Transcript The Drug Harm Index

Collecting and interpreting data
on the outcome/impact in the
UK
Anna Richardson
Crime Drugs and Alcohol Research
Home Office
June 2010
Overview
 Context
-
UK election 6 May
-
Coalition government
-
What next?
 Measurement of the previous Drug Strategy
 Evaluating the Drug Strategy through performance
monitoring
 Evaluating the Drug Strategy through the research
framework
Drug Strategy 2002
 4 Strands:
Prevent more young people from
becoming problematic drug users
Reduce the supply of Class A drugs
Reduce drug-related crime
Increase the number of problematic
drug users retained in / completing
treatment
Measuring the 2002 Drug Strategy
 PSA 4: Reduce the harm caused by illegal drugs (as
measured by the Drug Harm Index encompassing
measures of the availability of Class A drugs and drug
related crime) including substantially increasing the
number of drug misusing offenders entering treatment
through the Criminal Justice System.
Rationale for a drug harm index
 Overall aim of Drug Strategy:
“Reduce the harm that drugs cause to society - communities,
individuals and their families”
 Measures for individual strands of the strategy might not capture
reductions in all harms and might move in different directions
 The Drug Harm Index aims to capture in one simple measure the
various harms that society suffers as a consequence of problematic
drug use…
 …and therefore provides a measure of progress against the
overarching aim of the Strategy
Overview of the Drug Harm Index
 Captures a substantial basket of harms  all harm caused by illegal
drugs (e.g. productivity impact)
 Key Inclusion Criteria• Is it something the Drug Strategy can impact upon?
• Are all dimensions of the Strategy reflected?
Conceptual
model
• Does a data series already exist (to provide 1998 baseline)?
• Are the data timely/reliable/frequent?
Indicator
list
 Social and economic costs used to convert diverse indicators into a
common “currency” and show relative importance/provide basis for
index – May show this
Conceptual Model
DRUG HARMS
Drug Trafficking
INTERVENTIONS
DIP
YP PREVENTION
SUPPLY
REDUCTION
OUTPUTS
 #s in treatment
Fewer future
PDUs
 Availability
OUTCOMES
Fear of Drug-Crime
Less crime/safer
communities
Acquisitive Crime
Fewer health
problems
New HIV
New Hepatitis B
Drug Deaths
Mental Health
Overdose
Choice of indicators
 Criteria
• Harms must be drug-related
• Expected to be influenced by policy
• Cover all strands of policy
• Robust national indicators
• Available at least annually
• Time series back to 1998
List of harms included in the DHI
 Drug-related acquisitive crime
•
•
•
•
•
•
•
Burglary (domestic & commercial)
Theft of vehicle (domestic &
commercial)
Theft from vehicle (domestic &
commercial)
Bike theft
Other theft
Robbery
Shoplifting
 Community harms
•
•
Community perceptions of drug
use/dealing
Drug dealing offences
 Health impacts
•
New HIV cases due to intravenous
drug use (IDU), including those
infected through heterosexual sex with
someone who contracted the disease
through IDU
•
New Hepatitis B cases due to
intravenous drug use
•
New Hepatitis C cases due to
intravenous drug use
•
Drug-related deaths
•
Drug-related mental health and
behavioural problems
•
Drug overdoses
•
Drug-related neonatal problems
How do we combine data series with
different units/orders of magnitude into a single
index value?
1994 1995 1996 1997 1998 1999 2000 2001 2002
Perceptions of drug nuisance (%) 22
21
21
23
25
29
33
31
32
Recorded Trafficking offences
18340 21272 22122 23153 21788 19956 19820 19583 22324
Drug-related crime - OI
7916 11285 16110 19636 26055 26822 28755 30895
Drug Deaths
1002 1100 1247 1302 1459 1568 1663 1623
Mental & behavioural problems
4166 4638 5315 5529 5234 5151 5041 4651
Overdoses
4967 6866 7429 7505 7849 7956 7536 8167
HIV cases due to IDU
167
183
173
168
130
112
109
128
98
Hepatitis B
102
147
166
192
251
231
201
103
136
Hepatitis C
389
800
1198 1414 1761 1999 1454 963
963
Drug-related neonatal problems
1063 1075 1146 1136
Total harm
# ‘95
# ‘96
# ‘97
# ‘98
# ‘99
# ‘00
# ‘01
# ‘02
Creating the Index
 The variables could be statistically ‘normalised’ and
added together, but this would not reflect the relative
importance that society places on different harms
 We need an objective means of weighting the drug
harms
 We chose to use the estimates of the economic and
social costs of class A drug use
 Each harm can be valued in terms of the total costs
incurred by society, and these costs are used to
weight the individual harms within the overall index.
Constructing the index
The index combines two types of data:
• Unit social & economic costs for each harm
• Volume series for each harm (baseline 1998)
Steps:
1) Determine total costs (harm) for each year
= sum over j [volume of harm j * unit cost of harm j]
2) Use total cost to construct a weight for each harm in each year
= (total cost for harm j in year i) / (total cost in year i)
3) Determine the annual growth in each harm
= Ln(volume of harm year i) - Ln(volume of harm year i-1)
4) Determine the weighted growth in each harm in each year
= sum over j [weight (j) * volume growth series (j)]
5) Construct final index (year 1 = 100)
= Antilog growth series and make relative to base year
Results
The Drug Harm Index
120
112.0
DHI value (1998 = 100)
100
117.0
115.8
110.3
104.8
100.0
87.9
80
60
40
20
0
1998
1999
2000
2001
2002
2003
Year
2004
2005
2006
2007
2008
Results
The Drug Harm Index
120
112.0
DHI value (1998 = 100)
100
117.0
115.8
110.3
104.8
100.0
87.9
80
60
Updated Drug
Strategy
December 2002
Drug Strategy
April 1998
Drug Interventions
Programme
April 2003
40
20
0
1998
1999
2000
2001
2002
2003
Year
2004
2005
2006
2007
2008
Understanding changes in the DHI
Example:
- The DHI changed by -5.9% (2004-2005)
Main contributors (growth rate * weight)
Drug Deaths
Robbery
Commercial Burglary
Domestic Burglary
Theft from vehicle
Hepatitis C
Growth Weight
0.07 27.9%
0.10 13.3%
-0.20 12.7%
-0.19 13.0%
-0.24
4.6%
-0.29
3.2%
Impact
2.0%
1.3%
-2.6%
-2.5%
-1.1%
-0.9%
The Drug Harm Index …
Cannot
 Measure total harms
 Indicate which
interventions have had an
impact
 Provide an overall national
 Predict changes in harms
picture
 Provide a local picture
Can
 Monitor change over time
 Show which harms are
changing the most
Issues / Limitations
 Can only include harms for which there are robust
indicators
 Drug-related crime has a large weighting within the DHI, but
data for this element must be estimated
 Some of the indicators are only available with long time lags
and may be subject to change
 How quickly can an impact be expected? e.g. prevention
interventions aiming to reduce harm in the future
 A black box - any change is a combination of many small
changes and can relate to change in either.
Further information can be found at:

MacDonald, Z. et al (2005) Measuring the harm from illegal drugs using the Drug Harm
Index Home Office OLR 24/05 (http://www.homeoffice.gov.uk/rds/pdfs05/rdsolr2405.pdf)

MacDonald, Z. et al (2006) Measuring the harm from illegal drugs using the Drug Harm
Index – an Update Home Office OLR 08/06
(http://www.homeoffice.gov.uk/rds/pdfs06/rdsolr0806.pdf)

Godfrey, C. et al (2002) The economic and social cost of Class A drug use in England
and Wales Home Office Research Study 249
(http://www.homeoffice.gov.uk/rds/pdfs2/hors249.pdf)

Gordon, L. et al (2006) The economic and social costs of Class A drug use in England
and Wales, 2003/04 in Singleton, N et al (eds) Measuring different aspects of problem
drug use: methodological developments Home Office OLR 16/06
(http://www.homeoffice.gov.uk/rds/pdfs06/rdsolr1606.pdf)

Crime and Drugs Analysis and Research (2009) Measuring the harm from illegal drugs:
a summary of the Drug Harm Index 2006 Home Office Research Report 13
http://rds.homeoffice.gov.uk/rds/pdfs09/horr13.pdf
Drug Strategy 2008
 4 strands:
Protecting communities through robust enforcement to
tackle drug supply, drug-related crime and anti-social
behaviour
Preventing harm to children, young people and families
affected by drug misuse
Delivering new approaches to drug treatment and social
re-integration
Public information campaigns, communications and
community engagement
Measuring the 2008 Drug Strategy
 Strand 1: Protecting communities through robust enforcement to
tackle drug supply, drug-related crime and anti-social behaviour.
 PSA 25 indicator 3:
• The rate of drug-related offending
• % of public who perceive drug use/ dealing to be a problem in their area
 National Indicators
• NI16 – serious acquisitive crime rate;
• NI17 – perceptions of asb;
• NI18 – adult re-offending rates for those under probation supervision;
• NI21 – dealing with local concerns about asb and crime by local council
and police;
• NI30 – re-offending rate of PPOs;
• NI38 – drug-related (Class A) offending rate
Strand 2: Preventing harm to children, young
people and families affected by drug misuse.
 PSA 14 indicator 3:
• the proportion of young people frequently using illicit drugs, alcohol
or volatile substances;
 PSA 25 indicator 1:
• the number of drug users in effective treatment
 National Indicators
•
•
•
•
•
NI110 – young people’s participation in positive activities;
NI111 – first-time entrants to the YJS aged 10–17;
NI114 – rate of permanent exclusions from school;
NI115 – substance misuse by young people; and
NI117 – 16–18-year-olds who are not in education, training or
employment.
Strand 3: Delivering new approaches to drug
treatment and social re-integration.
 PSA 25 indicator 1:
• The number of drug users in effective treatment
 National indicators:
•
•
•
•
NI40 – drug users in effective treatment;
NI120 – all-age all-cause mortality rate;
NI141 – number of vulnerable people achieving independent living;
NI143, 145, 147, 149 – socially excluded adults living in settled and
suitable accommodation;
• NI144, 146, 148, 150 – socially excluded adults in employment,
education or training; and
• NI152 – working-age people on out-of-work benefits.
Strand 4: Public information campaigns,
communications and community engagement.
 PSA 14 indicator 3:
• the proportion of young people frequently using illicit drugs, alcohol
or volatile substances;
And,
 PSA 25 indicator 1:
• the number of drug users in effective treatment
Further information can be found at:
 Home Office 2010 Drug-misusing offenders: results from
the 2008 cohort for England and Wales, Home Office
http://rds.homeoffice.gov.uk/rds/pdfs10/misc0210.pdf
Research framework to support
evaluation – The CGRPD
 Established following DS commitment to improve drugs evidence
 Overall objective to:
• Provide a research strategy that will provide the foundations,
direction and guidance for collaboration within government, and
between government and other stakeholders, in the short and long
term.
 Sub-aims – to develop:
• a shared vision across government of the future direction of drugs
research;
• An overview of existing and planned government research into
drugs;
• A list of prioritised drug research needs for short and longer term;
• A delivery plan for future drugs research