Transcript “drug dependent” individuals.

SYMPTOMS OF DEPENDENCE AMONG SELF IDENTIFIED ADULT RECREATIONAL ILLICIT
DRUG USERS:
A SUB-ANALYSIS OF THE DRUGNET SURVEY
A graduate thesis by Shakti Nagarajan, (MPH)
Thesis Committee
Thomas Nicholson, PhD,
David Duncan, DrPH,
John B. White, PhD,
Lisa L. Lindley, DrPH,
Fred Stickle, PhD.
1
INTRODUCTION
 The use and sale of illicit drugs in America
continues unabated despite decades of the
“War on Drugs” and the “ Zero Tolerance”
drug policy.
 The 2002 National Household Survey on
Drug Use and Health (NHSDUH) findings
suggest that there has only been a marginal
decline in overall drug use.
2
NEGATIVE OUTCOMES OF THE DRUG WAR:
Black market for illicit drugs.
• Largest prison system in the world.
• Lots of money spent on law-enforcement
programs, and media campaigns.
• Artificial dichotomy of “good” and “bad” drugs-alcohol and tobacco effects are undermined.
• Rise in HIV/hepatitis C.
• Distorted public opinion about “illegal drugs”
and users.
Net result : Millions of Americans break federal
laws to use drugs in secrecy.
•
3
PUBLIC HEALTH APPROACH
Aim : Harm reduction and to raise the
standard of health of illicit drug users
Conventional methods of studying this
“hidden population” - Medical records,
surveys.
DRUGNET survey - An online, crosssectional, survey of nonabusive,
recreational adult drug users.
4
PURPOSE OF CURRENT STUDY
 It is imperative to study attributes of drug
use at the problematic end of consumption
spectrum i.e. Dependence and Abuse
 Was conducted to explore the possibility
drug dependence symptoms in a selfidentified population of non-abusers
5
NEED FOR THE STUDY
 Studies of recreational or occasional nondependent drug users are rare
 Study required to address the health needs of the
hidden population of healthy American,
recreational drug users
 Results will provide insights and information about
their drug use behaviors that can possibly have a
profound impact on conventional drug theories
6
RESEARCH QUESTIONS
1.Are there any symptoms of drug dependence
or abuse, in a self-identified population of
non-abusers?
2.If yes, how frequent are they?
3.Is there any pattern to them?
DELIMITATIONS
 U.S. citizens, aged 18 years and older
 took the survey on the Internet
 between December 1997 and June 1998
7
LIMITATIONS AND ASSUMPTIONS
LIMITATIONS
 Self-selected
subjects - Non-random sampling
 Self-reported behaviors. Validity of responses
cannot be verified.
ASSUMPTIONS
 Individuals
answered the questionnaire
openly and honestly and to the best of their
ability.
 Individuals were able to understand
directions for taking the survey.
8
REVIEW OF LITERATURE


Nicholson (1992) - objection to use of phrases
such as "drugs of abuse," and "war on drugs"
because they shape conceptual thinking and
attitudes.
Jonas (1994) - no scientific (toxicological,
pharmacological, epidemiological or medical)
basis whatsoever in segregating the legal drugs
from the illicit drugs.
Important to clarify nomenclature: use,
abuse, and dependence.
9
LITERATURE REVIEW contd.

World health organization (WHO) definition of dependence:
“it is a state of psychic or physical dependence, or both, on a drug, arising
in a person following administration of that drug on a periodic or
continuous basis. The characteristics of such a state will vary with the
agent involved, and these characteristics must always be made clear by
designating the particular type of drug dependence in each specific case;
For example, drug dependence of morphine type, of barbiturate type, of
amphetamine type, etc”. (WHO, 1965).


To make the term "drug dependence" more precise - a
cluster/collection of events that occurred in relation to the
dependence pattern in a drug user and called it the “drug
dependence syndrome”. (WHO 1981)
Dependence concept defined in DSM III, DSM III-R, and
DSM IV
10
Past Year Any Illicit Drug or Alcohol Substance Dependence
or Abuse among Persons Aged 12 or Older: 2002
Note. From “Substance Dependence, Abuse, and Treatment,” by Substance Abuse and
Mental Health Services Administration (SAMHSA), 2003, Results from the 200211
National Household Survey on Drug Use and Health (NHSDUH), Chapter 8.
METHODS
 Purpose - to explore the presence of drug
dependence symptoms in a non-random
sample of recreational drug users from the
DRUGNET study. If such symptoms are
present, the study will examine their
frequency, and patterns
 Results compared to National Household
Survey on Drug Use and Health (2002) and
the Epidemiologic Catchment Area study
(1980-1985)
12
RESEARCH QUESTIONS
 Are there any symptoms of drug dependence?
 If yes, how frequent are they?
 Is there any pattern to them?
Population - Non-clinical, adult recreational
drug users having access to the Internet.
Sample Selection - American citizens aged 18
years and older, who completed the
DRUGNET survey on the Internet between
December, 1997 through June, 1998.
13
PROCEDURES
Sub analysis of the DRUGNET survey.
Respondents were solicited via mailing
lists
Anonymizer link was provided,
informed consent taken
Respondents were allowed to choose
between ranges of coded responses.
14
DESIGN - CROSS-SECTIONAL STUDY
 Sub-analysis of previously collected
DRUGNET survey data (March, 1997 June, 1998).
 Questions pertaining to drug dependence
based on the DSM-III-R drug dependence
criteria were formulated and incorporated
into the survey in December 1997.
 541 respondents
15
INSTRUMENTATION
Four major subdivisions :
Demography and lifestyle indices
Experiences with drugs
Past legal history and attitudes regarding drug use
And,
The general well –being schedule (GWBS)
Focus of this study - items 26 through 33 (based on
DSM III-R), - demographics subdivision.
A subject who responded positively to three or more
of the drug dependence questions would qualify as
a potential “drug-dependent individual.”
16
DATA ANALYSIS
Input Variables:
1)Problems in life from drug use
2)Using drugs dangerously
3)Problems with legal system
4)Argument with family and friends about use of drugs
5)Withdrawal symptoms
6)Problems stopping use
7)Obtaining drugs consumes a lot of time
8)Been in treatment for dependence
Coding of responses:
Responses coded as follows:


Yes - 1
No – 0
Data analyzed using SPSS.
17
DATA ANALYSIS contd.
• First two research questions –
Answered by examining responses of items 26-33
Utilizes the Frequencies subprogram of SPSS.
Reported using regular frequency and cumulative
frequency tables.
• Third research question –
Utilizes the Principal Components Analysis (PCA)
subprogram of SPSS.
Initial tests of assumptions
 Kaier-Meyer-Okin Measures of Sampling Adequacy
 Bartlett ‘s Test of Sphericity
18
PCA
• is a statistical technique that can be applied to a
single set of variables to discover “which variables
in the set form coherent subsets that are relatively
independent of each other” (Tabachnik and Fidell,
1989).
• summarizes patterns of correlation among
observed variables and reduce it to a small number
of components to provide an operational definition
(in terms of Eigen values) of underlying processes
and test theories that will explain the underlying
processes.
19
 Scree plots will be graphed to determine the
components that will account for meaningful
amounts of shared variance in the sample.
 Further analysis of patterns will be done using
Rotated Component Matrix technique to redefine
factors of importance by enhancing the variance
produced by the weighted factors.
 The results will then be interpreted and discussed
accordingly.
20
RESULTS - SUMMARY
Sample Selection - 541 United States
citizens, aged 18 years or older who
responded to the DRUGNET survey,
between December 1997 and June
1998.
21
Demographics of Study Sample:
•
•
•
•
•
88.9% - White
Average age 31.8 yrs.
Men outnumbered women 3:1.
Ninety-four percent employed.
Approximately one-third of the subjects were married, and
43.9% had never been married.
• More than 9 out of 10 individuals were happy with their
marital status.
• The majority of the survey participants had a college
education and one-fourths of them were currently
attending college.
• Average GPA 3.4
• Median income of $50,000 - $69,999
22
LIFESTYLE AND BEHAVIORAL INDICES
• 88.6% of the survey respondents did not attend religious services
regularly.
• Almost half of the subjects reported strong feelings of
spirituality.
• 39.6% respondents reported that they participated in community
activities.
• 72.8%, (n = 394) voted regularly in the general elections.
• More than 9 out of 10 respondents felt that they were in good
health.
• 97.2% engaged in non-drug related recreation.
• One-third had regular parental childcare responsibilities,
• 4 out of 10 individuals reported that their children knew about
their use of illicit drugs.
23
Frequency of Drug Dependence Symptoms
Yes
Missing
Data
No
n
%
n
%
n
158
29.2
383
70.8
--
358
66.7
179
33.3
4
Problems with legal system due to use of illicit
drugs
101
18.8
437
81.2
3
Argument with family and friends over use of illicit
drugs
168
31.3
369
68.7
4
78
14.5
459
85.5
4
67
12.5
471
87.5
3
Procuring and using illicit drug takes a lot of time
in the life of the respondent
36
6.7
499
93.3
6
Has ever been in treatment for drug abuse or
dependence
30
5.6
507
94.4
4
Drug Dependence Symptom
Problems in life from use of illicit drugs
Using illicit drugs under dangerous circumstances
Withdrawal symptoms on stopping the use of an
illicit drug
Problems stopping use of illicit drugs
24
Total Problem Score (showing reported number of drug dependent
symptoms).
Number of drug
dependent
symptoms
Frequency
Valid %
Cumulative Percent
None
119
22
22
One
168
31
53
Two
106
19.6
72.6
Three
66
12.2
84.8
Four
42
7.8
92.6
Five
22
4.1
96.7
Six
11
2
98.7
Seven
5
0.9
99.6
Eight
2
0.4
100
Total
541
100
25
Using the frequency of these
symptoms, it was ascertained that
27.4% (n = 148) of the respondents
(i.e. roughly every one in four
respondents) were potentially “drug
dependent” individuals.
26
The following criterion tests were applied:
(a) The Kaiser-Meyer-Olkin Measure of
Sampling Adequacy – Our study sample
achieved a value of 0.763
(b) Bartlett’s Test of Sphericity – An
approximate Chi-Square value of 454.525
with a 28 degree of freedom, and a
significance value of 0.000 was achieved
27
Total Variance Explained in Principal Component
Analysis
Initial Eigenvalues
Rotation Sums of Squared
Loadings
Component
Total
% of
Variance
Cumulative
%
Total
% of
Variance
Cumulative
%
1
2.422
30.275
30.275
1.951
24.391
24.391
2
1.039
12.983
43.258
1.509
18.867
43.258
3
0.97
12.128
55.387
4
0.89
11.128
66.515
5
0.824
10.302
76.817
6
0.689
8.607
85.425
7
0.61
7.626
93.051
8
0.556
6.949
100
28
SCREE PLOT
3
Eigenvalue
2.5
2
1.5
Eigenvalue
1
0.5
0
1
2
3
4
5
6
7
8
Component Number
First two components were extracted.
29
Rotated Component Matrix1 for Principal Component Analysis
Input Variables
1
2
0.491
0.509
0.3
0.254
Problems with legal system due to use of illicit drugs
-0.039
0.841
Argument with family and friends over use of illicit drugs
0.544
0.134
Withdrawal symptoms on stopping the use of an illicit drug
0.731
0.047
Problems stopping use of illicit drugs
0.732
0.115
Procuring and using illicit drug takes a lot of time in the life of the
respondent
0.464
0.092
Has ever been in treatment for drug abuse or dependence
0.193
0.661
Problems in life from use of illicit drugs
Using illicit drugs under dangerous circumstances
1Rotation
Method: Varimax with Kaiser Normalization
30
NET RESULT
 The first two components account for approximately 43% of
the total variance.
 The Varimax rotation helped ascertain that physical
symptoms of drug dependence like withdrawal illness, and
difficulty stopping use loaded high on the first component,
while problems with the legal system and having been in
treatment loaded high on the second component.
 This meant that the first component comprised of a pattern of
individuals suffering from physical symptoms of withdrawal
illness. These individuals constitute a potential target
population for treatment and rehabilitation interventions. On
the other hand, the second component reflects a pattern of
individuals who have had legal problems due to the use of
illicit drugs and have received treatment although they report
less physical symptoms.
31
COMPARISON OF RESULTS WITH THE ECA AND THE
NHSDUH 2002 FINDINGS
 A direct comparison of the results of the DRUGNET
survey to the results of the ECA or NHSDUS
cannot be done because the samples surveyed are
different.
 The ECA found that in the total sampled
population of 3,925 individuals, 6.2% had a history
of abuse/dependence syndrome (Anthony & Helzer,
1991).
 According to the NHSDUH, among those who had
ever used illicit drugs, 4.3% were classified as
being substance dependent in the past year.
 Our study ascertained that 27.4% (n = 148) of the
respondents (i.e. roughly every one in four
respondents) were potentially “drug dependent”
individuals.
32
DISCUSSION
 Study provided an adequate demographic and
behavioral profile of the survey participants.
33
DISCUSSION
 Do 1 in 4 respondents really suffer from drug
dependence? Further clinical evaluation would
give the right picture.
 Our survey had close-ended questions – hence no
qualitative assessment of responses. Instrument
bias possible.
 Lack of a specific time frame for the persistence of
drug dependence symptoms in the survey
instrument. Study projects lifetime prevalence of
drug dependence in the sample.
34
IMPLICATION OF VARIMAX ROTATION OF PCA
RESULTS
 Encounters with law enforcement were driving
treatment among the study sample.
 This also indicates that scarce treatment
resources are being wasted on the wrong subset
of the population
 Alternatively, non-abusive drug users may not
need treatment for dependence but may be
engaging in risky behaviors that can be
addressed through harm reduction
interventions.
35
THE RESULTS
 Validates the existence of a hidden population of recreational
drug users.
 Demonstrates value of an on-line computer based tool in
accessing such populations.
 Provided a broad profile of recreational drug users.
 Validates the arguments put forth by opponents of the “drug
war.”
That is, although prevention of drug dependence is a much
talked about issue in this country, it is still the legal system
that drives the clinical system to implement treatment and
rehabilitative measures on those individuals that it feels right
for such measures. The majority of individuals who actually
report symptoms of dependence have been left to fend for
themselves.
36
RECOMMENDATIONS
 Drug policy makers need to consider the
ramifications of studies like the DRUGNET survey
and other such community studies that differ from
conventional concepts.
 More tolerant and more practical drug policy.
 Further research that explores patterns of drug
dependence in non-biased, non-clinical
populations should be encouraged
 Essay questions should be used in the
questionnaire format of studies like DRUGNET,
this will allow for qualitative research.
 Studies like the DRUGNET survey need to be put
back on line to access populations with sensitive
issues.
37
For copies of survey instrument, references and
resources used in this study address all
correspondence to [email protected]
or [email protected].
Thank you
38