Two decades of empirical research has documented the continuing risk of inpatient assaults on staff. Traditionally, the characteristics of assaultive patients have included male gender, diagnosis of psychosis or neurological abnormality, and a history of substance abuse and of violence toward others (1,2,3,4). Recent studies have expanded this list of characteristics to include younger age, increased likelihood of female gender, a diagnosis of personality disorder, and a history of being a victim of violence (5,6).
As the health care system has moved toward deinstitutionalization and privatization, community-based services have begun to report patient assaults on staff (7,8). Whether these assaults are a temporary phenomenon that is associated with patients' transfer to a new environment and that will attenuate over time or are one that will remain a serious risk to community staff has not been examined.
This report presents preliminary data on the characteristics of assaultive patients in community residences who had been discharged from Massachusetts state hospitals in the early 1990s.
The study population consisted of 554 male and 472 female outpatients served by the Massachusetts Department of Mental Health who were residents of community-based housing. Primarily Caucasian (87 percent), they ranged in age from 20 to 80 years, with a mean±SD age of 41± 11.01 years. All had a primary diagnosis of serious mental illness or personality disorder. Many were unemployed, and 90 percent were uninsured. Subjects resided in residential services operated by or under contract with the department of mental health. The residential facilities ranged in type from group homes to supported independent apartments.
Four types of assaults were examined in this study. Physical assaults were defined as unwanted contact with another person with intent to harm, including punching, kicking, slapping, biting, spitting, and throwing objects directly at staff. Sexual assaults were unwanted sexual contacts, including rape, attempted rape, fondling, forced kissing, and exposing. Nonverbal intimidation consisted of actions to threaten and frighten staff, such as pounding on the staff office door, randomly throwing objects, and destroying property. Verbal threats were statements meant to frighten or threaten staff; such statements included threats against life and property as well as racial slurs and other derogatory comments.
Data on assaults of staff by patients were gathered for two different time intervals: January 1991 to December 1992, and January 1994 to June 1998. For the two years between January 1991 and December 1992, assault data were obtained retrospectively from incident reports, chart reviews, and discussions with assaulted staff by the house managers of two agencies that operated 26 community residences. Because staff in community-based programs may not report all assaults (9), assaultive events for this period may have been underreported.
For the years between 1994 and 1998, assault data were gathered prospectively after the assaulted staff action program (ASAP) (2) was established in five agencies with 70 community-based residential programs. ASAP is a crisis intervention program to help staff victims cope with the psychological sequelae of assaults by patients. ASAP team members were residential staff who volunteered to be trained in standard crisis intervention procedures and methods for reporting incidents so that underreporting would be minimized (9). The charge nurse on each unit was mandated to report each incident to the ASAP team leader, complete a formal department of mental health incident report, and orally present an account of each incident at the daily staff meeting. ASAP teams practiced these reporting procedures until acceptable levels of agreement were obtained.
This paper reports data on assaults by individuals discharged from state hospitals during a 12-month period after discharge (1991–1992) and during a later 42-month period (1994–1998). Patient identifiers were not available for this study, and it was not possible to determine the extent to which the two samples overlapped. The frequencies of assaults during the two periods are based on episodes per person-year to adjust for differences in the two samples with regard to time at risk for assaults. The analysis assumes all individuals were at risk for the full period, although some brief hospitalizations during which the patient was away from the residential facility may have occurred.
Eighty assaults on staff were reported for 1991–1992, and 39 were reported for 1994–1998. No differences in type or frequency of assault were found by type of residential site, and no staff victim refused to participate in data collection for this study. Data on the age of staff members were not available.
Characteristics of assailants
In 1991–1992, in a population of 242 male patients and 175 female patients with an average age of 39±11.03 years, 57 male patients, or 71.2 percent, and 23 female patients, or 29.8 percent, assaulted staff. The patients committed a total of 80 assaults, including 30 physical assaults (37.5 percent of the assaults), seven sexual assaults (9.7 percent), one nonverbal assault (1.3 percent), and 42 verbal assaults (52.5 percent). The average age of the assaultive patients was 37±14.85 years. Schizophrenia was the most common diagnosis among the assaultive patients; 73 percent of the assaultive patients had this diagnosis.
In 1994–1998, in a population of 312 male patients and 297 female patients with an average of 43±11 years, 20 male patients, or 51 percent, and 19 female patients, or 49 percent, committed assaults. They committed a total of 39 assaults. Twenty-two assaults, or 57.1 percent, were physical assaults; four, or 10.2 percent, were sexual assaults; four, or 10.2 percent, were nonverbal assaults; and nine, or 23.1 percent, were verbal assaults. The average age of the assaultive patients was 37±7.3 years. The most common diagnosis was schizophrenia; 54 percent of the assaultive patients had this diagnosis.
Characteristics of staff victims
In 1991–1992, in a population of 136 male and 231 female residential staff, 32 male staff members (40 percent) and 48 female staff members (60 percent) were victims. Eighteen staff members, or 17.9 percent, had swollen bruises; 37, or 42 percent, had some degree of acute stress disorder; and 29, or 44 percent, had symptoms of hypervigilance, sleep disturbance, and intrusive memories.
In 1994–1998, in a population of 660 male and 1,073 female residential staff, 39 staff members were victims of assaults by patients. Fourteen male staff members, or 44.3 percent, and 25 female staff members, or 65.7 percent, were victims. Nine staff members, or 23.1 percent, had swollen bruises; 27, or 69.2 percent, had some degree of acute stress disorder; and 19, or 48.7 percent, had symptoms of hypervigilance, sleep disturbance, and instrusive memories.
Front-line staff were the targets of assault in all 80 incidents of assault reported in 1991–1992. They were the target in 26 of the incidents reported in 1994–1998, or 66.6 percent.
Patients' overall rate of assaultive incidents per person-year declined by 61 percent between the two time periods, from 2,131.5 to 834. In 1991–1992, the rate of assaults per person-year for male patients was 1.8 times higher than for female patients (.118 versus .066). Between 1991–1992 and 1994–1998, the rate per person-year for male patients declined, and in 1994–1998 the rate per person-year for male and female patients was equal (.009 for both male and female patients).
This study's results are consistent with previous findings about assaults by clients on staff members in community settings (7,8). The characteristics of the assaultive patients were partly consistent with earlier findings for inpatient settings (1,2,3,4,5,6). Many of the assaultive patients in this study had a diagnosis of schizophrenia, as did assaultive patients in previous studies (1,2,3,4). In the second time period in this study, male and female patients were equally likely to be assaultive, as has been reported in previous studies (5,6). However, unlike some recent findings of younger age and a high frequency of personality disorder among assaultive patients (5,6), the assaultive patients in this study were on average older, and most did not have a personality disorder diagnosis.
The findings on staff victims are consistent with previously reported research (1,2,3,4,5,6). Staff victims most at risk were direct care workers with less formal training and experience. Few instances of sexual assault were reported in this study; however, the findings on the frequency of sexual assaults differ from those of earlier studies (1,2,3,4,5,6). Although data on staff members' ages were not available, it is our impression that the higher frequency of sexual assaults may be related to younger female residential staff caring for older male patients.
The data suggest a significant reduction in assaults by patients in residential settings as time from discharge increased. This study did not address possible mechanisms for this decline. However, at least three hypotheses for future inquiry may explain this outcome. More experienced staff may have learned how to better prevent assault, the most chronically violent individuals may have been rehospitalized, or patients may have undergone a socialization process that acclimated them to the community residences.
The study has several methodological limitations, including possible underreporting (9), the possible role of ASAP in the decline in assaults between the two study periods, lack of a standardized assessment protocol, lack of procedures to ensure nonduplicative cases, nonrandom sampling, the culture of facilities, and nonspecific factors.
As state mental health agencies increase efforts to reduce the size of institutional populations, the process that generated the residential population described here will be replicated across the country. Further research appears warranted so that this process is safe and clinically efficacious (10).
The authors are affiliated with the Massachusetts Department of Mental Health and the University of Massachusetts Medical School. Address correspondence to Dr. Flannery at the Department of Psychiatry, Cambridge Hospital, 1493 Cambridge Street, Cambridge, Massachusetts 02139 (e-mail, firstname.lastname@example.org).