Hilliard G. Foster

Clinical predictors of self-mutilation in hospitalized forensic patients

This study evaluated the clinical correlates and inpatient course of self-mutilation in a diagnostically diverse sample of hospitalized forensic patients. Fifty-three male forensic inpatients, treated in a maximum-security hospital, who engaged in at least one instance of self-mutilation during a 2-year period, were studied and compared with 50 male forensic patients at the same hospital who had not engaged in self-mutilation. Self-mutilating patients were younger, more likely to carry a diagnosis of personality disorder or mental retardation, engaged in more outwardly directed aggressive behavior as assessed by the Overt Aggression Scale, were treated with substantially higher doses of neuroleptics, and were more likely to be civil or correctional patients than insanity acquittees. The two groups did not differ on variables such as history of suicide, history of violence, neurological characteristics, and other demographic variables. After an incident of self-mutilation, the probability of recurrence was high. The substantially higher level of outwardly directed aggression of self-mutilating patients, along with their higher apparent need for neuroleptization and the high risk of recurrence of the self-mutilation, suggest that they are a subset of violent individuals who are relatively unresponsive to treatment and who are dangerous to self and others.

Serum cholesterol and aggression in hospitalized male forensic patients

Human studies of the link between serum cholesterol and aggression have yielded equivocal results. Depending on the type of aggression studied (e. g., criminal violence or Type A hostility), investigators have found either a negative or a positive association between cholesterol and aggressive behavior. We conducted a retrospective analysis of aggressive incidents in a sample of hospitalized male forensic patients. The whole sample had lower cholesterol levels than the general population. Patients with low cholesterol levels (<200 mg/dl) engaged in more frequent aggressive behavior but showed no difference in severity of aggression. They also showed no difference in verbal vs physical aggression. The relationship between cholesterol and frequency of aggression was curvilinear, with the most frequent acts of aggression committed by patients with moderately low cholesterol levels. Current research findings regarding the cholesterol-aggression association suggest the need for further clarification of the behavioral parameters under investigation.

Neuropsychological deficit and aggressive behavior: a prospective study.

1. Failures in higher cortical executive functions and dyscontrol of aggressive impulses were hypothesized to co-occur in aggressive men. 2. The performance of a sample of violent forensic patients on selected neuropsychological measures (Judgment of Line Orientation, Stroop Color Word Test, Emotion Perception Test, Test Of Nonverbal Intelligence, Wisconsin Card Sorting Test, and Symbol Digit Modalities Test) were used to predict their overt aggressive behavior during a one-year period using the Overt Aggression Scale. 3. Whole-set correlation analyses revealed that scores on the Judgment of Line Orientation, Stroop Color Word Test, and Emotion Perception Test accounted for 57% of the variance of aggression scores. 4. Frequency of aggression, but not severity of aggression, can be predicted using these three scores.

Efficacy of carbamazepine in assaultive patients with frontal lobe dysfunction

1. A double-blind pilot study using a reversal design was conducted to examine the efficacy of carbamazepine in four cases of assaultiveness associated with frontal lobe dysfunction compared to five control cases. 2. Frontal lobe patients who received carbamazepine demonstrated a pattern of improvement on all cognitive, affective, and behavioral measures. This was followed by return to pretreatment level (measures of assaultiveness, depression, and general cerebral efficiency) or sustained improvement (measures of frontal lobe efficiency) when carbamazepine was discontinued. Control subjects showed only minimal differences across testing.

SERUM CHOLESTEROL LEVELS AND SEVERITY OF AGGRESSION

Analysis of serum cholesterol levels of 50 men admitted to a forensic hospital for crimes of violence showed 21 more violent men had lower serum cholesterol levels and a smaller range than those for 29 less violent men.

Rapists and child molesters: psychometric comparisons

The availability of psychosocial treatment for sex offenders is influenced to a considerable extent by the process of adjudication. Convicted rapists are usually incarcerated, and thus receive treatment in a high-structure setting. Convicted child molesters are usually paroled and receive treatment in the low-structure setting of outpatient psychotherapy. In the present study, three types of sex offenders (child molesters, child rapists, and adult rapists) were compared to examine the validity of the match between type of sex offender and type of treatment modality. No differences were found on factoranalytically derived psychological and social historical factor scores between child molesters and child rapists. Adult rapists, however, demonstrated considerably higher levels of pathology characterized by dysphoria, subjective distress, and inhibition, than child molesters and child rapists. The egodystonic psychopathology of the rapists suggests a more favorable prognosis.

SERUM CHOLESTEROL LEVELS AND FREQUENCY OF AGGRESSION

Analysis of serum cholesterol levels of 106 male forensic patients showed significant differences in frequency of aggression, with low cholesterol levels predominant among those more frequently violent.

Variables Associated with Violence in a Forensic Population

Cautious optimism has returned among practitioners and researchers involved in the prediction of violent behavior as a consequence of what has been termed “second-generation thinking” on the prediction of violence. New emphases have been suggested on actuarial techniques, alternate research methodologies, and comparisons between various clinical populations. In the present study, four clinical variable domains were sampled in a forensic population (psychiatric diagnosis, electroencephalographic measures, psychological test data, and social history data) in order to examine the absolute and relative strengths of their relationship with acts of severe violence. A characterological type of dysphoria was found to have the strongest association with violence, followed by the absence of temporal lobe damage, and a passive style of response to threat.

Alcohol Abuse, Violence, and Neurological Impairment A Forensic Study

The relationship between the frequency and severity of violence and three parameters of alcohol abuse (acuteness, chronicity, and age of onset) was examined retrospectively in a sample of forensic psychiatric patients. Subjects were further compared with respect to electroencephalographic characteristics. Acuteness and chronicity of alcohol abuse were found to be related to severity but not to frequency of violence. Acute alcohol abusers appeared to suffer from more severe central nervous system impairment than did nonabusers. Age of onset of alcohol abuse was found to have little impact on the variables examined. Results are discussed in the framework of empirically derived typologies of psychopathology.

Creatine kinase elevations and aggressive behavior in hospitalized forensic patients.

The relationship between creatine kinase (CK) and aggressive behavior was tested in 195 males consecutively admitted to a forensic hospital. Among patients receiving antipsychotic medication, the most violent patients had higher CK levels than less violent patients. This was not the case in patients who did not receive antipsychotic medication. CK levels were not influenced by age, ethnicity, or clinical diagnosis. CK levels were however influenced by prior assaul-tiveness and restraints. When these two factors were controlled for, CK levels remained strongly associated with subsequent violence. CK appears to be a potential predictor of violent behavior. It has the advantage of easy availability in comparison to other biological markers of aggression (e.g., 5-H1AA). Prospective studies are needed to confirm the validity of this biobehavioral association.