![Sample our Health and Social Care journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5938/TOC-Subject-Sample-2021-Health-Social-Care.jpg"})
Abstract
Aim: Over a 10-year period from 2003, around 1.7 million arrests in England and Wales resulted in the suspect being exposed to mandatory drug testing and assessment processes. These provisions formed part of a wider drug interventions programme costing £1.3 billion. This study sought to assess the impact of compliance with these measures on treatment uptake and reoffending. Recidivism risk factors were also investigated. Methods: The use of linked administrative data relating to matched samples of recent heroin and/or cocaine (H/C) users identified in one English police force area over a 12-month period (N = 1630). Findings: There was no association between compliance with a compulsory model of drug diversion and subsequent engagement with structured treatment services, rates of treatment retention and successful discharge. Compliance was also not found to be associated with reductions in the rate and volume of reoffending after 12 months. The factor with the largest effect on risk of recidivism was poly use of H/C. Main offence, engagement with structured treatment, number of prior convictions and (younger) age were also identified as recidivism risk factors. Conclusions: These results are discussed in the context of subsequent legislation and policy which further expands the reach of mandatory testing and assessment measures as a form of constraints-based drug diversion.
Declaration of interest
The authors report that they have no conflicts of interest.
Notes
Notes
1. Single Convention on Narcotic Drugs (1961), Convention on Psychotropic Substances (1971) and the United Nations Convention against Illicit Traffic in Narcotic Drugs and Psychotropic Substances (1988).
2. The Act also sought to integrate this new provision with existing restrictions on bail (RoB) measures, which had first been placed on the statute book via the 2003 Criminal Justice Act (Section 19).
3. The Act also moved the point at which a drug test may be carried out to arrest rather than charge, as had been the case under previous legislation.
4. In line with Skodbo et al. (2007: p. 4) engagement with treatment was only considered to be linked to the index positive test where it occurred within 144 days following this event (i.e. allowing for 60 days from the date of test for an individual’s care plan to be agreed and a further 12 weeks following the formulation of a care plan for individuals to enter structured, care planned treatment).
5. Refers to services such as specialist and general practitioner-based community prescribing, structured care-planned counselling, structured day programmes, aftercare services, detoxification and residential rehabilitation.
6. In line with the definition deployed by the National Treatment Agency (NTA) during this period, successful discharge was defined as having been referred on, completing treatment, or completing treatment and no longer using illicit drugs.
7. More specifically these were the drug test recorder (DTR), the drug interventions record (DIR), the national drug treatment monitoring system (NDTMS), and an extract from the Police National Computer (PNC).
8. Treatment data were linked by an independent data custodian using these attributor details.
9. A Police National Computer Identification number (PNCID).
10. The rate of proven reoffending refers to the proportion for whom a further offence had been committed in the 12 months following the index positive test (i.e. within 365 days of it), which was proven by a caution or court conviction (either in the 12-month period, or in a further 6 months – to allow for more serious cases to be processed through the courts i.e. within 548 days of the index positive test) (Ministry of Justice, Citation2011: p. 45). While the MoJ excludes breach offences in its counting rules, both breach and non-breach offences are reported here. This is to reflect the central role which enforcement of non-compliance plays within the ‘Tough Choices’ policy, and the significant personal and justice system costs arising from such a conviction. Over the period examined, for instance, one in seven (14%) receptions into English and Welsh prison establishments was as a consequence of a breach offence (Ministry of Justice, Citation2009: p. 67).
11. We replicated the approach used by Skodbo et al. (2007) to estimate time at reduced risk due to imprisonment. This involved identifying any custodial sentences recorded within the convictions data as having been imposed during the 12 months from the date of the index positive test, and taking 50% of the longest prison penalty imposed during each sentencing ‘episode’ in order to estimate the total time (days) spent in custody during this period. (For determinate custodial sentences under a year, offenders in England and Wales were, during the period examined, automatically eligible for release after serving half their sentence.) Given the absence of reliable data on the extent to which prison sentences imposed by the courts are served consecutively (Hansard, Citation2011), it was assumed that in the event of multiple custodial penalties being imposed following a court conviction, these sentences were served concurrently. Importantly, these estimates do not account for any time spent on remand in custody awaiting trial, as such information is not recorded by the relevant dataset. The follow-up period, therefore, related to ‘elapsed time’ over 12 calendar months, rather than 12 months of ‘free time’ (i.e. excluding any time spent in custody).
12. Available assessment data for the compliant group indicated that around two-fifths (39.5%, n = 259) reported being engaged in some form of treatment at the time of the required assessment and over half (54.6%, n = 351) subsequently agreed a care plan with a DIP practitioner as a result of this process.
13. Excluding breach offences, there was again no association observed between compliance with a compulsory assessment under ‘Tough Choices’ and this ‘adjusted’ rate of proven reoffending in the 12 months following identification as a recent H/C user (56.4% vs. 56.4%; χ²(1, N = 1312) = 0.00, p = 1.000; RR = 1.00; CI = 0.88–1.13).
14. Excluding breaches the number of ‘adjusted’ proven offences committed over the 12-month follow-up period fell to 3329 with an average (mean) of 2.5 offences (Mdn = 1.0, SD = 3.8, R = 0–34). There were no differences observed between the compliant (M = 2.4, Mdn = 1.0 SD = 3.6, R = 0–27) and non-compliant groups in this regard (M = 2.7, Mdn = 1.0, SD = 4.0, R = 0–34) (t(1295) = 1.12, p = 0.262).
15. This was also the case when breach offences were excluded (IRR = 0.88; 95% CI = 0.74–1.04; z = −1.49; p = 0.137).
16. This was also the case when breach offences were excluded: the total number of ‘adjusted’ offences leading to conviction was higher in the 12 months post-identification as a user of H/C (3329 offences) when compared with the 2 (2296), 3 (2435), 4 (2292) and 5-year (2532) periods preceding this point.
17. Using ‘adjusted’ data (i.e. removing both an offence in the 12-month pre-period, thus minimising the potential for mistakenly attributing any natural variation in repeated offending data to an intervention effect, as has been proposed by Skodbo et al. (2007: p. 13), and all breach offences), there was a 1.3% overall reduction in the number of known offences committed in the 12 months following initial identification as a recent user of H/C, relative to the corresponding period before this (from a mean of 2.6 offences to 2.5) (z = −2.89; r = −0.06; p = 0.004). Again there was a non-significant overall increase in this ‘adjusted’ level of offending observed among those complying with the ‘Tough Choices’ policy (+2.5%; a mean of 2.4 offences in both periods) (z = −1.24; r = −0.03; p = 0.216) relative to a significant 4.5% reduction observed for the non-compliant group (from a mean of 2.8 offences to 2.7) (z = −2.83; r = −0.08; p = 0.005).