‘For the last two years, the female prison population has been consistently under 4,000 for the first time in a decade. I want to see still fewer women in custody, especially those who are primary carers of young children.’ – Caroline Dinenage MP, Parliamentary Under Secretary of State for Women, Equalities and Family Justice (Reference DinenageDinenage 2015).
‘We want to see more effective provision for women offenders, making it possible for there to be a substantial fall in the women's prison population in the coming months and years.’ – House of Commons Justice Committee (2015: p. 16).
‘The number of women offenders is comparatively small but the impact is not and it is obvious that within the criminal justice system we cannot simply replicate what we provide for men and hope it will work for women.’ – David Ford MLA, Northern Ireland Minister of Justice (Prison Reform Trust 2015).
According to the charity Women in Prison (2015) there were 3935 women imprisoned in England and Wales on 13 July 2015, and the average cost of keeping a woman in prison for a year is £44 329 (Ministry of Justice 2014a). The charity reports that almost half of women prisoners have suffered domestic violence and just over half have suffered emotional, physical or sexual abuse in childhood. It reports that more than half admit to having used cocaine, crack cocaine or heroin in the 4 weeks prior to imprisonment, and an earlier study (Reference Plugge, Douglas and FitzpatrickPlugge 2006) found that 75% of women prisoners had used drugs in the 6 months prior to imprisonment. The charity reports that at least one-fifth of women prisoners were lone parents before imprisonment and it quotes an estimate that in 2010 more than 17 240 children were separated from their mothers by imprisonment.
Women prisoners represent only a fraction of women offenders. Only 15% of women sentenced for indictable offences receive an immediate custodial sentence, and each year there are nearly 200 000 prosecutions of women for non-motoring summary (non-indictable) offences (Ministry of Justice 2014b), for which the majority receive a non-custodial sentence. So, women in prison are only the tip of the iceberg of female offending.
Given the size of the problem, not just in terms of the resources of the criminal courts and the cost to the prison estate, but most importantly in terms of the impact of drug use and offending on the health and well-being of the women themselves and on their partners and children, it is enormously disappointing that since Perry et al's previous review (Reference Perry, Coulton and GlanvillePerry 2006) there has been so little research of a sufficiently high quality.Footnote a As Reference CorstonCorston (2007) observed, female offenders represent an under-researched, vulnerable population with specific needs distinct from their male counterparts. Although the authors’ latest review (Reference Perry, Neilson and Martyn-St JamesPerry 2015), the summary of which appears in this month's Cochrane Corner, has identified a number of randomised controlled trials (RCTs) of interventions for female drug-using offenders published since 2006, none has been of high quality.
Out of more than 10 000 reports identified, the 2015 review is based on only 9 of the 72 empirical studies that made it to Phase II screening. Out of the 9 studies, the majority were rated as being at unclear risk of bias due to a lack of descriptive information, and they yielded only low- or moderate-quality evidence at the most. It is also disappointing that there were too few studies to evaluate whether the treatment setting, for example, court or community, had an impact on success. It is important to know whether efforts should be focused on women appearing before the courts or on those imprisoned. It is also disappointing that none of the studies provided evidence as to cost-effectiveness.
The review
Reference Perry, Neilson and Martyn-St JamesPerry et al (2015) identified and reviewed 9 RCTs of interventions that were designed, wholly or in part, to eliminate or prevent relapse to drug use or criminal activity among women referred by the criminal justice system at baseline and for whom the study reported pre- and post-programme measures of drug use and criminal behaviour and the same length of follow-up for the two groups. Box 1 shows the experimental interventions included in the review.
BOX 1 Experimental interventions included in the review
-
Any pharmacological intervention (e.g. buprenorphine, methadone)
-
Any psychosocial intervention (e.g. therapeutic community, case management, cognitive–behavioural therapy, interpersonal psychotherapy, motivational interviewing)
The results of the 9 eligible trials were published between 1996 and 2014. Eight of them were from the USA and one was from Spain. The average age of the participants was in the early- to mid-30s. According to the review's authors, in all but one study the participants were of White ethnic origin, but this seems to be an erroneous or misleading observation, as in 5 studies the percentage of what the Cochrane reviewers term ‘non-White’ participants was more than 50%.
Does psychosocial intervention make any difference?
Five studies permitted a comparison of any psychosocial intervention with treatment as usual.
Two examined drug use. Reference Johnson, Friedmann and GreenJohnson et al (2011) found no effect. Reference Lanza, García and LamelasLanza et al (2014), in a study comparing cognitive–behavioural therapy (CBT) with acceptance commitment therapy (ACT) (Box 2) and a control intervention, found no effect of the CBT on the abstinence level, but an abstinence level of 44% following ACT, compared with 18% in the control group.
BOX 2 Acceptance commitment therapy
-
16 weekly group sessions of 90 minutes led by a trained therapist
-
Seeks to undermine the grip of the literal verbal content of cognition that provokes avoidance behaviour and to construct an alternative context in which behaviour aligned with one's values is more likely to occur
-
Aims to increase substance use abstinence
-
Sessions involve experiential and didactic learning
-
Employs validation and empowerment to help participants respond to previously avoided events in new ways
Two studies (Reference Nielsen, Scarpitti and InciardiNielsen 1996; Reference Guydish, Chan and BostromGuydish 2011) examined re-arrest rates and found no difference between interventions. However, three (Reference Nielsen, Scarpitti and InciardiNielsen 1996; Reference Zlotnick, Johnson and NajavitsZlotnick 2009; Reference Johnson, Friedmann and GreenJohnson 2011), albeit only with moderate-quality evidence, found a reduction in imprisonment (effect size 0.46). Box 3 shows the psychosocial interventions that reduced imprisonment.
BOX 3 Psychosocial interventions that reduced imprisonment
Collaborative behavioural management (Reference Johnson, Friedmann and GreenJohnson 2011)
-
12-week intervention
-
Treatment sessions with offender, officer and substance use counsellor at least every 2 weeks, plus further officer/offender contacts
-
Premise: reinforcing desired behaviour is more likely to result in sustained positive change than punishing undesired behaviour
Intensive work release (Reference Nielsen, Scarpitti and InciardiNielsen 1996)
‘Continual Recovery through Education and Skills Training’ (CREST):
-
Therapeutic community programme
-
1 month orientation, 2 months primary treatment, 3 months work release
Cognitive–behavioural therapy combined with standard therapy (Reference Zlotnick, Johnson and NajavitsZlotnick 2009)
Cognitive–behavioural therapy – ‘Seeking Safety’ programme:
-
development of coping skills to attain safety from post-traumatic stress disorder and substance use disorders
-
focus on the present
-
abstinence oriented
-
emphasis on empowering, compassionate approach
-
group modality, 90 minutes, typically 3 times a week for 6–8 weeks
-
3–5 women per group
-
optional weekly booster session for 12 weeks after release
Standard therapy – substance misuse treatment programme:
-
abstinence oriented
-
psychoeducational large-group format
-
weekly individual case management and drug counselling
-
attention to women's health, domestic violence, affect management, relapse prevention, career exploration, anger management, parenting
-
no specific trauma treatment
-
referral for further substance misuse treatment on release
These results appear a little less disappointing when the nature of the control intervention is taken into account. ‘Treatment as usual’ is not ‘no treatment’ or ‘no intervention’. The comparison by Reference Johnson, Friedmann and GreenJohnson et al (2011) of community collaborative behavioural management was with standard parole supervision in the form of 1–4 contacts a month with a parole officer, who was affiliated to a substance misuse treatment programme, and drug testing. The control intervention employed by Reference Lanza, García and LamelasLanza et al (2014) was a mental health assessment and a re-education programme. Reference Nielsen, Scarpitti and InciardiNielsen et al (1996) used as a control intervention Delaware's 6-month conventional work-release programme and this was a low-quality study. Reference Zlotnick, Johnson and NajavitsZlotnick et al (2009) employed a ‘standard therapy’ control intervention (Box 3) comprising a substance use treatment programme that included group and individual casework delivered over a number of months and dealing comprehensively with a variety of issues pertinent to female offenders.
The only trial that found no benefit for the trial intervention compared with treatment as usual was that of Reference Guydish, Chan and BostromGuydish et al (2011), who simply compared intensive probation case management, which included therapeutic and advocacy orientation and training, with standard probation. But this was a low-quality study and, as has been observed elsewhere (Reference Sorenson, Dilley and LondonSorenson 2003), although probation case management is designed to be more engaging than standard probation, there is a low level of face-to-face contact.
Pharmacological treatment
There was a single study of pharmacological treatment (buprenorphine) (Reference Cropsey, Lane and HaleCropsey 2011). Although it prevented or delayed relapse to opioid use during the treatment phase, the results were not sustained post-treatment and most women had relapsed into active opioid use at the 3-month follow-up point.
The nature of the psychosocial treatment
Four studies compared different psychosocial interventions. Reference Johnson, Friedmann and GreenJohnson et al (2011) found interpersonal psychotherapy superior to psychoeducational attention in reducing levels of depression and substance misuse in women with major depression. Reference Lanza, García and LamelasLanza et al (2014) failed to demonstrate an effect of CBT but, as indicated above, did find an effect of ACT in terms of abstinence. Reference Messina, Grella and CartierMessina et al (2010) compared a gender-responsive therapeutic community programme with a standard therapeutic community programme and found that the gender-responsive treatment had a greater effect in terms of both subsequent drug use and re-imprisonment. The components of the gender-responsive treatment are shown in Box 4.
BOX 4 Gender-responsive treatment
Therapies
Cognitive–behavioural approaches, mindfulness meditation, experiential therapies (guided imagery, visualisation, art therapy, movement), psychoeducational, relational and expressive arts techniques
‘Helping Women Recover’ – 17-session programme with 4 modules
-
Self module
-
Discovering what the ‘self’ is
-
Learning that addiction can be understood as a disorder of the self
-
Learning the sources of self-esteem
-
Considering the effects of sexism, racism and stigma on a sense of self
-
-
Relationship module
-
Women explore their roles in their families of origin
-
Discussing myths and realities about motherhood and their relationships with their mothers
-
Reviewing the histories of their relationships
-
Considering how they can build healthy support systems
-
-
Sexuality module
-
Exploring the connections between addictions and sexuality, and discuss body image, sexual identity, sexual abuse and the fear of sex when sober
-
-
Spirituality module
-
Introduction to the concepts of spirituality, prayer and meditation
-
‘Beyond Trauma’ – 11 sessions focused on 3 areas
-
Teaching women what abuse and trauma are
-
Helping them to understand what typical reactions to trauma and abuse are
-
Developing coping skills
Reference Sacks, Sacks and McKendrickSacks et al (2008) found that both therapeutic community intervention and the CBT control intervention resulted in improvements in mental health, substance misuse, criminal behaviour and HIV risk.
Cost and cost-effectiveness
None of the studies included any cost information, so it was not possible to examine the cost-effectiveness of the interventions.
So what?
As none of the reviewed studies was of more than moderate GRADE quality, by definition further research is likely, if not very likely, to have an important impact on the confidence in the estimates of the effects of the treatments and may, or even is likely to, change the estimate. The implication of this review for researchers is that ‘high quality research is required to evaluate the effectiveness of different treatment options for female drug-using offenders’ (Reference Perry, Neilson and Martyn-St JamesPerry 2015). Of particular value would be trials of some of the interventions employed in these studies with greater attention to overcoming the obstacles, and avoiding the pitfalls, that reduced their quality and gave rise to concerns about reporting bias.
What are the implications for practitioners and policy makers? The review provides cautious support for incorporating psychosocial treatment approaches – such as elements of therapeutic community treatment, interpersonal psychotherapy, ACT, behaviour management and CBT – into offender management and substance misuse treatment programmes, and for doing so in a gender-responsive way with a recognition of the role that trauma and abuse play in the lives of these women. However, expectations need to be modest: there may be little or no effect on drug use and arrest rates. But even a small reduction in the numbers of women imprisoned makes the work worthwhile, albeit that the cost is unknown.
eLetters
No eLetters have been published for this article.