Another Tool in the Toolbox for Domestic Violence Pretrial Determinations

At the time of Laura’s death in 2013, her abusive ex-boyfriend, who was also her murderer, was out on bail awaiting trial for charges stemming from prior domestic violence attacks and threats. According to court records and family members, he previously had “allegedly beat her with a baseball bat, dragged her behind a car, strangled her until she blacked out on the floor and told her over and over how he would kill her if she ever left him.” This is one of various tragic cases we hear about each year. Alternatively, however, there have also been other cases where alleged abusers have been detained pretrial, have not been able to post bail either because it was denied or because they could not afford it, and after spending some time in jail had the charges dropped for a number of reasons.

Out of this wide range of cases there are special legal considerations at the pretrial stage for domestic violence cases due to the potential risk to the victim and the past or present association of the accused and the victim. Although our legal system in California recognizes that there must be extra precautions for the victims’ safety, in practice California does not have a validated evidence-based method to identify and manage the most dangerous domestic violence offenders while affording those others accused their due process rights during the pretrial stage. In this post I will describe the California procedures already in place when dealing with domestic violence cases, and recommend some evidence-based risk assessment tools that the courts and others involved can use.

In California the legal pretrial procedure with domestic violence offenses and charges is generally different than laws and procedure for other criminal and civil charges. Individuals arrested or detained for most misdemeanor offenses can be cited and released. Cal. Penal Code § 853.6(a)(1). However, individuals who are arrested for misdemeanor domestic violence-related offenses, including violations of protective court orders, are not eligible for immediate citation and release. Cal. Penal Code § 853.6(a)(2). Similarly, individuals who are accused of violating a civil temporary restraining order (Cal. Civ. Proc. Code § 527.6) will also be ineligible for citation and release “if the detained person made threats to kill or harm, has engaged in violence against, or has gone to the residence or workplace of, the protected party.” Cal. Penal Code § 853.6(a)(3)(6). To obtain pretrial release these individuals must either post the scheduled bail amount or wait until they appear in front of a judge (pdf, go to page 13). In rare circumstances, they might be released if the arresting officer concludes that there is no reasonable likelihood “that the offense will continue” . . . or “that the safety of persons or property would be imminently endangered by release of the person arrested.” Cal. Penal Code § 853.6(a)(2).

Individuals accused of more serious domestic violence offenses will be detained pending a hearing before they are eligible for bail or being released on their own recognizance (OR). Cal. Penal Code § 1270.1. At the hearing, the judge will consider past acts of violence and alleged threats against others to determine “the danger that may be posed to other persons if the detained person is released.” Cal. Penal Code § 1270.1(c). While those accused of most criminal charges are entitled to bail under the California Constitution, the judge may deny bail if the defendant is charged with a felony “involving acts of violence” or “sexual assault” on another person and the judge finds “based upon clear and convincing evidence that there is a substantial likelihood the person’s release would result in great bodily harm to others.” Cal. Const. art. I, § 12. Otherwise, California law allows judicial officers “to set bail in an amount that [they] deem[ ] sufficient … to ensure the protection of a victim … of domestic violence.” Cal. Penal Code § 1269c.

There may be situations, however, where no amount of money bail will ensure the protection of the victim since high-risk individuals might be able to afford the costs or obtain a surety bond. On the other hand, there might also be situations where information about the defendant is minimal and those in the judicial process will err on the side of caution and deny pretrial release to low-risk individuals. A validated-evidence based risk assessment tool can help to better ensure that these victims’ rights laws are effective and just. Although Laura’s case was outside of California, her case would have scored an 18 on a lethality-screening test, an indicator of “extreme danger.” According to Jacquelyn Campbell, a leading national expert in domestic violence, “the system might have worked best together to identify that perpetrator as high-risk and manage that case in a more proactive way.”

There are various risk assessment tools and other instruments specific to domestic violence that help determine which defendants are at high risk of reoffending as well as a victim’s risk of lethal assault. The general benefits in using these instruments include:

  • A more effective allocation of scarce resources (jail space, electronic GPS or other judicial monitoring, intervention and other programs).
  • Helping domestic violence workers and victims create responsive safety plans.
  • Avoiding the mixing of low-risk offenders with high-risk offenders, especially since certain intensive programs for the most dangerous offenders can actually increase recidivism rates among low-risk offenders.
  • Improving communication and providing a common understanding of risk for the different parties and pretrial decision makers involved.
  • Diminish reoffending by building case plans around certain risk factors.

Among the instruments that the Bail and Release Work Group in Santa Clara County, CA, deemed appropriate to assist in pretrial determinations were the Ontario Domestic Assault Risk Assessment (ODARA), the Spousal Assault Risk Assessment (SARA), and the Domestic Violence Screening Instrument (DVSI-R) (pdf, go to page 35).

The ODARA measures the likelihood of future domestic violence, as well as the severity and the time these assaults will occur; “no other DV risk instrument has a documented ability to do all this.” After a year long-study of risk assessment tools the Maine Coalition to End Domestic Violence (MCEDV) stated that ODARA’s “predictive accuracy of 77% is the best there is in validated, evidence-based DV risk assessment tools without involving tools requiring clinical expertise.” Since January 1, 2015, law enforcement officers in Maine are now required to complete the ODARA for domestic violence-related cases and have the results available to officials setting bail and prosecutors. Me. Rev. Stat. tit. 19-A, § 4012.

While most domestic violence instruments focus on male offenders and heterosexual relationships, recent research has validated the ODARA for female offenders. There is also ongoing research for validating it for use with same sex-couples. This development is groundbreaking since 1 in 4 men have been victims of domestic violence within their lifetime and same-sex couple domestic violence occurs at-the same rate or more often than opposite-sex couple cases.


Although this tool was originally developed for front-line police officers, it can be used when domestic violence is alleged by correctional services, probation, and pretrial services as well as other victim services. It takes approximately ten minutes to score the assessment and consists of gathering 13 yes-or-no questions from either the victim or records. According to Francine Garland Stark, Executive Director of MCEDV, “ODARA is an appropriate tool for the criminal justice system, because it focuses on the likelihood of re-offense, and officers can quite easily determine most of the 13 items from the offender’s criminal record, without having to rely extensively on the victim’s input; other tools require information from the victim and are tailored more for safety planning with the victim.” Stark telephone interview, 03/31/16. The questions range from the offender’s history of violence, to the victim’s personal circumstances and the details of the latest assault.

Maine entered into a formal agreement with the Waypoint Centre for Mental Health Care to implement the ODARA statewide, but money was not exchanged. Stark telephone interview, 03/31/16. There is a free electronic training program online, and neither fees nor professional restrictions are required to score the ODARA.

The SARA is an easy-to-use checklist that helps determine the “degree to which an individual poses a domestic violence threat” to his or her spouse, family member, or persons involved. The SARA is not an actuarial tool that provides a definite score on the level of risk (pdf, go to page 103). Instead, it is meant to guide professional judgment about risk while leaving room for external considerations, “such as the nature of the environment into which an individual is likely to be released” (pdf, go to page 103).

The checklist includes 20 items that cover criminal history, psychosocial adjustment, spousal assault history, and information about the alleged current offense (pdf, go to page 26). This information is obtained through interviews with the victim and the accused, a review of records, and standardized measures of physical, emotional and substance abuse. A range of professionals can use the SARA including correctional officers, law enforcement officers and professionals in government agencies. However, when the SARA is used to evaluate and make decisions about an individual the users must meet MHS b-level qualification. This requires a completion of “graduate-level courses in tests and measurement” or “equivalent documented training.”

The DVSI-R assesses the risk of re-offending in domestic violence-related cases. The DVSI-R is an efficient 12-item tool that primarily focuses on the offender’s criminal history, which can be obtained through databases and probation or court records (pdf, go to page 440). It also includes inquiring into recent separation, employment status and the presence of children during the incident (pdf, go to page 20). It was designed to help expedite the processing of DV cases in Colorado since Colorado law requires domestic violence offenders to remain in custody until they appear before a judge (pdf, go to page 440). The tool was meant to be readily available “to prosecutors, judges, and probation officers soon after an offender is arrested” (pdf, go to page 440). Although this tool functions as a quick screen, a disadvantage with it is that it should be followed by a more intensive evaluation with instruments such as the SARA when it proposes a high-level of risk (pdf, go to page 19).

Pretrial domestic violence determinations are complex; decision makers must balance the safety of the victim, the due process rights of the accused and the efficient use of scarce resources. The foregoing explains advantages as well as some disadvantages when considering which of these instruments to apply. These risk assessment instruments are not meant to displace experience, knowledge or skill, but to add an extra tool that will help make better and more efficient decisions.