This study analyzed data on 9,732 adults admitted to publicly-funded heroin treatment programs in Texas between 1997-2001 to compare the characteristics of heroin inhalers versus injectors. The results showed inhalers were younger at admission and first use, had used heroin for fewer years, and were more likely to be first-time admissions. Inhalers also had higher incomes, were more likely to have children at home, and less likely to be homeless or involved in the criminal justice system compared to injectors. Additional interviews were planned to further understand reasons for using different administration methods and not entering treatment.

1. Differences in Characteristics of Heroin Inhalers and Injectors
at Admission to Treatment
J. C. Maxwell, R. T. Spence, & T. M. Bohman
UT Center for Social Work Research
Introduction
 Purpose: to understand why heroin inhalers (“snorters”) in Texas enter treatment
sooner than injectors.
injectors
 This study extends earlier findings by comparing the characteristics of heroin
injectors versus inhalers at their first admission to publicly-funded treatment using
CODAP/TEDS data.
 This study will build on earlier studies, but looks at differences in patterns of
heroin use in a much larger sample characterized both by its race/ethnic diversity
and by the use of black tar heroin, a type of heroin primarily seen only in the
western United States and Mexico.
 This project has three phases:
1) analysis of CODAP/TEDS data on clients who had a primary problem with
heroin;
2) in-depth interviews with heroin users in public and private programs; and
3) interviews and focus groups with heroin inhalers and injectors who are not in
treatment.
 This poster reports on the first phase only.
Sample
 9,732 unduplicated adults who were first admissions to TCADA-funded
residential, outpatient drug-free, or outpatient methadone treatment programs
between January 1, 1997 and December 31, 2001.
 Clients had a primary problem with either inhaled or injected heroin which they
had used in the month prior to admission.
 While they were first admissions to TCADA programs, 64% had been treated
previously in non-TCADA-funded programs.
 92.5% were injectors and 7.5% were inhalers.
 Mean age at admission was 35.8 years (SD 10.6). Age of first use of heroin was
22.9 years (SD 8.0).
 The mean lag between first use and admission to TCADA-funded treatment was
13.5 years (SD 10.8).
 69% were male; 49% were Hispanic, 40% were Anglo, and 12% were African
American.
 Average annual legal income was $5,522; 17% were employed and education
level was 11.1 years.
 75% lived with their families, 28% were married or co-habiting with a partner,
11% were homeless, and 30% were involved with the criminal justice or legal
system.
 33% reported a secondary problem with powder cocaine, 12% with alcohol, 7%
with marijuana, and 4% with crack cocaine; 38% reported no problems with other
drugs.
Results
Comparisons of Heroin Injectors and Inhalers on Demographic,
Criminal Justice Status, Treatment, Drug Use, and Substance
Abuse Impairment Characteristics
Demographics
Age at Admission
Age of First Use
Lag—First Use to Entry to
Treatment (Years)
Injectors
36.0
22.8
S.E.
0.482
0.188
13.8
0.463
Percent Percent
Injectors Inhalers
70.0
61.7
40.7
29.2
49.6
34.5
9.7
36.2
74.9
80.6
11.4
5.0
1.58
2.18
32.1
43.3
11.0
11.3
16.9
23.3
$5,426
$6,694
Inhalers S.E.
31.0
0.614
23.5
0.347
8.1
Odds
Ratio*
.81
.43
1.15
2.82
1.23
.61
1.72
1.61
1.14
1.23
1.23
Pr> |Z|
<.0001
Confidence
Limits
0.70-0.94
0.31-0.61
0.94-1.40
1.83-4.33
1.01-1.51
0.50-0.73
1.09-2.72
1.31-1.98
0.87-1.49
1.03-1.47
0.99-1.53
% Male
.0068
% Anglo
<.0001
% Hispanic
.1716
% African American
<.0001
% Live with Family
.0430
% Homeless
<.0001
% Pregnant
.0206
%Minor Children in Home
<.0001
Years of Education
.3586
%Employed at Admission
.0253
Yearly Income
.0610
Criminal Justice Status
% Involved Criminal Justice 30.7
25.6
.69
<.0001
0.59-0.80
Legal System
% on Probation
10.8
11.1
.95
.7498
0.69-1.31
% on Parole
13.5
6.3
.36
<.0001
0.26-0.49
Treatment
%First TCADA Treatment
35.2
49.3
1.91 <.0001
1.68-2.18
%Admitted to Methadone
29.0
37.7
0.94
.3342
0.84-1.06
Secondary Drug Problem
%None
38.2
29.9
1.24
.0886
0.97-1.58
%Alcohol
12. 2
10.4
1.04
.6838
0.86-1.25
%Powder Cocaine
33.3
31.5
.51
<.0001
0.36-0.71
%Crack Cocaine
4.0
8.7
2.19 <.0001
1.82-2.63
%Marijuana
6.1
14.3
2.21 <.0001
1.73-2.82
Substance Abuse Impairment (% with 0 Days of Problems in Last Month)
Sickness/Health Problems
62.8
70.6
.68
.0598
0.46-1.02
Employment Problems
35.7
45.3
.66
.0257
0.46-0.95
Family/Marital Problems
43.3
46.7
.81
.2110
0.59-1.12
Peer/Social Problems
61.3
63.1
.88
.5198
0.60-1.29
Psychological Problems
41.8
45.2
.84
.2741
0.61-1.15
Drug/Alcohol Problems
14.0
14.8
.87
.5878
0.52-1.45
Multivariate Prediction of Heroin Inhaler versus Injector
Status from Demographic, Criminal Justice Status,
Treatment, Drug Use, Second Drug Problems and
Substance Abuse Impairment Characteristics
Odds
Ratio
Logistic Regression Model
Estimate Pr> |Z|
Demographics
Male versus Female
-0.10
0.43
0.91
African American versus Anglo
2.51
0.00
12.32
Hispanic versus Anglo
0.55
0.00
1.74
Age at Admission
-0.10
0.00
0.91
Age of First Use
0.06
0.00
1.06
Live with Family
-0.09
0.64
0.91
Homeless
-0.26
0.18
0.77
Pregnant
0.13
0.67
1.14
Minor Children in Home
0.31
0.02
1.36
Years of Education
0.05
0.52
1.05
Employed at Admission
0.20
0.10
1.22
Yearly Income
0.11
0.02
1.11
Criminal Justice Status
Criminal Justice/Legal Involvement
0.00
0.99
1.00
On Probation
-0.26
0.62
0.77
On Parole
-0.82
0.09
0.44
Treatment
First TCADA Treatment
0.39
0.00
1.47
Admitted to Methadone
0.23
0.57
1.26
Second Drug Problem
None
0.17
0.42
1.18
Alcohol
0.37
0.20
1.45
Powder Cocaine
-0.51
0.13
0.60
Crack Cocaine
0.65
0.01
1.91
Marijuana
0.59
0.08
1.80
Substance Abuse Impairment (% with 0 Days of Problems in Last
Month)
Sickness/Health Problems
-0.19
0.06
0.83
Employment Problems
-0.01
0.85
0.99
Family/Marital Problems
0.06
0.56
1.06
Peer/Social Problems in
-0.07
0.58
0.93
Psychological Problems
0.00
0.95
1.00
Drug/Alcohol Problems
-0.04
0.64
0.96
*Odds ratios greater than 1 indicate inhalers are more likely to have this characteristic present while
odds ratios less than 1 indicate inhalers are less likely to have this characteristic present
Demographics of Injectors & Inhalers at
Admission to Treatment
Analysis
40
Injectors
 Statistical models included a generalized linear model with the outcome
modeled as a binomial and multinomial distribution and a hierarchical linear
model for continuous outcomes.
 The Generalized Estimating Equation (GEE) model was used to account for the
variation in heroin user characteristics due to treatment program.
 All analyses were conducted using SAS v8.
0.604
Pr> |t]
<.0001
<.0001
Discussion
Inhalers
80%
 Consistent with other studies, we found that there are significant differences
between injectors and inhalers at admission to treatment. Inhalers are younger,
more likely to have used heroin for fewer years, to be first admissions to
treatment, to have children in the home, and to have higher annual incomes.
 Texas injectors and inhalers reported no difference in number of days that heroin
was used in the month prior to admission. We would speculate that some of the
inhalers had reached a level of dependence where their habits could no longer
be maintained through daily inhaling and that faced with the need to shift to
injecting, they had, instead, chosen to enter treatment.
 Our study found that there were very large differences by race/ethnicity, with
African Americans in the sample being much more likely to be inhalers than
injectors. This finding is not well understood and will be the focus of additional
study.
 While the bivariate analysis showed that injectors were more likely to have
problems with the legal system and to be on parole, which would be a reflection
of their longer crime careers, these patterns were not significant in the
multivariate prediction model, which controlled for age.
 Having minor children in the household was a significant predictor of being an
inhaler.
 Only higher income was a significant economic predictor of inhaling when age,
gender, and race/ethnicity were controlled. However, findings about the
economic status may be confounded because TCADA programs give priority to
poverty-level clients.
Limitations/Further Study
 The findings from this first phase of the study are limited by the variables collected
in CODAP/TEDS and they cannot be generalized to all heroin users in treatment in
Texas since only 9 of the 69 methadone programs in the state report CODAP/TEDS
data. Collecting comparable data on clients in private programs is part of Phase 2 of
this research study.
 Phase 2 will be in-depth face-to-face interviews with heroin users to determine
reasons for shifting or not shifting to needles. Data will be collected on patterns of
use of other drugs, attitudes towards needle use (fear of needles or “needle
fixation”), prior sexual and physical abuse, mental and physical status, childhood
and family history, criminal history, and reasons for remaining an inhaler or for
shifting to injection.
 Phase 3 will be interviews and focus groups with heroin users who have never been
in treatment.
Demographics of Injectors & Inhalers at
Admission to Treatment
Demographics of Injectors & Inhalers at
Admission to Treatment
70%
35
Injectors
30
60%
25
50%
70%
20
40%
30%
10
20%
60%
50%
40%
30%
Acknowledgements
20%
5
Funding from NIDA (R21 DA014744)
Cooperating treatment providers and study participants
Accepted for publication in Substance Use and Misuse.
Contact Jane Maxwell
512 232-0610
jcmaxwell@mail.utexas.edu
Inhalers
80%
15




Injectors
90%
Inhalers
10%
0
10%
0%
0%
Admission Age*
Age 1st Use*
Lag*
# Days Used
*p <0.05
Male*
Anglo*
Hispanic
Black*
Live with Family*
*p <0.05
*p <0.05
Homeless*
Pregnant*
Kids at Home*
Employed*