The majority of those detained in prisons will eventually return to the community. While no routine data collection currently reports the number of individuals released from custody in Australia, the most recent estimate suggests that there are approximately 54,751 prison separations annually [1] and that each of these instances of re-entry is associated with heightened all-cause mortality rates [2], most often as a result of fatal overdose [3,4]. These high mortality rates following a period of incarceration have been documented in several countries, including Australia, the United States, the United Kingdom and France [2-5].

Beyond the elevated mortality risk, re-entry into the community following a period of incarceration is typically a stressful transition. Released prisoners face the task of attempting to secure housing and employment, to re-establish connections with family and friends, often in combination with coping with substance use problems and mental health disorders [6]. Binswanger et al. [5] identified that during this transitional period, released prisoners were much more likely to engage in high-risk behaviours, particularly unsafe sexual practices and illicit substance use. They also found that former prisoners were more likely to die from any cause in the first two weeks following release and were as much as 129 times more likely to die from drug overdose than the general population. Additionally, [7] found that the highest rates of all-cause mortality following release from prison was associated with prisoners who had a history of violent criminal offences, multiple incarcerations, and psychiatric hospital admission during incarceration.

The same factors that increase the post-release stress of former prisoners may also contribute to their risk of reoffending and reincarceration. Reoffending has been associated with personal factors such as criminal history, criminogenic needs and social achievement [8], as well as situational factors such as lack of employment, living in low socioeconomic areas, and having limited social support and family attachments [9]. Other significant contributors to the likelihood of reincarceration include instability in the family home and homelessness [10].

Substance abuse has also been found to be a significant contributor to reoffending; however rates of recidivism vary depending on the type of drug and frequency of use [11,12]. For example, in the Drug Use Monitoring Australia (DUMA) study, 45% of the participants reported that drug and alcohol abuse had contributed to their current offence [13]. Among juveniles, drug and alcohol use is also a significant contributor to offending behaviour. For example, Putnins [11] identified a significant relationship between alcohol and recidivism. It was found that six months following release from juvenile detention, those who frequently engaged in alcohol use were up to 77% more likely to reoffend compared to those whose alcohol use was less frequent.

Practical and logistical factors may also contribute to reoffending post-release.For example, Borzycki [14] highlighted that released prisoners often have inadequate personal identification, which results in delays accessing welfare benefits and other public services. Exprisoners also often have limited access to financial resources, a problem further compounded by their inadequate personal identification and inability to access welfare. Furthermore, they often lack information regarding available support services that could assist them in dealing with transitional issues post-release. Inability to access support, financial or otherwise, is likely to exacerbate the stress of release into the community, hinder the process of reintegration and increase the likelihood of reoffending.

Indeed, available evidence suggests that rehabilitation and reintegration of prisoners is challenging, with many prisoners experiencing multiple separations from correctional centres across the life course. For example, data indicates that 55% of those inmates in custody in Australia on the 30th of June 2011 have served at least one previous prison sentence [15].

In this context, throughcare and aftercare models of rehabilitation and reintegration post-release have been gaining support [14,16-19]. These models provide services throughout the transition from custody to the community in order to maintaintreatment and support during this difficult period. Evidence suggests that throughcare models can be effective in maintaining rehabilitative gains and reducing the risk of reoffending post-release [20,21], particularly among substance using offenders [22].

The Connections Program is based on such a throughcare model. It was established in 2007, and is operated by Justice Health& Forensic Mental Health Network, a division of the New South Wales Ministry of Health that provides health services to those in contact with the NSW criminal justice system¹. The program provides statewide coordinated release planning and linkage with health and welfare services for patients in custody with problematic drug use, aiming to facilitate re-entry, reduce recidivism and prevent drug-related post-release mortality. This is achieved through a thorough assessment of the needs of referred patients, which constitutes the basis of a release plan addressing the patients’ identified needs including but not limited to accommodation, employment, medical treatment and access to social services.

The aim of this paper is to examine the data collected over the first 14 months of the Connections Program in order to provide a snapshot of its participants and their transition needs. The patient data collected from interviews before (at assessment) and after their participation in the program (one-month post-release) offers rich details of the characteristics of the patients, their behaviour post-release, the challenges they face upon their release into the community, and the way their transitional needs change over time. Data of this kind is vital to inform and prioritise the provision of services based on these presenting needs.

Connections program

The Connections Program targets adults in correctional centres in NSW, who have a demonstrable illicit drug problem, and are eligible for release at least one month after their referral date. Entry into the program is voluntary, with Connections accepting self-referrals as well as those made by health personnel, Probation and Parole staff and others. The expected program duration is set to allow an initial assessment of patients’ health and social functioning, as well as financial and vocational needs pre-release, the development of a release plan, and a minimum follow-up period of one month post-release. However, extended post-release follow-up periods are also available in cases of demonstrated need; for example, in cases where a patient was being released after a long period of incarceration.

Data collection and sources

After providing informed consent, each prospective patient completed assessment interview gathering demographic information and investigating health and psychosocial functioning in the following domains: employment, education, alcohol and other drugs (AOD), physical and mental health, social support, relationships, accommodation, incarceration status and identification. Patients were required to complete this assessment upon agreeing to participate in the Connections Program, and then again either one month after being released (i.e., at program completion) or upon return to custody. Questions were presented to patients in various formats throughout the questionnaire including: yes/no, forced-choice categorical, multipleresponse, and free recall response types.

All patients completing an assessment interview were asked to provide their informed consent for their de-identified data to be used for research and evaluation purposes. Only data from those patients who provided informed consent have been included in the analyses reported here. Justice Health and Forensic Mental Health Network, and the University of New South Wales Human Research Ethics Committees provided ethical approval for the analysis of Connections Program data.

Participants

Information regarding 1,035 potential patients was entered into the Connections Database from September 3, 2007 to November 15, 2008. Of these, 50-declined consent to participate and 156 had not yet completed an assessment interview. Data is presentedfrom the assessment interviews conducted with the remaining 829 patients. Additionally, we report information regarding the post-release behaviours either of 285 patients who completed an assessment one month following their release or upon return to custody. Overall, 32.5% of all patients assessed completed post-release interviews.

The majority of patients participating in Connections were male (82%) with an average age of 34 years (SD=7.8 years, range 18.7 to 60.5 years). The largest proportion of patients were in the age group of 25-29 (25%), followed by the age group 30-34 (22.8%) and 35-39 (18%). Most were born in Australia (89.1% of 818 respondents) and indicated that they identified with the Australian culture alone (55.3% of 667 respondents). Twenty-nine per cent identified as Aboriginal or Torres Strait Islander alone, while 2.2% identified as Vietnamese, 0.7% identified as Arabic, 0.3% identified as Islander and 9.7% identified with another unspecified culture. A majority of patients reported never having been married (58%), and a majority (63.8%) had children.

Patient characteristics at initial assessment

Attainment, social & financial functioning: Connections patients were most likely to have ‘never been married’ (58%), or to be in a married or de-facto relationship (22.8%). Approximately equal proportions had a regular partner (6.8%), were separated (6.6%), or were divorced (5.8%). When asked, 18.4% of the sample indicated that they were experiencing difficulties keeping in contact with their family or friends while in gaol, and 5.9% anticipated that communication difficulties would persist once they were released. The majority of Connections patients reported being parents (63.8%). The majority of patients (81.7%) left school before completing the full six years of their secondary education (Table 1). Furthermore, almost 14% of all Connections patients reported experiencing literacy or numeracy problems.

Table 1: Connections patients’ education and qualifications and accommodation status following release

Almost one-fifth of patients (17%) had no formal identification documents at the time of interview, and one in three patients (34.9%) did not have a Medicare card. Nine in ten patients were receiving government benefits prior to incarceration, most often in the form of employment support benefit (63.4%) or disability support pensions (23.7%). A majority (78.3%) of patients indicated that they had experienced financial problems the last time they were released from custody.

While 82.8% of Connections patients reported having been employed at some time in the past, only 10% were working during the six months prior to incarceration and 36% had not worked for more than 5 years before coming to custody. The majority of Connections patients indicated that they were employed in labouring or related occupations (40.9%) or were trades people (24.7%), followed by sales or hospitality (13.1%) and plant/machine/driver (7.1%).Just under one-fifth (18.7%) of all patients had a job to return to post-release and less than half (40%) had participated in training or education programs during any custodial episode. The largest proportion (34.1%) of participants reported planning to reside in a public rental accommodation postrelease.

Health status

Physical health: The majority of Connections patients (70.1%) reported existing physical health problems, with an average of 1.45 different concerns (SD=0.74). Blood borne viruses and sexually transmitted infections (BBV/STI) were the most prevalent type of physical illness (36%), followed by musculoskeletal (21%), ‘other’ problems (20%), respiratory (17%), cardiac (6%) and reproductive problems (2%). Half (49%) of 554 respondents had not previously received any treatment for these problems.

Half of patients (50.9%) had had a head injury in their lifetime, and reported an average of 3.37 separate incidents (SD = 4.1). Reported residual effects of these injuries were most often headaches or migraines (34.9%), followed by brain injury or memory problems (19.9%). These effects were proximal to the incident and in some cases may have caused on-going problems but this was not explored further.

Mental health: More than half of patients (56.1%) reported having received treatment for an average of 2.19 mental health problems in their lifetime (SD = 1.29), with 61.7% of these patients indicating that they were taking psychiatric medication at the time of their assessment. Connections patients most frequently reported having been diagnosed with depression (35.3%), followed by anxiety (19.6%), substance dependence (17.9%), and schizophrenia (12.0%).

Over a quarter (28%) of patients reported having considered or attempted suicide and 13.7% reported having considered or attempted self-harm at some time. However, less than 1% of all patients were considered a current suicide risk at the time of their assessment interview.

Substance use & treatment: The majority (85.7%) of Connections patients reported having a drug problem immediately prior tocoming into custody, using an average of 2.03 different drug types (SD= 1.16). Heroin was used by the greatest number of Connections patients (51.9%), followed by amphetamines (35.9%) and cannabis (29%). Almost three-quarters of patients (71.7%) reported that the offence for which they were incarcerated was drug related, and 17.8% of patients admitted continuing to use drugs whilst in custody. The majority of Connections patients (91%) had previously participated in at least one form of AOD treatment and had accessed an average of 3.1 different treatment types. Pharmacological treatment (methadone, buprenorphine and naltrexone) was accessed by most patients(71.8%), followed by residential rehabilitation (40.6%) and counselling or psychotherapy services (38.7%)

Additionally 92% of patients had participated in at least one form of AOD treatment (M=1.25, SD=0.59) at some point while in custody, most often methadone (77.2%). A clear majority of the Connections patients (88%) also indicated that they would need to continue with AOD treatment post release, with half indicating that they would like access to additional AOD treatments. Of those requesting new treatments most sought counselling (67.6%), followed by relapse prevention (31.1%) and ‘other’ treatments (28.1%), while residential rehabilitation programs were the least popular option (8.3%).

Offending and incarceration: For the majority of patients (68.4%) entering into custody was the consequence of a new charge, whereas for 15.5% of them it was a consequence of breaching their parole. A further 10% were incarcerated for being convicted of new charges and subsequently had their parole revoked. Of those patients who entered custody on new charges, property offences such as breaking and entering, stealing and shoplifting accounted for the largest proportion of custodial sentences (49.6%). This was followed by violent offences, such as assault and grievous bodily harm (16.9%), and drug offences including possession and supply of prohibited substances (11.2%).

The majority of patients (73.7%) would have spent less than 12 months in custody by the time they were to be released, with 21.8% serving between one and four years, and 4.3% serving between 4 and 10 years. Approximately 9% of Connections patients were serving their first custodial sentence.Almost one third of patients (30.3%) reported participating in a course to address their offending behaviour at some point while in custody, and 22.2% indicated that they would like additional assistance to prevent future offending.

Previous transitional issues: When asked 85.6% of Connections participants indicated that they had faced an average of 3.16 different problems when they had last been released from custody (SD= 2.06). Overall problems with drug and alcohol use were cited most frequently (55.3%), followed by housing (35.8%) and financial difficulties (29%). Connections patients were least likely to report having had gambling problems (5.3%) (Figure1).

Figure 1: Post release problems encountered by gender

Treatment planning: On average Connections staff identified 4.81 treatment issues for each patient (SD=2.99). Staff most frequently planned for issues relating to alcohol and other drug use (85.5% of cases), followed by physical and mental health (75%) and finally financial problems (66%) (Figure 2).

Figure 2: Issues targeted in pre-release treatment plans by gender

Follow-up interviews

The following analyses explore the post-release behaviours of patients who completed interviews one-month after their release from custody with the support of Connections (N = 285). Thirty-nine (13.7%) of these interviews were conducted upon return to custody if that occurred within one-month (RTC group), while the remainder were conducted in the community approximately one-month postrelease (i.e., Community Follow-up; CFU, n = 246). A majority of patients in the CFU group were male (84.6%) and on average were 34.5 years of age (SD = 8.11). Similarly, 82.9% of those in the RTC group were male, with an average age of 32.7 years (SD = 8.11).

Of those in the RTC group, half returned for a breach of parole either by committing a new offence (33.3%) or by breaching other conditions of their parole orders (20.5%). Patients in this group reported an average of 1.28 new offences (SD = 1.33) with a range of 0-3. RTC patients were most likely to commit property offences (51.2%), followed by violent offences (22%).

Self-reported post-release behaviours: The post-release behaviour of those participants interviewed in the community significantly differed from behaviour reported by those interviewed upon return to custody in a number of ways. Those who remained in the community were more likely than those returning to custody have been engaged in education or training; remained for longer in the accommodation arranged for them before moving; and continued their required physical and mental health treatment. They were also less likely to have used drugs, reported using fewer types of drugs on average and experienced fewer transitional issues than those who returned to custody within the first month post-release (Table 2).

Table 2: Post-release Behaviour of CFU and RTC Groups

The transitional problems identified by patients before and after involvement with Connections are shown for the CFU (Figure 3) and RTC groups below (Figure 4). Post-Connections the CFU group more often reported physical, financial and ‘other’ problems and were less likely to report housing, AOD use, peer and identification problems. Members of the RTC group on the other hand more often reported facing mental health issues, AOD use and ‘other’ factors, and less frequently reported physical health, housing, financial, family/peer, employment and identification problems pre- to post-Connections.

Figure 3: Transitional issues identified by CFU patients pre- and post-Connections

Figure 4: Transitional issues identified by RTC patients pre- and post-Connections

Connections patients are a group with significant levels of disadvantage across many domains. Most had not completed their secondary education (81.7%), almost 14% of the sample indicated that they had literacy or numeracy problems, and at the time of assessment 89.8% indicated that they had been in receipt of Centrelink benefits before coming to custody.

A substantial majority of those interviewed (70.1%) indicated that they had existing physical health problems, half (50.9%) reported ever having had a head injury (average of 3.37) and more than half of the sample (56.1%) reported a mental health diagnosis; most frequently depression (35.3%) or anxiety (19.6%). Almost 30% reported having considered or attempted suicide in their lifetime. Over 85% of patients perceived their own drug use as problematic, with the greatest number reporting heroin use (51.9%). Nearly all Connections patients had engaged in at least one form of AOD treatment prior to joining Connections, and 92.4% reported having undergone some form of treatment while in custody at some time; most frequently Opioid Substitution Treatment (OST; 85.6%).

The majority of patients were serving sentences of less than 12 months (73.7%) and most often the Connections incarceration episode was not their first time in custody (91%). At some point while in custody, almost one third of patients (30.3%) had engaged in a program to address their offending behaviour. Of those with a history of prior incarceration, most had encountered transitional issues post-release (85.6%), frequently citing problems with AOD (55.3%), housing (35.8%) and financial difficulties (29%).These patterns echo the existing literature on transitional difficulties [6,10,14]. Connections staffs were most likely to provide care plans addressing AOD (85.5%), physical or mental health (75%), and financial problems (66%).

Follow-up interviews were completed by 285 patients, 39 upon return to custody, and 246 in the community one-month post-release. When considering the post-release behaviours of these individuals, patients returning to custody within the first 30 days differed significantly from patients who remained in the community for at least this period. Consistent with Gendreau, Little and Goggins’ review [8], those those returning to custody displayed lower levels of social achievement post-release in that they were significantly less likely to engage in training or education post-release. Similarly, in keeping with the findings of [10] housing instability was more common among those who returned to custody by follow-up, as was a tendency to disengage from required physical and mental health treatments. Our results also support the existence of a link between substance use and reoffending [11-13], with those returning to custody being significantly more likely to use drugs and to use more different drugs than those who did not return to custody prior to follow-up. Indeed, overall, the data suggest a relationship between short term reintegration success and the number of transitional problems experienced to the extent that those returning to custody reported experiencing a greater number of transitional problems-most frequently in the domains of substance use (61.5%), housing (35.9%) and relationships (25.6%). While it is important to note that the fact of return to custody may have led those in the return to custody group to feel more negative about their transition experience and therefore report having more transitional problems, it is interesting that a different profile of transition issues emerged for those who remained in the community; Instead, placing an emphasis on financial problems (31.7%) and ‘other’ transitional issues (19.5%), in addition to the common concern related to housing (26.4%).

Overall then, these prisoners and Connections patients represent a group with high service provision needs [6,14,17,23] who are generally serving short sentences (less than 12 months). As a consequence, throughcare initiatives and other support agencies have only a limited window of opportunity to engage with these patients to effect positive change. Specifically, just one-month prior to release almost 17% of those interviewed at assessment had none of the forms of identification required for the purposes of securing benefits or accessing services upon release in the community (e.g., a birth certificate or a Medicare card). A clear majority (81.3%) anticipated that they would be unemployed upon release, of which less than 40% reported having engaged in vocational training or education programs while in custody. Over 78% had previously encountered financial problems post-release and roughly, one third of those with existing health problems at assessment reported that their condition was untreated. Just 30% of the sample had completed a program to address their offending behaviour during a custodial episode, and 28% of the group reported a history of suicide attempts or ideation. This evidence of the substantial and serious needs of those patients accessing Connections highlights the necessity for appropriately tailored and targeted service provision that can be delivered within the available time frame and throughout the re/integration period [14]. Given the scope and nature of the issues most often encountered by patients (AOD, housing and financial problems) and targeted by Connections staff (AOD use, physical and mental health, and financial problems) the Connections Program has the potential to provide appropriately targeted transitional assistance to improve reintegration success, however further research assessing the efficacy of this intervention for reducing recidivism is required.

Finally, although causation cannot be inferred given the data, information regarding the differences in behaviours between those returning to custody within 30 days and those remaining in the community could also be instructive. Patients who remained in the community were also those who refrained from or limited their drug use (compared to the RTC group), who were more likely to be compliant with their psychiatric medications and other health treatments, who had more stable housing and who were more likely to be engaged with training or education; These findings are consistent with the existing literature [11,12,24-26]. Indeed, Bradley and colleagues [24] suggest that housing stability and compliance with medical treatment are likely to be related. Consequently, it may be that patients will benefit from interventions and support prioritising these types of transitional concerns as opposed to some of the others assessed here [27,28]. For example, transitional programs providing explicit support linking prisoners with external health care providers, and providing practical assistance (e.g., transport) to appointments may be integral to maintaining stable community engagement and reducing re-incarceration. Further research investigating the extent to which these behavioural differences predict re-incarceration would therefore be invaluable for guiding throughcare resource allocation and prioritisation into the future.

The authors would like to thank the Connections Program Operations Manager, Steven Ward and Data Managers Wilson Lai and Anna Le, for their invaluable assistance and support extracting and interpreting the data. Sincere thanks also go to the Connections Program Clinical Support Workers and to the Connections patients.

¹Because the Connections Program operates under the auspices of a health service, those participating in the program are referred to as patients rather than participants henceforth.