Research Reports

FEMALE OFFENDERS WITH AND WITHOUT MAJOR MENTAL HEALTH PROBLEMS: A COMPARATIVE INVESTIGATION

R-46
Kelley Blanchette and Laurence L. Motiuk
Research Division
Correctional Service Canada
May, 1996

Table of Contents

Acknowledgments

Introduction

Method

Participants
Instrumentation
Procedure

Results

Mental Health
Case Characteristics
Criminal Background
Prison Incident(s)
Temporary Absence(s)
Prison Release(s)
Post-release Outcome(s)
Other Correlates and Re-admission

Discussion

References

ACKNOWLEDGEMENTS

We would like to extend our sincere thanks to Patricia Sadoway who had assisted us in gathering information at the Prison for Women for this investigation. Ray Belcourt of the Research Division also deserves credit for the valuable information he provided. In addition, we wish to express our appreciation to the Federally Sentenced Women who participated in this project. Without the aforementioned assistance and co-operation, this research could never have been completed.

Introduction

There is currently little research on female offender assessment with virtually none focused on the issues of criminal history, mental health and recidivism. However, the number of North American women charged criminally is increasing (Boe, 1992), illustrating the importance of obtaining more information about the criminal behaviour patterns of this group.

Unfortunately, most research attempting to isolate factors associated with criminality have used male samples, and results tend to be generalized to the female offender population. This occurs despite evidence that there are gender differences in the predictors of criminal recidivism.

For example, female offenders are much less likely than male offenders to have a previous criminal history (although they are more likely than male offenders to have a partner with a criminal history). Female offenders are also more likely to have been physically or sexually abused, are more likely to offend against someone they know, and tend to commit less violent crimes than men (Bureau of Justice Statistics, March 1994). Finally, although little is known about female recidivism, there is evidence that instruments that reliably predict recidivism in male offender samples are not as successful with female offenders (Bonta, Pang, & Wallace-Capretta, 1995; Hann & Harman, 1989).

A partial explanation for these differences is that for every six Canadian men suspected of and prosecuted for a criminal offence only one woman is subject to the same. This ratio increases to 9:1 for those incarcerated for their offence(s) (United Nations Survey of Crime Trends, 1994). Women also tend to receive shorter sentences than men. The great majority of women sentenced to provincial terms of incarceration serve six months or less, and almost 40% of these women serve 14 days or less (Shaw, 1994).

There are two plausible explanations for these findings. The first is that women are committing relatively fewer serious offences, and are therefore less likely to be sentenced to incarceration. Alternatively, women may be filtered out of the criminal justice system at all stages of the process. A statistical review of Canadian female offenders (Hatch & Faith, 1989) indicates that women probably commit about 10% of all serious (violent) crime in Canada. Therefore, despite the fact that women commit fewer violent crimes, there is still a discernible under-representation of female offenders in custody.

In contrast, two-thirds of those who seek psychiatric help are women (Michell, 1988). This over-representation cannot be attributed solely to the notion that women are more likely to seek help because results from several household surveys have shown that women are, in fact, more likely to be diagnosed with a psychiatric disorder (D’Arcy, 1982; Eaton et al., 1989; Leaf, Weissman, Myers, Tischler, & Holzer, 1984). Accordingly, Statistics Canada (1988 cited in Burstow, 1992) reported that about two-thirds of Canadian women have mental health problems compared with just one-third of men.

Further, Rogers and Bagby (1992) have determined that four times as many women were involuntarily hospitalized for mental disorders as men (38% versus 10%). The authors suggest that the propensity to involuntarily hospitalize women is based on paternalistic goals of "protect" and / or forcing them to receive treatment (p. 414).

It is arguable that behaviour defined as "criminal" or as "mentally disordered" are manifestations of the same phenomenon: the practice of labeling and attempting to fix behaviour that is socially undesirable. For example, a common social definition of criminal behaviour refers to actions that violate customary and traditional norms (Andrews & Bonta, 1994, p. 23). Others have argued that this same terminology can be used to define mentally disordered behaviour as it serves to preserve the status quo and to impose patriarchy (Burstow, 1992; Caplan, 1992; Larkin & Caplan, 1992). Whether an act is labeled as "criminal" or "mentally disordered," it is a part of a broader category of behaviour that society sees as deviant. As such, the mental health and criminal justice systems are deeply intertwined.

It is well-documented that those with mental disorders are over-represented in prison populations (Abram, 1990; Hodgins & Côté, 1993a; Neighbors et al., 1987). It has also been demonstrated that, within the penal system, the prevalence of mental disorder varies with institutional security level. Offenders in treatment centres have more mental health problems than offenders in security units (Motiuk & Porporino, 1991), while inmates in maximum-security facilities tend to have higher disorder rates than inmates in medium-security facilities who tend to have higher disorder rates than inmates in minimum-security facilities (Neighbors et al., 1987). Although the over-representation of those with mental disorders in the penal system has been clearly established, the reasons for this are ambiguous. Evidence does suggest that offenders with manifest psychiatric symptoms are more vulnerable to arrest (Teplin, 1984) and spend disproportionately more time on remand and awaiting sentencing (Gignell, 1990; cited in Porporino & Motiuk, 1995). Further, offenders with mental disorders are perceived more negatively by correctional officers (Kropp, Cox, Roesch, & Eaves, 1989), are sentenced to longer periods of incarceration (and serve longer proportions of their sentences), and are held in higher security institutions (Porporino & Motiuk, 1995). Finally, offenders with major mental (psychotic) disorders are less likely to be granted temporary absences and discretionary release, and are more likely to have their conditional release revoked for minor technical violations and are less likely than those without major disorders to commit new offences (Porporino & Motiuk, 1995).

It is important to note this was the case despite evidence that the types of offences committed and institutional behaviour did not differ significantly between offenders with and without psychiatric problems. This supports the suggestion that people with mental disorders are criminalized (Teplin, 1984; 1985). As a result, many individuals with mental health problems are trapped in the criminal justice system, serving intermittent terms of probation and imprisonment for petty crimes.

Research comparing the institutional behaviour of offenders with mental health problems to those without such problems has yielded inconsistent results. Some studies have documented higher rates of rule infractions and disruptive episodes among offenders with mental health problems (Toch & Adams, 1989; Hodgins & Côté, 1991), while others have postulated that there are no differences in institutional adjustment and those without disorders are actually more likely to re-offend (Porporino & Motiuk, 1995).

One reason for this disparity comes from the various ways in which investigators classify disordered offender groups. Typically, the presence of mental disorder is defined by contact with mental health services prior to or during the individual’s term of imprisonment. But, this definition might inflate this group’s rate of misbehaviour as individuals who have been referred for psychiatric intervention represent a subgroup of offenders with disorders who might be particularly prone to disruptive behaviour (Porporino & Motiuk, 1995). Further, a mental disorder diagnosis might influence correctional staff to be particularly vigilant for inappropriate behaviour.

A less biased comparison would involve the application of an objective diagnostic tool to a large group of offenders to classify them into disordered and non-disordered groups. A between-groups comparison could then determine whether or not offenders with mental disorder respond differently to incarceration. This would allow for an unbiased estimate of whether or not the correctional system responds in a biased manner to offenders with mental disorders.

The granting of temporary absences and / or discretionary release (such as parole) represents the correctional system’s response to perceived risk. More specifically, federal offenders considered to be low-risk are more likely to be granted temporary absences and are more likely to be released on parole than statutory release. Full parole is generally granted after having served one-third of a sentence, while statutory release only occurs after having served two-thirds of the sentence.

There is some indication that offenders with major mental disorders are less likely to be granted temporary absences or discretionary release. Yet, it has not been conclusively determined that such offenders are at higher risk of institutional infractions or of re-offending post-release. While diagnosis of a major mental disorder is a questionable risk predictor, other factors such as criminal history and antisocial personality have demonstrated consistent and robust predictive accuracy (Gendreau, Little, & Goggin, 1995). It has been suggested that these same factors are predictive of risk for offenders with mental disorders (Hodgins & Côté, 1993b).

This present study followed a sample of federally incarcerated female offenders who were incarcerated at Prison for Women in 1989. At that time, Prison for Women was the only federal institution in Canada for female inmates. In Canada, offenders are sentenced to two years or longer fall under federal jurisdiction. This means that all offenders who participated in the study were serving fairly lengthy prison terms.

An objective diagnostic instrument was used to determine the prevalence of a variety of psychiatric diagnoses. Offenders with major mental health problems were then compared to their non-disordered counterparts as to case history, criminal history, institutional adjustment and post-release outcome.

A review of the published research has indicated that the results of post-release outcome studies of offenders with mental disorders depend, in part, on how the researchers define disorder, and recidivism. While the present research was concerned primarily with whether or not a diagnosis of a major mental disorder could aid in the prediction of post-release outcome, it was expected that manipulation of the operational definitions of mental disorder and recidivism would produce different post-release outcome results.

In light of the fact that so many offenders with mental health problems are currently being held in penal institutions, it is important to understand how criminal offence patterns might be influenced by the presence or absence of particular disorders. It is also important to ensure that the correctional system is responding to these offenders in a manner appropriate with level of risk. This information will aid in the management and rehabilitation of female offenders which, in the long term, helps to protect society through a greater ability to predict risk.

Method

Case Characteristics of Study Participants and Population at Prison for Women

Participants
Instrumentation
Procedure

Participants

The study focused on a sample of 76 federally sentenced adult female offenders who were incarcerated at Prison for Women in Kingston, Ontario in 1989. Each of these women agreed to participate, and each completed an objective mental health survey at some point during the course of their incarceration. These women represent a cross-section (57%) of the population at Prison for Women during at that time. At the time of the mental health survey, the participants ranged in age from 20 to 54 years (with an average age of 32.9). Follow-up data was collected for these women, although recidivism data was unavailable for 10 participants who, as of May 1995, had not yet been released.

A comparison between the study participants and the overall Prison for Women population (Correctional Service Canada, 1989) is provided in Table 1.

Table 1

Variable	Study Sample (N = 76)	Population (N = 152)
Mean Age (years)	32.9 (SD 8.2)	33.6 (SD 8.9)
	% (n/76)	% (n/N)
Race
Caucasian	74.0 (54)a	74.3 (113/152)
Aboriginal	13.7 (10)	9.8 (15)
Black	4.1 (3)	7.2 (11)
Other	8.2 (6)	7.8 (13)
Marital Status
Single	39.2 (29)b	55.2 (84/150)
Married / Common law	37.9 (28)	28.2 (43)
Sep. / Div. / Widowed	23.0 (17)	14.9 (23)
Aggregate Sentence
< 3 years	15.8 (12)	21.6 (33/152)
3 < 6 years	30.2 (23)	30.2 (46)
6 < 10 years	15.8 (12)	15.0 (23)
10 < 20 years	10.5 (8)	8.4 (13)
Life	27.6 (21)	24.3 (37)
# of Previous Federal Terms
None	82.9 (63)	80.2 (144/152)
One	13.2 (10)	13.8 (24)
Two or more	3.9 (3)	5.7 (12)

Note. ^a There were 3 missing cases for variable "race", therefore percentages are based on a sample size of 73. ^b There were 2 missing cases for variable "marital status", therefore percentages are based on a sample size of 74.

Based on the data presented in Table 1, it appears that the study sample is relatively representative of the 1989 Prison for Women population.

Instrumentation

The instrument used to assess mental health was version III-A of the Diagnostic Interview Schedule (DIS), developed by Robins and Helzer (1985). The DIS was designed for research with large samples and provides current, two-week, six-month, and lifetime prevalence figures for various diagnoses. Studies that have tested the DIS have reported that it is a reliable and valid instrument (Helzer et al., 1985; Helzer, Spitznagel, & McEvoy, 1987; Wittchen et al., 1989).

The DIS is a structured interview that can be administered by properly trained lay interviewers within one to two hours, depending on the degree of mental health difficulties evidenced by the respondent. All diagnoses provided by the DIS are made on a lifetime basis first, followed by interview probes into the recency of the last symptom experienced. As such, the interview can distinguish whether the disorder is current, or has been present within the last two weeks, the last month, the last six months or the last year. It also provides the opportunity to either use or ignore diagnostic hierarchies, making it possible to compare different methods for establishing diagnoses.

Data for the follow-up portion of the study was obtained from several different sources: official criminal history records, automated databases, and institutional files which include case management and program participation information.

Procedure

The investigation involved four stages of data collection. The first wave of data are the results of the mental health survey conducted at Prison for Women in 1989. Only diagnoses based on lifetime prevalence estimates, using stringent diagnostic criteria, were used.

A comprehensive coding manual was prepared for the second, third and fourth waves of the study. This involved data collection as to criminal history and case management / file information (such as demographic data and case history), institutional adjustment and in-program performance, and post-release outcome.

Information for the second and third waves of the study (pertaining to criminal and case history, institutional adjustment and in-program performance) was obtained through a systematic Prison for Women file review. All case file reviews were conducted using a structured case-file review protocol.

The fourth and final wave of the study involved data collection data relating to post-release outcome. Objective information on release status, type, and suspension and revocation was obtained through CSC’s automated data bases. Return to custody for any reason, revocation of conditional release for minor technical violations, conviction for a new non-violent offence(s), and convictions for a new violent offence(s) were used as post-release outcome measures.

The offenders were followed from first release after the DIS survey (September, 1989) until May 1, 1995. Of the 76 participants who completed the DIS interview, 10 had not been released from the institution as of that date

Results

Prevalence of DIS-DSM Disorders Using Stringent Criteria

Mental Health
Case Characteristics
Criminal Background
Prison Incident(s)
Temporary Absence(s)
Prison Release(s)
Post-release Outcome(s)
Other Correlates and Re-admission

Mental Health

It is important to note that "any major mental disorder" includes only those participants who met stringent (lifetime) diagnostic criteria for either schizophrenia, schizophreniform disorder or mania. The frequency of mental disorder in this sample is much higher than in the general population, and is even higher than other offender populations. Table 2 outlines the distribution of selected disorders within the sample (with the exclusion criteria applied).

Table 2

Disorder	% (n / 76)
Any major mental disorder	17.1 (13)
Major depression	32.9 (25)
Generalized anxiety disorder	19.7 (15)
Psychosexual dysfunction	34.2 (26)
Antisocial personality	36.8 (28)
Alcohol use / dependence	63.2 (48)
Drug use / dependence	50.0 (38)

As shown in Table 2, over 17% of the sample met the criteria for a major mental disorder. The sample was therefore divided into two groups: a "with" major mental health problem group, made up of the 13 offenders who met the criteria for any major mental disorder, and a "without" group made up of the rest of the sample.

Case Characteristics

To determine whether the groups differed from each other in terms of case characteristics, variance and chi-square statistics were analyzed. Table 3 sets out the results of these analyses.

Table 3

	Without (n = 63)		With (n = 13)
Variable	M	SD	M	SD	F
Age (years): at admission	31.5	8.2	27.2	6.9	3.01
at survey	33.6	8.3	29.8	6.9	2.36
	n	%	n	%	X2
Admitting Offence:
Violent	51	80.9	10	76.9
Non-violent	12	19.1	3	23.1	0.11
Federal Term:
First	53	84.1	10	76.9
Second +	10	15.9	3	23.1	0.39
	M	SD	M	SD	F
Sentence Length (months) a	61.5	38.6	78.9	46.0	1.77
Convictions (#):
Prior Non-violent	10.5	13.7	7.2	8.3	0.73
Prior Violent	0.9	1.5	3.2	5.1	9.00**

Note: ^a Life sentences excluded. ** p < .01

As demonstrated in Table 3, female offenders with a major mental health problem did not differ significantly from those without as to age, nature of admitting offence or previous incarceration in federal custody. Sentence lengths did not, therefore, differ significantly between the two groups. However, there was a small trend for those with a major mental disorder to have received longer sentences, despite collateral evidence showing that they were slightly less likely to have been admitted for a violent offence.

There were significant differences between the groups as to the number of prior violent convictions. The female offenders with a major mental health problem had, on average, more than five times the number of previous violent offences as those without such difficulties. It is, however, important to note that due to the small number of women in the with disorders group (and in the sample as a whole), this finding should be interpreted with caution. Variance analysis compares group averages, so one extreme score in such a small sample could potentially inflate any differences between groups.

Despite this caution, the fact that Table 3 indicates differences between the groups as to past violent convictions is an important consideration when interpreting the remaining results. 

Criminal Background

To examine offender criminal history in more detail, previous offences were broken down by type, with averages computed for each offence type for each group. Again, variance analyses were performed to determine how those with a major mental health problem differed from those without such difficulties as to variables associated with criminal history (see Table 4).

Table 4

Type of Conviction	M	SD	M	SD	F
Robbery	0.32	0.86	1.62	4.37	4.89*
Assault	0.44	0.91	1.00	1.41	3.26
Property	4.10	7.92	2.46	3.76	0.52
Fraud	2.00	5.30	1.31	2.78	0.21
Drug Offences	0.67	1.85	0.62	1.04	0.01
Public Morals	1.56	5.17	0.85	1.46	0.24

Note: * p < .05

Examination of the data in Table 4 reveals that the discrepancy between groups as to number of previous violent offences is due to the higher frequency of robbery offences among female offenders with a major mental health problem. Female offenders with a major mental health problem also tended to have more previous convictions for assault, although this finding was not statistically significant.

A final set of analyses involved comparing the groups as to previous court dispositions. Table 5 sets out the total volume of court dispositions recorded for both groups.

Table 5

Type of Disposition	M	SD	M	SD	F
Absolute / Conditional Discharges	0.11	0.32	0.08	0.28	0.13
Suspended Sentences	0.84	1.45	1.00	1.35	0.71
Fines	1.33	1.95	1.84	3.02	0.61
Probation Terms	1.35	1.90	1.62	1.85	0.21
Incarcerations	0.30	0.59	0.38	0.65	0.21

There were no significant differences between the groups as to the numbers or types of past court dispositions. This is particularly noteworthy when considering the fact that female offenders with a major mental health problem in this study had committed more violent criminal offences.

Prison Incident(s)

Involvement in institutional misconduct was assessed by classifying all recorded charges during the period under review into five categories: any (includes all), violent (assault/ sexual assault/hostage taking), escape, contraband (drugs/ alcohol/ weapons), and behavioural. Institutional adjustment data was not available for six offenders. Table 6 sets out the average number of each of these types of incidents for both the with and without major mental disorder groups.

Table 6 

			Without		With
Variable			n	%	n	%	X2
Incident		Yes	33	56.9	6	46.1
		No	25	43.1	7	53.9	0.50
Violent		Yes	16	28.1	4	30.8
		No	41	71.9	9	69.2	0.04
Escape /		Yes	2	3.5	1	7.7
Attempt		No	55	96.5	12	92.3	0.45
Contraband		Yes	8	14.0	3	23.1
		No	49	86.0	10	76.9	0.65
Behavioural		Yes	9	15.8	4	30.8
		No	48	84.2	9	69.2	1.57
			M	SD	M	SD	F
Number of Incidents			4.52	9.99	5.77	11.12	0.16

Anova and chi-square analyses revealed no significant differences between the groups as to the frequency or type of institutional charges incurred. However, Table 6 does indicate that, on average, female offenders with major mental health problems tend to have more institutional incidents, which appears to be accounted for by an increase in behavioural infractions.

Temporary Absence(s)

Temporary absences are granted to offenders for a number of reasons: medical, rehabilitative, administrative (such as a court appearance), compassionate, and family / community contact. While escorted temporary absences (ETAs) are normally granted for essential medical or administrative reasons, unescorted temporary absences (UTAs) are routinely granted as a stepping stone to release. UTAs provide an opportunity for the offender to gradually reintegrate into the community and offer a means for the correctional system to assess the offender’s ability to operate successfully outside the institution. Offenders perceived as "high-risk" are generally not granted temporary absences, except under escort and for absolutely essential reasons. As such, the granting of temporary absences reflects the correctional system’s response to perceived risk.

Table 7 outlines the temporary absence history of the offenders with and without major mental disorders. Although this data was unavailable for some participants, analyses revealed no differences in the proportions of offenders in the two groups who were granted various forms of temporary absences. It is, however, surprising to note, that there was some evidence that those without major mental health problems received more ETAs, while those with a major mental health problems received more UTAs. However, these differences were not statistically significant.

Table 7

			Without		With
Variable			n	%	n	%	X2
Temporary		Yes	46	90.2	8	80.0
Absences		No	5	9.8	2	20.0	0.86
ETAs		Yes	48	90.6	8	80.0
		No	5	9.4	2	20.0	0.95
UTAs		Yes	27	50.0	6	60.0
		No	27	50.0	4	50.0	0.34
Medical		Yes	37	71.2	6	66.7
		No	15	28.8	3	33.3	0.07
Administrative		Yes	12	23.1	2	22.2
		No	40	76.9	7	77.8	0.00
Community		Yes	3	5.8	0	0.0
Service		No	49	94.2	9	100.0	0.55
Family Contact		Yes	23	44.2	2	22.2
		No	29	55.8	7	77.8	1.54
Compassionate		Yes	10	19.2	1	11.1
		No	42	80.8	8	88.9	0.34
Rehabilitative		Yes	21	40.4	4	44.4
		No	31	59.6	5	55.6	0.05
			M	SD	M	SD	F
# of Temporary Absences			14.9	15.5	11.0	10.0	0.58
# of ETAs			12.8	13.7	6.7	9.0	1.85
# of UTAs			2.4	3.7	4.3	5.8	1.95

Prison Release(s)

In Canada, conditional release from federal custody occurs either in the form of parole (day or full) or statutory release. Eligibility for full parole is contingent upon having served one-third of your sentence, and is granted at the discretion of the National Parole Board. Offenders not granted parole are typically released on statutory release, after having served two-thirds of their sentence. Parole, like the granting of temporary absences, can be viewed as the correctional system’s response to perceived risk. If an offender is perceived to be high-risk, they are not likely to be granted any form of parole.

Table 8 sets out the patterns of release from federal custody for both groups. Those with a major mental health problem were just as likely to be released as those without such difficulties. Similarly, there were no significant differences between the groups as to release type, time served or proportion of time served.

Table 8

				Without		With
Variable				n	%	n	%	X2
Released			Yes	54	85.7	12	92.3
			No	9	14.3	1	7.7	0.41
Release Type:		Parole		44	81.5	10	91.7
	Statutory Release			11	18.5	1	8.3	0.73
				M	SD	M	SD	F
Time Served (days)				998	832.3	850	733.2	0.32
Proportion of Time Served				.39	.21	.35	.17	0.40

Post-release Outcome(s)

Four outcome measures were used to compare the groups as to success/failure after release: return to custody for any reason, conditional release revocation for a technical violation, conviction for a new offence, and conviction for a new violent offence. The follow-up period varied by offender (as a function of time at risk in the community). Time-at-risk was defined as time between release and first revocation / return to custody (for those who had their release revoked or were convicted of a new offence), or time between release and May 1, 1995 (for those who did not return to custody). Time-at-risk ranged from 4 days to 5.8 years, with an average of 24.3 months.

The analyses presented in Table 9 reveal no significant differences between the two groups as to post-release outcome. However, those with a major mental health problem tend to be more likely to return to custody (for any reason) and to incur more technical violations.

Table 9

		Without		With
Outcome Measure		n	%	n	%	X2
Any return to custody	Yes	29	53.7	10	83.3
	No	25	46.3	2	16.7	3.57
Technical violation	Yes	21	38.9	8	66.7
	No	33	61.1	4	33.3	0.08
New non-violent	Yes	21	38.9	6	50.0
offence	No	33	61.1	6	50.0	0.48
New violent offence	Yes	9	16.7	2	16.7
	No	45	83.3	10	83.3	1.00

Other Correlates and Re-admission

Ten potential predictors were selected and correlated with the broadest measure of post-release outcome to examine which factors might be predictive. The correlational analyses were run separately for participants with and without a major mental disorder. This allows for an examination of whether the same factors are associated with outcome for both groups. Table 10 sets out the results of these analyses.

Table 10

Other Correlates of Re-admission: Without and With Major Mental Health Problems

	Without (n = 54)	With (n = 12)
Measure	r	r
Age at admission	-.40**	.27
Age at first non-violent offence	-.43**	-.02
Age at first violent offence	-.39**	-.12
# past non-violent convictions	.67****	.63*
# past violent convictions	.26	.53
Major Depression	.26	.20
Generalized Anxiety Disorder	-.18	.20
Psychosexual Dysfunction	-.01	-.38
Antisocial Personality Disorder	.42**	.13
Alcohol Use / Dependence	.29*	.40
Drug Use / Dependence	.37**	.26

Note: * p < .05; ** p< .01; *** p < .001; **** p < .0001.

Table 10 reveals that the best recidivism predictor (as defined by return to custody) is the number of past non-violent convictions. This is not surprising as there is a large body of research that substantiates this finding. It is noteworthy that the variables related to age are highly associated with post-release outcome for female offenders without major mental health problems, but have virtually no relationship with post-release outcome for those with such difficulties.

It is also interesting to note that the presence of other particular disorders (such as antisocial personality, alcohol or drug abuse / dependence) is significantly associated with post-release failure for female offenders without major mental health problems, but not for those with such difficulties. There was, however, a trend in the same direction that did not reach statistical significance. This trend was particularly evident when alcohol use and major mental disorder were both present

Discussion

This investigation has suggested that the criminal justice system does not respond differently to female offenders with major mental health problems. Female offenders with major mental disorders (psychoses) were not more likely to serve longer proportions of their sentences, or to be denied temporary absences or discretionary release than those without such a diagnosis. In examining the various ways in which the correctional system responds to real or perceived risk, no significant differences were found between groups with and without a major mental disorder. This was the case despite the fact that the groups were not matched for criminal history and that the offenders with a major mental health problem had significantly more prior violent offences.

A similar investigation with a sample of federal male offenders reported that there was evidence for correctional over-response the management of offenders with a major mental disorder (Porporino & Motiuk, 1995). This data indicated that male offenders with major mental health prroblems were given fewer opportunities for early release on full parole and, when released, were significantly more likely to have their release revoked without a new offence. In addition, those with such difficulties were less likely to gain access to temporary absences prior to release. The fact that the investigators matched those with a major mental disorder to those without such a disorder (as to age and criminal history) adds credence to the notion that the correctional system might have a biased response to disordered individuals.

This study’s findings appear to counter those reported by Porporino and Motiuk (1995). However, the with and without major mental health problem groups were not matched for criminal history or age, and the offenders with a major mental disorder had significantly more prior violent offences. There was also some indication that those with a major mental health problem were more likely to engage in behavioural disruptions while incarcerated. However, despite these facts, there was no evidence of systemic bias in responding to federally sentenced women with psychotic disorders.

Although a number of studies have demonstrated that offenders with mental disorders are treated unjustly at all stages of the criminal justice process, there was no evidence of this occurring in the present sample. The discrepancy between this and previous research might be accounted for by the fact that the participants in the current investigation were female offenders, whereas much, or all, of the previous research has been conducted with male samples.

Perhaps female offenders with major disorders are not perceived as presenting as much of a threat as male offenders with the same mental health problems. The correctional system may not exercise the same degree of caution when dealing with female offenders with major mental disorders that they might with male offenders. More research with female offenders is needed before firm conclusions can be taken. Hopefully, future research will address this study’s limitations by increasing the sample sizes and, if possible, matching participants with a major mental health problem to those without such difficulties on a number of relevant variables such as age, admitting offence and criminal history

References

Abram, K. (1990). The problem of co-occurring disorders among jail detainees: antisocial disorder, alcoholism, drug abuse, and depression. Law and Human Behavior, 14(4), 33-345.

Andrews, D.A., & Bonta, J. (1994). The Psychology of Criminal Conduct. Cincinnati: Anderson Publishing Co..

Boe, R. (1992). CSC Offender Population Forecast for 1993 to 2002. Ottawa: Correctional Service of Canada.

Bonta, J., Pang, B., & Wallace-Capretta, S. (1995). Predictors of recidivism among incarcerated female offenders. The Prison Journal, 75(3), 277-293.

Bureau of Justice Statistics (March, 1994). Special Report: Women in Prison. (Publication No. NCJ - 145321). Washington, DC: U.S. Government Printing Office.

Burstow, B. (1992). Psychiatry. In: Radical Feminist Therapy: Working in the Context of Violence (pp. 21-39).

Caplan, P. (1992). Gender issues in the diagnosis of mental disorder. Women & Therapy, 12(4), 71-83.

Correctional Service of Canada (1989). Population Profile Report. Ottawa: Author.

D’Arcy, C. (1982). Prevalence and correlates of nonpsychotic psychiatric symptoms in the general population. Canadian Journal of Psychiatry, 27(4), 316-324.

Eaton, W., Kramer, M., Anthony, J., Dryman, A., Shapiro, S., & Locke, B. (1989). The incidence of DIS/DSM-III mental disorders: data from the NIMH epidemiologic catchment area program. Acta Psychiatrica Scandinavica, 79(2), 163-178.

Hann, R., & Harman, W. (1989). Release Risk Prediction: A Test of the Nuffield Scoring System for Native and Female Inmates. Ottawa: Ministry of the Solicitor General of Canada.

Hatch, A., & Faith, K. (1989). The female offender in Canada: a statistical profile. Canadian Journal of Women and the Law, 3(2), 432-456.

Helzer, J., Robins, L., McEvoy, L., Spitznagel, E., Stoltzman, R., Farmer, A., & Brockington, I. (1985). A comparison of clinical and Diagnostic Interview Schedule: physician reexamination of lay-interviewed cases in the general population. Archives of General Psychiatry, 42, 657-665.

Helzer, J., Spitznagel, E., & McEvoy, L. (1987). The predictive validity of lay Diagnostic Interview Schedule in the general population: Comparison with physician examiners. Archives of General Psychiatry, 44, 1069-1077.

Hodgins, S., & Côté, G. (1991). The mental health of penitentiary inmates in isolation. Canadian Journal of Criminology, 172-182.

Hodgins, S., & Côté, G. (1993). The criminality of mentally disordered offenders. Criminal Justice and Behavior, 20(2), 115-129.

Kropp, R., Cox, D., Roesch, R., & Eaves, D. (1989). The perceptions of correctional officers toward mentally disordered offenders. International Journal of Law and Psychiatry, 12, 181-188.

Larkin, J., & Caplan, P. (1992). The gatekeeping process of the DSM. Canadian Journal of Community Mental Health, 11(1), 17-28.

Leaf, P., Weissman, M., Myers, J., Tischler, G., & Holzer, C. (1984). Social factors related to psychiatric disorder: the Yale Epidemiologic Catchment Area Study. Social Psychiatry, 19, 53-61.

Michell, G. (1988). The reproduction of narcissism. Women & Therapy, 7, 35-52.

Motiuk, L.L. (1991). Antecedents and Consequences of Prison Adjustment: A Systematic Assessment and Reassessment Approach. Unpublished doctoral dissertation, Carleton University, Ottawa, Ontario, Canada.

Motiuk, L.L., & Porporino, F.J. (1991). The prevalence, nature, and severity of mental health problems among federal male inmates in Canadian penitentiaries (Report No. R-24). Ottawa: Correctional Service of Canada.

Neighbors, H., Williams, D., Gunnings, T., Lipscomb, W., Broman, C., & Lepkowski, J. (1987). Prevalence of Mental Disorder in Michigan Prisons. Final Report submitted to the Michigan Department of Corrections. University of Michigan and Michigan State University.

Porporino, F.J., & Motiuk, L.L. (1995). The prison careers of offenders with mental disorders. International Journal of Law and Psychiatry, 18(1), 1-16.

Robins, L., & Helzer, J. (1985). Diagnostic Interview Schedule (DIS) Version III-A. Department of Psychiatry, Washington University School of Medicine.

Rogers, R., & Bagby, M. (1992). Diversion of mentally disordered offenders: a legitimate role for clinicians? Behavioral Sciences and the Law, 10, 407-418.

Shaw, M. (1994). Women in prison: a literature review. Forum on Corrections Research, 6(1), 13-18.

Teplin, L. (1984). Criminalizing mental disorder: the comparative arrest rate of the mentally ill. American Psychologist, 39(7), 794-803.

Teplin, L. (1985). The criminality of the mentally ill: a dangerous misconception. American Journal of Psychiatry, 142(5), 593-599.

Toch, H., & Adams, K. (1989). The disturbed violent offender. London: Yale University Press.

United Nations (1994). Crime Trends and Criminal Justice Operations at the Regional and Interregional Levels. Results of the 3^rd U.N. survey of Crime Trends, Operations of Criminal Justice Systems and Crime Prevention Strategies.

Wittchen, H., Burke, J., Semler, G., Pfister, H., Von Cranach, M., & Zaudig, M. (1989). Recall and dating of psychiatric symptoms: test-retest reliability of time-related symptom questions in a standard psychiatric interview. Archives of General Psychiatry, 46, 437-443.