Alemi et al., 2006
US | Intervention: Seamless combination of probation and treatment
Comparator: Traditional probation | Clients on probation and requiring substance misuse treatment in Northern Virginia and Maryland
Study design: RCT plus decision-analytic modelling
Source of effectiveness data: single study (N = 272)
Source of resource use data: Self-report and official probation office records
Source of unit cost data: National estimates | Cost-analysis | Costs:
Treatment; arrest and court processing; incarceration; homeless shelter; hospitalisation
Cost results:
Seamless probation: $38.84 per follow-up day per client
Traditional probation: $21.60 per follow-up day per client | Seamless probation was $6,293 more expensive than traditional probation per client per year | Perspective: US tax payers
Currency: US $
Cost year: 2004
Time horizon: 2.75 years
Discounting: Not reported
Internal validity: 17/4/2 |
Barnoski, 2004
US | Intervention: Functional family therapy for juvenile offenders
Comparator: Untreated control group | Moderate or high-risk juvenile offenders (age 13–17)
Study design: Prospective observational study
Source of effectiveness data: Single study (N = 700)
Source of resource use data: Washington State Juvenile Court Assessment Programme
Source of unit cost data: NA | Cost-benefit analysis | Costs:
Functional family therapy treatment programme; criminal justice costs
Cost results:
Functional family therapy: $2,100 per family
Outcomes:
18-month recidivism rates Total taxpayer and crime victim costs avoided
Effectiveness results:
Functional family therapy: 17% recidivism rate
Control: 32% recidivism rate $22,448 costs avoided | Benefit-cost ratio of $10.69 | Perspective: societal and criminal justice system
Currency: US $
Cost year: Not reported
Time horizon: 18 months
Discounting: Not conducted
Internal validity: 21/6/8 |
Caldwell et al., 2006
US | Intervention: Intensive juvenile corrective service treatment programme
Comparator: Usual juvenile corrective service intervention | Unmanageable incarcerated delinquent boys
Study design: Quasi- experimental design
Source of clinical effectiveness: single study (N = 202)
Source of resource use: database of public circuit court records
Source of unit cost: published literature | Cost-benefit analysis | Costs: Cost of intervention, juvenile institution care, arrest, prosecution and defence.
Treatment group cost: $173,012/youth
Comparison group cost: $216,388/youth (P <0.05)
Outcomes: All offences, felony offences, violent offences
Number of offences charged:
Treatment group: 1.09
Comparison group: 2.49 (p <0.05)
Violent offence: Treatment group: 0.25 Comparison group: 0.85 (p < 0.001)
Felony offence: Treatment group: 0.48 Comparison group: 0.89 (p < 0.05) | Intensive juvenile treatment dominated the usual treatment of juvenile corrective service
Cost-benefit ratio: 1 to 7.18 | Perspective: Public sector
Currency: US$
Cost year: 2001
Time horizon: 4.5 years
Discounting: not conducted
Internal validity: 22/1/12 |
Crane et al., 2005
US | Interventions:
In-office family therapy In-home family therapy
Comparator: Untreated control group | Youths who had received services for conduct disorder between May and October 1999
Study design: Retrospective longitudinal study of Kansas Medicaid claims forms
Source of effectiveness data: single study (N = 3,753)
Source of resource use data: Medicaid claims records
Source of unit cost data: Medicaid | Cost-analysis | Costs: Intervention treatments; pharmacy; hospital and professional services
Cost results:
Interventions:
$11,116 per child $1,622 per child
Control: $16,260 per child | Both interventions resulted in significant net savings (p < .0001) | Perspective: Health Insurance
Currency: US $
Cost year: NA
Time horizon: 30 months
Discounting: No
Internal validity: 8/10/5 |
Davidson et al., 2008
UK | Intervention: CBT plus treatment as usual (TAU) in a community setting
Comparator: TAU alone | Adult men with diagnosis of antisocial personality disorder and acts of aggression 6 months before study
Study design: RCT
Source of effectiveness data: Single study (N = 52)
Source of resource use data: Directly from study case-notes
Source of unit cost data: National sources | Cost-analysis | Costs: Psychiatric, A&E, primary care and social services
Cost results: CBT: £1,295 per participant (+£1,300 per participant for CBT sessions)
TAU: £1,133 per participant | CBT costs more per participant than TAU alone over 12 months | Perspective: NHS
Currency: £
Cost year: 2007
Time horizon: 12 months
Discounting: N/A
Internal validity: 9/7/7 |
Dembo et al., 2000a
US | Intervention: Family empowerment intervention – families receive home-based meetings from a clinically trained paraprofessional
Comparator: Extended services intervention – families receive monthly phone contacts | Juvenile offenders (11–18 years) and their families
Study design: Prospective longitudinal study
Source of effectiveness data: Single study (N = 303)
Source of resource use data: Florida
Department of Juvenile Justice
Source of unit cost data: Local estimates | Cost-analysis | Costs: Interventions; recidivism (arrests, state attorney, public defender, judicial and department of juvenile justice costs)
Cost results (based on 3,600 diversion cases):
Initial year costs:
Family empowerment intervention: $5,295,600
Extended services intervention: $6,980,400
New arrest costs:
Family empowerment intervention: $4,956,084
Extended services intervention: $7,957,656 | Net saving of $4,686,372 per 3,600 youths ($1,302 per case) | Perspective: Criminal justice system
Currency: US $
Cost year: Not reported
Time horizon: 2 years
Discounting: No
Internal validity: 6/10/7 |
Dretzke et al., 2005
UK | Intervention:
Three types of parent-training/education programmes:
Group community-based Group clinic-based Individual home- based
Comparator:
No treatment | Children with conduct disorder aged up to 18 years
Study design: Decision-analytic modelling
Source of clinical effectiveness data:
Systematic review and meta-analysis (clinical effectiveness between parent-training/education programmes); hypothetical rates (parent-training/education programmes versus no treatment)
Source of resource use data: Expert opinion supported by published literature
Source of unit costs: national sources | Cost-minimisation analysis (comparison across the three types of parent-training/education programmes) and secondary cost-effectiveness analysis (all parent-training/education programmes versus no treatment) | Costs:
Intervention costs: staff, super- vision, travelling, crèche, course packs, room hire
Cost results:
Cost per family: Group community-based parent-training/education programmes: £899 (assuming eight families per group)
Group clinic-based parent-training/education programmes: £629 (assuming eight families per group)
Individual home-based parent- training/education programmes: £3,839
No treatment: 0
Outcomes:
Child behaviour- related measures (Hypothetical) levels of response to treatment and improvement in children’s HRQoL expressed in QALYs
Effectiveness results:
No significant differences in outcome between the three types of parent-training/education programmes
Hypothetical 5%, 10% and 50% response rates; hypothetical 0.01, 0.025%, 0.1 and 0.2 improvement in QALYs | Group clinic- based parent- training/education programmes dominate the two other types of parent-training/education programmes
ICERs of parent-training/education programmes versus no treatment assuming a 80% uptake:
50% response rate
Group community-based parent-training/education programmes: £1,438 per responder
Group clinic-based parent-training/education programmes: £1,006 per responder
Individual home-based parent-training/education programmes: £6,143 per responder
0.2 improvement in QALYs
Group community-based parent-training/education programmes: £4,495/QALY
Group clinic-based parent- training/education programmes: £3,144/QALY
Individual home-based parent-training/education programmes: £19,196/QALY | Perspective: NHS
Currency: UK £
Cost year: 2003
Time horizon: 10 weeks
Discounting: N/A
Internal validity: 20/6/9 |
Foster et al., 2006
US | Intervention: Fast-track multi-component intervention
Comparator: Untreated control group | Youths screened for classroom conduct problems in four sites in the US
Study design: Multi-centre RCT
Source of effectiveness data: Single study (N = 891)
Source of resource use data: Annual budget records
Source of unit cost data: Annual budget records | Cost-effectiveness analysis | Costs: Fast-track intervention
Cost results:
Intervention: $58,283 per child
Control: $0 per child
Outcomes:
Diagnosis of conduct disorder – cases of conduct disorder averted Index Criminal Offences scale – index criminal offence avoided Interpersonal Violence scale – acts of interpersonal violence avoided
Effectiveness results: Not reported |
ICER of $3,481,433 per case of conduct disorder averted ICER of $423,480 per index crime averted ICER of $736,010 per act of interpersonal violence averted
| Perspective: State department of mental health
Currency: US $
Cost year: 2004
Time horizon: 10 years
Discounting: Yes
Internal validity: 20/9/6 |
Griffith et al., 1999
US | Intervention: In-prison therapeutic community treatment for drug misuse
Comparator: Untreated comparison group | Male prisoners with history of substance misuse
Study design: Retrospective observational study
Source of effectiveness data: Single study (N = 394)
Source of resource use data: Texas Department of Criminal Justice Institutional Division database
Source of unit cost data: Published literature | Cost- effectiveness analysis | Costs: In-prison therapeutic community treatment; parole and aftercare
Cost results:
Low-risk aftercare completers: $21,860 per patient
High-risk aftercare completers: $22,208
High-risk untreated comparison group: $18,402 per patient
Low-risk untreated comparison group: $17,928 per patient
Outcomes: % not reincarcerated
Effectiveness results:
Low-risk aftercare completers: 78%
High-risk aftercare completers: 74%
Low-risk untreated comparison group: 71%
High-risk untreated comparison group: 45% | Low-risk comparison: ICER of $494 per 1% reduction in reincarceration
High-risk comparison: ICER of $165 per 1% reduction in reincarceration | Perspective: criminal justice system
Currency: US $
Cost year: 1999
Time horizon: 3 years
Discounting: Not conducted
Internal validity: 21/6/8 |
Masse & Barnett, 2002
US | Intervention: Provision of intensive preschool services to children in low-income families
Comparator: Untreated control group | 1–5 year olds (born between 1972 and 1977) at risk of retarded intellectual and social development and followed-up at age 21
Study design: RCT
Source of effectiveness data: Single study (N = 112)
Source of resource use data: Programme sponsor records and national statistics
Source of unit cost data: local and national estimates | Cost-benefit analysis | Costs: Programme costs in child development centre setting and public school setting
Cost results:
Child development centre: $35,864
Public school: $41,916
Outcomes (programme differentials/net benefits):
Future earnings; maternal earnings; education costs; healthcare costs
Effectiveness results (discount rate of 3%):
Total benefits: $135,546 | Net benefits:
Child development centre: $99,682 per child
Public school: $93,630 per child | Perspective: Societal
Currency: US $
Cost year: 2002
Time horizon: 16–20 years
Discounting: Yes
Internal validity: 15/11/9 |
McCollister et al., 2003
US | Intervention: Work release therapeutic community and aftercare programme for criminal offenders (Delaware CREST Outreach Centre)
Comparator: No standard work release group | Male prisoners with history of substance misuse
Study design: Randomised ITT study
Source of effectiveness data: Single study (N = 836)
Source of resource use data: Drug Abuse Treatment Cost Analysis Program (DATCAP)
Source of unit cost data: Published literature | Cost-effectiveness analysis | Costs: Intervention costs including work release and aftercare programmes
Cost results:
Intervention: $1,937
Control: $0
Outcomes: Number of incarceration days avoided during follow-up
Effectiveness results:
Intervention: 74 days incarcerated
Control: 104 days incarcerated | ICER of $65 per avoided incarceration day | Perspective: prison sector
Currency: US $
Cost year: NA
Time horizon: 18 months
Discounting: Not conducted
Internal validity: 16/9/10 |
McCollister et al., 2003a; 2004
US | Intervention: In-prison therapeutic community and aftercare programme (Amity programme, California)
Comparator: No treatment control group | Male prisoners with history of substance misuse
Study design: Randomised ITT study
Source of effectiveness data: Single study (N = 576)
Source of resource use data: Drug Abuse Treatment Cost Analysis Program (DATCAP)
Source of unit cost data: Published literature | Cost- effectiveness analysis | Costs: Intervention costs; hospital inpatient and outpatient visits; methadone/detoxification drug maintenance; self-help/12-step programmes
Cost results: 12 months
Intervention: $4,112 per
patient Control: $0 per patient
5 years
Intervention: $7,041 per patient
Control: $1,731 per patient
Outcomes: Number of incarceration days avoided during follow-up
Effectiveness results: 12 months
Intervention: 91 days incarcerated
Control: 142 days incarcerated
5 years
Intervention: 544 days incarcerated
Control: 626 days incarcerated | 12-month results: ICER of $80 per avoided incarceration day
5-year results: ICER of $65 per avoided incarceration day | Perspective: health service and prison sector
Currency: US $
Cost year: 2000
Time horizon: 12 months and 5 years
Discounting: Not conducted
Internal validity: 19/8/8 |
| Myers et al., 2000 US | Intervention: Project Back-on- Track multi-component intervention
Comparator: Untreated community control group | Early-career juvenile offenders (9–17 years)
Study design: Observational case-control
Source of effectiveness data: Single study (N = 60)
Source of resource use data: Not reported
Source of unit cost data: Published estimates | Cost-analysis | Costs: Back-on-Track treatment; criminal offences
Cost results:
Intervention: $600 per youth
Control: $600 per youth
Outcomes: Total number of crimes (total costs of crimes)
Effectiveness results:
Intervention: 3 ($9,000)
Control: 21 ($63,000) | Net saving of $1,800 per youth receiving Back-on-Track treatment | Perspective: Criminal justice system
Currency: US $
Cost year: Not reported
Time horizon: 12 months
Discounting: N/A
Internal validity: 8/8/7 |
Nores et al., 2005
US | Intervention: High/Scope Perry preschool programme
Comparator: Untreated control group | 3–4 year old African-American children from Michigan in 1960s followed up to age 40 years
Study design: RCT
Source of effectiveness data: Single study (N = 123)
Source of resource use data: Interviews and official records
Source of unit cost data: Published national estimates | Cost-benefit analysis | Costs: Programme costs
Cost results: Intervention: $15,166
Outcomes (programme differentials/net benefits): Child care costs; education costs; tax contributions; crime costs; welfare payments (receipts)
Effectiveness results (discount rate of 3%):
Participant perspective: $49,190 per participant General public perspective: $195,621 per participant Societal perspective: $244,645 per participant
| Net benefits:
$49,190 per participant $180,455 per participant $229,645 per participant
| Perspective: Participants/general public/society
Currency: US $
Cost year: 2000
Time horizon: 36–37 years
Discounting: Yes
Internal validity: 23/4/8 |
Reynolds et al., 2002
US | Intervention: Chicago Child-Parent centres
Comparator: TAU comparison group | Low income children (3–9 years) followed up to age 20 years
Study design: Longitudinal cohort study
Source of effectiveness data: Single study (N = 1,539)
Source of resource use data: Chicago Public Schools budget
Source of unit cost data: Local estimates | Cost-benefit analysis | Costs: Chicago Child-Parent centre staff; administration; operations and maintenance; family and community support; transportation and community services; school district services
Cost results: Chicago Child-Parent programme: $6,692 per child
Outcomes (programme differentials/net benefits): Child care costs; special education costs; child welfare savings; abuse/neglect victim savings; juvenile justice/crime victim savings
Effectiveness results (discount rate of 3%):
Participant perspective: $21,988 per child Taxpayer/crime victim perspective: $19,079 per child Societal perspective: $41,067 per child
| Net benefits:
$15,296 per child $12,389 per child $34,375 per child
| Perspective: Participants/taxpayer/crime victim/society
Currency: US $
Cost year: 1998
Time horizon: 15–17 years
Discounting: Yes
Internal validity: 22/6/7 |
Robertson et al., 2001
USA | Intervention: Intensive supervision and monitoring CBT
Comparator: Regular probation | Children aged 11–17 years who committed delinquent activity or status offences
Study design: quasi- experimental design (N = 293)
Source of data for clinical outcomes: Patients’ data (N = 153)
Source of resource use data: Patient questionnaires and court records
Source of unit cost: Not reported | Cost-benefit analysis | Costs: Increase in spending (within justice system) necessary to support and maintain the intervention programmes
Intensive supervision and monitoring : NA
CBT: $1,493
Outcomes: Expected short-run reduction in justice system expenditures due to intervention
Intensive supervision and monitoring : NA
CBT: $2,928
Cost/patient:
Intensive supervision and monitoring: $927 CBT: −$2927 | CBT programme resulted in net reduction in local justice system expenditures of $1,435 per offender
Cost-benefit ratio: 1 to 1.96
Intensive supervision and monitoring programme did not result in significant difference in justice system expenditures | Perspective: public sector
Currency: US $
Cost year: 2001
Time horizon: 18 months
Discounting: NA
Internal validity: 12/5/18 |
Zhang et al., 2006
US | Intervention: Preventing Parolee Crime Programme -multiple community–based services for parolees
Comparator: Untreated control group | Californian parolees within 12–36 months of release
Study design: Observational study
Source of effectiveness data: Single study (N = 239,919)
Source of resource use data: California Department of Corrections’ database
Source of unit cost data: California Department of Corrections’ internal accounting documents | Cost-analysis | Costs: Preventing Parolee Crime Programme; parole supervision; daily incarceration costs
Cost results: Cost-saving factors (incarceration costs avoided): $66,166,198 Parole/Preventing Parolee Crime Programme expenditures: $45,087,182 | Net savings: $21,079,016
Cost-benefit ratio: $1:$1.47 | Perspective: societal and criminal justice system
Currency: US $
Cost year: 2000/01 and 2001/02
Time horizon: 2 years
Discounting: Not conducted
Internal validity: 11/7/5 |
