Research and analysis

Technical Guide - Criminogenic needs of prolific offenders

Published 8 May 2025

Applies to England and Wales

1. Data sources

These statistics draw from data from administrative databases across His Majesty’s Prisons and Probation service (HMPPS), His Majesty’s Courts and Tribunals service (HMCTS) as well as the Police. The specific data sources used were:

1. The Offender Assessment System (OASys)
2. MoJ Magistrates’ Courts case management system
3. MoJ Crown Court case management system
4. MoJ extract of the Police National Computer (PNC)

1.1 Offender Assessment System (OASys)

The Offender Assessment System (OASys) is used by His Majesty’s Prison and Probation Service (HMPPS) in England and Wales to measure and record the risks and needs of offenders in custody or under supervision in the community. Offender assessments are conducted on individuals serving sentences in custody and in the community and include a wide range of information on offending-related risks and needs.

For OASys we extract all full (Layer 3) assessments for offenders in our cohort.

Prior to the reunification of the probation service on 26 June 2021, several Community Rehabilitation Companies (CRCs) used alternate risk assessment systems, for which data are not available. As such, OASys coverage in these areas, such as London, will be low, with assessments only likely to be matched for those on the National Probation Service caseload and others on licence from longer custodial sentences. However, the overall assessment coverage of 78% of the prolific offender cohort and 65% of the non-prolific offender cohort, is a reasonable level of coverage on which to base this analysis.

1.2 MoJ Magistrates’ Courts case management system

The MoJ Magistrates’ Courts dataset contains a record of all defendants, cases and appearances within the Magistrates’ Court system in England and Wales. It has been constructed using weekly extracts from LIBRA (the case management system for Magistrates’ Court cases), originally used for Home Office Court Appearance Statistics (HOCAS). The data held on the Libra system is generally deemed to be of good quality.

From this system we extract records of the offenders and offences for any offence committed from 1 September 2017 and convicted by 31 March 2020.

1.3 MoJ Crown Court case management system

The MoJ Crown Court dataset has been extracted from the XHIBIT management information system, used by His Majesty’s Courts and Tribunals Service (HMCTS) to manage cases within the Crown Court. This dataset provides data on defendants’ appearances in criminal cases before the Crown Court in England and Wales before August 2023 (when Crown Court case management transitioned from XHIBIT to a new system – Common Platform). The data held on the XHIBIT system is generally deemed to be of good quality.

From this system we extract records of the offenders and offences for any offence committed from 1 September 2017 and convicted by 31 March 2020.

1.4 Police National Computer (PNC)

The Police National Computer (PNC) is an administrative database used by law enforcement agencies. The analysis presented in this statistical release uses information drawn from the MoJ’s extract of data from the PNC. This extract of the PNC includes data on individuals with convictions and cautions.

Like any large-scale recording system, the PNC is subject to errors with data entry and recording. The PNC is regularly updated such that further analysis at a later date may generate revised figures. The quality of the information recorded on the PNC is generally assumed to be relatively high, as it is an operational system on which the police depend, but analysis can reveal errors that are typical when handling administrative datasets of this scale.

From the PNC we extract the flag that identifies which offenders meet the definition of prolific offender, based on their entire offending history (which is contained only in PNC, as courts databases do not record out-of-court disposals).

1.5 Use of administrative data

A common limitation of data from large-scale administrative record systems, is the possibility of mistakes with entering and processing the data. The quality of the information held in these operational databases, while generally high, does differ by data field, depending on the field’s frequency of use and importance in the day-to-day running of the courts, prisons and probation systems.

Details of all administrative data sources used in the production of this release can be found in the MoJ Statement of Administrative Sources.

2. Methodology

2.1 Data linking

The three HMPPS and HMCTS databases (OASys, LIBRA and XHIBIT) set out in section 1 do not have a common personal identifier to enable the same individual to be identified across the systems. This means the data cannot be linked in a straightforward way. In addition, multiple records may be associated with the same individual in one database. There is no unique identifier available to reliably link records for the same person from within and between the databases.

Splink (an open source data linking programme) has therefore been used to identify unique, deduplicated offender records and link the data across datasets. Splink is based on probabilistic linkage and was developed internally by MoJ’s data linking team. Further details about Splink are available in the Splink Guidance on GOV.UK.

Linkage to records from the Police National Computer was then achieved through deterministic matching, a method that identifies exact matches, using the individual’s PNC number (an identifier that is recorded in the PNC and OASys).

2.2 Definition of a prolific offender

An individual is defined as a young adult prolific offender if, on their last appearance in the Criminal Justice System (CJS), (i) they were aged 18 to 20, had a total of 8 or more previous convictions or cautions, and had at least 4 previous convictions or cautions when aged 18 to 20, and (ii) they were aged 21 or older, had a total of 16 or more previous convictions or cautions, had less than 8 previous convictions or cautions when aged 21 or older, and had 4 or more previous convictions or cautions when aged between 18 and 20.

An individual is defined as an adult prolific offender if, on their last appearance in the CJS, they were aged 21 or older, had a total of 16 or more previous convictions or cautions, and had 8 or more previous convictions or cautions when aged 21 or older.

See Official Statistics on the characteristics of prolific offenders for further details.

2.3 Offender assessments and criminogenic needs

2.3.1 OASys assessments

Offender assessments are conducted on individuals serving sentences in custody and in the community and include a wide range of information on offending-related risks and needs. When conducting an OASys assessment, assessors can choose which layer OASys assessment they create – a basic assessment (Layer 1), or a full assessment (Layer 3).

A basic Layer 1 assessment is a reduced version of an assessment. It does not include criminogenic needs sections and contains a basic sentence plan.

A Layer 3 assessment (herein referred to as a Layer 3) contains all sections of OASys including predictors, criminogenic needs, risks, a risk management plan and a sentence plan.

2.3.2 Criminogenic needs

Criminogenic needs are dynamic risk factors that are directly linked to criminal behaviour. There are eight criminogenic needs linked to offending behaviour which are assessed using OASys: Accommodation, Employability, Relationships, Lifestyle & associates, Drug misuse, Alcohol misuse, Thinking & Behaviour, and Attitudes.

Those needs are determined using a system of scored questions. Each question is scored on a scale from 0-2 (some being binary scored 0 or 2, and others 0, 1 or 2). 0 denotes ‘no need’, 1 denotes ‘some need’ and 2 denotes ‘significant need’. The scores within each set of questions are summed, and a need is identified if the score equals or exceeds a threshold or cut-off value for that set of questions (see Table 1).

Table 1: Criminogenic needs and relevant scored questions in OASys

Criminogenic needs	Scored questions	Scale range	Cut-off
Accommodation	3.3, 3.4, 3.5, 3.6	0-8	2+
Employment	4.2, 4.3, 4.4, 4.5	0-8	3+
Relationships	6.1, 6.3, 6.6	0-6	2+
Lifestyle & associates	7.2, 7.3, 7.5	0-6	2+
Drug misuse	8.4, 8.5, 8.6, 8.8, 8.9	0-10	2+
Alcohol misuse	9.1, 9.2, 9.3, 9.5	0-8	4+
Thinking & behaviour	11.5, 11.6, 11.7, 11.9	0-8	4+
Attitudes	12.1, 12.4, 12.5, 12.8	0-8	2+

For each of the categories of criminogenic need, assessment questions cover the following topics:

Accommodation need: No Fixed Abode (NFA) or transient, suitability of accommodation, permanence of accommodation, suitability of location of accommodation, accommodation linked to offending behaviour and accommodation linked to risk of serious harm. A score of 2 on Question 3.3 of an OASys assessment (relating to No Fixed Abode or Transient accommodation), causes the other questions on Accommodation to be scored 2 as well to give a maximum score of 8. This would be an automatic flag of an accommodation need. There is no equivalent to this in any other questions.

Employability need: Unemployment, employment history, work-related skills, attitude to employment, reading, writing and numeracy, employment related to offending behaviour.

Relationships need: Relationships with family, childhood experience, relationship with partner, previous relationship experience, domestic violence: perpetrator or victim, parental responsibilities, relationships related to offending behaviour.

Lifestyle & associates need: Activities that encourage offending, influence by criminal peers, recklessness/risk taking behaviour, lifestyle and associates linked to offending behaviour.

Drug misuse need: Current drug misuse, level of use of main drug, injecting drugs, motivation to tackle drugs, drugs major part of lifestyle, drugs linked to offending behaviour.

Alcohol misuse need: Current alcohol misuse, binge drinking, previous alcohol use, motivation to tackle alcohol use, alcohol linked to offending behaviour.

Thinking & behaviour need: Interpersonal skills, impulsivity, temper control, problem recognition, problem solving, awareness of consequences, understanding of others’ views, thinking and behaviour linked to offending behaviour.

Attitudes need: Pro-criminal attitudes, attitude to supervisions, attitude to community/society, motivation to reduce offending, attitudes linked to offending behaviour.

2.3.3 Selection of OASys assessments

Within our selected cohort of offenders who committed and were convicted of offences between 1 September 2017 and 31 March 2020, we selected all Layer 3 assessments on record for those offenders. We then matched Layer 3 criminogenic needs assessments to each offence within the reporting period (offences committed from 1 September 2017 and disposed of by 31 March 2020), with the intention of associating an offender’s needs at the time of an offence. The matching methodology selected the chronologically closest Layer 3 assessment to each offence, requiring that the assessment was at most 6 months prior to or following the date of the offence. This serves to identify, as best as possible, an offender’s needs at the time of an offence. The proportions of offenders who have an available OASys assessment after this matching are shown in Table 2.

Table 2: OASys assessment availability for prolific and non-prolific cohorts

	Number of offenders	Number of offenders with a matched Layer 3 OASys assessment	Percentage of offenders with a matched Layer 3 OASys assessment
Prolific offenders	21,553	17,082	79.3%
Non-prolific offenders	14,942	10,445	69.9%

3. Data governance

The BOLD programme has established procedures for the effective governance of the data it uses across all pilot projects. You can find BOLD’s Privacy Notice on GOV.UK.

3.1 Governance

Analysis and research using data collected in operational systems across HMPPS is covered by pre-existing Data Protection Impact Assessments (DPIAs). This work is covered by these DPIAs under the research and analysis purpose. Statistics are in aggregate form only for the purposes of understanding offenders. Information about any specific individual is not of interest and not disclosed.

3.2 Confidentiality

There are arrangements in place for protecting individuals’ confidential data when statistics are published or otherwise released into the public domain. The Code of Practice for Statistics in section T6 states that:

Organisations should look after people’s information securely and manage data in ways that are consistent with relevant legislation and serve the public good.

To comply with this, as well as with the Data Protection Act of 2018, and to maintain the trust and co-operation of those who use these ad hoc statistics, the following provisions have been put in place:

• Private information collected by MoJ is stored in line with our data security policies.
• Electronic data is held on password-protected networks.
• All new staff undergo security vetting before receiving access to data systems and all staff undertake mandatory training on information responsibility annually.

3.3 Engaging the public

Public trust around how data is shared is critical for BOLD, and we partnered with the Centre for Data Ethics & Innovation (CDEI), and the research company Britain Thinks, to undertake extensive engagement with affected groups, trusted intermediaries, and the general public. The results of this exercise, and what we have learnt from listening to the public, have tangibly informed the design of the BOLD programme and have been published by the CDEI.

4. Users

The contents of this report will be of interest to government policy makers, the agencies responsible for offender management at both national and local levels, providers, practitioners, as well as others who want to understand more about prolific offenders.

Government policymakers may also use these statistics to inform key elements of government policies. Offender management agencies may use these statistics to gain a better understanding of the needs of prolific offenders. Key agencies include: HMPPS, private and voluntary sector providers of prison and probation services and local authorities.

5. Contact details and feedback

These statistics have been produced by the Ministry of Justice: Better Outcomes through Linked Data (BOLD) - GOV.UK (www.gov.uk).

Any enquiries and feedback on these statistics can be sent to MoJ at RR-pilot-BOLD@justice.gov.uk.