Research and analysis

Criminogenic needs of prolific offenders

Published 8 May 2025

Applies to England and Wales

1. Executive summary

1.1 Overview

This report sets out the criminogenic needs profile of prolific offenders. It focuses on examining key differences in relation to the type, prevalence and severity of these needs between prolific offenders and non-prolific offenders.

The Ministry of Justice (MoJ) definition of prolific offenders^{[footnote 1]} has been used for this analysis and examines offenders who are prolific at 18 years of age and above (adult prolific offenders and young adult prolific offenders).

Previous analyses of prolific offenders have focussed on their characteristics rather than their identified criminogenic needs.^{[footnote 1]} These statistics aim to provide a better understanding of the criminogenic needs of prolific offenders as a group in comparison to non-prolific offenders so as to inform policies aimed at addressing these needs and reducing reoffending.

This work has been completed by the Ministry of Justice: Better Outcomes through Linked Data (BOLD) - GOV.UK (www.gov.uk).

1.2 Methodology

This report presents findings based on the criminogenic needs scores of all offenders who committed and were convicted of offences between 1 September 2017 and 31 March 2020.

The analysis is based on the linking of data across the following data sources which capture information in England and Wales: The Offender Assessment System (OASys), MoJ Magistrates’ Courts case management system, MoJ Crown Court case management system, and the Police National Computer (PNC).

These datasets do not contain a consistent unique identifier to enable linkage between them. To address this, records were linked at person level using Splink, a probabilistic matching tool developed by MoJ for deduplicating and linking large datasets as well as using deterministic matching.

Limitations outlined in section 3.5 of this report should be considered when interpreting the findings.

1.3 Headline findings

• Broadly, prolific offenders have similar needs to other offenders but in greater prevalence and severity.

• Prolific offenders are typified by having multiple needs, with 73% having 6 to 8 (out of 8) identified criminogenic needs, compared to 49% for non-prolific offenders.

Prevalence

• Prolific offenders have a higher prevalence of criminogenic needs than non-prolific offenders.
• The difference in prevalence of need between prolific and non-prolific is most pronounced for drug misuse (21 percentage points) as well as accommodation (16 percentage points) and employment (16 percentage points).

Severity

• Of those who have an identified need, prolific offenders have a higher severity of that need than non-prolific offenders.

• The contrast in severity of need between prolific and non-prolific offenders is most pronounced for attitudes (8 percentage points), lifestyle & associates (7 percentage points), drug misuse (7 percentage points), and accommodation (6 percentage points).

• The severity of employment need moderately correlates with the severity of lifestyle & associates need (correlation coefficient of 0.32). Of the 92% of prolific offenders who have a lifestyle & associates need, 77% also have an employment need.

Changing needs

• Over the reporting period, almost half (47%) of prolific offenders’ needs remained unchanged. Needs increased for 37% of prolific offenders, while 16% of prolific offenders’ needs decreased.

2. Introduction

2.1 The Better Outcomes through Linked Data (BOLD) programme

BOLD is a cross-government programme led by the Ministry of Justice (MoJ). It was created to show how people with complex needs can be better supported by linking and improving the government data held on them in a safe and secure way.

The BOLD programme demonstrates the value of linking data to improve our evidence on what works in four main areas: supporting victims, reducing homelessness, tackling substance misuse, and reducing reoffending.

You can find more information on the BOLD programme at Ministry of Justice: Better Outcomes through Linked Data (BOLD) - GOV.UK (www.gov.uk).

This publication has been produced as part of the BOLD programme’s Reducing Reoffending demonstrator pilot.

2.2 Background

Prolific offenders represent a minority of offenders (9% between 2000 and 2021) but a majority of convictions (52%).^{[footnote 1]} They therefore contribute disproportionately to reoffending and its associated costs. Given that prolific offenders are a major driver in repeat offending, a better understanding of their complex criminogenic needs can help inform approaches to reduce reoffending for this cohort.

 Previous analyses of prolific offenders have concentrated on their characteristics (demographic, offence- and disposal-related) rather than their identified criminogenic needs.^{[footnote 2][footnote 3]} Other analyses of the identified criminogenic needs of offenders have generally concentrated on all offenders rather than prolific offenders specifically.^{[footnote 4]} These statistics bridge that gap by examining the criminogenic needs of prolific offenders as a group in comparison to non-prolific offenders.

3. Methodology

This report presents findings based on the criminogenic needs scores of all offenders who committed and were convicted of offences between 1 September 2017 and 31 March 2020. Considerations of the reporting period are discussed in section 3.5.1 Dataset timeframe.

3.1 Data sources

The analysis is based on the linking of data across the following data sources which capture information in England and Wales:

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)

Further details on these data sources can be found in the accompanying technical guide.

3.2 Data linking

The datasets used for this analysis do not contain a consistent unique identifier to enable linkage between them. To address this, records from OASys and the courts data sources were linked at person level using Splink, a probabilistic matching tool developed by MoJ for deduplicating and linking large datasets. This makes it possible to track individuals within and across datasets allowing for the identification of repeat offenders and providing a clearer picture of their interactions across the justice system.

We matched each offence within the recording period (1 September 2017 to 31 March 2020) to an OASys criminogenic needs assessment that occurred within 6 months of the offence. This served as an approximate measure to identify the needs at the time of an offence for each offender in our cohort. Refer to the technical guide for further details on the selection of OASys assessments.

Linkage to records from the Police National Computer was then achieved through deterministic matching with the individual’s PNC number (an identifier that is recorded in the PNC and OASys). Offences with out-of-court disposals are recorded in the PNC but are not recorded within the courts databases. Therefore, linkage to PNC data was crucial, as the definition of prolific offending depends on all offending history including cautions and out-of-court disposals. More details on the data linking process and use of Splink can be found in the technical guide. 

3.3 Definition of a prolific offender

This report follows the MoJ definition of prolific offenders^{[footnote 1]} and examines offenders who are prolific at 18 years of age and above (adult prolific offenders and young adult prolific offenders).

3.4 Offender assessments

Data on criminogenic needs are recorded in the Offender Assessment System (OASys) which is an operational database used to assess the risks and needs of eligible offenders in prisons and on probation across England and Wales.

There are eight criminogenic needs linked to offending behaviour which are recorded in a full (Layer 3) OASys assessment. These are accommodation, employability, relationships, lifestyle & associates, drug misuse, alcohol misuse, thinking & behaviour, and attitudes (refer to the technical guide for further details).^{[footnote 4]} For this analysis, assessments of criminogenic needs were available for 79% of the prolific offender cohort and 70% of the non-prolific offender cohort.

3.5 Limitations

There are limitations and assumptions that need to be considered when interpreting the findings included in this report, as follows:

3.5.1 Dataset timeframe

The reporting period covers two and a half years (1 September 2017 and 31 March 2020). The start of the reporting period was constrained by data quality factors. The end of the reporting period was selected in order to exclude impacts associated with both the reunification of probation services in 2021 and the Covid-19 pandemic on data recording and the criminal justice system (CJS) more broadly. This means the most recent needs assessments of prolific offenders are not reported in this publication. However, given existing evidence over recent years indicates that the characteristics of prolific offenders^{[footnote 1][footnote 2][footnote 3]} have remained relatively stable, the assumption here is that these descriptive statistics are unlikely to have changed for more recent years.

3.5.2 OASys coverage bias

We expect that the needs of the non-prolific offenders used in our report are higher than the average non-prolific offender. This is because we have only included non-prolific offenders who have full criminogenic needs assessments for comparative purposes. Completion of full assessments is biased towards offenders with more complex issues, rather than all offenders. Consequently, because the non-prolific population has a lower assessment coverage (70%) than the prolific population (79%), the differences between the needs of prolific offenders and non-prolific offenders presented in this report are conservative and may be an underestimate of the differences.

3.5.3 Coverage of offender contact with the CJS

The findings presented in this report are based on all offenders who committed and were convicted of offences between 1 September 2017 and 31 March 2020 and were assessed as having a criminogenic need within 6 months of these offences. As such, the analysis does not take into account any OASys assessments that offenders may have had prior to and after this period.

4. Findings

4.1 Multiple criminogenic needs

As demonstrated in Figure 1, prolific offenders more commonly have a greater total number of needs than non-prolific offenders.

73% of prolific offenders have 6 or more needs, while 49% of non-prolific offenders have 6 or more needs. The vertical dotted line identifies the region of higher number of needs (6+) where prolific offenders outnumber non-prolific offenders.

Figure 1: Percentages of offenders with different numbers of criminogenic needs

4.2 Prevalence of criminogenic needs

Figure 2 below shows the percentage of offenders who were identified as having each criminogenic need. Results show that every criminogenic need is more prevalent among prolific offenders than among non-prolific offenders.  

The difference in prevalence of need between prolific and non-prolific is most pronounced for drug misuse (21 percentage points), as well as accommodation (16 percentage points) and employment (16 percentage points) needs.

Figure 2: Percentage of offenders who are identified as having a criminogenic need

4.3 Severity of criminogenic needs

Having an identified criminogenic need means that an offender’s score for that need’s set of OASys questions has met a threshold value (detailed in the technical guide). Figure 3 shows that for offenders who have an identified criminogenic need, the scores are higher for prolific offenders than non-prolific offenders. This indicates that prolific offenders have a higher severity of need than non-prolific offenders.

The contrast in severity of need between prolific and non-prolific offenders is most pronounced for attitudes (8 percentage points), lifestyle & associates (7 percentage points), drug misuse (7 percentage points), and accommodation (6 percentage points).

Figure 3: Severity of need (average score as a % of maximum score), where offenders have a defined criminogenic need

4.4 Patterns of need severity

Figure 4 displays the proportion of offenders with each need score across the 8 criminogenic needs. Having an identified criminogenic need (i.e. a yes/no classification) means that an offender’s score for that need’s particular set of OASys questions has met a threshold value (detailed in the technical guide). The vertical dotted line indicates the score threshold at which offenders are defined as having the identified need.

There is notable variation in the way in which severity of offender needs is distributed. Severity of needs tends to be higher for prolific offenders compared to non-prolific offenders. In particular:

• The level of employment need was fairly spread out for both prolific and non-prolific offenders. A larger proportion of non-prolific offenders (36%) have no employment need (score of 0-2) compared to prolific offenders (20%), who tend to have greater levels of employment needs.

• Most prolific and non-prolific offenders either have no accommodation need or the most severe accommodation need. This is likely down to the scoring rule where a default score of 8 is given if an individual is of no fixed abode (NFA) or in transient accommodation. A higher proportion of non-prolific offenders (44%) have no accommodation need (score of 0-1) compared to prolific offenders (29%). In contrast, a greater proportion of prolific offenders (43%) compared to non-prolific offenders (28%) are of no fixed abode or in transient accommodation (score of 8).

• A large proportion of offenders had no drug misuse need (score of 0-1), with the percentage being higher for non-prolific offenders (47%) compared to prolific offenders (26%). The proportion of offenders with intermediate to high severity need (scores of 4+) is higher for prolific offenders (58%) than non-prolific offenders (35%).

• The level of alcohol misuse is similar for both prolific and non-prolific offenders, with the greatest proportion having a need score of zero (33% for prolific and 42% for non-prolific offenders). The remaining offenders have varying levels of need spread out from low to high.

• For both prolific and non-prolific offenders, relationships, lifestyle & associates and attitudes needs tended to be of intermediate severity.

• Needs in relation to thinking and behaviour tend to be intermediate to high in their severity amongst the offender cohort overall. A higher proportion of non-prolific offenders (26%) have intermediate levels of need (score of 4) compared to prolific offenders (22%). In contrast, a higher proportion of prolific offenders (61%) have higher severity of need (score of 5-8) compared to non-prolific offenders (50%).

Figure 4: Distribution of need scores - Percentage of prolific and non-prolific offenders with each score for the eight criminogenic needs. The dotted line indicates the score threshold above which offenders are identified as having that criminogenic need.

4.5 Combinations of criminogenic needs among prolific offenders

The Kendall’s Tau correlation coefficient was used to quantify the correlation between need scores for pairs of criminogenic needs, as displayed in figure 5. A maximum correlation coefficient of 1 (or -1) represents perfect correlation (or perfect inverse correlation) which is the closest possible association between need scores. A correlation coefficient of zero represents no correlation and no association between the needs.

Figure 5: Heat map of the strength of correlation between pairs of criminogenic needs

Most pairings show weak to moderate strength of correlation. However, of those where the correlation is more moderate, several needs pairs stand out. The pairings with a correlation of 0.32 and greater are listed in Table 1.

Employment need is moderately associated with the lifestyle & associates need (correlation coefficient of 0.32). 92% of prolific offenders have a need in lifestyle & associates, 80% have a need in employment, and 77% have both criminogenic needs.

The pairings between attitudes, thinking and behaviour, and lifestyle & associates are also moderately correlated (correlation coefficients ranging from 0.37 to 0.47). 78% of prolific offenders have all three of these needs.

Table 1: Correlation coefficients between pairings of criminogenic need

Need 1	Need 2	Kendall’s Tau Correlation Coefficient
Employment	Lifestyle & Associates	0.32
Attitudes	Lifestyle & Associates	0.40
Thinking & behaviour	Lifestyle & Associates	0.37
Attitudes	Thinking & behaviour	0.47

4.6 Changes in prolific offenders’ needs over time

We measure the change in the criminogenic needs of prolific offenders over time by comparing each offender’s latest needs assessment to their first during the reporting period. 40% of prolific offenders in the data only have one OASys assessment and are thus excluded from this analysis. Furthermore, the analysis does not take into account any assessments which may have taken place outside of the reporting period.

Table 2 describes how prolific offenders’ latest overall needs scores have changed relative to their first assessments. The most common outcome was that prolific offenders experienced no change (47%) in their overall needs score. More prolific offenders experienced an increase (37%) than a decrease (16%) in score.

Table 2 Change of overall need for prolific offenders

Change	Number of offenders	Percentage of offenders (%)
Increase	3,800	36.8
Decrease	1,631	15.8
No change	4,885	47.4

Table 3 breaks down the change in overall need by offenders’ sex. The same proportion of females and males (47%) have no change in overall need.

Table 3 Change of overall need by prolific offenders’ sex

Change	Female (%)	Male (%)
Increase	35.5	37.0
Decrease	17.1	15.6
No change	47.4	47.4

When broken down by need category (Table 4), we observe that accommodation need changes more (changes – either increases or decreases – for 32% of prolific offenders) than other needs, and relationships need changes the least (changes for 17% of prolific offenders).

Table 4: Change of each need for prolific offenders

OASys category	Increase (%)	Decrease (%)	No change (%)
Accommodation	18.7	12.7	68.6
Alcohol misuse	11.7	7.4	80.9
Attitudes	21.7	7.0	71.2
Drug misuse	18.5	8.8	72.7
Employment	12.1	6.4	81.5
Lifestyle & associates	15.8	4.4	79.8
Relationships	13.5	3.6	82.9
Thinking & behaviour	14.5	4.7	80.8

5. Contacts

Media contacts

Press enquiries should be directed to the Ministry of Justice press office.

020 3334 3536

Statistical contacts

Other enquiries about these statistics should be directed to Data at the Ministry of Justice: RR-pilot-BOLD@justice.gov.uk  

Ministry of Justice Data Directorate: BOLD Programme

10th Floor

102 Petty France

London SW1H 9AJ

---

1. Official Statistics - Characteristics of Prolific Offenders, 2000-2021 - GOV.UK (2023) ↩ ↩² ↩³ ↩⁴ ↩⁵

2. Official Statistics - Characteristics of prolific offenders 2000-2016 - GOV.UK (2019) ↩ ↩²

3. Experimental Statistics - Criminal pathways of prolific offenders – part 1 (2017), part 2 (Feb 2019), part 3 (May 2019) – GOV.UK (these contain some limited information on prolific offenders’ educational and employment needs, but not criminogenic needs). ↩ ↩²

4. Official Statistics - Identified needs of offenders in custody and the community from the Offender Assessment System, 30 June 2021 - GOV.UK ↩ ↩²