Transcript No Slide Title

IS THERE A CAUSAL RELATIONSHIP BETWEEN
SUBSTANCE ABUSE AND PSYCHIATRIC ILLNESS?
Dr Martin Frisher
Department of Medicines Management
Keele University, Staffordshire, UK
[email protected]
Manchester, Tuesday 24th November 2005
SLIDE 1 of 34
UK Department of Health
Comorbidity in Primary Care




Study 1: Estimate the occurrence of comorbidity in
primary care in England and Wales.
Study 2: Analyse the time sequence between mono
and dual diagnosis.
Study 3: Compare health care utilisation by mono,
comorbid and control patients.
Study 4: Confirm whether secondary care diagnoses
are recorded in primary care.
SLIDE 2 of 34
Papers from the study
Frisher M, Collins C, Millson D, Crome I, Croft P.
Prevalence of comorbid psychiatric illness and
substance misuse in primary care in England and
Wales Journal of Epidemiology and Community Health
2004; 58:1036-1041
Frisher M, Crome I, Macleod J, Millson D, Croft P.
Substance misuse and psychiatric illness:
prospective observational study using the general
practice research database. Journal of Epidemiology
and Community Health 2005; 59:847–850
SLIDE 3 of 34
First presentation to health service
Addicts’ mental illness being being ignored?
George Harrison’s attacker Michael Abram’s schizophrenia was not
diagnosed - because of heroin abuse?


On the other hand, among psychiatric patients…..“addiction
being ignored”
“Urgent action is needed to help mentally-ill people who are
addicted to drink or drugs, an influential group of MPs has
warned”. (All-Party Parliamentary Drugs Misuse Group, 2000)
SLIDE 4 of 34
Comorbidity: US Research


50% of schizophrenic patients have substance
misuse disorder
Such co-morbidity is associated with:
heavy use of psychiatric inpatient care, poor treatment
compliance, poor prognosis and high offending rates

Services combining psychiatric and substance
misuse treatments may be more effective than
management by separate services
SLIDE 5 of 34
Pathways to Comorbidity
A: Substance use
B: Psychiatric illness
•psychoses
•schizophrenia
•paranoia
•neurosis
•personality
disorders
•single use
•harmful use
•dependence
•withdrawal
C: Diagnosed Substance
Misuse Disorder
COMORBIDITY
A
B B
A B
C
C
B
Crome I. (1999). Substance misuse and psychiatric comorbidity:
towards improved service provision. Drugs: Education, Prevention and
Policy. 6 (2) pp151-174
SLIDE 6 of 34
Type of Comorbidity Study



Prevalence
 General population, primary care, specialist
services
Development of comorbidity
 Clinical, epidemiological
Treatment
 Standard or specialised
SLIDE 7 of 34
Studying Co-morbidity using the GPRD

Data collected from about 1.8 million
patients (230/370 practices)

258 substance misuse diagnoses and 1,693
psychiatric diagnoses (based on ICD 9 and
Read Codes)
SLIDE 8 of 34
Classification of Psychiatric and
Substance Abuse Diagnoses
Substance
Abuse
Psychoses
Schizophrenia
Paranoia
Neurosis
Personality
Disorders
Other Disorders
Drug Misuse
Drug Misuse &
Psychoses
Drug Misuse &
Schizophrenia
Drug Misuse &
Paranoia
Drug Misuse &
Neurosis
Drug Misuse &
Personality
Disorders
Drug Misuse &
Other Disorders
Drug Dependence & Drug Dependence & Drug Dependence & Drug Dependence &
Drug
Psychoses
Schizophrenia
Paranoia
Neurosis
Dependence
Licit
Dependence
Licit dependence
&Psychoses
Licit dependence & Licit dependence & Licit dependence &
Schizophrenia
Paranoia
Neurosis
Drug Dependence &
Drug Dependence &
Personality
Other Disorders
Disorders
Licit dependence &
Licit dependence &
Personality
Other disorders
Disorders
SLIDE 9 of 34
Patients included in the study
Patient group
Number of cases
Total PYE
1993-1998
Av. Annual PYE
All comorbid cases
4,612
18,688
3,115
B
All mono substance abuse cases
5,285
15,055
2,509
5,281
925,293
154,216
C
Matched sample of mono psychiatric
cases
9,260
5,243,047
873,841
D
Matched sample of patients
(no psychiatric or drug diagnoses)
TOTAL
24,438
6,202,083
1,033,681
Group
A
SLIDE 10 of 34
Changes in Psychiatric Illness and Substance Abuse Use
Published data, Extrapolated to E&W
1994
1998
condition
number of
patients
% of
population
number of
patients
% of
population
% change
depression
1,726,323
3.5
2,428,851
5
40
anxiety
1,457,802
3
1,915,458
3.9
31
schizophrenia
87,167
0.17
89,529
0.18
2
substance
abuse
37,790
0.07
52,243
0.11
38
SLIDE 11 of 34
Annual Comorbidity Prevalence Rate
Year
1993
1994
1995
1996
1997
1998
Comorbid
within year
492
612
665
702
780
872
GPRD
PYE
988,735
1,002,276
1,028,414
1,045,543
1,056,767
1,080,348
Rate per
lower upper Number of cases in
100,000 PYE 95% CI 95% CI E&W population
24,436
50
48
51
29,985
61
60
63
31,754
65
63
66
32,971
67
66
69
36,246
74
72
75
39,637
81
79
82
SLIDE 12 of 34
Incident and Prevalent Cases
% of annual cases .
incident cases
prevalent cases
100
80
60
40
20
0
1
2
3
4
5
prevalent cases
11.6
19.5
19.1
21.8
22.2
incident cases
88.4
80.5
80.9
78.2
77.8
SLIDE 13 of 34
Annual comorbidity rates per 100,000
PYE by ageband and gender
FEMALES
Drug Misuse
Drug Dependence
Licit Dependence
100
Drug Misuse
av. annual comorbidity rate
av. annual comorbidity rate
MALES
80
60
40
20
Drug Dependence
Licit Dependence
100
80
60
40
20
0
0
5-15 16-24 25-34 35-44 45-54 55-64 65-74 75-84
age group
85+
5-15 16-24 25-34 35-44 45-54 55-64 65-74 75-84 85+
age group
SLIDE 14 of 34
Rate of psychiatric comorbidity per 100,000 PYE
Percentage change in psychiatric comorbidity 1993-1998
Year
1993
1994
1995
1996
1997
1998
% rate change 93/98
Estimated number of
patients in 1998
Psychoses
5.1
3.9
5.1
5.3
9.0
12.5
147.1
6,138
Schizophrenia Paranoia
1.0
0.6
1.7
1.5
2.2
1.6
1.8
1.6
2.1
2.5
2.3
1.5
128.8
144.1
1,134
727
Neurosis
43.1
54.6
57.6
60.8
65.4
69.6
61.6
Personality
disorder
4.3
5.3
6.9
5.6
4.9
6.6
54.7
Other
disorders
3.2
3.4
3.8
3.5
4.5
5.5
68.7
34,183
3,226
2,681
SLIDE 15 of 34
Illicit Drug Use and Mental Health
‘Evidence establishes clear
link between use of
cannabis and mental health’
BMJ Editorial 2002
SLIDE 16 of 34
Cannabis and Mental Health
‘Cannabis use does not appear to be
causally related to the incidence of
schizophrenia’
‘It may precipitate disorders in persons who are
vulnerable to developing psychosis’
Degenhardt et al Drug and Alcohol Dependence 2003
SLIDE 17 of 34
The move to causality



Use of cannabis increases the risk of schizophrenia and
depression but baseline psychiatric illness is not associated with an
elevated risk of substance use (Rey J and Tennant C. Cannabis
and Mental Health. BMJ 2002; 325: 1183-4 2002).
“Use of cannabis will contribute to more episodes or
new cases of (psychiatric) illness” (assumes that cannabis use
is increasing and/or) the level of cannabis use among users is
increasing).
“Surprising that, given this association, there does not
appear to increasing incidence of schizophrenia in the
population”.
SLIDE 18 of 34
Does Substance Misuse Cause Mental Illness?
Consensus Pre BMJ



Substance misuse can have a negative impact upon the
course of psychotic disorders and can exacerbate
symptoms
Some substances, particularly alcohol, cannabinoids,
hallucinogens and stimulants can produce psychotic
symptoms directly without coincidental mental illness.
Although these substances may trigger psychotic
disorders amongst people with a predisposition, the
existence of any stronger causal relationship remains
controversial.
SLIDE 19 of 34
Is the evidence clear?

Some reports suggest cannabis exacerbates
symptoms of schizophrenia, but two clinical
studies suggest cannabis use alleviates
symptoms.

Cannabis use is not associated with the
development of psychosis in an 'ultra' highrisk group Aust N Z J Psychiatry 2002
SLIDE 20 of 34
Dunedin Birth Cohort Study
N
cannabis
at 15
y
n
N
cannabis
at 18
y
n
outcome
schizophreniform disorder
y
n
3
26
29
22
708
730
25
734
759
% who develop
schizophreniform
y
n
10.3
89.7
3.0
97.0
outcome
schizophreniform disorder
y
n
10
226
236
15
508
523
25
734
759
% who develop
schizophreniform
y
n
4.2
95.8
2.9
97.1
Association between cannabis use in adolescence and schizophrenia and depressive
symptoms and disorders at age 26 (n=759), controlling for childhood psychotic
symptoms and use of other drugs in adolescence
Cannabis at 15, OR = 4.50 (1.11 to 18.21)
Cannabis at 18, OR = 1.65 (0.65 to 4.18)
SLIDE 21 of 34
Population Attributable Risk:
Smoking, lung cancer and CHD
Cigarette smokers
Non smokers
Relative risk
Attributable risk
Annual mortality rate per 100,000 population
Lung
Coronary Heart
cancer
Disease
140
669
10
413
14
1.6
130
256
Source: Epidemiology in Medicine Little Brown, Toronto,
Hennekens & Buring (1987), p 94
SLIDE 22 of 34
GPRD Study questions
1.
2.
3.
What is the relative risk of psychiatric illness
among substance abusers compared to non
substance abusers?
What is the relative risk of substance abuse among
psychiatric cases compared to non psychiatric
cases?
How has the proportion of comorbid patients
with a primary substance abuse diagnosis changed
from 1993 to 1998?
SLIDE 23 of 34
Time sequence between mono and dual diagnosis
Entry into cohort (N= 1.8 million patients)
No substance abuse or psychiatric illness in previous year
1 No diagnosis of substance abuse or psychiatric illness
2 Diagnosis of ONLY a) substance abuse or b) psychiatric illness
3 Diagnosis of substance abuse then psychiatric illness
4 Diagnosis of psychiatric illness then substance abuse
SLIDE 24 of 34
Study Cohorts
Group
N
%
1 no substance or psychiatric diagnosis
1,557,137
84.5
2 substance only
5,228
0.3
3 psychiatric only
275,820
15.0
4 comorbid
3,969
0.2
a) substance then psychatric
1,588
40.0
b) psychatric then substance
2,162
54.5
c) same day
219
5.5
All cases
1,842,162
100.0
SLIDE 25 of 34
Ordering of first diagnoses among
comorbid patients
First Diagnosis
substance abuse
psychiatric
both types on same day
Total
number of cases
1,615
2,150
212
3,977
% of cases
41
54
5
100
SLIDE 26 of 34
Median interval from first diagnosis to
comorbid diagnosis
substance first
Psychiatric Diagnosis (n of cases)
Psychoses
85
Schizophrenia
4
Paranoia
17
Neurosis
1320
Personality Disorders
111
Other Disorders
78
All
1615
months
(median)
9.9
2.7
6.2
7.3
5.7
5.7
7.1
psychiatric first
(n of cases)
102
31
12
1866
61
78
2150
months
(median)
9.9
7.0
6.3
11.8
8.2
10.2
11.4
SLIDE 27 of 34
Time ordering of comorbidity in relation
to psychiatric diagnoses
% psychiatric first
% substance first
% same day
% of cases
80
70
60
50
40
30
20
10
0
Psychoses
Schizophrenia
Paranoia
Neurosis
Personality
Disorders
Other
Disorders
All
SLIDE 28 of 34
Risk Exposure and Disease Outcome
exposure
substance
abuse
outcome
psychiatric diagnosis
RR
yes
no
yes
1,588
5,227
6,815
0.23
no
276,954
1,558,393
1,835,347
0.15
exposure
psychiatric
yes
no
outcome
substance abuse
yes
no
2,162
277,982
5,779
1,556,238
1.54
RR
280,144
0.008
1,562,017
0.004
2.09
SLIDE 29 of 34
Relative risk of psychiatric illness and
substance abuse
% general population exposed to psychiatric illness 1993-1998
15.09
% exposed substance abuse who develop psychiatric illness 1993-1998
23.3
Relative risk of psychiatric illness (substance vs non substance abusers)
1.54 (1.47-1.62)
% general population exposed to substance illness 1993-1998
0.37
% exposed psychiatric who develop substance abuse 1993-1998
0.77
Relative risk of substance abuse (psychiatric vs non psychiatric cases)
2.08 (1.98-2.19)
SLIDE 30 of 34
POPULATION ATTRIBUTABLE RISK
Proportion of illness in the population
attributable potentially explained by the exposure
Exposure
Illness
% of cases
95% CI
Substance abuse
psychiatric illness (all)
0.2
0.17-0.22
Substance abuse
schizophrenia and psychoses
0.11
0.01-0.21
Psychiatric illness
substance abuse
14.2
13.0-15.3
Schizophrenia and psychoses
substance abuse
0.76
0.47-1.05
SLIDE 31 of 34
Changes in baseline conditions
% of annual incident cases
baseline substance abuse
baseline psychiatric illness
baseline comorbid
60
50
40
30
20
10
0
1993
1994
1995
1996
1997
1998
SLIDE 32 of 34
Summary (1)
 The
numbers of individuals newly
developing comorbidity in primary care
is increasing year-on-year.
 Does the increase in comorbidity
reported here represent an actual
increase in the prevalence of comorbid
conditions or the medicalisation of
social, economic or personal problems?
SLIDE 33 of 34
Summary (2)


Does active early recognition of comorbidity
lead to better outcomes?
The findings of the present study do not
contradict studies in different populations which
have found a link between certain forms of drug
use and psychiatric illness. However they
suggest that attempts to prevent comorbidity by
focusing on substance abuse will meet with only
very limited success.
SLIDE 34 of 34