Commentary: Stalking Risk Profile

The first anti‐stalking legislation appeared in the United States 16 years ago.¹ Before anti‐stalking legislation was enacted, little was known about stalkers. After the initial laws were passed, stalkers and their victims became a focus of research. A Medscape² search conducted for this publication revealed 110 articles about stalkers.

To date, research has supplied consistent demographic data about stalkers. They are older than other offenders, are typically male, and are unmarried at the time of offending.^3–5 Now that we are entering the second decade of research on stalking, various demographic data about subsets of stalkers are being compiled as they relate to predictors of violence.⁶

Research reveals stable victim characteristics. The typical stalking victim is female⁷ and a former intimate of the stalker.^8,9 Epidemiological studies have been conducted^7,10 and demonstrate that between 8 and 15 percent of women and 2 percent of men report having been stalked. Stalking remains underreported.¹⁰ Only one half to one third of stalking victims reported such crimes.^7,11

Epidemiological studies have stimulated researchers to study nonforensic populations of stalkers. Concern has arisen among researchers that forensic stalkers differ from nonforensic stalkers. Attempts have been made to characterize these groups better by studying college students. Roberts¹² compared the experiences of female students whose relationships ended with stalking, with harassment, and with no harassment. The study concluded that the longer the relationship, the more the risk of stalking.

Epidemiological studies have also been useful in delineating the natural course of stalking. Purcell et al.¹³ conducted an epidemiological study that examined the nature and prevalence of stalking in an Australian community. Of 1844 survey respondents, 23 percent met the legal criteria for stalking. In almost half of the cases, stalking abated within two weeks. After two weeks, stalking persisted for a median of six months. Strangers typically stalked their victims for less than two weeks. Someone known to the victim typically engaged in stalking that persisted more than two weeks. Harassment became more intrusive after two weeks. Threats, as well as assaults, were more likely to occur after a two‐week period. Subjects harassed less than two weeks did not differ psychiatrically from a group of people who were never harassed.

Although his was not an epidemiological study, Rosenfeld¹⁴ found that 49 percent of stalkers recidivate, and 80 percent of those reoffend in the first year. Recidivism was associated with substance abuse and a personality disorder.

While demographic data and epidemiological studies are somewhat consistent, there has been no agreement concerning classification. Some classification systems are being incorporated into research models, although there is still no uniform classification model.

To date, there have been several classification systems. Some describe the nature of the attachment of the stalker to his victim, while others rely on characteristics of the stalker himself. As recently as January 2006, a new classification system was proposed.¹⁵

Presently, research is most focused on the relationship between stalking and violence. Studies correlate violence with the demographic variables of victims, stalkers, and the nature of the relationship between the stalkers and victims. The risk of lethal violence and stalking has been researched. A recent study by McFarlane et al.¹⁶ reported that 76 percent of femicide victims were stalked by their assailants in the 12 months prior to their murders.

Harmon et al.⁶ found that stalkers who were former intimates were violent, had an Axis II disorder, abused substances, and threatened their victims. Roberts¹⁷ studied 220 female undergraduates who were stalking victims. He found that violence was associated with a direct threat made during stalking, with the stalker's abuse of prescription drugs, or with the former partner's being jealous.

While violence and its association with stalking are vital, Purcell and her colleagues¹³ point out that most stalking does not result in violence. The area of stalking research that has received the least attention is risk assessment. Yet, forensic psychiatrists and psychologists routinely perform risk assessments of individuals charged with stalking.

A risk assessment has been proposed by James and Farnham¹⁸ for assessing stalkers for serious violence. They studied 85 stalkers in London and found that previous risk factors associated with violence did not apply to a subgroup of stalkers who were seriously violent. They did not find that substance abuse, the presence of a personality disorder, or a history of previous convictions for violence were associated with serious violence. Serious violence was found to be associated with prior visits to the victim's home and former sexual intimacy.

There have been no models for assessing stalkers for less serious violence, although Mullen et al.¹⁹ clearly discussed and outlined such factors in a study of 145 stalkers in 1999. Studies have addressed violence and stalking but have not delineated degrees of violence as separate factors.

A recent risk assessment model proposed by Kropp and colleagues²⁰ described the difficulty with assessments and offered a model that relies on yet another method of classifying stalkers. More important, these investigators noted that actuarial assessments are not useful because stalking is targeted violence, the stalking behavior is not always violent, and the behavior can persist for long periods.

Morrison²¹ studied 100 aggressive stalkers in Canada and found that previous violent behavior, strong negative emotions, and obsessional tendencies toward the victim were the strongest predictors of violence.

Meloy²² proposed a different model of risk management dependent on whether the stalker was harassing a public figure or a former intimate. While the risk assessments mentioned herein focus on stalker characteristics, a recent attempt was made to associate victimization with a risk of violence. Brewster²³ focused on former intimate victims instead of current intimate victims. A moderate correlation between verbal threats and physical violence was found among surveyed victims. Threats of violence were found to be better predictors of violence during stalking than was past violence.

Investigators in a few studies have used statistical measures to quantify the psychological effects of stalking on victims. Blaauw et al.²⁴ studied stalking victims and compared their symptoms of psychopathology and features of the stalking. They found that symptoms reported by stalking victims were more comparable with those of psychiatric outpatients than with those in the general population. They also found that 75 percent of stalking victims displayed symptoms consistent with a psychiatric disorder. Some studies have focused on specific psychiatric syndromes among victims, such as post‐traumatic stress disorder.²⁵

Davis and colleagues²⁶ examined the relationship of stalking victimization, level of fear, and health outcomes in men. They found that women were more likely to report fear than were men, men were more likely to be stalked by a stranger, stalking behaviors were similar across gender, and poor current health for women especially was most highly associated with level of fear. Both sexes reported a greater likelihood of having developed a chronic disease and having been injured.

From a review of the literature just discussed, it is clear that more data are needed on assessing and managing risk among stalkers. Prospective studies are necessary, and a uniform classification system would be helpful. Warren et al.²⁷ recently proposed a problem behavior model that recommends specifics for assessment and individualized treatment through the use of clinics that target specific behavior. Treatment included components such as cognitive therapy, victim empathy, substance abuse treatment, social skills, and a combination of specific therapeutic interventions coupled with the imposition of judicial sanctions.²⁸

In this significant addition to the stalking research literature, Mullen and colleagues²⁹ have developed a Stalking Risk Profile that incorporates the nature of the relationship between stalker and victim; the stalker's motivations; the general psychological, psychopathological, and social realities of the stalker; the circumstances of the victim; and the legal and mental health context. This model is not an actuarial scale, but rather an assessment to be used on a case‐by‐case basis.

Included in this profile is a recommended assessment of the stalker that features standardized tests of cognitive function, experience and expression of anger, personality traits, self‐image, acceptance of responsibility for behavior and interpersonal attachment style. The authors also use the HCR‐20³⁰ and Spousal Assault Risk Assessment³¹ when indicated.

The risk assessment model advanced by Mullen and associates²⁸ includes victim variables and provides specifics for assessment and treatment of the stalker. The authors discuss in less detail the assessment and treatment of victims.

• American Academy of Psychiatry and the Law