment and management 4 The RVD-BCN: guide and risk factors 4.1 Risk factors of the RVD-BCN 4.2 Other risk factors 5 Technical specifications of the RVD-BCN 6 The RVD-BCN Short User Manual 7 Conclusions and recommendations for use 8 References 27 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 1 Introduction The professional activity of practitioners dealing with women experiencing gender-based violence calls for cons- tant attention to predicting new incidents of violence. Concern over the possibility of mistreatment, violence or abuse of the victim, particularly the most severe and irreversible abuse, continuing or worsening is one of the overarching reasons for intervening and thereby ensuring the safety and well-being of the victim. We present the RVD-BCN Guide and its user manual, which is to facilitate the making of predictions on the occurrence or recurrence of severe violent incidents against women perpetrated by their partner (current or former partner). The RVD-BCN is used to conduct risk assessments of intimate partner violence. It constitutes a procedure which establishes guidelines and norms to facilitate the decision-making of professionals that work with women expe- riencing situations of violence throughout the stages of their intervention. This guide has been designed for use primarily with female victims but can also be used, by extension, with offenders. Violence against women in the context of an intimate relationship, which shall be identified with the acronym IPV (intimate partner violence), is a complex and repetitive act that is difficult to eradicate and hinges on many elements, noteworthy among which are certain risk factors of the offender, the victim and realities that surround the relationship held with the partner and also with the former partner. Identifying the risk of IPV emerging is the first step towards preventing it, and must be personalised and designed according to the dynamic reality of IPV itself. To effectively assess the risk of intimate partner violence, the following are required: a) knowledge of the phenomenon of violence against women; b) skills in assessment techniques and the evaluation of relevant information in conflict situations; c) ultimately, knowing how to make decisions and communicating them to the agents involved in those situa- tions. The RVD-BCN puts forward a guide procedure to contribute –in the violence risk assessment process– towards decision-making on the most probable future violent acts, based on so-called structured professional judge- ment. This manual presents the RVD-BCN guide, the instructions on how to use it as well as other relevant infor- mation in order to guarantee effective use thereof. As explained in the first part of this document, the origin, creation and development of RVD-BCN were an initia- tive of the Barcelona Network to Fight Violence Against Women. The Network is a structure that oversees the technical cooperation of the services and public institutions involved in the fight towards eradicating gender- based violence. It was created in 2001 and promoted by Barcelona City Council and the Barcelona Health Con- sortium. It encompasses the social, health, police, educational and judicial services that work towards gender- based violence prevention and care. The RVD-BCN was designed so that all the practitioners that work in these services could benefit form its use. The characteristics and features of the RVD-BCN allow different professionals – whatever their training and qualifications – to have a guide for assessing the risk of intimate partner violence, which is objective, reliable and is endowed with a proven predictive power. It will no doubt be of enormous help in their day-to-day work in favour of the safety and well-being of the victims of intimate partner violence. 29 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 2 Specific aspects of intimate partner violence (IPV) In 1991, the United Nations Draft Declaration on Violence Against Women described violence against women as “any act, omission, controlling behaviour or threat that results in, or is likely to result in, physical, sexual or psychological injury to women”. Under what the World Health Organisation (WHO) calls “violence against wo- men” is intimate partner violence, which is one of the most common forms of such violence. In both scientific and professional literature, violence against women in which the offender is or has been the intimate partner of the victim is mostly called “intimate partner violence” (IPV). It is sometimes called “gender-based violence, even “domestic violence”, but it must be borne in mind that IPV is a specific form of violence and is not exactly the same as the aforementioned ones. The legal designation of “gender-based violence” as defined in current Spanish legislation is equivalent to the term IPV we employ in this manual. IPV is not restricted to the violence in which the victim is always a woman and the offender is always a man; in fact, in its original and strict sense, it is used irrespective of the gender of the offender and the victim. The term partner refers specifically to the “intimate couple”, comprising two people, men or women of legal age or adolescents that have had consensual intimate relations over a minimum period of several weeks, whether they have cohabited or not continuously at the same address. Therefore, it includes spouses and former spouses, those in current or former de facto relationships, fiancés, and also more casual intimate partners. In the latter case, according to the professional’s judgement, it must be borne in mind, in order to consider the IPV, if there has been a consensual, relatively stable and lasting intimate relationship between the members of the couple. Needless to say, this definition excludes fraternal, work or family relations whose relationship is not based on the existence of current or past sexual and intimate relations. Intimate partner violence was defined in a strict sense by the WHO in 2005 as “the range of sexually, psychologi- cally and physically coercive acts used against adult and adolescent women by current or former male intimate partners”. In this sense, in Spain and Catalonia, gender-based violence, the general term based on the definition of Organic Law 1/2004 on Comprehensive Protection Measures against Gender-Based Violence, corresponds to a form of partner violence in which the offender is a man and the victim is a woman. For this reason, the full description of the object of analysis of the RVD-BCN Guide is “violence against women in heterosexual partner relationships”. Throughout this manual, we will call simply call it “intimate partner violence” (IPV) for the sake of simplicity and realism. The reality of intimate partner violence reveals to us that the frequency of severe violent acts perpetrated by the man on his female partner (or former partner) is the highest (almost over 90%) of the most severe forms of violence (Andrés-Pueyo, 2010)1 .On the basis of this epidemiological observation, the term IPV has been generalised to shorten the term “violence against women in heterosexual partner relationships”. IPV represents a constellation of abusive and violent acts that men inflict on women, current or former intimate partners, and which range form severe violent acts, murders, kidnappings and severe physical assaults to non- sexual harassment, humiliation, coercion, threats, extortion, sexual abuse and economic violence. Therefore, it is a complex phenomenon that covers different forms of violent conduct, attitudes, feelings, practices, experien- ces and relationship styles between members of an intimate couple (or former couple) that causes harm, malai- se and serious personal losses to the victim. It is a varied and complex pattern of violent and coercive behaviour, 30 1.Physical or sexual violence, threats of physical harm or death and stalking. (Andrés-Pueyo, 2010). CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES which is generally repetitive and includes different acts (sometimes intentions, such as in the case of threats) of intimate partner violence. All these acts, which can be combined and spread over time in a chronic manner, seek to subject the victim to the power and control of the offender through fear, insecurity, suffering, pain and physical or psychological harm. In general, and without intervention, IPV is recurrent and repetitive, very difficult to eradicate because sometimes it is a constituent part of a couple’s relationship. In some cases, the process escalates in the frequency and severity of violent incidents that cause serious harm and consequences to the victim, which can even end up causing personal or social incapacitation or death. The WHO defines four forms of violence: physical, sexual, psychological and that arising from controlling be- haviour on the part of partners or former partners (husbands, boyfriends, etc.). The RVD-BCN is built around assessing the risk of severe violence including physical, sexual and certain threats and controlling behaviour and stalking that, due to their peculiarities, may become physical and/or real sexual acts of violence and that may become a reality in the future. Physical violence2 impliesany intentional act of force that causes or could cause bodily harm to the victim assaul- ted (bruises, injuries, burns, fractures, beatings, etc.). It includes any act involving force against a woman’s body as well as any failure to assist a woman, intentionally or negligently, with the result or risk of causing her physical injury or bodily harm. Physical violence is identified empirically with acts such as being punched or receiving the impact of an object thrown with the deliberate intention of inflicting injury, being pushed or having hair pulled, being punched with the fist or hit with any other object that could cause injury, being kicked, dragged or beaten, strangled or deliberately burned and threatened with a gun, knife or another weapon, attempting to knock a person down with a vehicle, etc. Sexual violence describes a broad range of specific acts against the freedom and rights of sexual intimacy, exe- cuted against the will of the woman through threats, intimidation, coercion or when the woman is in a state of unconsciousness or helplessness. It covers any act of a sexual nature that does not have the consent of the woman, including exhibition, observation and the imposition of sexual relations, by means of violence, intimida- tion, domination or emotional manipulation. It also includes different forms of sexual abuse with psychological and/or physical coercion to obtain sex, forcing unwanted sexual practices on the woman, forcing the woman to have sexual relations when she is ill, pregnant or postpartum, forcing her to swap partners, to have sexual rela- tions with animals, the use of pornography or sexual games without consent, the filming of intercourse sessions and the undue use of new technologies, etc. The difficulty the woman has in identifying this form of violence in- flicted by her partner is often related to beliefs on the obligation and right to maintain intimate relations within the couple. Sexual violence is identified empirically according to the following three elements: a) being forced to have sexual relations against the woman’s will; b) having sexual relations for fear of what her partner might do, and c) being forced to perform a sexual act she considers degrading or humiliating. 2. Violence: strategy to inflict deliberate harm. Mistreatment: frequent violence. Abuse: violence inflicted in the context of a trusting relationship. 31 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Psychological violence, also sometimes called emotional abuse and/or verbal violence, refers to abusive acts or behaviour that may cause feelings of belittlement, intimidation, guilt, suffering and undermining or loss of self- esteem in the victim. Some examples would include humiliating, insulting, belittling, shouting, ridiculing, accu- sing, social or economic isolation, threatening, abandoning, destroying or damaging personal objects. However, it also includes denying the violence, blaming or attributing the responsibility for the violent acts to the victim. Psychological violence covers any behaviour or deliberate omission that causes degradation or suffering of the woman as a result of threats, humiliation, vexation, the demand for obedience or submission, verbal coercion, insults, isolation or any other limitation of her freedom. It includes acts such as insulting and foul language, de- grading, ridiculing, humiliating, undermining self-esteem, guilt, blaming and morbid distrust towards her. It also includes threatening postures and gestures (for instance, threatening violence or to take away the children) and restrictive behaviour (for instance, controlling friendships, withholding money, limiting the woman’s freedom to leave the home). Grouped under this form of violence is certain destructive behaviour (for example, towards objects of economic or sentimental value to the victim or the maltreatment of the victim’s pets) that seeks to frighten the partner and make them suffer emotionally. Also part of psychological violence is the woman bla- ming herself for the man’s violent conduct. Though apparently certain actions such as breaking or destroying objects, spaces or animals belonging to the woman may be considered physical violence, for the purposes of IPV risk assessment they are deemed to be psychological violence. This form of violence may have a symbolic value, for instance, tearing up photographs or letters, or a material value, such as making jewellery disappear, taking over the car, breaking furniture, torturing or even killing pets, burning the house down, etc. Threats and stalking. These two forms of violence, very common in intimate partner violence, deserve to be mentioned separately and specifically. Firstly, on account of the form of behaviour involved in these two forms of assault, we consider them psychological (there is no direct physical contact) but they are specific because they show, when they are genuine, the offender’s intention to carry out certain forms of physical violence. In the first case, threats, especially when they refer to anticipating physical assaults and actions, are very similar to physical violence not so much because of their consequences but rather because they indicate the motivation and anticipate future assaults. Generally speaking, it is appropriate to consider them as more a form of physical than psychological violence when they are not fleeting and very circumstantial (Kroop, 2008). It must be remem- bered that certain threats, especially those that cause a firm belief and/or uncontrollable fear, inflicted on the woman by her partner, should be considered analogous to other acts of physical violence (Andrés-Pueyo, 2009). With regard to harassment, it should be mentioned that this form of violence that generally involves harassing one’s partner is usually identified as non-sexual harassment or stalking to distinguish it from other forms of harassment with a clear sexual intent inflicted by men on women. It is a phenomenon that has acquired much prevalence since legal measures imposed on the abuser force him to stay away from his partner. Acts of stalking include following or spying on the woman, sending letters, emails or making direct or indirect phonecalls, ma- king themselves visible in places frequented by their partner, stealing their post and not giving them the post that corresponds to them, sending them anonymous letters with threats and instructions on what to do, etc. 32 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Also included in this section is behaviour that does not respect the privacy of the woman, for instance, contro- lling phonecalls, email, correspondence, etc. The last category of violence against women which has been distinguished is what the WHO calls controlling behaviour and it is almost always characteristic of violence inflicted in the domestic or intimate partner context. Controlling behaviour includes acts such as keeping the woman from seeing her friends (threatening them), res- tricting the woman’s contact with her family of birth, insisting on knowing where she is at all times, ignoring or treating her indifferently, getting angry with her if she speaks with other men, constantly accusing her of being unfaithful, controlling her access to healthcare and other similar behaviour, economic control, etc. In fact, it is a mixed category which encompasses many forms of behaviour and attitudes on the part of the offender that are already identified to an extent in psychological violence, threats and stalking. All these forms of violence are common in IPV. Some new and important elements must be added to have a full picture of the reality of this phenomenon. These are features of IPV which are common to other forms of vio- lence against women. Three of them are set apart: severity, recurrence and a combination of forms of violence. • Severity. Although all violence, even low-intensity violence, inevitably causes harm, and the effects are subjective, we can distinguish different levels. There is a graduation, which is difficult to quantify, which ranges from milder incidents of psychological abuse to severe and recurring trauma, or even to assaults that result in death (lethality). The severity of physical violence is an essential feature in IPV assessments. For instance, physical violence is stabbing a knife in the abdomen and also kicking or punching. In fact, the risk of death, the most severe form of physical violence, must be considered separately due to its great specificity (Campbell, 1985, 2003), especially if we wish to assess the risk this action may cause. The RVD-BCN Guide is not created to specifically assess the risk of murder of the woman (Andrés Pueyo, 2009, 2010). For this form of violence, there is the Danger Assessment Tool (DA) (Andrés-Pueyo et al., 2008). Differences in the degree of violent conduct are highly important. Sexual violence can vary from forcing unwanted sexual relations to forcing the woman to engage in sexual relations with third parties or in public. Psychological violence can vary between occasional humiliation, coercion and the threat of assault and death on children or parents of the victim. Kidnapping, telephone stalking, etc., any form of violence can by considered in levels of severity. When it is a question of anticipating the likelihood of future violence, this is a parameter that must not be overlooked, and the RVD-BCN has very much taken this into account in its design. This ranking according to the severity of the violence applies to threats, that range from general and unspecific statements to other more specific and explicit ones, such as threatening to kill the victim or their family members, and may also include the offender’s suicide. The same can be said of stalking, which may be mild, such as a sporadic phonecall or an anonymous letter in the letter box, or very serious, such as permanently following the victim on the street, constantly entering the home of the partner or former partner illegally when they no longer live together, etc. The casuistry is very extensive but the severity of the incidents and their potential consequences should be distinguished and, in addition, they should be broken down individually according to the vulnerability of the victim. 33 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Physical Sexual Psychological Threats Stalking Injuries which Imposing violent Public humiliation Public threats making Constant physical require medical and dangerous and coercion reference to how following, which attention and/or sexual relations and when they will is very apparent Severe hospitalisation be carried out and intrusive Assaults that do not Demanding sexual Insults and occa- Expressions of a Occasional pho- cause injuries or relations as a sional emotional generic nature in necalls with which are of little marital duty belittlement arguments and hurtful content Mild medical importance situations of anger Table 1. Forms of violence and severity of violent actions • Recurrence, frequency and chronicity. IPV is a repetitive phenomenon, which is very often chronic and is sometimes called “habitual abuse”. In fact, the models that speak of the cycle of violence, such as that of L. Walker (Walker, 1999) show this recurrence, which in many cases conceals a very long history span- ning five to ten years on average. • The combination of forms of violence. Very often more than one form of violence presents itself; spe- cifically, the most common combination is physical and sexual violence (which does not occur in other forms of violent crimes). Psychological violence is omnipresent and appears together with the above. This combined manifestation may be lasting or not, according to how the couple’s relationship evolves. In intimate partner violence, the three forms of abuse – physical, sexual and psychological – generally occur together. IPV, in addition to the components and characteristics described above, and which are common to violence against women, on account of its specific make-up (surrounding the intimate relations between members of the couple), has a series of specific features that must be understood to prevent it. They are: 1. High recurrence. For two reasons, IPV has a high level of recurrence: a) it occurs between people that have (or have had) a close and ongoing interpersonal relationship over time, and b) due to the very fact that the relationship is close, it is more likely that conflicts and violent acts will arise. 2. There is marked inequality in the relationship between offender and victim; this inequality can be identified by comparing ages, resources, power relations, etc. 34 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 3. Generally, IPV occurs in very much frequented private surroundings (within the home, the family, etc.). Therefore, it can very easily go unnoticed, and can even never be discovered unless the victim reports it. 4. The forms of violence against women are often varied and can occur successively or simultaneously in tme. In fact, changing the use of violent tactics depending on the course of the chro- nicity of the abuse is very common. 5. IPV, especially that which occurs within the family, takes place in a context in which the separation and break-up of the offender and the victim are very difficult, sometimes impossible, which facilitates the recurrence of IPV. 6. The victim’s feelings towards the offender tend to be contradictory and much more complicated that those of a victim towards an unknown offender, therefore future incidents may suddenly occur as a result of changes in feelings and the relations of both members of the couple. 7. The pattern of violence inflicted by the partner on the woman, far from corresponding to an accidental incident, usually follows a pattern of ongoing abuse. Finally, in order to get a full picture of intimate partner violence, we will briefly outline the consequences of IPV actions. Like any violent act, the consequences have a certain nature depending on the form of violence and, naturally, will be the central aspect of prevention since it is these that determine the malaise and severity of the incidents. We will summarise the consequences of IPV based on a WHO report (2002). They can be grouped according to their severity, who they affect (besides women) and whether they affect physical and/or mental health. The main effects categorised as fatal include: immediate death (homicide or suicide) and death due to late or chronic consequences of IPV (injuries, suicide, HIV/AIDS, etc.). In physical health, it is common to find: various injuries (bruises, traumatisms, wounds, burns, etc., that can end up causing incapacity), functional im- pairment, non-specific physical symptoms (for instance, headaches), worse subjective health, obesity or extre- me thinness. In chronic health effects: chronic pain, irritable bowel syndrome, other gastrointestinal disorders, somatic complaints, cardiovascular disorders, metabolic or endocrine disorders, the failure to follow health treatment and disinterest in personal care. With regard to sexual and reproductive health: loss of sexual desire, menstrual disorders, sexually transmitted diseases including HIV/AIDS, vaginal bleeding and fibrosis, chronic pelvic pain, urinary infection, unwanted pregnancy, miscarriage, etc. In addition, and for violence during preg- nancy, the following may appear: vaginal bleeding, the threat of miscarriage, a risk pregnancy, premature birth, low birth weight of baby, etc. 35 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES In the area of mental health, the appearance of depression, anxiety and sleep disorders, post-traumatic stress disorder, eating disorders, suicide attempts and alcohol, drug and substance abuse are common. With regard to the relationship, common are social isolation, job loss and absenteeism at work. In this context lies the problem of “mental pain and suffering”, which is the most complex legal aspect of psychological violence to assess and identify. Emotional distress, the legal equivalent of so-called “psychological damage” is a real phenomenon but the difficulty of recording it objectively is huge, and therefore the motive behind major litigations. Like all other psychological phenomena, it cannot be observed directly. Its existence can be confirmed by way of analysing the presence of certain indicators in the field of motor behaviour, verbal behaviour, physiological responses to emotional changes associated with psychological events as well as their effects. The most widely accepted em- pirical reference of“ emotional damage” in the forensic context is a psychopathological condition that is called post-traumatic stress disorder (PTSD). PTSD corresponds to the “psychological imprint” caused by actions or events experienced by the victim at the hands of their offender. PTSD is considered to be similar to the physical injuries resulting from different acts grouped under the label of physical violence (punching, pushing, stabbing, etc.). For the same features of PTSD, this is a transient phenomenon and therefore the precise evaluation must be done in a limited time span (con- cerning the actions they caused). It is considered that once the PTSD has disappeared (at least the symptoms that characterise it), certain symptomatological disorders of an emotional nature such as depression, dysthymia and anxiety disorders can last. Following the analogy with physical violence, these secondary symptoms are considered chronic or permanent consequences of emotional damage (similar to the scars produced by healed wounds). Indirectly on the health of children, alterations in integral development, feelings of threat, learning and sociali- sation difficulties, adoption of violent conduct with classmates, and the increase in the frequency of psychoso- matic illnesses may appear. Violence may also affect other dependents of the woman and those living with her. 36 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 3 IPV prevention: violence risk assessment and management A conclusion mostly accepted by all IPV specialists is that the best way to FIGHT violence and its effects is pre- vention (WHO, 2006). Noteworthy among the strategies to diminish and eradicate violence against women are prevention campaigns, early detection, the enactment of laws against IPV, the allocation of resources to avoid revictimisation of abused women, etc. Violence risk assessment, for which the RVD-BCN was designed and built, is a cross-cutting element in many prevention strategies. The prediction of future violence is in itself one of the main IPV prevention measures since, with its use, new assaults, and even the woman’s death, can be avoided (Dutton and Kropp, 2000; Zoe Hilton and Harris, 2005). For prevention to be effective, there are many and varied strategies. All of them converge on the fact that to reduce IPV, the risk factors must be reduced and the victim and offender protection factors must be increased. The environmental conditions that hinder violent actions must also be improved. Thus, providing protection resources to known or potential victims (empowerment) and reducing the violent potential of offenders may be complementary objectives to the fight against IPV. The prevention of violence involves those sectors related to criminal justice, education, social welfare, security, transport, housing and the media, as well as associations that represent victims and other social groups. Prevention involves integrated community action aimed spe- cifically at achieving this objective. The RVD-BCN Guide is therefore another element in IPV prevention and is designed so that all professionals can use it on a regular basis. Primary prevention policies target the entire population, meaning that its effects are raising the awareness of possible future victims, reporting possible abusers, making the resources to FIGHT IPV known, etc. However, prevention policies stretch well beyond that, especially when we speak of secondary and tertiary prevention. Se- condary prevention is done with subjects that are members of populations or groups that are likely to be victims or offenders, thus it is much more selective and more effective at the same time. The third type of prevention is on cases that have already been identified as gender-based violence victims or offenders. Both in secondary and tertiary prevention, the involvement of professionals is direct and personalised. These professionals belong to groups such as social workers, security agents, justice officials and professionals, health workers, etc. All these professionals at some point in their task of assisting women may encounter cases of intimate partner violence and have to act to prevent its consequences. In any prevention programme of a secondary or tertiary nature, it is necessary to ascertain a priori the magnitude, type and risk of the act we wish to prevent, naturally to act both accordingly and proportionately to the risk detec- ted. This observation is fundamental because prevention requires efforts, which are very often costly and whose benefits are spaced out over time, to be modulated according to the real or estimated risks. For example, let us consider for a minute, the launch of emergency services when a major storm is approaching or when an infection is detected that poses the threat of an epidemic; the actions that are decided (even including evacuation of the po- pulation at risk or detention in cases of quarantine) depend on the assessment that has been made of the meteo- rological or medical risk. This assessment is still a prognosis of what could happen and the first step in prevention. 37 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Secondly, this anticipation or prediction must be carried out efficiently, objectively and guaranteeing, in the case of wishing to predict future violence, the rights of the people we are assessing. The prediction of violence is es- sential in the prevention of its effects. Let us consider the situation of a female victim of gender-based violence when her partner is released from prison with the explicit threat of making his threats of revenge a reality. The precise assessment of the risk run by this woman is the key to preventing future assaults and safeguarding her safety and well-being. This risk assessment – anticipating what can happen with a certain realistic estimation of the risk – with this offender in the future may explicitly fall on a prison officer, a social worker, a policeman/ woman or a psychologist, or any other professional that deals with the woman at risk. Seen in an orderly fashion over time, prevention first calls for explicit and careful risk assessments. The most routine interventions do not always act in this manner, but we believe that this task should be generalised to other situations in which risk assessment may seem unnecessary, since the reason behind the intervention of a practitioner with a woman does not aim to directly protect her as a known victim of IPV. However, when the- se professionals intervene, they make a prediction. Generally speaking, it is an informal or implicit prediction but which bears an influence on the intervention. When a woman seeks advice to find gainful employment in order to begin the separation process in the future, she may be in a situation at risk of violence that must be explored. Other times, a consultation with a physician for chronic and extensive health conditions may be the result of an active IPV process. In many situations of consultation and requests for assistance made by women, implicit estimates of the risk of future violence are required of the professional. In these situations in which the prediction is implicit, the use of appropriate procedures is not usual and the prognosis is a “by-product” of the general treatment of the case managed by the practitioners themselves. In other cases, the making of predic- tions is much more formal and explicit, for instance when a forensic psychologist submits a report to the judge for penitentiary issues on the risk of recidivism of an inmate convicted of IPV for whom a leave permit is being considered. The prevention of violence against women involves the four following phases: a) Identification of the problem: defining, assessing and monitoring it. We have already seen that violence against women has many facets and this distinction is paramount in this first phase of preventive action. It answers the question: what do we wish to prevent? b) Identification of the risk and protection factors for violence against women that are characteristic of different forms of violence. It is in this context of prevention that violence risk assessment takes on a very special and important significance. It answers the question: which risk factors may be present in the im- mediate future that facilitate the emergence of violence? c) The development of intervention strategies to act in risk situations. It answers the question: how can we reduce or eliminate risk factors or render the consequences of violence less serious? 38 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES d) Broad implementation of intervention and action processes in situations of violence. It answers the question: what resources does the victim have to stop being one? Violence risk assessment specifically deals with the first two phases and contributes towards developing the following stages. The RVD-BCN Guide is designed to meet the needs of experts in the initial two phases and to orientate the third. Naturally, the fourth phase will depend on the professional context in which the victim is found. The risk of violence and its prediction Violence is a complex issue and, fortunately, severe violence is less frequent; this, in itself, hinders the precision of prognoses. We must consider that the frequency (in epidemiological terms) of very severe abuse stands at a very small proportion per hundred thousand (the risk of murder is a very small number per million), which make them events that are extremely difficult to predict. One of the reasons for the complexity of violence, and their prediction, is their multi-causal nature. A single and exclusive reason that fully justifies (which explains in a cau- sal sense) violent acts or behaviour does not exist. But the significance of violent conduct and its consequences is the main reason behind the need to predict it. Violence and violent acts are somewhat uncertain events akin to other phenomena such as atmospheric, technological or economic changes. Knowing which are the so-called risk factors and the triggers of violent conduct constitutes the conceptual core of prevention and the prediction of violent conduct. Predicting or anticipating a violent incident requires, above all, knowledge of the past history of this pheno- menon. Therein lies the foundation, the basis, for predicting future violence. In addition, very often precise knowledge of the causes and processes that produce a phenomenon does not guarantee the capacity to effecti- vely predict it. For instance, we know why an earthquake occurs, but this knowledge is not sufficient to predict it effectively. The reason is simple. An individual event, such as an earthquake or a sharp drop in the stock market, as well as an assault, are determined by many different elements which, regardless of their origin, act together and therefore there is no way to ascertain the isolated effects of each one of them and less so to determine the result of their multiple interactions at a given time. Therefore, a sporadic act, a violent action, cannot be predic- ted. We can only estimate the probability of that act occurring. This prediction is always a relative estimation of the risk of that specific act occurring. This subtle change is essential for understanding the limits and strength of the predictive power, and obviously affects the RVD-BCN, which is not like a crystal ball or a deck of cards to predict the future. The RVD-BCN is a guide to estimate the risk of severe intimate partner violence. Researchers and professionals working in the domain of violence have worked intensely to ascertain the risk factors (and, to a lesser extent, protection factors) which, whether they are sociodemographic, family, individual or community factors, or factors of another nature are associated probabilistically with violent acts. Risk factors 39 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES are characteristics associated with a greater likelihood of a given act occurring, in this case an act of physical or sexual violence or of another nature. They constitute the core element of risk assessments and, though the presence of one or more risk factors does not necessarily indicate that a certain causal relationship exists, the probability of an event associated with risk factors does increase. Much of the research motivated by the desire to prevent and assess the risk that may occur – in this case, violence against women – is the result of epidemio- logical and criminological research. A common result is to find a store of these factors that sometimes must be treated according to the weight or quantitative influence they exert on the probability of a violent act emerging (Andrews and Bonta, 2003). Risk factors do not always have to be related with the causality of IPV because they can be “triggers” or “modulators”, at other times they are combinations of causes, etc. It is important, in the task of risk assessment, to constantly avoid analysing the effect of each risk factor as if it were the “sole cause”. What does the risk assessment of future violent conduct entail? Estimating the risk of violence is nothing more than a procedure, a technique, to identify the probability of violent conduct appearing in a limited time period and under conditions that are more or less defined (we call them future “scenarios”). Therefore, and from this point forth, we must bear in mind: a) that the risk of IPV is changeable and the changes are associated with the variable dynamic of the couple’s circumstances (co-habitation, separation, custody, etc.) and b) that this variability makes it necessary to make repeated risk estimations, since the estimation made at a given time (for instance, in the formal separation of a couple) may be different to the estimation made a few months later (for instance, when the woman enters a new intimate relationship). It is possible to predict the risk of violent conduct more accurately than by simple chance (Webster, 1997; Hart, 2001; Andrés-Pueyo, 2009). In addition, risk estimation that follow guidelines and guides heightens the preci- sion of predictions over intuition or clinical assessment (Hart, 2001). Moreover, the use of a guide, such as the RVD-BCN, facilitates the transparency of reviews of any decision regarding the prognosis and its fulfilment or not. The risk of any choice or decision can be predicted if we know the determining factors of this choice and we have information on similar choices made previously and whose results are known. For this reason, we must analyse the past carefully and rigorously. The problem of IPV can be analysed with these coordinates. Violent conduct, in whichever form, such as that of an abuser on their partner, is intentional behaviour; it is true that this behaviour is highly “automated or uncontrolled” but, in the end, it is always an act led by the man to control and harm their partner. Violent conduct is always the result of the decision to act in this manner. Violence risk assessment is essentially a procedure that enables us to understand the danger of violent conduct appearing in the future with the objective of eliminating this appearance or limiting its potential adverse effects. To undertake this process successfully, we are interested in recording information on different aspects of the 40 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES violence: what type of violent conduct may occur?, how often?, under what conditions or in what scenarios?, what conditions are present?, has it happened before?, what intervention was undertaken?, what happened next?, etc. In recent years, new methods of violence prediction have been developed based on the assessment of risk fac- tors that anticipate violent conduct (Hart, 2001). These methods have specialised in assessing specific forms of violence and have emerged from the collaboration between researchers and professionals to resolve, first of all, the practical requirement of making effective predictions (Andrés-Pueyo, 2009, Andrés-Pueyo and Echeburúa, 2010). We will briefly define what we understand by risk, we will describe the main IPV risk factors, and how a risk assessment procedure is prepared (concisely) on the basis of the structured professional judgement approach (Arbach and Andrés-Pueyo, 2007). Risk is a “danger that may occur with certain probability in the future and of which we do not fully understand the causes, or they cannot be fully controlled” (Hart, 2001). This concept is applicable to violence like other natural or human phenomena that have harmful consequences. We speak of the “risk of violence” as an alter- native concept to “dangerousness” (Andrés-Pueyo and Redondo, 2007). Predicting, anticipating the risk of violence against women in the context of intimate relationships is different to explaining and understanding the causes of this form of violence. Risk is our goal, not violence, nor the state of the offender or victim; in risk assessment this element becomes our concern. Violence risk assessment and management is capable of reducing the risk without having a full and valid theory on why it happens, what are its causes and what are its triggering mechanisms. Predicting and explaining are not always the same. The search for risk factors in the context of intimate partner violence, needless to say, has been the object of many systematic studies for the last twenty years. Most of them have focussed on identifying the factors that make the man a possible offender and, in particular, taking into account physical or sexual violence as a criterion to be predicted. Less studies have dealt with analysing the variables associated with victimisation and psycholo- gical violence. In addition, in the case of psychological violence, studies have not proven successful. It is very important to remember there are risk factors for specific forms of violence. Among the risk factors for violence against women, three groups of criteria must be set apart: those characteristic of violence risk pre- diction in general (history of violence, age, poor social adaptation, irritability and impulsivity, drug and alcohol use), those specific to violence against woman (traditional gender role attitudes, beliefs of male superiority and tolerance of violence, partner dissatisfaction, etc.), and those that are more typical of intimate partner violence, such as rape or a previous history of sexual abuse, breaches of the woman’s security measures, uxoricide or ho- micidal and/or suicide ideation (Campbell, 1995; Dutton, 1995). There is major consensus on the lists of these factors and most prediction instruments take them into consideration when assessing the risk of future violence. 41 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Among the risk factors, three types can be distinguished. The first are so-called “historic or static” risk factors (for instance: youth violence in adults, history of childhood abuse, etc.), since the time in which they occurred (the past) cannot be modified and they play an important role as long-term predictors. There are also “dynamic” factors which, unlike historic factors, are changeable and have a relatively short duration. These dynamic fac- tors, in turn, are divided into “acute”, which have a very short duration and are changeable (for instance, alcohol poisoning or separation or divorce of the couple), and “chronic” factors, which are changeable but occur repea- tedly and permanently (alcoholism or a chronic severe mental disorder, etc.). Of course, the effects of dynamic factors, which may be very significant, are limited in time and are manageable (to a certain degree), unlike static factors. This characteristic, which is based on estimating the probability of risk factors present in the past (in a specific situational context) continuing or emerging in the future, renders risk assessment variable over time. Among the experts in future violence risk assessment, there is agreement that a relatively small number of pre- dictors, combined in a certain manner, are sufficient to make good group predictions, and if applied rigorously are also effective at an individual level. The most important risk factors in IPV risk assessment are those which reveal in which direction and what we have to look for in the previous history of the couple, the respective mem- bers thereof and previous violence. Risk factors point us towards choosing the “minimum” relevant information. The remaining steps in risk assessment are easier and more flexible to perform. RVD-BCN risk factors are descri- bed later on, as is their means of assessment. General IPV risk assessment procedure Risk assessment is not a technique exclusive to psychology or the health sciences. It possesses sophisticated techniques of great applicability in many areas of science and technology: meteorology, seismology, medici- ne, economics and system engineering, among others. In the area of penitentiary techniques, violence risk assessment has arisen from the problem of violence recidivism, particularly in cases such as sexual assaults and intimate partner violence (Andrés-Pueyo, 2008). The case of IPV assessment is a further application of this technology. The effective assessment of the risk of violence is, in essential terms, a logical process that aims to estimate the probability of certain events (naturally, violent conduct) occurring in the future. Risk assessment does not seek to predict the future, describing in advance what, how and when a future crime will occur (as in the fantasy described in the film Minority Report). Violence risk assessment can be defined operationally as “the formulation of a risk level in a range from low to high and which is imminent, the result of the weighted combination of certain risk and protection factors as- sociated empirically with the risk of a specific form of (violent) behaviour occurring in a given future scenario”. These parameters are easily applied to the problem of IPV assessment and prevention. 42 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES To make successful predictions, data and information on various biographical aspects of the alleged offender, the relationship of the couple, etc., are required. We should mention that the effectiveness of risk assessment directly depends on the quality of the information on the risk factors we are able to collect and analyse. The general violence risk assessment procedure, especially when using guides such as the RVD-BCN, comprises four phases that, in sequential order, are as follows: Fist phase: the collection of information. This is an initial phase that depends most particularly on the availability of detailed biographical information on the persons involved in violent acts, especially the offender. This information generally comes from interviews, questionnaires and information compiled in files (from the social services, prisons, doctors, etc.), which are increasingly common in our professional environment. In addition, much of this information can be compared, reviewed, denied or incorpora- ted from collateral information (work files, verbal reports from family members, neighbours, etc.). As- sessments hinge on the “faithfulness” of this data, and it is for this reason that rigour and caution must be exercised when compiling all this information. It must be pointed out that in most cases, especially in the context of violence against women – mostly occurring in couples or within the family – there is a great deal of information on some cases that is distributed by social, judicial, police and medical servi- ces, which must be combined; professionals must not be required to repeat interviews, examinations or studies that have already been undertaken on the same persons, because eventually this ends up being prejudicial. It is true that in this regard, good coordination is required between the services inter- vening in violence against women, but it is necessary to achieve that and it is no easy feat. It is strongly recommended that two tasks of enormous benefit be coordinated: a) preparing written information to be stored, and b) facilitating the access of all professionals to said information (for instance, by means of a shared database system). The second phase of the assessment process is deciding on the presence or absence of risk factors in each form of violence, and their importance. The RVD-BCN Guide contains a limited serious of items that correspond to risk factors of a different nature, which predict future violent conduct. It must be decided, following the analysis of the available information, if these factors are present or absent in each case and point in time of the assessment. Generally speaking –for this reason that these items have been included in guides and assessment guides – it is not difficult to decide on the presence or not of factors in the history or current situation of the assessed subject, but this decision sometimes presents queries. Hence, these guides outline precise and replicable criteria in writing that contribute to the reliability of decisions. In addition to the presence or absence of the assessed item, in the third phase of violence risk as- sessment, the assessor, not the victim, is asked to assess the importance of certain items in the case under evaluation, which may prove critical to deciding on, above all, the imminence of the risk of vio- 43 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES lence or the unexpected appearance of an element (which may be a threat to children in the context of assaults on partners). Once this detailed decision-making stage has been undertaken around each risk factor, the final assessment must be made, which entails deciding on the overall risk of violence of the case. Two previous elements that frame the outcome of the assessment must be pointed out. The first is that every assessment is limited to a time period specific to the case (sometimes, six months or a year, but in the case of RVD-BCN between three to six months). It should be noted that decisions of indefinite validity in time are not taken. The second is that every assessment concerns a given form of violence and cannot be generalised indiscriminately. In the case of RVD-BCN, it is severe physical and sexual violence. That is to say, if we assess the risk of sexual violence, we cannot assume that it can be used to predict the psycho- logical abuse of partners or the risk of suicide. The task of summarising violence risk assessment by way of “guides” is not done in a regulated manner in some of them (Spousal Assault Risk Assessment, SARA), with the exception of some guides such as the RVD-BCN, which have some quantitative guidelines and ranges of response to contribute to the final definition of the level of risk declared. The final assessment is divided into three (or four) levels, which are: low, moderate or high (and sometimes imminent), very common in all risk assessment procedures as diverse as meteorology, economics or natural hazards. The final phase in the risk assessment process corresponds to the communication and dissemination of the assessment findings. It should be noted that generally violence risk assessments are part of different processes, such as: decisions on court or prison proceedings, forensic reviews of personal situations of offenders and/or victims, victim care services, etc. Therefore, they are assessments sometimes geared towards responding to questions posed by external agents to those in charge of the assessments. It is worth noting that the information afforded by these assessments, on account of their very nature, depends on the duration of the prognosis and their probabilistic relativity, since, as we have insisted, the prediction of the risk of violence does not determine the guarantee of occurrence of a specific act. At other times, these assessments are to commence a programme of specific protection measures and interventions or also to make the victim aware of the situation that may arise in their reality of violence in the future. This entire process allows the assessor to deduce elements to manage the risk of future violent conduct. The ex- haustive analysis of the victim’s resources, the history of the offender and the victim, the in-depth examination of the clinical status at the time of the assessment, and speculation on the future of the subject under different conditions and/or scenarios, enables highly individual proposals to be put forward, and therefore practices for all those responsible for violence against women. 44 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES The target audience of the RVD-BCN are all those professionals from the public and private sector, involved in violence risk management: prison officers, care services for women, agents involved in juvenile justice, foren- sics, etc. More specifically, with the proper preparation and training, it may be used by forensic psychologists and criminologists, psychologists that work with prison officers, clinical psychologists, clinical and forensic psy- chiatrists, forensic pathologists, criminologists, police and State security bodies, other justice staff, social wor- kers, primary care physicians, nursing staff, etc. These guides can be employed in various population sectors, noteworthy among which are people with criminal records, psychiatric history, as well as individuals accused or implicated in acts of domestic violence whether or not they have criminal or psychiatric records. They are desig- ned to be applied to those over 18 years of age (minimum age varies according to the partner role considered in the analysed case) and therefore, with the exception of some very rare situations, are not suitable for predicting the risk of violence in adolescents. Meaning of the Risk Levels Low Communicate the findings of the assessment to the woman without exaggerating the situation. Information regarding the resources available to her must be provided and the autonomy and safety of the woman must be promoted. Moderate Warn of the risk of future violence, and engage in follow-up combining the resources at hand with the autonomy measures available to the woman. High Alert the victim to the high risk to which she is exposed and work closely with the woman to ensure her safety. The first consequence of risk assessment – therefore, the first management action – is related to risk measu- rement. As we mentioned previously, three or four levels of risk are generally used: imminent, high, moderate or medium, and low. In fact, imminent and high constitute the same level of risk, only the first indicates the immediacy of the risk emerging and calls for an extreme emergency intervention. In the case of low risk, the risk management action is characterised by not having to take specific measures or follow-up measures in the case. If the risk level is rated as moderate or medium, action at two levels is recommended: monitor and follow up on the case, intervene if necessary, and reassess, within a relatively short period, the risk to ascertain whether it increases or decreases over the course of weeks or months following the previous assessment. If the outcome of the assessment revealed a high risk, then the situation should be treated as an emergency: intense and priority action. This priority becomes very urgent when the level of risk is imminent.3 3. Imminent: It signifies a high and valid risk within 24 to 48 hours. 45 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Needless to say, there are many measures to be taken to intervene in cases of violence against women, which affect many agents within whose (sometimes exclusive) remit of powers they fall, as set forth by law. All these agents may play a role in violence risk management. By and large, management strategies are associated with reducing and/or eliminating risk factors (alcohol consumption) or increasing those protection factors that may mitigate the effects of unmodifiable risk factors. Risk management is a balance between what we can do and what we cannot change, which affects the likelihood of violence occurring. On account of the very repetitive nature of IPV, risk management, which is derived from assessment, is very dynamic and changeable, as much as the risk of violence itself. We must insist that the efficiency of violence risk prediction processes does not so much depend on the as- sessment procedure as on the quality of the information at hand to make the appropriate decisions, and which must be obtained – as much as possible – from different information sources and by way of different instru- ments. Moreover, the accuracy of violence risk predictions depends, in part, on the prevalence of what we wish to predict, since this prevalence determines the likelihood of making true or false positives. Therefore, intimate partner violence, as we have seen, has a prevalence of around 2% to 5% in cases of severe violence, 10% to 20% in cases of not so severe violence, and this facilitates the making of accurate and reasonable predictions. We will not make predictions of future behaviour but estimations of the probability of certain violent incidents with a time validity that is also limited. In the case of physical intimate partner violence, having information regarding the personality, biography and recent state of the offender that may be provided by the victim, is of great help for the professional to predict the likelihood of the offender continuing the progression of assaults or not. 46 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 4 The RVD-BCN: guide and risk factors The RVD-BCN is a tool for assessing the risk of severe violence against women perpetrated by her current or former partner that may occur in the near future (between three and six months). It is based on the structured professional judgement approach (Hart, 2001; Arbach and Andrés-Pueyo, 2007), whereby the final decision on risk arising from the assessment made using this guide falls on the practitioner using it. The guide provides information and resources to facilitate this decision. The RVD-BCN does not automatically detect the level of risk, detection is transferred to the practitioner as a fixed and invariable value. Responsibility for the decision to predict the probability of new incidents of intimate partner violence falls on the practitioner, and the use of the RVD-BCN heightens the accuracy and transparency of this decision and acts as the first step in the future intervention with the woman being assessed. The RVD-BCN contains a set of elements which require the assessor to undertake a series of steps to turn the information referring to a specific intimate partner violence case into a prognosis. It also outlines guidelines for action to prevent the recurrence of severe violence against women perpetrated by the current alleged offender in the future. The guide collects initial information on the case, which is to be assessed as well as the date of assessment and details concerning the person conducting the assessment. Secondly, it includes a list of risk factors, which amount to sixteen, which are of mandatory assessment for all IPV cases. The guide is presented in a table with the name of the factor and the response options, of which there are three: presence, absence of the risk factor, or “information not available to grade the risk factor”. Following the list of risk factors, the professional acquires a first risk estimation by way of counting the risk factors that are “present” in the case assessment. This initial assessment must be graded according to the RVD- BCN’s three levels of risk: low, moderate and high. To perform said codification, cut-off points are outlined in the guide. The following element in the RVD-BCN Guide is a control to ascertain how many risk factors in the previous list could not be assessed due to a lack of credible information. This control allows us to validate the aforemen- tioned codification, since it indicates the number of risk factors upon which the initial assessment is based. A minimum of seven factors must be assessed in the RVD-BCN. Before concluding the risk assessment process, the RVD-BCN provides for the possibility of including other fac- tors the professional considers key at that time and in that particular case to assess the risk experienced by the woman and which are missing in the previous list. It concerns the introduction in the case assessment of any element as a risk factor which is present and which could not be assessed in the initial list of the previous sixteen factors. On the basis of this new exploration of the woman’s risk situation and previous provisional and quanti- tative assessment, the assessor must indicate, in the following section, the final assessment of the prognosis of 47 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES future severe violence towards the woman in the boxes designated for this purpose. There are three categories: low, moderate and high risk as the final outcome. Two further elements constitute the full RVD-BCN Guide: the description of the action undertaken or to be un- dertaken according to the current existing risk, and the indication, where necessary, of foreseeable circumstan- ces that may heighten the risk of violence in the future and which would act as a warning to act if that situation were to actually occur. The Guide RVD-BCN form is found in the second part of this document. We will now describe in detail the nature, definition and other details concerning the sixteen risk factors of the RVD-BCN. Risk factors of the RVD-BCN Los dieciséis factores de riesgo(FR) del RVD-BCN se agrupan en cinco categorías: 1. History of violent conduct towards the partner or former partner These five risk factors concern the history of violence of the alleged offender (partner or former partner) of the victim. They must be interpreted, in this regard, as something that happened in the past even though it could be very recent (even the very day the assessment is made). It is interesting to consider the nature, variety, inten- sity and frequency of the violent acts but also the intentions to act violently (which for various reasons has not come to pass), as well as in what situations it has occurred and the time dynamic (increase or decrease) of this type of behaviour. They constitute the most powerful group of risk factors for making a prediction of violence and for the duration of the validity of the prognosis, because they reflect the history of partner violence (with the current victim or other potentially earlier victims) and anticipate what might happen. They are all static risk factors and reflect the potential lasting effect of future and long-term violence. RF1: ASSAULTS OR PHYSICAL AND/OR SEXUAL VIOLENCE TOWARDS THE WOMAN OR FORMER PARTNERS IN THE LAST EIGHTEEN MONTHS Men who have committed physical assaults on their partner (or on other partners) in the past have a greater risk of repeating this form of violence in the near future (Kropp, 2008). The rate of recidivism is estimated to range between 30% and 70% in a two-year period (Dutton, 1995). 48 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES It is important to bear in mind that we are making a retrospective estimation over eighteen months, which are approximate and flexible. It is important to consider here the history of severe physical and sexual violence (which had consequen- ces that require medical attention or care for injuries sustained from the assaults). Brief definition: Non-accidental action that during the last eighteen months has led to physical and/or sexual harm. Physical violence is understood to be any assault (punching, slapping, scratching, pushing, kicking, etc.) with the result or risk of causing physical injury or harm. Sexual violence covers any act of a sexual nature without the consent of the woman, including exhibition, observation and the imposition of sexual relations by means of violence, intimidation, abuse of the offender’s position or manipulation. Example: “The offender gets angry with his partner because she does not know how to make a potato omelette. He grabs her by the hair, shakes her and closes the door of the fridge tying her hands together.” RF2: ASSAULTS OR VIOLENCE AGAINST THIRD PARTIES WHETHER OR NOT FAMILY MEMBERS The habitual use of violent conduct (generally physical) not restricted to the partner or former partner as a victim is relevant because it indicates the alleged offender’s repertoire of conflict resolution strategies. One of the most common research findings is that offenders with a previous history of violence against people are more likely to become violent again in the future, and this probability is higher that in those that do not have a history of violence in their personal life. If this risk factor is present and is relevant (in terms of severity and frequency), the probability of future assaults continuing or which may affect the victim’s family members (parents, children, etc.) should be strongly considered in the risk assessment outcome. It is a factor concerning the generalisation of the risk of violence towards people surrounding the potential victim (particularly her family and domestic environment). Brief definition: The alleged offender committed physical/sexual assault or attempted assault on family members (whether blood or political ties), and/or acquaintances (male or female friends, work collea- gues or neighbours) and/or strangers. Example: “The couple’s youngest child claims that his father always slaps his bottom with the cable of the iron when he does something he does not like.” 49 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES RF3: ASSAULTS ON THE WOMAN DURING PREGNANCY The most common reason given for the increased risk of violence against women during pregnancy is that the father/male partner feels a greater sense of stress over the impending birth. The stress mani- fests itself as frustration, which is directed at the mother and their future child. Women are four times more likely to suffer abuse as a result of an unintended or unwanted pregnancy (Heise, 1993). It also indicates the alleged offender’s lack of capacity to inhibit violent conduct in light of the weakness and vulnerability of their partner and the foetus. It is not a common factor but has much prospective inferen- ce. Here we will not limit ourselves to assessing physical and/or sexual assaults and we will considerable plausible threats, extreme control of the victim, pressure to abort, etc. Brief definition: Physical and/or sexual assault during the woman’s pregnancy by her partner or former partner. It also includes severe and persistent psychological abuse. Example: “When the woman is in the first month of pregnancy, the offender grabs her by the neck, throws her to the ground and insults her while he kicks her in the back. She sustains severe injuries: collarbone strain and multiple bruises.” “Following a physical assault, the woman suffers placenta abruption and as she was seven and a half months pregnant, her daughter was born premature.” RF4: POLICE/CRIMINAL RECORD OF VIOLENCE AGAINST THE PARTNER OR FORMER PARTNER (CURRENT PARTNER OR OTHER PARTNERS IN PREVIOUS INCIDENTS) This item refers to the existence, in the previous biography of the alleged perpetrator of the violence, of sentences, arrests, accusations, applied security measures, etc., and related to acts of violence against the current partner (or previous partners). The forms of violence that gave rise to these measures inclu- ded psychological or sexual abuse, threats, bodily harm, etc. The previous violence, as we have said, is always the most important reference point when it comes to the possibility of future violence. If, in addi- tion, this violence has been inflicted on a woman that has been a partner of the offender, this indicates a consistent pattern of relationship behaviour that presents a poor prognosis. If the alleged offender has had previous related convictions or sanctions, it indicates the resistance of his conduct to court procee- dings, the low dissuasive capacity of official and punitive interventions, as well as the seriousness of the previous incidents, which are probably not exaggerations by the current partner. When this refers to violence inflicted upon former partners, it is very likely that the current partner does not know anything about it. 50 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Brief definition: The criminal and/or police record of the alleged offender shows a history of physical, sexual or emotional abuse of a current or former intimate partner. Example: “The offender has two convictions because he was involved in a big fight with the use of fi- rearms and because he sexually assaulted his former partner. He therefore served a five-year sentence in prison.” RF5: THE OFFENDER HAS BREACHED THE WOMAN’S COURT PROTECTION MEASURES The habit and frequency of court order violations is worrying in IPV, because it can soon be “imposed” by the alleged offender as a “facility” of the victim, and these situations should be distinguished not so much for legal reasons (in risk assessment) but because, for whatever reason, they significantly increase the risk of violence. Severe assaults and abuse sometimes occur in these situations even if the alleged offender is subject to a court restraining order or similar order. The importance of this risk factor must be noted as it informs us of the low dissuasive capacity of legal and coercive measures previously applied to the abuser. In these cases, it is worth accessing official reports in which this type of information is stated (if possible). Consideration must be given to the greater likelihood of severe or very severe violence occurring when the offender violates restraining orders (Campbell, 1995). Brief definition: The offender has breached a protection measure. He has been subject to convictions or police investigations as a result of having breached a protection measure for his partner or former partner. Example: “Despite the existence of a restraining order, the woman sees the offender approaching her on the street, shouting; she locks herself in an ATM booth and calls emergency services. When a patrol of the Mossos d’Esquadra [Catalan Police Force] gets there; they arrest the offender and find a hammer, large knife and axe in his backpack.” 51 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 2. Threats and/or severe abuse of the woman These two risk factors refer to the reality at the time the assessment is conducted; they are also considered a few months earlier to assess, simultaneously, one of the main IPV risk factors: verbal and emotional abuse (Stith et al., 2004). This section includes threats as a noteworthy factor that reveals the future intentions of the offen- der and therefore must be very much borne in mind as dynamic risk factors, valid for short-term prognoses. Both are useful factors for considering the evolution of the risk of violence. Another important criterion is assessing the severity, frequency and also dynamic changes of the factors included here. RF6: THE WOMAN HAS RECEIVED SERIOUS AND CREDIBLE THREATS TO HER PHYSICAL INTEGRITY, AND/OR WITH THE USE OF WEAPONS Threats always, or almost always, appear in cases of violence against women. They constitute a relevant form of violence in IPV and accompany other assaults (usually physical or sexual). They are sometimes very generic (they seem more like an insult) and others are very specific and repeated, with detailed accounts of what the man wishes to do to his partner (“if you file for separation, I will kill you and I will kill myself”). They may arise in highly conflictive situations and in the midst of a fight, even in the pre- sence of family members, neighbours or the police, or just as secret and persistent telephone messages. Three reference points allow us to assess the importance of these threats: the fear they generate in the victim, the use of a weapon (a hammer, a rope or any bladed weapon or firearm) and the likelihood of the threats (especially when they include assault and the ensuing consequences). Conducting this as- sessment is always complex but it is very important to confirm or rule out its presence. The future development of violent conduct is very much associated with earlier threats, whether verbal, death or suicide-related. The risk is accentuated when these threats are made in the presence of other people, when the threats are made with dangerous objects or different types of weapons and when conduct corresponds to a clear intention to cause harm (Corral, Echeburúa i Fernández-Montalvo, 2009). Brief definition: The alleged offender threatens, intimidates or coerces the woman with or without the use of weapons. Example: “I want you to die, and you do not have much time left because you are soon going to die in a car accident.” “Tell your mother that from now on she will have one child less because I am going to come and kill you.” “I will burn down the flat with you in it, nobody will come in here because I am going to kill you.” 52 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES RF7: THE WOMAN HAS SUFFERED SEVERE EMOTIONAL AND VERBAL ABUSE IN THE LAST SIX MONTHS A recent history of severe emotional and verbal abuse is an indicator of how the abuse has occurred in recent months, including at the time of assessment. Roughly six months is the approximate reference point and an examination is required of the reality of said abuse with regard to its frequency (every week, every day, etc.), as well as its increase (the woman may see the doctor when it has subsided and she feels more confident, etc.), since these two elements are predictors of the immediate future of IPV. In Stith’s meta-analysis (2004), it appeared as one of the factors most closely associated with IPV than any other kind. Brief definition: Non-accidental action that has caused emotional harm in the last six months. Severe emotional and verbal abuse covers degrading, humiliating, insulting, threatening, criticising, belittling, ridiculing, vexing, demanding obedience and submission, verbal coercion or any other limitation of their freedom. Example: “The offender constantly reproaches her for not knowing how to do the housework and for her physical appearance, saying she is repulsive and is not at all physically attractive. He forces her to wear shoes two sizes too small.” 3. Aggravating circumstances These risk factors, which amount to seven, cover a wide range of risk factors that increase the probability of future intimate partner violence. Most of them are dynamic, both chronic and acute, and the presence of many of these risk factors suggests a possible variability of future IPV. They do not exclusively refer to the alleged offender (he has access to weapons, has suffered from a mental disorder, etc.) but includes others associated with the couple such as “de facto separation or the intention to separate” or “the increase in the frequency or severity of IPV”. The time reference is mainly from the time of assessment, though some factors, such as “suici- dality” (RF13) or the history of mental disorder (RF11) have a much longer time reference (even at any stage of the life of the alleged offender). RF8: THE WOMAN INFORMS THE ALLEGED OFFENDER OF HER WISH TO SEPARATE OR THE SEPARATION OCCURRED LESS THAN SIX MONTHS AGO Several studies have shown that separation and estrangement exert a very important influence, in a pro- portion that varies between 25% and 52% in the homicide of partners (Kroop, 2008). Incidents of intima- 53 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES te partner violence are more serious after the respective members have separated or around the time of separation (Belfrage et al., 2011). Thus, the most serious violence occurs when the couple separates or are in the process of separating, especially if this occurs at the initiative of the victim and is accompanied by harassment, because in these cases the final separation of the couple is more complicated (Amor, Echeburúa, Corral, Zubizarreta and Sarasua, 2002). Brief definition: At the time of assessment and/or the incident, the woman informed the alleged offen- der of her intention to end the intimate relationship or they have been separated for less than six mon- ths. Example: “The offender finds out from a friend that his partner has met with a solicitor to initiate the separation process.” RF9: INCREASE IN THE FREQUENCY OR SEVERITY OF THE VIOLENT INCIDENTS IN THE LAST SIX MONTHS In general, the repetitive nature of IPV allows us to observe the frequency and/or severity of violent incidents in the past. Here it is especially interesting to recognise the reality of the last six months with regard to whether the violence has reduced, remained the same or increased in terms of the number of incidents as well as the seriousness and triggers present. It may be considered that a recent rise or esca- lation foresees a high risk of recurrence, especially in the short term and in relation to possible conflicts (trial for divorce, separation, police report, etc.). Analysing this risk factor may also serve to identify a possible “cycle of violence” and to determine its state with regard to the couple’s relationship. Episodes of violence may be varied (physical, sexual, etc.), what is of interest here is whether there are changes in the pattern and history of abuse. Brief definition: increase in the frequency and/or severity of the aggressive conduct towards the partner or former partner over the last six months. Example: “The woman claims that the assaults endured previously were more sporadic but this week she has borne the brunt of physical assaults virtually every day; finally, she cannot tolerate the situation to which she is subject any longer. In one such assault, her eardrum burst and, as a result of the serious injury sustained, lost her auditory capacity.” 54 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES RF10: THE ALLEGED OFFENDER ABUSES DRUGS AND/OR ALCOHOL The discussion on the effect of alcohol and drugs in IPV, like in most forms of violence, is indisputable (Felson, 2007), although many publications and public discourses insist on considering that IPV is not caused by the use of said substances. A discussion on the “causality” or “responsibility” of drug use/ abuse in IPV is not of interest here. For the prediction, this discussion is irrelevant. Without a doubt, the association between violence and drug use/abuse is very evident and should always be considered in the sense of analysis of past contingencies in which violent events occurred (with or without drug or alcohol abuse) in order to reasonably speculate about the future. It has been demonstrated that alcohol use and abuse, and to a lesser extent, use of other drugs is associated with all forms of assault against partners (Felson, 2007). International research constantly indicates that alcohol consumption is a risk factor for intimate partner violence. Among the drugs most particularly associated with serious intimate partner violence are alcohol and cocaine. The three clinical categories related to drugs must also be differentiated: occasional or social consump- tion, abuse and dependence, which show us the chronic nature of probable consumption in the future since they have direct but different relations with IPV. However, this is the major excuse of many victims to reject the aggressive nature of their partner or former partner: they usually associate violence with drug use and they trust that, when this problem disappears, IPV will also vanish from the reality of the mistreated woman. Although drugs have different effects on users, as a risk factor they are equivalent. The most relevant as regards the association with IPV are alcohol and cocaine. Here the risk factor concerns drug use by the offender, but, if it were necessary to assess the case, the victim’s problems with substance use should be included in the additional factors of the RVD-BCN and considered in the final RVD-BCN assessment. Brief definition: The alleged offender is an addict or abuser of legal and/or illegal substances (cocaine, alcohol, hashish, heroin, psychotropic drugs, etc.). Example: “The offender consumes large amounts of alcohol at the weekend and the woman hides in a wardrobe in the home until he leaves again, for fear of assault when he is under the effects of alcohol.” 55 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES RF11: DIAGNOSIS OR HISTORY OF SEVERE MENTAL DISORDER IN THE ALLEGED Offender There is much debate on the role played by mental disorders in violent action and, especially, in the field of IPV (Kroop, 2008). In the event of severe violence, we often consider the plausible hypothesis of the presence of a mental or personality disorder in the offender; however, many legal and sociological studies, or even health studies, emphasise the idea that this is not the case, that attitude and social role underlie IPV, and that association with the mental illness of the offender is a myth. Empirical evidence tells us that among the group of those who perpetrate violence on their partner, mental difficulties (mild, moderate and severe) are common (Elbogen, 2004; Kroop, 2008) though they do not constitute the only risk factor nor are they probably the most important. In the enormous com- plexity of mental disorder, of particular interest are those problems that present an acute level of seve- rity and cause the offender to strongly disconnect from reality (hallucinations, delirium, hyperactivity, very strange or dangerous behaviour, pathological jealousy, etc.), or chronic difficulties in social adapta- tion (like some personality disorders characterised by outbursts of anger or chronic affective disorders). Individuals that have a mental or personality disorder have greater propensity towards behaving and taking inappropriate decisions in situations of real or imagined conflict with their partner (Arbach and Andrés-Pueyo, 2007) and violent incidents can arise therefrom, especially with the combination of alco- hol or drug use. Evidence of the presence or history of mental or personality disorder does not always require a specific diagnosis to complete the assessment and may be useful indirect information for this risk factor. If infor- mation is not available, it is advisable to request it from experts before making unfounded assumptions on the basis of inaccurate information. Brief definition: Diagnosis or history of severe mental disorder in the alleged offender, considering as such, for instance: • Personality disorder with anger, impulsivity and emotional instability. • Schizophrenic disorder, major depression, bipolar disorder, paranoid disorder and similar disorders. Example: “The offender is diagnosed with borderline personality disorder for which he is currently not receiving any psychological or psychiatric treatment, because he believes that in the past he did not see any improvement by following medical guidelines.” 56 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES RF 12: POSSESSION OR EASY ACCESS TO WEAPONS BY THE ALLEGED Offender Offenders that, in their history of violence with the partner, have used weapons (firearms, knives, sticks, tools, etc.), if only in a threatening manner and have not actually used them, generally have a greater risk of recidivism. Obviously, the ease of use and access to weapons (such as professional criminals, law enforcement staff, hunters, sportsmen, etc.) may represent a higher risk of IPV and greatly increase the severity of violence, especially firearms. Reality shows that there are many and varied forms of using im- plements of all kinds (cooking utensils, ropes, etc.) to commit serious assaults. However, the fundamen- tal concern to be considered in the evaluation of this risk factor are firearms and so-called bladed we- apons, which present a greater risk than the other objects with the possibility of using them as weapons. Brief definition: The alleged offender has an arms license, is in possession of weapons or has a real op- portunity to access them. Example: “The offender is an expert hunter with a firearms license and a collection of guns inherited from his father.” RF 13: SUICIDE ATTEMPTS OR IDEATION ON THE PART OF THE ALLEGED Offender Suicide ideation, like homicide, is a powerful predictor of very serious violent incidents (among which there is suicide following the murder, which is quite common in intimate partner violence). In fact, sui- cidal behaviour is very often associated with femicide. This risk factor must also be assessed since there is a very strong association between affective disorders and suicide attempts or ideation, therefore it is interesting to ascertain the antecedents of this risk factor. In general, the time interval of this risk factor covers the offender’s entire life, and the presence is confirmed though the person has been clinically treated for problems of this nature. Investigation should be made into whether suicide attempts or idea- tion have intensified in the last six months and, above all, whether threats of suicide in relation to the future dynamic of the couple have been verbalised. Brief definition: The alleged offender has attempted suicide, has had or has thoughts about suicide. Ejemplo: “The offender, during the arrest, self-inflicted injuries in the police car by banging his head against the glass, then attempting to hang himself with his jacket from the prison door.” “During the court proceedings, the offender grabs his partner, goes onto the balcony and acts as if they are about to jump off together, asking the woman: will we kill ourselves?” 57 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES RF 14: EXTREME CONTROL OF THE WOMAN’S CONDUCT OUT OF JEALOUSY OR SIMILAR FEELINGS It is one of the most common and powerful violence risk factors and, in extreme cases, may be one of the most noteworthy with regard to fatal violence (sometimes associated with an intense possessive feeling over the woman). Under this risk factor lie a series of motives related to a strong feeling of ownership and very disturbing emotion that destabilises the offender emotionally to a high degree. It is mainly ba- sed on jealousy of a sexual nature, characteristic of intimate relationships and that brings the offender to develop a series of different violent tactics, all with the same ultimate goal, i.e. controlling their partner (Echeburúa, 2009). Studies suggest that jealousy is present in half of these violent incidents (O’Leary, 2007; Fagan and Browne, 1994). It has been estimated that sexual jealousy triggers between 7% and 41% of intimate partner violence incidents (Block, 2001). The difficulty in assessing this risk factor is related to its level of seriousness, a risk factor is considered serious when its presence renders the social and emotional life of the victim uncomfortable and difficult. Brief definition: The alleged offender manifests controlling behaviour, stalking and restriction of the woman’s freedom motivated by jealousy or similar feelings (for instance, cultural beliefs or beliefs of another kind). Example: “You will not go out with your friend. You are a lesbian. What you want is to go to bed with her.” 4. Vulnerability of the woman The offender’s risk factors, as perpetrator of the violent act are generally considered in IPV theories, especially in ethical and legal terms. It is not common to take risk factors concerning the abused woman into account. Howe- ver, reality shows that there are women that, for different reasons, are more victimised and vulnerable than others (Tjaden and Thoennes, 2000), without this implying that the woman is considered as the cause of her victim status, but the presence of her vulnerability factors must be taken into account for two reasons: the in- crease in the predictive power that may be added to these factors and the greater individualisation of strategies to prevent violence recidivism. Obviously, we can consider many factors concerning the victim, but traditionally speaking, empirical research has identified a number of them such as age (especially age difference between the respective members of the couple, education, belonging to minority and marginalised groups, etc.). 58 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES RF 15: WOMAN’S SOCIAL ISOLATION AND/OR LACK OF PERSONAL RESOURCES, AND/OR THE WOMAN’S JUSTIFICATION OF THE VIOLENCE INFLICTED BY THE ALLEGED Offender, AND/OR THE PRESENCE OF THE WOMAN’S CHILDREN THAT ARE MINORS AND/OR DEPENDENTS Although research into women’s vulnerability factors that render them more susceptible to abuse and violence is very much debated, it seems that both researchers and practitioners agree on their impor- tance as an IPV risk factor. This consideration does not mean that the victim may be responsible for their situation, but it is relevant to observe if risk factors that are controllable arise in certain circumstances of the victim that are key to prevention. Regarding the profile of the victims of violence, they tend to be more vulnerable on account of age, illness, loneliness or dependence (Corral, Echeburúa and Fernández- Montalvo, 2009). There are many risk factors grouped under this label (hence the length of the name of the factor). Situa- tions or incidents such as the following are included: absence of personal resources (of all kinds) that make her dependent on her partner, attitudes that justify the situation of partner violence and the pre- sence of dependents (children or parents) in her care. Brief definition: The woman does not possess or have personal, social, family, economic and/or work re- sources at hand, or they are very scarce. Isolation occurs in one or several of the aforementioned areas, and/or The woman justifies, minimises or denies that the alleged offender has inflicted a form of violence on their person, and/or The woman has children that are minors and/or dependents in her care, which may or may not be com- mon to the alleged offender. The risk factor is understood to be present should any of the aforementioned indicators exist. Example: “He is a good person but when he drinks he loses control and becomes violent; he does not want to do it but sometimes he gets nervous; the thing is that I do not do things well and he gets angry, he acts in accordance with is customs and beliefs, etc. [justifications].” 59 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 5. Woman’s perception of the risk situation This last risk factor is essentially a kind of screening or sampling of how the woman feels at the time of the assessment and with regard to the future in the most serious contingency that may occur, a fatal assault. It is an overall approximation of the victim’s perception at the time of assessment and can be very subjective and variable, but relevant due to the fact that it is the victim herself who provides the information. RF 16: THE WOMAN BELIEVES THAT THE ALLEGED Offender IS CAPABLE OF KILLING HER PERSONALLY OR HAVING HER KILLED BY THIRD PARTIES In the RVD-BCN Guide, this risk factor is derived from the idea that the victim herself is aware (or not) of the seriousness of the violent situation being experienced and her future in this situation. There are authors and experts that consider that, due to the fact that women are the only (or almost the only) observers (together with the abuser), obviously they know what has happened and it is them that can really anticipate what can happen. However, there are many opinions and incidents that refute this con- sideration and affirm that, in general, female victims of violence tend to underestimate the risk to which they are exposed by their partner (Kroop, 2008). Studies previously undertaken to verify women’s predictive power of the risk of being physically or psy- chologically assaulted again, indicate that the prediction is higher than chance. Thus, her risk assessment and anticipation can be a useful source of information for professionals. In general, if the history of abu- se goes back a long way, especially physical assaults, the general assessment made by the victim on their future risk is more biased (overestimating or underestimating the risk) than when the history of abuse is more recent. In any case, the woman’s explicit assessment of whether her partner or former partner could kill her is a direct way of assessing very extraordinary circumstances but which reflects the victim’s perception of the perpetrator’s capacity to harm them. Brief definition: The woman has a real perception of the high risk to which she is exposed and is convin- ced that the alleged offender is capable of killing her or fulfilling his death threats through third parties (family members, friends, hitmen, etc.). Example: “The couple live in the same home and are in the process of separation. The woman manifests great fear for her physical integrity and blocks the door of the room, where she lives, with a chair so the offender cannot get in. The woman never leaves the home alone, has the blinds drawn during the day and has changed the lock on the door for fear of the constant death threats made by her former partner.” 60 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Other risk factors Next we will consider other risk factors that are not part of RVD-BCN but which may be present in given cases and which increase the risk assessment conducted by the professional. The association between possible IPV risk factors is very extensive since they are connected with the personal and relationship circumstances of the offender and victim. Such a complex reality by heterogeneity makes it difficult to have a single and valid list for every victim. The individualisation of assessments, which is one of the strong arguments of clinical assessment advocates, requires the RVD-BCN list to be complemented almost compulsorily (especially in moderate- and high-risk cases) with risk factors which take on great importance in the specific case. Among the most common are: 1. The offender was the victim or witness of domestic abuse in their childhood or adolescence. 2. The offender experiences situations of significant emotional and/or personal crisis that, despite not being related to a psychiatric disorder, may be similar to the effects of the loss of emotional control. 3. The victim has experienced recent incidents of stalking by the offender. 4. Existence of a history of sadistic (sexual) behaviour or extreme violence (not necessarily partner violence). The list of RVD-BCN risk factors constitutes the basic core of the risk assessment, but is not the entire RVD-BCN. It is simply to offer actuarial guidelines in order to conduct a final risk assessment, which will be a decision fundamentally determined by the person conducting the assessment subsequent to following some necessary steps to complete the prognosis. These steps include the verification of a minimum number of assessed factors and the inclusion, where appropriate, of new specific risk factors in certain cases that heighten the risk of vio- lence being assessed. 61 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 5 Technical specifications of the RVD-BCN The RVD-BCN Guide, especially the list of risk factors that comprise its main structure, has been rigorously as- sessed before formalising the guide as a useful professional instrument. By way of a prospective longitudinal study, a series of metric properties were compared that are summarised below. Firstly, it must be pointed out that given the objective of RVD-BCN and its logic, based on the structured professional judgement approach, it is not a psychometric test similar to those used in the context of psychology and which psychological tests have traditionally represented. Despite the appearance of the RVD-BCN and other violent risk prediction instruments, which may be reminiscent of a psychological test, they are not and do not work as such. Generally, psychological tests are used to assess capacities, personality traits, symptoms and other clinical problems, etc. They usually have scales that allow people to be awarded grades that psychologists use in their diagnosis, selection tasks, etc. Psychometric properties guarantee the quality of the tests but cannot be transferred in an equal manner to violence risk assessment guides, which have their own quantitative parameters of quality. We avail of this comment on psychological tests to point out that risk assessment guides are not a technique specific and exclusive to psychology, but which all practitioners can use when making decisions that affect the risk of violence recidivism by offenders or criminals. They are not designed to perform dangerousness, clinical or criminological assessments, they simply allow us to review the risk factors that must essentially be assessed to make a prediction of violence that is superior to a random prediction. However, the RVD-BCN, especially its combination of risk factors, does not cease to be a risk estimation instru- ment based on quantitative procedures and must acquire properties that prove its quality for use in the profes- sional environment. Among the most important features, which will be outlined briefly below, are: reliability, convergent validity and predictive validity. Before describing these indices, we will briefly describe the work that has been carried out. All the data is taken from a longitudinal study undertaken in Barcelona, during 2010 and 2011, which entailed the application of the RVD-BCN Guide to a sample of 216 women that are users of services associated with the Barcelona Network to Fight Violence Against Women and following up on these women over three or six mon- ths. The longitudinal study4 was carried out at three different points in time (with a retention rate of 80%). In the first phase, the RVD-BCN was administered to all the women participating. A group of practitioners from the security sector (police force and Catalan regional police force), social services (social workers, psychologists, etc.), legal services (psychologists, solicitors and criminologists) and health services (doctors and nurses) applied the guide, all of whom had professional training and experience in the field of IPV and previous training in the use of the guide. Once the RVD-BCN was administered and before quantitatively assessing the outcome of the sum of risk factors present in each case, the practitioners were asked their opinion on the overall risk assessment of each of the women participating at that point in time. This variable, assessed by means of clinical judgement, will be used to compare the convergent validity of the RVD-BCN. 4. The entire study was drawn up in accordance with the ethical and professional standards applicable to this type of research and the project was approved by the Ethics and Clinical Research Committee of the IDIAP Jordi Gol i Gurina (Primary Care Research Institute). The University of Barcelona’s Group of Advanced Studies on Violence designed and developed the study. It should be noted that all the WHO recommendations concerning studies of domestic violence victims were followed since the study itself could increase the risk of 62 victimisation of the women participating (WHO, 2005).. CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Following an interval of three months (T2) and three months after that (T3), the same assessors contacted the women participating to ascertain whether they had been victims of violent acts by their partners or former partners over that period. In this follow-up, a series of 21 questions were posed with regard to violent incidents grouped into five categories: physical violence, sexual violence, psychological violence, threats and stalking. The following table illustrates the main descriptors of the RVD-BCN validation sample per sector and in their entirety. Security Social Services Justice Health Total RVD-BCN N 44 106 16 50 216 Mean age 39,25 37,88 36,13 44,65 39,09 SD age 10,06 10,83 9,97 16,07 12,31 Total mean 5,7 7,82 6,87 7,02 7,93 SD total 3,01 2,49 2,78 2,98 2,84 Nivell del risc Low (%) 77,3 45,3 56,3 46,01 52,8 Moderate (%) 13,6 31,1 25,01 38,01 28,7 High (%) 9,1 23,6 18,7 15,98 18,5 T2 (%) No. 40 96 9 41 186 Physical violence 10 4,1 nse 24,3 9,3 Sexual violence 5 3,3 nse 19,4 6,8 Psychological violence 22,5 37,1 44,4 76,3 42,4 Threats 7,5 19,6 11,1 44,7 20,8 Stalking 12,8 24,2 11,1 44,7 25,4 T3 (%) N 39 83 10 36 168 Physical violence 7,7 10,1 10 12,5 10 Sexual violence 2,6 5,2 nse 13,3 5,8 Psychological violence 15,4 38,5 20 41,2 32,3 Threats 12,8 22,8 nse 26,5 19,8 Stalking 7,7 23,8 10 38,2 22,1 Table 2. Descriptors of the RVD-BCN validation study per sector and follow-up period Note: T2: assessment after three months; T3: reassessment three months later; SD: Standard deviation; NS: Not specified 63 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Several important consequences can be deduced from the methodological strength of the RVD-BCN, and espe- cially due to the possibility of generalising results to the general population of women that are victims of intimate partner violence that resort to the public services to resolve their problems. Firstly, noteworthy is the continuity of the participants that were analysed. Between T1 and T2, continuity was 86.11% and between T1 and T3, it was 77,78%. Secondly, the prevalence of violent incidents in T2 and T3 is, generally speaking and for each specific case and form of violence, similar to that described in the epidemiological studies of IPV in Catalonia and Spain (see the Survey on Domestic Violence in Catalonia, 2010, and the Macro-Survey on Gender-Based Violence in Spain, 2006). Thirdly, worthy of mention is that the RVD-BCN quantitative assessment of the various sectors only shows a sig- nificant difference (p>0,005) between the values obtained between the police services and social services, such that the former are lower. Significant differences did not appear between the remaining services. The graph presented below illustrates the distribution of the total grading of the RVD-BCN scale, in which a bias can be seen towards the low values of distribution. Histogram 40 Mean = 7,13 Standard devioation = 2,836 N = 216 30 20 10 0 0,00 5,00 10,00 15,00 20,00 Figure 1. Distribution of the total scores of the RVD-BCN Guide An important element in the use of the RVD-BCN refers to the accessibility of the information concerning the different risk factors by the assessor, since most of the information employed comes from the woman that is the IPV victim, but also from the offender. The table presented below shows the manner in which most of the risk factors have been evaluated. In all the cases, information on the risk factors was obtained and is relatively similar to that found in prevalence studies of IPV risk factors on an epidemiological scale. 64 Frequency CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES No. RISK FACTORS PRESENCE % 15 Vulnerability of the woman 88,4 7 Emotional abuse 71,8 14 Extreme control 66,7 1 Intimate partner violence 63,8 6 Serious threats 60,6 10 Alcohol/drugs 58,3 2 Non-partner violence 53,2 8 Announcement of separation 45,4 3 Violence during pregnancy 41,2 16 Risk perception (of the woman) 37,0 9 Increase in severity 32,9 4 Criminal record 31,0 13 Suicidality 22,7 5 Breach of protection measures 18,1 12 Access to weapons 14,8 11 Mental disorder 8,3 Table 3. Prevalence of the sixteen RVD-BCN risk factors in the validation study With regard to the number of risk factors assessed in the T1 of RVD-BCN, in 82.8% of cases, answers can be given to the risk factors in each case. In fact, in 98.1% of cases, a minimum of seven risk factors were able to be comple- ted, which will be the minimum advised to validate a prediction based on RVD-BCN. Naturally, it is ideal to obtain information on all the risk factors, but any assessment made of less than seven assessed risk factors is to be considered invalid. It is strongly recommended that, in addition to the seven risk factors, endeavours be made to extend the information that allows the sixteen RVD-BCN risk factors to be com- plemented. 65 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Reliability The study analysed the reliability of the RVD-BCN with regard to the internal consistency and inter-observer relia- bilitys.5 The internal consistency was obtained using the Cronbach alpha technique and afforded a result of 0,716. The inter-observer reliability, calculated using the Kendall Tau test, gave a result of 0,729. In light of the internal consistency value, of a moderate level, and the inter-observer reliability value, of a good level for this type of test, we can consider that the RVD-BCN is a reliable test for assessing the risk of violence. It is noteworthy that it is inter-observer reliability that affords the better result, which implies that the assessments conducted by indepen- dent practitioners on the same case are sufficiently reliable for us to consider those made by other professionals using the RVD-BCN valid. Validity As for validity, two different estimations which were employed in the study of the RVD-BCN’s properties must be set apart. The predictive validity was assessed, that is to say, the correspondence between the risk assessment made using the RVD-BCN and the incidents which occurred after three and six months, and the convergent vali- dity, which indicates the extent to which risk assessments conducted using the quantitative estimation resulting from adding the presence of risk factors together (between a minimum of 0 and maximum of 16) are similar to clinical judgement (the overall assessment by experts) in each case. We will begin the description with predictive validity. In order to assess the predictive validity, the procedure was conducted as follows, on the basis of the data from the longitudinal study. Firstly, the data concerning violent incidents were categorised under T2 and T3. Secondly, the predictive validity was calculated using the logistic regression technique, in which the predictive variable was the total score of the RVD-BCN scale of 16 risk factors. For each dependent variable (physical violence, sexual vio- lence, psychological violence, threats and stalking), the AUC (area under curve) value was obtained, which repre- sents a global estimate of predictive efficiency, as well as the odds ratio (“relative opportunity”), which provides a simple approximation on the probability of getting the prognosis right on the basis of 1 and reveals the increase in predictive power of the test and for each form of violence that must be anticipated. 5. The inter-observer reliability was analysed on the basis of the comparison of thirteen cases in which two independent observers as- 66 sessed the sixteen RVD-BCN risk factors. CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 3 months 6 months Criterion OR AUC OR AUC Physical violence 1,28* 0,72 1,33** 0,72 Threats 1,28** 0,68 1,11 0,59 Psychological violence 1,12* 0,59 1,08 0,56 Stalking 1,05 0,55 1,17* 0,63 Sexual violence 1,04 0,54 0,98 0,50 *p< 0,05; **p< 0,01 Table 4. Parameters of the predictive validity of the RVD-BCN according to the form of violence and time interval of the prediction Note: OR: odds ratio; AUC: area under curve As can be seen, the predictive power is significant at three months for physical violence and threats (significant AUC), and at six months for physical violence and stalking. As for the odds ratio, it can be understood that the predictive power is really important when we deal with severe intimate partner violence. The objective of the RVD-BCN is to assess the risk of severe intimate partner violence. In terms of definition, this traditionally includes physical and sexual violence as well as some types of threats (of physical/sexual harm) and stalking (the most chronic and intrusive). A new variable was created which combined incidents of this kind in a single variable, severe violence, which was also broken down to compare the overall characteristics of the RVD- BCN and, also especially to find approximate cut-off points for the levels of risk to report (low, moderate and high). To obtain the predictive power result for the new variable, severe violence – which adds data on forms of violen- ce, all of which are important – logistic regression was also used and an AUC value of 0,72 resulted. In general, a moderate-medium value of predictive power is considered acceptable if we consider that the mean AUC values of most of the existing violence risk assessment tools is 0,70 (Fazel and Jay, 2010). From this calculation, cut-off po- ints were put forward, calculated on the basis of the ROC curve, to divide the risk levels of the RVD-BCN (of which there are three: low, moderate and high) so that we maintain the best balance between correct and incorrect an- swers. We must take into account that errors may be of two types: false positives (violent incidents did not occur in cases rated high risk) and false negatives (violent incidents occurred in cases rated low risk). The cut-off points proposed are: low risk, 1 to 7 points; moderate risk, 8 and 9 points, and high risk, 10 to 16 points. In the following graph, the distribution of risk levels is shown based on these cut-off points. 67 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 100 90 80 70 60 52,8 50 40 28,7 30 18,5 20 10 0 Low Moderate High Figure 2. Distribution of the RVD-BCN risk levels With these new cut-off points, the properties of the RVD-BCN were assessed in terms of their predictive power taking into account the probability of two types of errors: false positives and false negatives. These two types of errors are inversely proportional, since the total number of errors in the predictions correspond to the sum of the two aforementioned errors, but the assessor, in managing the guide’s cut-off points, can decide which of the two types of error they prefer next to one another. As is evident, in violence risk assessment, we must always seek to reduce the number of false negatives as much as possible but maintaining a level of false positives as low as possible. There are statistical parameters that inform us which is the best decision on this balance. These indices are called “sensitivity” and “specificity”. The first refers to false negatives and the second to false positives. Two further indices also appear, the PPV (positive predictive value) and NPV (negative predictive value). The table below shows the values of these indices in the RVC-BCN in percentages. 68 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES RVD-BCN 3 months 6 months Prevalence 9,3 10 Sensibility 88,2 81,3 Specificity 57,2 59,7 PPV 17,4 18,3 NPV 97,4 96,6 Table 5. Predictive power parameters of the RVD-BCN Guide (values in %) Note: PPV: positive predictive value; NPV: negative predictive value The general interpretation of the values that appear in the previous table help enormously to understand the usefulness of the RVD-BCN in predictive terms. Firstly, it should be noted that the prevalence of severe violence at three and six months, as can be seen, is rather significant and relatively stable, 9,3% and 10% respectively. As far as the predictive power is concerned, four possibilities must be set apart: sensitivity, which has a value between 88,2% and 81,3%, which represents the level of correct identification of the existing risk of violence; the specificity value ranges between 57,2% and 59,7% at three and six months respectively, which indicates the capacity to reject the risk when it does not exist; the PPV index confirms the likelihood of the positive result, and ranges between 17,4% and 18,3% (let us recall that prevalence is 9,3% and 10% respectively). The likelihood of NPV, that is to say, to render a true negative is between 97.4% and 96.6%. In brief, the RVD-BCN is a very sensi- tive guide for detecting the presence of the risk of violence though at the expense of a somewhat relevant level of false positives. Let us remember that we had assessed the predictive power as moderate-medium (AUC = 0,74). In any case, with the use of the RVD-BCN, we can improve the prognosis with regard to the victim’s safety in exchange for a level of error which affects the alleged offenders in statistical terms and on a group rather than on an individual scale. In this regard, the parameters obtained are similar to other guides with the same purpose (Andrés-Pueyo and Echeburúa, 2010). Finally, what remains is to describe the level of convergent validity between the RVD-BCN and global assessments conducted by experts in T1. In this case, a correlation between the expert’s level of risk allocation – also in three categories: low, moderate and high – with the grading of the level of risk assigned by calculating the aforemen- tioned cut-off points, and the result was a correlation (according to Spearman’s technique) of a value of 0,534 (p<0.001), which guarantees a good level of convergent validity between these two criteria: the expert opinion of practitioners and the conversion into three levels of the sum of the presence of risk factors according to the RVD-BCN. 69 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 6 The RVD-BCN Short User Manual Now that we have presented the guide and its features, we shall describe the RVD-BCN application procedure. To conduct a risk assessment, it is imperative to use all the available sources of information. It is advisable that all assessors analysis all information, oral or written, as thoroughly as possible, to complete the information upon which the risk assessment shall be based, as well as consider the limitations, if necessary, related to the omission of information or absence of relevant data. The sequence of operations to complete the risk assessment using the RVD-BCN involves: 1. Completion of the particulars of the woman to be assessed and the date of the assessment. This section, seemingly unimportant, is of great significance. First we must have the accurate details concerning the person should any incident occur (such as uncommon foreign names, women that move within the municipality, that are dealt with by different services, etc.). The accuracy of the particulars is of paramount importance. The date of the assessment is also of interest if we wish to compare it with previous assessments. 2. Exploration of the presence or absence of the sixteen RVD risk factors. The information to conduct these assessment can be obtained from different sources, which include: the interview with the victim, reports and information taken from files or official documentation, informa- tion provided by other people and services, etc. Once evidence characteristic of each risk factor is obtained, their presence or absence is determined. In the event of the absence of evidence, additional research is advised (if possible) in order to make the decision. Let us remember that the validity of the prognosis hinges first and foremost on the correct identification of the risk factors. 3. Measurement and grading of the level of initial risk. It is done by adding all the risk factors present (each is the equivalent of one point) after deciding on the presence or absence of the sixteen risk factors. It should be borne in mind there are three risk factors responses: YES, it is present; NO, it is not present – but in both cases the judgment must be based on ha- ving information that allows the opinion to be made. If no information is available, the box ND (no data) is ticked. The RVD-BCN Guide will be deemed invalid if a minimum of seven risk factors are not assessed. Once the sum total of present risk factors is obtained, the risk level is graded in one of three categories: low, if the sum is equal to or less than 7; moderate, if the sum is equal to 8 or 9, and high, if the sum is equal to or higher than 10 points. 70 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 4. Inclusion of new risk factors specific to the case in the assessment. In certain cases, the professional can find risk factors that do not appear in the list of the sixteen assessed in the RVD-BCN and which are considered to heighten the level of existing risk. When the first quantita- tive risk assessment has been performed, it is analysed whether other risk factors must be incorporated in that case and which, under the current and reasonably foreseeable circumstances, in a short interval of days or weeks, could increase the level of risk or render it very immediate. It is essential to make a re- latively detailed record, in writing, of this inclusion of new risk elements in the assessment, and thereby guarantee the information we will combine with the initial assessment to formulate the final assessment to be made by the practitioner in the following phase. 5. Final assessment and proposed risk level. The final risk assessment will be made taking into account the initial assessment guided by a score of sixteen risk factors and the inclusion, if necessary, of specific information that justifies the increase in the existing risk. Based on these sources of information, the practitioner will formulate their final as- sessment in the three categories defined in the guide. 6. Description of the action undertaken or to be undertaken based on the currently existing risk assessment. The practitioner will briefly record what actions have been implemented and/or are being carried out as a direct consequence of the currently existing risk assessment. 7. Foreseeable circumstances that may increase the level of risk in the future (alert). This section of the RVD-BCN has the task of alerting practitioners on the increase of risk caused if certain foreseeable circumstances become a reality. This alert function of the RVD-BCN allows preventive mea- sures to be undertaken to prevent new acts of violence against the woman from occurring. It is important to highlight that risk assessment is not a static process. The level of risk may change and fluctuate in time due to the circumstances of the offender (and also the victim). Therefore, it is advised that assessments be repeated at least every three months. In addition, the assessments must be reviewed in the event of changes of the critical circumstances capable of generating new conflicts or a resurgence of the violence, for example, if there is an increase in the likelihood of encounters and renewed contact between the offender and the victim. 71 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 7 Conclusions and recommendations for use A multitude of varied motives and reasons underpin the use of violence risk assessment and it involves a great deal of practitioners. In a brief overview, we can grasp the basis of the aforementioned statement. Risk assessment may be conducted for reasons related to the victim, their safety and well-being. It is used to formulate a plan which considers a certain level of protection according to the risk to which the victim is exposed and which is limited in time, since the risk is a changing phenomenon that varies over time. In this sense, a second reason may be the planning of protection measures. The third reason is related to the consideration of the risks presented and how they can be eradicated/reduced as well as one of the most im- portant reasons, raising the awareness of the woman by way of “her own perception” of the risk she is exposed to and a more objective and expert assessment, to strike an essential compromise between the victim and the practitioner on the appropriate steps to take to protect the victim. With regard to the offender, risk assessment has other uses (which are only mentioned here): adjustment of control measures, assessment of the efficacy of treatment and the identification of risk factors susceptible to change and intervention. Finally, in the professional domain, the reasons for employing risk assessment are very important: firstly, it im- plies an increase in the predictive power that practitioners already possess on account of their experience and training, help in decision-making which is always welcome due to the responsibility and complexity of this type of decision and, in our opinion, one of the most valuable reasons: subsequent transparency of the reasons be- hind the decisions made. Risk assessment which follows a guide or guide always allow a written record of the work done to be produced in order to anticipate what might happen to the IPV victim. This motivation is of paramount importance because it guarantees a historic record of the aspects assessed to make the prognosis. In this regard, risk assessment with guides and formally designed guides is hugely beneficial to clinical assessments, which are often not done explicitly, but as an intuitive or implicit result of other intervention processes with the victim or offender. 72 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES To assess the risk of violence, several recommendations should be borne in mind: 1. Identify which form of violence we wish to prevent and anticipate. Specifying which threat the woman faces is fundamental for deciding on the intervention strategy. Obviously, attempting to assess the risk of murder is not the same as assessing the risk of stalking or psychological violence. The reason behind this observation is not only the severity of the consequences but the prevalence and likelihood of each of them. Murder is impossible to predict due to its very low probability (3 x 1.000.000) and assessing the risk of murder is also very difficult for the same reason; however we have appropriate techniques for this task such as the Danger Assessment Tool by J. K. Cam- pbell (Campbell et al., 2009; Campbell, 1995), an actuarial guide specific to this task. 2. Identify the lethality of the threat. The probability of threats becoming a reality must be considered; it is somewhat similar to threats of suicide, not all of them are real and plausible, but the degree of likelihood must be decided upon. Death threats made by an offender that is detained and cannot access the victim have a different value to when the offender is about to be released or charged in court proceedings. 3. Recognise the victim to whom the threat is addressed and other potential victims. Generally speaking, in IPV, the current or former partner is the possible victim, but sometimes the chil- dren or parents and new partner of the victim are also the target of the offender’s violent conduct. It is especially important to be aware of who may be a direct or indirect target of the violence in this domain. 4. Assess the possibility of the threat being imminent. This is highly relevant as it determines the precedence and urgency given to the assessment and the intervention. Under these circumstances, it is essential to warn the victim and mobilise the resources available to prevent the action that we consider immediate. 5. This may be the last step according to the service and the practitioner conducting the risk assessment, but it would involve direct intervention in controlling the offender through the available resources (incapacita- tion, arrest, etc.). 73 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Is it possible to predict and therefore prevent intimate partner violence? Let us consider that the answer, in glo- bal terms, is yes. However, specific episodes of future violent conduct are difficult to predict with the precision desirable. This task is especially complicated when individuals do not present personal traits that have connota- tions of propensity towards violence or a history of violence. Nevertheless, it is easier to anticipate future vio- lence when it concerns potentially violent people, who have a previous history of repetitive violent conduct, or a personality that shows propensity to violent conduct or, if applicable, that suffer a disorder closely associated with violent conduct. In conclusion, we could say that we will never be able to know if a person will commit a given violent act in the future, acts such as violent assaults against partners cannot be anticipated. However, the likelihood of them oc- curring can be, and this is the objective sought by risk assessment techniques: to gauge the probability of violent conduct arising in a given context (especially) if it is serious. We cannot guess the future, we can only assess the likelihood of violence emerging on certain occasions (after a few months, days, etc.) and under specific conditions (in a family, school environment, etc.). Therefore, the prediction of violence becomes an estimation of the relative risk of violent conduct by a person occurring in a given environment and for a relatively exact time period. The RVD-BCN Guide is effective in assessing the risk of future violence, but it is not a tool for predicting future violent conduct; a distinction must be made between these categories since we shall always work with risk es- timations sufficiently in order to arbitrate security measures and risk management strategies that allow the risk of violence to be eradicated in a reasonable and practical manner. 74 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES In summary, in addition to making better predictions, the advantages of using the RVD-BCN are as follows: 1. Making appropriate decisions. This means that guides help improve the consistency of decisions, increase the accuracy and establish a norm for individualised interventions, since risk assessment implies a meticulous process of analysis of the strengths and weaknesses of the subject and their clinical state. They help steer the intervention of professionals in prediction tasks and do not leave the risk assessment process up to their free judge- ment, since this method has proven to be unreliable and of questionable validity. 2. Increase the rigour and transparency of decisions. The assessment procedure forces information sources, which generate the data that form part of each of the judgements on the items in the guides, to be compared. In this manner, users’ rights are protected when reviewing what research and professional practice has shown to be relevant to the prognosis. The guide contributes to enhancing the consistency of decisions since compared data collection systems are taken into account which contain relevant and significant data on the biographical history of the person assessed, the variables of their clinical condition and situation (risk/protection factors) and upon which a prognosis of future conduct must be made. 3. Safety management. Violence risk assessment that follows these procedures generates many ideas around how to protect victims, since they have a time perspective relative to the risk of violence that allows authorities and the people involved (offenders and victims) to adopt precise measures for action related to the foreseen risks. In this regard, natural and social risk management experiences can afford us many strategies for action applicable to the safety of victims. 75 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES GENERAL RECOMMENDATIONS FOR THE RVD USER 1 Obtain information on the guide’s factors through available or potential sources. Though the victim is usually the main and most common provider of information, it must be borne in mind that information from the police and other services that have dealt with the woman’s case under assessment, including solicitors, clinical staff, family members, witnesses, etc., is valid. 2 Consider that the basis for risk assessment is the information that allows us to decide whether the risk factor is present or absent. It is highly desirable to have information that is as accurate as possible. This information may be required for subsequent verification, for which it is re- commended that written records be produced and the information used to make decisions be archived. This information may also prove useful for analysing the quality of the assessment. 3 Do not obtain information on the risk factors (and other security issues) in situations in which the victim has no guarantees of talking without being afraid (presence of the partner, in a public place, etc.). Concerns for her safety in the interview situation or contact with women’s services may affect the quality and credibility of the information given by the victim. 4 It must be borne in mind that the guide’s list of risk factors in not “unique and exclusive”. It is a list referring to the risk factors that are “more frequent in cases of severe intimate partner violence” and which have been included in the guide to assess them at all times and in all cases. These are the factors that must always be evaluated, but the list can be extended by virtue of each individual case, and it is the practitioner that must include these factors in the final case assessment. 5 It should be noted that the victim may have to face very complicated or even dangerous situa- tions as a result of the appraisals, and must be capable of managing these new threats posed by the assessments. 6 It must be remembered that the risk is a “transient” and “variable” condition (sometimes changes can be very fast and sudden). Generally, this implies the days and weeks subsequent to the formal complaint, intervention, etc., hence the importance of ensuring the intervention (the very act of assessing the risk) does not heighten the risk of assault against the woman. 76 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 8 References Amor, P.J.; Echeburúa, E.; Corral, P.; Zubizarreta, I; Sarasúa, B. (2002) Repercusiones psicopatológicas de la violencia doméstica en la mujer en función de las circunstancias del maltrato. International Journal of Clinical and Health Psychology, 2, 227-246 Andrés-Pueyo, A.; López, S.; Alvarez, E. (2008) Valoración del riesgo de violencia contra la pareja por medio de la SARA. Papeles del Psicólogo, 29 (1): 107-122 Andrés-Pueyo, A. (2009) Violència contra la pareja: definición, prevalencia y factores de riesgo. Valencia, Centro Reina Sofía Andrés-Pueyo, A.; Echeburúa, E. (2010) Valoración del riesgo de violencia: instrumentos disponibles e indicaciones de aplicación. Psicothema 2010, Vol. 22, 3, p. 403-409 Andrews, D.; Bonta, J. (2006) The psychology of criminal conduct. New York: Lexis Nexis Belfrage, H.; S. Strand, et al. (2011) Assessment and Management of Risk for Intimate Partner Violence by Police Officers Using the Spousal Assault Risk Assessment Guide. Law and Human Behavior, 10 Campbell, J. C. (ed.) (1995) Assessing dangerousness. Violence by sexual offenders, batterers, and child abusers. Thousand Oaks: SAGE Publications. Campbell, J.; Webster, D.; Glass, N. (2009) The danger assessment: validation of a lethality risk assessment instrument for intimate partner femicide. Journal of Interpersonal Violence, 24, p. 653-674 77 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Dutton, D.G.; Golant, S.K. (1997) El golpeador. Un perfil psicológico. Buenos Aires: Paidós Dutton, D. G.; Kropp, P. R. (2000) A review of domestic violence risk instruments. Trauma Violence & Abuse, 1 (2): 171-181 Echeburúa, E.; Fernández, J. et al. (2009) Predicción del riesgo de homicidio y de violencia grave en la relación de pareja: intrumentos de evalua- ción del riesgo y adopción de medidas de protección. Valencia, Centro Reina Sofía Echeburúa, E.; Amor, P. et al. (2010) Escala de predicción del riesgo de violencia grave contra la pareja. Psicothema 22: 1054-1060 Singh, J.; Fazel, S. (2010) Forensic risk assessment. Criminal Justice & Behavior, 37: 965-988. Felson, R. et al. (2008) Is the Association between Alcohol Use and Delinquency Causal or Spurious? Criminology, 46: 301-324 Hart, S. D.; Michie, C. et al. (2007) Precision of actuarial risk assessment instruments: Evaluating the ‘margins of error’ of group v. individual predictions of violence. British Journal of Psychiatry, 190 (Supl 49): s60-s65 Hilton, Z.; Harris, G. T. (2005) Predicting wife assault: a critical review and implications for policy and practice. Trauma, Violence & Abuse, 6 (1): 3-23 78 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES Kropp, R.; Hart, S.; Weabster, C.; Eaves, D. (1995/2005) SARA Manual para la valoración del riesgo de violencia contra la pareja. Barcelona: Publicacions i Edicions de la Universitat de Barcelona Kropp, R. (2008) Intimate partner violence risk assessment and management. Violence and Victims, 23 (2), 202-222 OMS (2006) Researching violence against women. A practical guide for researchers and activist. WHO Publication Redondo, S.; Andrés-Pueyo, A. (2007) La psicologia de la delincuencia. Papeles del Psicólogo, 28 (3): 147-156 Stith, S. M.; Smith, D. B. et al. (2004) Intimate partner physical abuse perpetration and victimization risk factors: A meta-analytic review. Aggression and Violent Behaviour 10: 65-98 Tjaden, P.; Thoennes, N. (2000) Extent, Nature, and Consequences of Intimate Partner Violence, Findings From the National Violence Against Women Survey, 1-57 US Health Department, CDC Walker, L. E. (1999) Psychology and domestic violence around the world American Psychologist, 54, 21-29 Webster, C. D.; Douglas, K. S. et al. (1997) HCR-20. Assessing risk for violence. Version 2. Vancouver, Canada: Mental Health, Law and Policy Institute, Simon Fraser University 79 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES 80 CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES CIRCUIT BARCELONA CONTRA LA VIOLÈNCIA VERS LES DONES THE RVD-BCN RISK ASSESSMENT GUIDE FOR VIOLENCE AGAINST WOMEN PERPETRATED BY THEIR PARTNER OR FORMER PARTNER Barcelona, 15 december 011 The authors will be stated in any exploitation of the work authorised under the license.