Turning Point NW-London: A replication randomised trial of police-led diversion in North West London

Randomised Control Trial Status


Trial Hypothesis

​First time offenders not previously convicted at court, or repeat offenders with no more than three previous court convictions (in the last two years if aged 22 and above, or one year if 21 and under), whom police would otherwise caution or charge for prosecution, can be more effectively dealt with by police-led offender management than formal criminal justice processing, if subject to a condition of the certainty of  prosecution in the event of reoffending or breaking an agreed “contract” about their conduct. 

Geographical area

​London boroughs of Barnet, Brent and Harrow, covered by the North West Basic Command Unit (NW BCU) of the Metropolitan Police Service.

Research Institution / Organisation

University of Cambridge

In Collaboration With

Metropolitan Police Service

Project Start Date

July 2018

Participants - inclusion criteria

​Offenders eligible for caution or a charge that would not receive a custodial sentence, have no more than three previous court convictions in the last 2 years (aged 22 years and above) or 1 year (aged 21 years and below) and have not committed an excluded offence e.g. DA, hate crime, domestic burglary, weapons.  No formal admission of guilt is required.

Target Sample Size


Study Design

​Two arm RCT. Randomisation of offenders comparing treatment to control. Random assignment stratified by three age cohorts (under 18s, 18-21, 22 and over).


​Treatment subjects attend a criminogenic needs assessment with a police offender manager within 48-60 hours of being offered Turning Point.  They must adhere to a 4 month, tailored, contract of conditions (restorative, rehabilitative and if necessary prohibitive) designed to make amends with victims and address root causes of offending. E.g. Apology letters, financial compensation, attendance at; RJ conference, Victims Awareness Course, drug, alcohol, mental health, anger management, employment and training support services and volunteer work.  Victims are consulted on contract content. Subjects are swiftly prosecuted for their original offence if they breach or reoffend but are not criminalised if they successfully complete their contract (i.e. not charged or cautioned; recorded as Outcome 22 for HOCR).  Control subjects are cautioned/charged as normal.

Outcome Measures

  1. Reoffending: Prevalence, frequency, time-to-failure and Cambridge Harm Index level of rearrests and reconvictions as compared between the treatment and control group, six months, 12 months and 24 months post randomisation.  Sub-group analysis by age, gender, ethnicity, trigger offence type, 1st time or repeat offender.

  2. Victim satisfaction: Interview surveys with victims of offenders in both treatment and control groups.

  3. Process analysis:  Detailed participant-observer case study reporting covering the RCT’s implementation and contextual issues, including analysis of treatment integrity, fidelity and dosage.

  4. Production of a ‘How-to-do-it Manual’ for further replication or introduction as policy.

Date due for completion

July 2024

Date last updated

Thursday 04 March 2021
Return to Research Map