Keywords

Relative to men, women are less likely to commit crime; statistics supporting this statement are robust and internationally applicable and women represent, on average, about 5% of those incarcerated (Blanchette & Brown, 2006). The gender gap is largest when comparing males to females on data pertaining to violent crimes (Federal Bureau of Investigations, 2011; Statistics Canada, n.d.). Researchers have reported that these gender differences hold true, regardless of whether the evidence is gleaned from official statistics, self-report surveys, or victimization studies (e.g., Blanchette & Brown, 2006).

Women also reoffend at much lower rates than their male counterparts (Florida Dept. of Corrections, 2017; National Resource Center on Justice Involved Women, 2016). This is commensurate with their lower assessed risk relative to men. Although women differ from men in their criminal offending behaviour (e.g., criminal history, offence types, relationship to victims), many of the same factors reliably predict offending and reoffending for males and females (Andrews et al., 2012; Andrews & Bonta, 2010). This will be discussed in more detail later in this chapter. It merits mention here, however, that there is some empirical evidence for ‘gender specific’ risk factors as well (e.g., Blanchette & Brown, 2006). For example, while mainstream research suggests that mental health problems are not predictive of criminal offending generally (Andrews & Bonta, 2010; Bonta, Blais, & Wilson, 2013), a few studies have indicated that some mental health diagnoses are associated with criminality for women in particular (e.g., Salisbury, Van Voorhis, & Spiropoulos, 2009).

5.1 Frequency and Prevalence of Mental Disorder

Establishing prevalence rates for mental health problems is a complicated task. Issues such as the scope and definition of what is meant by a ‘mental health problem’, a ‘mental disorder’, or a ‘mental illness’ are further complicated by determinations and parameters regarding seriousness (e.g., serious or severe mental illnesses), means of assessment (e.g., self-report, endorsement of symptoms, psychiatric diagnosis), type of service provision and patterns of help seeking (e.g., hospital admissions) and sample selection (e.g., gender, age, race, culture and socioeconomic considerations). Generally speaking, mental disorders or illnesses are characterized by any combination of clinically significant disturbances in thought, emotions and behaviour that reflect a dysfunction in the psychological, biological or developmental processes underlying mental functioning (American Psychiatric Association, 2013).

Gender differences in the lifetime prevalence of mental disorders have long been recognized worldwide (Turcotte, 2011; World Health Organization, WHO, 2006). Gender is also associated with differences in susceptibility, expression, comorbidity and course of illness, diagnosis, treatment and adjustment to mental disorder (Muenzenmaier et al., 2015; WHO, 2002, 2006). Women with serious mental illness experience elevated rates of victimization, trauma, poverty, and homelessness (Padgett, Hawkins, Abrams, & Davis, 2006; WHO, 2006), and the context of women’s traditionally disadvantaged social status and vulnerability figures prominently in the feminist analysis of etiology and rates of psychopathology, which also includes analyses of the intersectionality of diverse backgrounds with respect to race, ethnicity, culture, disability and sexual orientation.

It is well established that both men and women with mental disorders are over-represented in criminal justice systems internationally (Brink, 2005; Bronson & Berzofsky, 2017; Collier & Friedman, 2016; Fazel & Danesh, 2002; Fazel, Hayes, Bartellas, Clerici, & Trestman, 2016; Fazel & Seewald, 2012; Prins, 2014). Regardless of methodological complications including sample selection, statistical models, assessment measures, patterns of incarceration and other factors influencing variations in prevalence rates (Fazel et al., 2016), the rates of documented mental health problems and the prevalence of mental disorders is significantly higher than general population comparisons (Brink, 2005; Fazel & Danesh, 2002; Fazel et al., 2016; Prins, 2014; Steadman, Osher, Robbins, Case, & Samuels, 2009). This global reality is often attributed to the inadequacy and/or decline of appropriate mental health resources in the community over time (Chaimowitz, 2012; Munetz, Grande, & Chambers, 2001).

Insofar as most studies of mental illness within prisons are cross-sectional and thus only collect data at one point in time, it is difficult to assess the degree to which mentally ill individuals are more likely to end up in prisons or whether imprisonment leads to more mental health issues. Research does indicate an increased risk for offenders with mental illness having multiple incarcerations (Baillargeon, Binswanger, Penn, Williams, & Murray, 2009). Confinement is a stressful event in itself. Incarcerated individuals experience stress in reaction to the transition from the outside world to prison life, as evidenced by an increase in blood pressure, anxiety, and depression (Islam-Zwart, Vik, & Rawlins, 2007). Within corrections, offenders with mental illness present challenges to correctional management while incarcerated, for example they average more disciplinary infractions per year than offenders without mental illness (O’keefe & Schnell, 2007). Thus, in treating offenders with mental illness, it is important to integrate the principles of effective corrections (e.g., Risk, Need, and Responsivity (RNR model, discussed later) with the principles of effective mental health treatment.

Women offenders report poorer mental health status than women in the general population (Tye & Mullen, 2006) and poorer mental health than incarcerated men (Marcus-Mendoza, 2010; Steadman et al., 2009; Warren et al., 2002). For example, in a seminal prevalence study in the United States, while approximately 12% of women in the general population had symptoms of a mental disorder, it was 60–75% among women prisoners (James & Glaze, 2006). Mental disorders are reported as extremely common among women offenders, with research studies estimating between 30% and 84% of incarcerated women suffering from mental health disorders (Drapalski, Youman, Stuewig, & Tangney, 2009; Steadman et al., 2009; Tye & Mullen, 2006). In England and Wales, one study suggested that 90% of women in prison have at least one of neurosis, psychosis, or personality disorder, alcohol abuse or drug dependence (Palmer in Møller, Stöver, Jürgens, Gatherer, & Nikogosian, 2007). In a systematic review of 62 surveys of prisoners in 12 countries, Fazel and Danesh (2002) reported that of 4260 women, 4% had psychotic illnesses, 12% major depression and 42% had at least one personality disorder (25% had Borderline Personality Disorder and 21% had Antisocial Personality Disorder).

Mental health problems reported by women offenders include, but are not limited to, depression (Fazel & Danesh, 2002; Fazel et al., 2016; James & Glaze, 2006; Steadman et al., 2009), anxiety (Kubiak, Beeble, & Bybee, 2009; Steadman et al., 2009), suicidal thinking and/or self-injurious behaviour (Charles, Abram, McClelland, & Teplin, 2003), Borderline Personality Disorder (Drapalski et al., 2009; Fazel & Danesh, 2002), intellectual disabilities (Lindsay et al., 2004) and Post Traumatic Stress Disorder (PTSD; Kubiak et al., 2009; Lynch et al., 2014; Steadman et al., 2009). Furthermore, comorbidity of mental disorders and substance abuse is especially prevalent among women offenders (James & Glaze, 2006; Nowotny, Belknap, Lynch, & Dehart, 2014; Saxena, Messina, & Grella, 2014).

5.2 Theoretical Models Relevant to Service Delivery

From “nothing works” (Martinson, 1974) to “what works” (Andrews, Bonta, & Hoge, 1990; Andrews, Zinger, et al., 1990) to “but does it work for women” (Blanchette & Brown, 2006) is one way to describe the evolution of thought that has driven theoretical models relevant to service delivery for female offenders. In their award-winning book, Blanchette and Brown (2006), outline the existing theoretical paradigms that represent the integrated perspectives crossing multiple disciplinary boundaries. They frame this review by gender-neutral, female-centred, and hybrid theories and ultimately conclude the following:

  1. 1.

    Women are no longer ‘theoretical afterthoughts’;

  2. 2.

    Work still needs to be done to adequately explain the base rate differential in offending between men and women;

  3. 3.

    Our capacity to explain female criminal conduct is enhanced when considering gender-informed as opposed to gender-neutral theories;

  4. 4.

    Female-centred theories have not been studied to the same degree as those that are gender-neutral;

  5. 5.

    There is variability in the extent to which theory has been translated into practice; and

  6. 6.

    Seemingly divergent theoretical perspectives which are often highly debated in the literature are in fact complementary to one another.

For an in-depth history of theory driving service delivery for female offenders, it is recommended that the work of Blanchette and Brown be examined. For the purposes of this chapter, the focus will remain on the Risk-Need-Responsivity Model (RNR —emanating from the Personal, Interpersonal and Community-Reinforcement Theory (PIC-R), Andrews, 1982; Andrews & Bonta, 2010) and other prominent female-centred theories including Relational Theory (Miller, 1986), and Feminist Pathways perspectives (e.g., Belknap, 2007; Belknap & Holsinger, 1998; Daly, 1992). Furthermore, some emphasis will be placed on strengths-based perspectives and their applicability with female offenders.

The Risk, Need, and Responsivity Principles (Andrews & Bonta, 2010; Andrews, Bonta, & Hoge, 1990; Andrews, Zinger, et al., 1990) are an output of the PIC-R theory and play a prominent role in treatment efforts in Canadian, American, and European jurisdictions, among others. In brief, the risk principle states that those offenders exhibiting the highest levels of risk/the highest likelihood of reoffending should receive the most intensive levels of intervention. The need principle states that treatment should target those dynamic needs that have been empirically assessed, and are linked to, reductions in criminal recidivism. Finally, the responsivity principle places emphasis on how the intervention should be delivered (e.g., positive reinforcement, prosocial modelling, prosocial skills acquisition, extinction, and cognitive restructuring) and more specifically, interventions need to match the learning style, motivation, aptitude, and abilities of the offender in question. It further outlines the importance of structured behavioural interventions in a warm and empathic manner while simultaneously adopting a firm but fair approach (e.g., Gendreau, French, & Gionet, 2004). Importantly, despite vigorous debate about the applicability of these principles for female offenders, there is substantial theoretical evidence to support their use with this group (e.g., Blanchette & Brown, 2006; Dowden & Andrews, 1999).

In considering the need principle, it is important to recognize that researchers have provided evidence to suggest that some criminogenic needs (i.e., dynamic needs that are empirically linked to criminal behavior) emerge as particularly relevant for women. Hollin and Palmer (2006) provide a critique of the literature noting that common criminogenic needs do not imply that the etiology or importance is the same for men and women but maintain that some factors such as experience of physical or sexual abuse are arguably criminogenic needs for women. Personal/emotional factors (e.g., Bell, Trevethan, & Allegri, 2004; Robinson, Porporino, & Beal, 1998), employment (e.g., Greiner, Law, & Brown, 2015), and substance abuse (e.g., Saxena et al., 2014) all have empirical evidence to support this contention. Furthermore, other researchers have raised mental health, parenting, victimization/abuse, and adverse social conditions as female-focused factors that should be considered, and integrated within interventions for female offenders (e.g., Blanchette & Brown, 2006; Derkzen, Booth, McConnell, & Taylor, 2012; Derkzen, Harris, Wardrop, & Thompson, 2017).

As outlined above, in the application of the Risk, Need and Responsivity principles, and in considering the best treatment strategies for application of the responsivity principle in particular, cognitive behavioural therapy (CBT ) and skills acquisition have been emphasized as playing a particularly important role in intervention efforts. For example, in examining treatment programs targeting substance abuse and posttraumatic stress disorder, Zlotnick, Johnson, and Najavits (2009) demonstrated that incarcerated women following CBT driven programs demonstrated improvements on clinician rated PTSD symptoms and continued improvement on psychopathology targets.

Relational Theory (Miller, 1986) argues that healthy human development necessitates that individuals feel connected to one another and that this need is particularly critical in women. Healthy relationships are defined as being empathic, empowering, and mutually influential. This theory has been critical to informing women-centred intervention strategies (as discussed below) but has not focused on explaining female offending behavior. Nevertheless, there is emerging evidence to support relational theory and its impact on recidivism outcomes for women (e.g., Benda, 2005). Related constructs such as social bonds have also been examined in relation to recidivism outcomes providing evidence to suggest that the impact varies by gender (Cobbina, Huebner, & Berg, 2010).

Evidence and literature to date do suggest significant alignment between RNR and relational theory perspectives (Blanchette & Brown, 2006) and there is increasing evidence to support arguments are aligning between gender-responsive and gender-neutral theories sometimes arguing that gender-specific concerns may be best viewed as specific responsivity factors for women (e.g., Rettinger and Andrews, 2010). Furthermore, there is some evidence to support the validity of empowerment as a responsivity factor that assists in developing competencies and enables women to achieve independence (Blanchette & Eldjupovic-Guzina, 1998).

Originating with Daly (1992), Feminist Pathways posits that childhood victimization (e.g., abuse, neglect) plays a central role in girls’ criminal trajectories. The theory maintains that the voices of girls and women are critical to our comprehensive understanding of criminal pathways. The theory contends that victimization is a significant contributor to the eventual use of drugs (and ultimately drug abuse) as a coping mechanism. Furthermore, involvement in selling drugs, prostitution and robbery are mechanisms for street survival after girls and women escape these abusive situations. Ultimately, theorists ascribing to this theory argue that women may be ‘criminalized’ for their survival strategies (Chesney-Lind, 1998) and that such cycles result in emotional distress, low self-esteem, anxiety, depression and aggressive/impulsive behaviours (Zaplin, 2008). Since the original pathways work, other pathway models have been proposed implicating abusive male partners who negatively coerce women into lives of crime (e.g., Belknap & Holsinger, 1998) and a variety of other research supports the relevance, and interest in, pathways perspectives (e.g., Brennan, Breitenbach, Dieterich, Salisbury, & Van Voorhis, 2012; Gannon, Rose, & Ward, 2010; Reisig, Holtfreter, & Morash, 2006; Salisbury & Van Voorhis, 2009; Simpson, Yahner, & Dugan, 2008).

In considering trauma as a contributor to long-term negative outcomes, Messina and Grella (2006) examined childhood trauma and women’s health outcomes in a California prison population. Their data suggested that childhood traumatic events have strong and cumulative negative outcomes on health. More specifically, their results suggested that as exposure to childhood traumatic events increased the likelihood of negative health related outcomes increased. They point to early prevention and intervention, along with appropriate trauma treatment, as being critical within correctional treatment settings. Saxena et al. (2014) also provide evidence to support gender-responsive substance abuse treatment (GRT) and its effectiveness with women who have experienced prior abuse given that GRT maximizes the benefits of the trauma-informed, gender-sensitive intervention. As noted by Covington and Bloom (2006), all of the above supports the proposition that the integration of substance abuse treatment and trauma services is critical in the consideration of treatment elements for female offenders.

Strengths-based approaches, such as those proposed by Van Wormer (2001), suggest that the client’s strengths need to be recognized and integrated into assessments and interventions in corrections. For example, when developing treatment plans, outcome reports, and risk assessments for girls or women, assessors should consider and leverage the offender’s strengths in order to help her heal and re-integrate into the community. Some proponents of strengths-based approaches argue that traditional intervention with incarcerated girls and women is complicated by the oppressive patriarchal structure of the jail/prison system, clients’ victimization histories and the various psychosocial problems frequently presented by female clients (Mahoney & Daniel, 2006). Accordingly, strength-based approaches may be particularly salient in the treatment of female correctional clients.

Despite advancements in our theoretical knowledge, it is still valid to argue that integrating women-specific factors only enhances theory and service delivery for women offenders and despite on-going debate around the application of gender-neutral theory, there is overwhelming evidence to support its relevance. In fact, upon in-depth examination and more collaborative approaches in treatment design (see below), it becomes abundantly apparent that these theories are complimentary and collectively build on our capacity to better support female offenders when considered holistically as opposed to independently or antagonistically.

5.3 Diagnosis and Assessment

5.3.1 Risk Assessment

Offender risk assessment has evolved considerably over the past 30 years, as paradigms have moved from ‘first generation’ assessments to ‘third generation’ assessments; some even reference ‘fourth generation’ assessments (Bonta & Wormith, 2008). In brief, first generation assessment relied on unstructured clinical judgement. Second generation assessment improved predictive accuracy by standardizing consideration of static risk factors that have been empirically linked to re-offending. Third generation tools included offenders needs (dynamic risk factors) that are empirically linked to re-offending. Third generation tools have the advantage of considering changes in risk as a result of interventions (e.g., correctional programs). Finally, fourth generation assessment instruments have been described as those that integrate case planning with risk/needs assessment. Few jurisdictions continue to use the first generation assessments, as those using mathematical/actuarial methods (second through fourth) have demonstrated superiority in terms of predictive accuracy (Grove, Zald, Lebow, Snitz, & Nelson, 2000; Swets, Dawes, & Monahan, 2000).

Many risk assessment instruments have been studied and validated with robust empirical results supporting their use. Unfortunately, with few exceptions, the research is based on samples of male offenders (Blanchette & Brown, 2006) and critics argue that the failure to consider gender and diversity issues in risk assessment results in inequitable practices of classification for women and other minority offender populations. Accordingly, they argue that these biases result in the systemic discrimination of these groups, ranging from over classification to failure to provide appropriate services (Bloom & Covington, 2000; Hannah-Moffat & Shaw, 2001).

While there is no widely used (cross-jurisdictional) risk assessment tool developed specifically for women, some measures, although developed as ‘gender neutral’ tools, show promise in terms of their predictive accuracy for women. Examples include the Level of Service/Case Management Inventory (LS/CMI) and its predecessors (see studies by Andrews et al., 2012; Geraghty & Woodhams, 2015), the HCR-20 (see studies by Coid et al., 2009; Strub, Douglas, & Nicholls, 2016), and the VRAG (see studies by Coid et al., 2009). Notwithstanding these promising results, some still suggest that actuarial tools that are gender-informed and developed from the ground up will bring additional relevance and predictive power to assessments for women (Blanchette & Brown, 2006).

5.3.2 Clinical Diagnosis and Assessment

A significant proportion of incarcerated women has been exposed to trauma and victimization that often began in childhood or adolescence with neglect or physical and sexual abuse and continued into adulthood with intimate partner abuse and sexual assaults (Aday, Dye, & Kaiser, 2014; Clements-Nolle, Wolden, & Bargmann-Losche, 2009; Dehart, Lynch, Belknap, Dass-Brailsford & Green, 2014; Hollin & Palmer, 2006; Kimonis et al., 2010; Nowotny et al., 2014; Messina & Grella, 2006; Warren et al., 2002). When combined with the experience of current or past trauma, and/or substance abuse, mental illness functions to increase a woman’s involvement in criminal activity and thus the likelihood of incarceration (Kubiak, Fedock, Kim, & Bybee, 2017; Lewis, 2006; Lynch, DeHart, Belknap, & Green, 2012). These factors can also create additional problems for a woman offender by exerting an effect upon behaviour, reasoning, memory, social and adaptive functioning, and motivation. Further, these issues may lead to difficulties in adjusting to incarceration and have been found to be related to higher rates of prison misconduct (O’keefe & Schnell, 2007). In spite of this research demonstrating the prevalence of cumulative and complex mental health needs, women offenders generally encounter more barriers to accessing services in the community (Staton, Leukefeld, & Logan, 2001).

The WHO describes mental health as more than the absence of mental illness, considering it a “state of well-being” that allows individuals to realize their own abilities, cope with daily life stresses, and make a contribution to their community. Critical tenets of mental health include perceived feelings of well-being, self-efficacy, autonomy and competence as well as the recognition of one’s ability to realize their intellectual and emotional potential (WHO, 2016). Not surprisingly, the actualization of such a conceptualization poses significant challenges in the context of incarceration in general, and with incarcerated women in particular.

Given the high rates of incarcerated women’s mental health issues and the high comorbidity of these with substance use disorders and histories of trauma, clinical assessment is of critical importance. Within the criminal justice system, assessments essentially fall into two categories: those that address interventions and those that address classification/risk. Blanchette (2002) suggests that factors commonly cited as women-specific criminogenic needs generally fall into the ‘personal/emotional’ domain, and include low self-esteem, histories of trauma and victimization, and self-injury/attempted suicide. These factors also figure prominently into conceptualizations of mental health. Since criminogenic and mental health needs of women offenders are distinct, it is essential that the approach to assessment and treatment be integrated so that needs are addressed in a way that can assist both the women as well as staff tasked with supporting and managing offenders.

For women offenders, the clinical assessment must take into account gender specific issues, including the assessment of mental health problems, substance abuse, histories of trauma and victimization, self-injurious behaviour and suicidality, violence risk, alongside criminogenic need areas. Indeed, failing to do this compromises the effectiveness of any subsequent planned intervention. Further, if these underlying issues are not addressed and stabilized, mental health problems can be exacerbated and women will be subject to more institutional charges, disciplinary infractions and may be further disadvantaged regarding release (Houser & Belenko, 2015).

We recommend that a comprehensive clinical assessment for women offenders include:

  • Encouraging collaboration in the process (e.g., the woman should know the purpose of the screening/assessment).

  • Attending to contextual factors of women’s lives (inclusive of an exploration of trauma, parental responsibilities, disability, poverty and economic marginalization, and intersections of race, culture, ethnicity and sexual orientation) and incorporating a strength-based approach that considers factors associated with mental wellness.

  • Explicitly acknowledging the role of trauma when assigning psychiatric diagnoses; this includes exploring current trauma-related symptoms and functional impairment.

  • Use of standardized clinical instruments; particularly for the determination of psychopathology, cognitive capacity and suicidality. Choosing standardized clinical instruments that have been developed for women (preferred) or adapted and tested on women, and validated for race, if relevant.

  • Incorporating various methods of gathering information: assessment tools, clinical interview, collateral information, retrospective data including previous evaluations, community assessments, etc.

  • Providing a rational framework and formulation for understanding complex needs and strengths associated with mental illness and trauma; highlighting the links between these, criminogenic factors, and a range of emotional/behavioural issues that may otherwise be targeted in isolation.

As an entry point to the clinical assessment process, many correctional jurisdictions (e.g., Canada, United States) start with a mental health screening (Every-Palmer et al., 2014). The purpose of the screening processes is to determine which offenders require further assessment and possible referral to mental health services. The screening protocol should detail the results of the screening, the action taken for positive scores and what (if any) further assessment is required. Mental health screening in a correctional context should include general psychopathology, depression, suicidality, substance abuse and cognitive capacity. Screening should occur early in the correctional system, preferably within weeks of intake (Krespi-Boothby, Mullholland, Cases, Carrington, & Bolger, 2010). Commonly used measures that have demonstrated applicability for women offenders include the eight-item Brief Jail Mental Health Screen (BJMHS; Steadman, Scott, Osher, Agnese, & Robbins, 2005); the Jail Screening Assessment Tool (JSAT; Nicholls, Roesch, Olley, Ogloff, & Hemphill, 2005), which includes the completion of a brief semi-structured mental status interview and a revised version of the Brief Psychiatric Rating Scale (BPRS; Overall & Gorham, 1962); the Kessler Psychological Distress Scale (K6; Kessler et al., 2003); and the Computerized Mental Health Intake Screening System (CoMHISS; Correctional Service of Canada, 2010). Although mental health screening can be indicative of potential mental health problems, it does not result in diagnosis.

A clinical assessment differs from screening in that it is a much more detailed and extensive process for defining the nature of the problem identified, determining a diagnosis, and developing specific treatment recommendations to address the problem. Generally, a clinical assessment delves into the individual’s current experiences and her physical, psychological and sociocultural history. A thorough and comprehensive clinical assessment requires multiple avenues to obtain the necessary clinical information, including self-assessment instruments, clinical records, structured clinical interviews, standardized assessment measures/tools, and collateral information.

Given the prevalence rates of certain mental health issues for women offenders, for a clinical assessment to be viewed as robust it should include standardized measures that consider: (1) psychometric measures to assess clinical syndromes (e.g, Millon Clinical Multiaxial Inventory-III (MCMI-III; Millon, 1997); Minnesota Multiphasic Personality Inventory—Second Revised Edition (MMPI-2; Butcher et al., 2001); Personality Assessment Inventory, (PAI; Morey, 2007); Basic Personality Inventory, (BPI; Jackson, 1996) and personality disorders (e.g., Structured Clinical Interview for DSM-IV Axis II Personality Disorders, (SCID-II; First, Gibon, Spitzer, Williams, & Benjamin, 1997); (2) depression and anxiety measures (e.g., Beck Depression Inventory—II, (BDI-II; Beck, Steer, & Brown, 1996); Beck Anxiety Inventory (BAI; Beck & Steer, 1993); State-Trait Anxiety Inventory (STAI; Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983); (3) suicide risk factors (e.g., Beck Hopelessness Scale (BHS; Beck & Steer, 1988); Depression Hopelessness and Suicide Screening Form (DHS; Mills & Kroner, 2004); and (4) actuarial risk measures (e.g., Level of Service/Case Management Inventory (LSI/CMI (Andrews, Bonta, & Wormith, 2004). As well, the process should include a comprehensive clinical interview that explores trauma, posttraumatic stress disorder, history of substance abuse and interpersonal violence (per the Diagnostic and Statistical Manual of Disorders (DSM-V; American Psychiatric Association, 2013), a diagnosis of PTSD requires a history of exposure to a traumatic event). If cognitive impairments are indicated (either through screening or presentation), further intellectual testing should be conducted. Assessments of intellectual functioning should be obtained using an individually administered, reliable and valid standardized test, such as the Weschler Adult Intelligence Scale (WAIS-III, Wechsler, 1997). Consistent with the WHO perspective on mental wellness, the interview should also focus on determining a woman’s strengths and protective factors; familiarity with these can assist in directing treatment to optimize desired outcomes. For instance, research has demonstrated that factors such as relationships with prosocial community supports, involvement in structured activities, accessing mental health services, and personal motivation may help to promote criminal desistance among high risk, high need mentally-disordered offenders (Stewart, Brine, Wilton, Power, & Hnain, 2015).

Finally, in light of the importance of the Risk, Need and Responsivity model with women offenders (Blanchette, 2000), it is essential that characteristics of treatment responsivity be thoroughly examined (e.g., intelligence; learning style; cultural considerations; treatment readiness and motivation; emotional disorder). Furthermore, mental illnesses frequently cause functional impairments that may seriously impact an individual’s responsivity to interventions targeting criminogenic risk factors. These need to be identified and addressed. For example, a woman with PTSD may not benefit from participating in treatment or programming for substance abuse until the symptoms of PTSD—such as, depression, excessive worrying, lack of motivation, difficulty concentrating, decreased energy, mood swings—are addressed.

Once the comprehensive assessment material is gathered, the essential task of the clinician is to develop a clinical formulation focused toward treatment and intervention that includes an analysis of the extent to which symptoms of mental illness may be relevant to the understanding and prediction of risk. Mental disorders for women offenders, per se, have not been conclusively linked to recidivism statistics. This could be in part a function of low base rates overall, but is more likely similar to the research on men offenders where aggregate data give very little information to clinicians faced with individual risk assessment. For example, according to Pilgrim (2010, p. 282), “a diagnosis such as ‘schizophrenia’ tells us virtually nothing about risk to others. It is only by using multifactorial formulations specific to the offender that we move towards improved risk assessment.”

Toward that end, symptoms of mental illness that may manifest across a variety of the psychiatric diagnoses common to women offenders (i.e., depression, anxiety, personality disorders, PTSD) include: impulsivity, emotional dysregulation, self-injurious behaviours, difficulties with anger and hostility, pessimism, difficulty concentrating, low self-esteem and problems with self-image, which in turn correspond, in essence, to criminogenic need variables. Given personal histories that include multiple marginalizations, trauma, and substance abuse, and frequent co-morbidities, it may be far more useful to investigate the relationship between psychological variables and offending rather than the relationship of specific diagnoses to recidivism per se. By identifying these variables and addressing them in treatment, with particular attention to stabilization and behavioural change, service providers are better able to treat both the underlying issues and address criminogenic factors. This holistic approach can serve the goals of influencing behaviour while incarcerated as well as improve reintegration success for women.

5.4 Interventions: What Works, What Might Work, and What Doesn’t Matter

As referenced above, the “what works” literature is quite comprehensive and there is ample theory to draw from in our efforts to better apply these theoretical constructs within different treatment and service delivery models. Nevertheless, it is important to distinguish the goals of the intervention under examination. An intervention that is successful in addressing symptoms of trauma or mental illness, for example, may not prove effective in efforts to reduce recidivism. In turn, as a starting point, it is important to emphasize that no single intervention is a panacea.

Interventions, and their related goals, may pertain to personality (e.g., addressing antisocial and/or borderline personality disorder), mental health, physical health, cognitive ability, or motivation, to name only a few. As above, the strategy ascribed to may vary in approach and success as a function of the target in question. In turn, there is recognition that although there are common principles that align with all service delivery strategies, there is acceptance and acknowledgement that there is no “one size fits all” assumption while working with female offenders.

Gender-informed programs were first promulgated only about 10–15 years ago; however, meta-analytic studies are now emerging, examining their effectiveness in terms of recidivism reduction. Most recently, Gobeil, Blanchette, and Stewart (2016) were interested in determining whether gender-informed and gender-neutral interventions promote similar treatment effects for women. Despite variability in the treatment targets examined (e.g., substance abuse, self-esteem, anger management), the authors concluded that there is preliminary evidence from high-quality studies that gender-informed programs are more effective at reducing recidivism than gender-neutral approaches. Their results further suggested that interventions focusing primarily on substance use had significantly larger effect sizes than did those focusing in other areas. Equally important, interventions offered in the context of a therapeutic community also demonstrated larger effect sizes. Findings also support the need for interventions that bridge institution and community treatment elements. There is also some research evidence to suggest that girls who follow gendered pathways to crime may be more likely to benefit from the relational approach used in gender-informed programs as compared to girls who did not demonstrate these gendered pathways (Day, Zahn, & Tichavsky, 2014).

In considering the effectiveness of interventions respecting theory on gender-informed programming, the Correctional Service of Canada has provided recent evidence around the effectiveness of Women Offender Correctional Program (WOCP) and Aboriginal Women Offender Correctional Program (AWOCP),Footnote 1 both rooted in culture and gender responsive approaches. These programs are also trauma-informed while recognizing factors more prevalent in female offender such as parenting stress and adverse social conditions (Derkzen et al., 2017). Results from Derkzen et al. (2017) suggest that women successfully complete these programs, with recognition that the level of risk, need, and histories of violence do have a negative impact on completion rates. Furthermore, for those women who complete the programs, more positive discretionary release rates are achieved.

Stewart and Gobeil (2015) conducted a rapid evidence assessment which examined features of programs providing the strongest outcomes for female offenders arguing that three key areas contribute to the strongest outcomes for this population: (1) substance abuse treatment provided in-custody or therapeutic community programs; (2) gender-responsive programs that emphasize strengths and competencies, as well as skills acquisition; and (3) following in-custody treatment with participation in community follow-up sessions (i.e., continuum of care). Ultimately, the authors suggest that these results are critical for guiding program designers and administrators interested in effectively promoting public safety goals and female offender reintegration. The importance of a continuum of care is further echoed by Sacks, McKendrick, and Hamilton (2012) who argue that the ability to sustain, and even improve, behaviour change after a women leaves prison relies heavily on access to community-based continuity of mental health and substance abuse services upon re-entry. Finally, Andrews et al. (2012) used the Level of Service/Case Management Inventory and the Youth version to generate risk need domains that are considered relevant for girls and women ultimately highlighting the exceptional validity of targeting substance abuse for females. Grella and Greenwell (2007) maintain that engaging substance-abusing women offenders in community treatment after parole improves their retention in treatment and reduces the likelihood of recidivism. Ultimately, substance abuse treatment both inside and outside of the institutional environment is demonstrated as critical to successful outcomes (e.g., Kassebaum, 1999).

Bloom and colleagues (e.g., Bloom, Owen, & Covington, 2004) have written extensively on gender-responsive programming, offering guiding principles, policy blueprints, and intervention practices that are critical to program design, interventions, and evaluation. Much of the emerging research supports their efforts and reinforces the need to ensure we continue to meet the unique needs of female offenders in our intervention efforts. It is also critical to ensure that as anticipated by the responsivity principle, we are closely monitoring, and adapting to, the responses of female offenders engaged in intervention efforts.

Earlier we provided a very high level and generic definition of the responsivity principle from the RNR model. In their analysis and overview of the assessment and treatment of female offenders, Blanchette and Brown (2006) proposed a gender-informed responsivity principle, as a tentative reformulation of the original work of Andrews, Bonta, and colleagues. Once again, it is recommended that readers with interests in this area refer to their comprehensive critique of this area; however, in sum, their expressed belief is that the “spirit” of this principle can readily accommodate concepts such as empowerment and mutuality (as advocated for within relational theory) thereby advocating for their inclusion in this reformulation as follows:

A gender-informed responsivity principle states that in general, optimal treatment response will be achieved when treatment providers deliver structured behavioural interventions (grounded in feminist philosophies as well as social learning theory) in an empathic and empowering manner (strengths-based model) while simultaneously adopting a firm but fair approach. (Blanchette & Brown, 2006, p. 126)

In considering “what might work”, it is important to cautiously, yet optimistically, continue to apply the risk principle. To date, research in this area exists with some flaws and there is a need to expand upon the research literature in this regard. Furthermore, there is some evidence to suggest that co-educational programs/interventions are of value; however, this literature is in its relative infancy and requires further validation. Finally, there is strong evidence for the prominence of mental health needs for female offenders; however, in terms of a capacity to contribute to reductions in recidivism, additional work is required before researchers can state with confidence that this falls into the “what works” domain.

In considering “what doesn’t work”, ample research exists to confirm that punishment (e.g. ‘tough on crime’ regimes) fails to influence desired outcomes in a favourable manner. For punishment to work it must be both swift and appropriate to the transgression, the former of which is rarely attainable in criminal justice systems that typically include lengthy court processes. Equally relevant, security responses to mental health needs have not in fact demonstrated to be an appropriate response. Certainly, Canada’s Office of the Correctional Investigator (2015) has highlighted significant concerns around the application of mental health strategies in Canadian Prisons. More specifically, this work highlights the challenges with providing effective therapeutic interventions within prisons, and maximum security environments in particular (e.g., John Service Consulting, 2010). These concerns have certainly been echoed in other international jurisdictions (e.g., Gonzalez & Connell, 2014; House of Commons Committee of Public Accounts, 2017).

Research to date has offered varieties of delivery specific elements relevant to effective treatment service delivery. Optimal treatment outcomes are argued to be most often achieved when community-based, as opposed to institutional based treatment is provided (e.g., Andrews, 2001; Kennedy, 2004); however, more recent evidence suggests that for women, there may be a weaker effect for community based treatment (in isolation), possibly related to dosage or format and of course with recognition that previous research was based primarily on male study samples (Gobeil et al., 2016). There is also some evidence to suggest that women are more successful in single- versus mixed-gender formats (e.g., Ashley, Marsden, & Brady, 2003; Lex, 1995). Therapeutic environment, in general, and characteristics of the therapist, in particular, have emerged as critical to consider in treatment efforts (e.g., Bloom, Owen, & Covington, 2003; Pollack, 1986). Client characteristics such as individual strengths, resiliency and/or protective factors should be integrated into offender rehabilitation strategies (e.g., Andrews, 2001; Andrews & Bonta, 2010; Bloom et al., 2003; Ward & Brown, 2004). Women-centred training is also emerging as an increasingly important consideration in the provision of effective support and service delivery to women offenders (e.g., Nolan, Harris, & Derkzen, 2017). Finally, there is increasing evidence to support the advantages of integrated intervention approaches/models (Blanchette & Brown, 2006). That is, interventions that address multiple needs at the same time, treatment that is capable of addressing both substance abuse and emotional regulation simultaneously, for example (e.g., Correctional Service of Canada’s Integrated Women Offender Correctional Program; WOCP).

Based on the evidence outlined, we would argue, “what works” in treatment and service delivery interventions include:

  1. 1.

    gender-informed interventions, including the integration of trauma-informed care;

  2. 2.

    targeting criminogenic needs;

  3. 3.

    holistic approaches;

  4. 4.

    highly structured, skills-focused and practical interventions;

  5. 5.

    women-centred training for staff;

  6. 6.

    recognizing the unique needs of women and the context in which offending occurs (i.e., trajectories/pathways); and,

  7. 7.

    the maintenance of prosocial family/community ties.

There have been notable advancements in women-centred corrections in Canada (Robeson Barrett, Allenby, & Taylor, 2010), the United States (Bloom et al., 2003), the United Kingdom (Ministry of Justice - NOMS Women and Equalities Group, 2012), and Australia (Salomone, n.d.; Howells, 2000). Almost 30 years ago, Correctional Service of Canada’s Task Force on Federally Sentenced Women (1990) provided recommendations for the improvement of correctional policy and practice for female offenders through basic principles that now guide correctional interventions for women. These principles include empowerment, meaningful and responsible choices, respect and dignity, supportive environments, and shared responsibility. They further spoke to the relevance of holistic programming interventions. As a result of a study conducted by the National Institute of Corrections (NIC) in collaboration with Bloom et al. (2003, 2004), the United States also has guiding principles to promote gender-responsive interventions. These include: gender matters, environment (safety, respect, and dignity), relationships (healthy connections), services and supervision (for substance abuse, trauma and mental health), socio-economic status (education and training), and community (re-entry and collaboration). South Australia fully supports the RNR principles and their application with female offenders. More specifically, Howells (2000) argued that these principles sharpen our thinking on what needs to be done in managing women in prison in a more coherent, effective, and humane way.

With the evolution and establishment of guiding principles around effective gender-specific interventions come advancements in treatment and programming options for female offenders. The Correctional Service of Canada has a long-standing history of programs for female offenders; however, as briefly noted above, their current model is integrated as it targets multiple need areas within one correctional program continuum.

Khilnani (2016) provides brief overviews of some gender-specific interventions such as Seeking Safety, a therapeutic program developed by Lisa Najavits. This intervention is designed to treat both post-traumatic stress disorder and substance abuse. Khilnani also discusses the Systems Training for Emotional Predictability and Problem Solving Program (STEPPS) for incarcerated women struggling with trauma and self-esteem issues related to histories of sexual abuse and unhealthy relationships. This program focuses on behavioural and emotions management regulation strategies and is offered in a psychoeducational group format for those women suffering with borderline personality disorder. Finally, the Ladies Empowerment and Action Program (LEAP) is a Miami-based initiative available to women seeking to improve the likelihood of successful release outcomes. LEAP is designed to empower incarcerated women to make positive life changes and uses a multi-disciplinary approach including entrepreneurship training, education, and mentorship. LEAP partners with a local university to offer business classes to women matching the selection criteria for the program.

5.5 Future Implications

The research evidence is clear: women offenders have different trajectories into the criminal justice system, and gender-informed interventions, including trauma-informed care, maximize the likelihood of successful reintegration. Both gender neutral (e.g., Andrews and colleagues, Andrews, 2001; Andrews & Bonta, 2010); and feminist (e.g., Belknap, 2007) perspectives acknowledge the importance of tailoring interventions to the individual client to capitalize on responsivity to services provided. Despite this, few jurisdictions offer training for front line staff on the provision of gender-responsive services for girls and women. As noted earlier, Correctional Service of Canada offers Women-Centred Training and a recent evaluation of this initiative yielded positive results. This is an important innovation that should be considered in other jurisdictions.

The research on ‘what works’ for women has been instructive with respect to the applicability of the RNR principles and other important considerations for women. Nonetheless, it is still in its infancy (relative to what is known for their male counterparts), and as more primary studies accumulate, prospective meta-analyses will further inform best practices in interventions for women.

Finally, in an era of quickly evolving technological solutions, it is hoped that correctional jurisdictions will be able to capitalize on innovations such as video visitation and telemedicine. These may be particularly beneficial to women, given that they are a small, often geographically dispersed population with unique needs in terms of family and community connections.

5.6 Technology and Innovation

Technology and innovation poses a very interesting challenge in certain correctional and community based environments where the resource constraints are sometimes quite significant; however, we are seeing some success in these areas and should continue to expand upon these positive outcomes. For example, where feasible, some service delivery environments are capitalizing on telehealth/telepsychiatry and the use of electronic medical records systems (see, for example https://www.techcareehr.com). Video visitation for women with children is proving beneficial to the well-being of women, allowing them to maintain a mother-child bond when in-person visitation is not feasible. For example, Correctional Service of Canada’s Mother-Child program now includes a non-residential component entitled ‘ChildLink’—a video visitation program which allows women inmates to communicate with their children in the community using video conferencing technology (e.g., WebEx). This is an important innovation that, in line with relational theory, will help women to manage the stress of incarceration and because there is empirical evidence demonstrating that the maintenance of prosocial family ties assists with women’s adjustment to incarceration (Blanchette, 2005; Jiang & Winfree, 2006).

Finally, through greater inter-agency collaboration and sharing of information, there is growing opportunity to leverage technology in support of reintegration efforts. Specifically, technological advances will assist with inter-agency collaboration so that important information can quickly and easily be shared between service providers as appropriate (e.g., police, courts, corrections, mental health agencies, and other support organizations).

5.7 Conclusions

Women offenders differ from their male counterparts in several important ways. The nature and prevalence of mental disorder varies by gender. The onset or ‘triggers’ of mental illness and offending behaviour seem to vary by gender as well; there is good emerging evidence to show that women’s pathways into the criminal justice system are gendered. Arguably, systemic responses to both mental illness and criminality should also be gender-informed to maximize wellness and desistence from crime. Assessment and treatment services for women should attend to important contextual factors of their lives (e.g., experiences of trauma, parental responsibilities, and economic marginalization) and leverage women’s strengths to promote healing and desistence from crime.

The very large proportion of women offenders with significant mental health needs underscores the importance of integrated mental health treatment and correctional case management. Ideally, multidisciplinary teams (including correctional officers, parole officers, health/mental health care providers) should be specially selected and trained in the fundamentals of mental illness and provision of gender-informed care. Effective clinical case coordination should incorporate some form of a dedicated staffing model that assigns particular staff members on a caseload basis. Case coordination in this manner serves to both acknowledge the importance of relational factors for women while also enhancing staff familiarity with the multidimensional needs of each woman and reinforcing an integrated approach that maintains the woman at the centre. Within an institutional context, correctional operations and/or security staff must work in close collaboration with mental health teams to maintain the safety and security of all, to and to optimize both correctional and mental health outcomes for women. In sum, we emphatically support the holistic approach to intervention for women, particularly those with mental health needs.