Alcohol, Tobacco, and Other Drug Use in Illinois ... Tobacco, and Other Drug Use in Illinois:...

Alcohol, Tobacco, and Other Drug Use in Illinois: Prevalence

and Treatment Need, 2003

U.S. Center for Substance Abuse Treatment

Funded by the U.S. Center for Substance Abuse Treatment under the State Needs Assessment Program.

Rod R. Blagojevich, Governor

Alcohol, Tobacco, and Other Drug Use in Illinois: Prevalence and Treatment Need, 2003 Timothy P. Johnson, Ph.D. Young Ik Cho, Ph.D.

Anthony Lerner Lillian Pickup, Administrator Lisa Cohen Division of Alcoholism and Substance Abuse Illinois Department of Human Services November 2004 DASA grants full permission to produce and distribute any part of this document. Citation of the source is required. As of the printing date, this document was pending review by the federal Office of Management and Budget.

Theodora Binion-Taylor, Associate Director 100 West Randolph, Suite 5600 Chicago, Illinois 60601 (312) 814-3840

Survey Research Laboratory College of Urban Planning and Public Affairs University of Illinois at Chicago

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003 iii

¿¿¿ CONTENTS

List of Tables ..................................................................................................................................... iv List of Figures.................................................................................................................................... vi Acknowledgments............................................................................................................................ vii Executive Summary .......................................................................................................................... ix Introduction and Overview .................................................................................................................1 Methodology Summary...................................................................................................................... 2 Results................................................................................................................................................ 3

Sample Characteristics................................................................................................................. 3 Substance Use Prevalence Estimates........................................................................................... 7

Cigarette Use.......................................................................................................................... 8 Alcohol Use........................................................................................................................... 10 Marijuana Use.......................................................................................................................13 Other Illicit Drug Use............................................................................................................15 Nonmedical Use of Prescription Drugs.................................................................................18 Any Drug Use ........................................................................................................................21 Age of First Use .................................................................................................................... 23

Substance Abuse Treatment Need Prevalence Estimates.......................................................... 25 Estimated Treatment Need.................................................................................................. 26 Abuse Treatment Need......................................................................................................... 29 Dependence Treatment Need .............................................................................................. 29 Unmet Treatment Need ....................................................................................................... 34

Prevalence and Treatment Need Estimates for Vulnerable Groups.......................................... 36 Persons Without Health Insurance Coverage...................................................................... 36 Persons with Physical Disabilities........................................................................................ 36 Persons with Mental Disorders............................................................................................ 39 Persons Ever Homeless........................................................................................................ 39 Recipients of Public Financial Support................................................................................ 42 Domestic Violence Victims................................................................................................... 42 Chronic Gamblers ................................................................................................................ 45 Estimated Treatment Need for Vulnerable Groups............................................................. 45

Discussion ........................................................................................................................................ 49 References.........................................................................................................................................51

Appendix A. Study Methodological Report Appendix B. Substance Use Definitions Appendix C. Definitions of Illinois Sample Areas Appendix D. Advance Letter Sent to Sampled Households Appendix E. Survey Limitations I: Sampling Errors Appendix F. Survey Limitations II: Coverage Errors Appendix G. Survey Limitations III: Nonresponse Errors Appendix H. Survey Limitations IV: Measurement Errors

iv Illinois Division of Alcoholism & Substance Abuse

¿¿¿ TABLES

1. Demographic Characteristics of Illinois Household Survey Sample, 2003................................5

2. Vulnerable Populations: Prevalence Estimates for Illinois, 2003..............................................6

3. Lifetime, Past-Year, and Past-Month ATOD Use in Illinois, 2003.............................................7

4. Cigarette Use in Illinois, by Demographic Characteristics, 2003...............................................9

5. Population Estimates of Cigarette Use in Illinois, by Gender, Age, and Race/Ethnicity, 2003 ..........................................................................................................................................10

6. Alcohol Use in Illinois, by Demographic Characteristics, 2003 ............................................... 11

7. Population Estimates of Alcohol Use in Illinois, by Gender, Age, and Race/Ethnicity, 2003 .......................................................................................................................................... 12

8. Past-Year Binge Drinking in Illinois, by Demographic Characteristics, 2003......................... 13

9. Marijuana Use in Illinois, by Demographic Characteristics, 2003...........................................14

10. Population Estimates of Marijuana Use in Illinois, by Gender, Age, and Race/Ethnicity, 2003 .......................................................................................................................................... 15

11. Lifetime Use of Crack/Cocaine, Heroin, Hallucinogens, Inhalants, and Club Drugs in Illinois, by Demographic Characteristics, 2003 ....................................................................... 16

12. Lifetime and Past-Year Illicit Drug Use in Illinois, by Demographic Characteristics, 2003.... 17

13. Population Estimates of Illicit Drug Use in Illinois, by Gender, Age, and Race/Ethnicity, 2003 .......................................................................................................................................... 18

14. Lifetime Nonmedical Use of Prescription Drugs in Illinois, by Demographic Characteristics, 2003 ................................................................................................................20

15. Population Estimates of Lifetime Nonmedical Use of Prescription Drugs in Illinois, by Gender, Age, and Race/Ethnicity, 2003................................................................................... 21

16. Use of Any Drug in Illinois, by Demographic Characteristics, 2003........................................22

17. Average Age of Drug Use Initiation in Illinois, by Demographic Characteristics, 2003..........24

18. Substance-Related DSM-IV Symptoms in Illinois, 2003 .........................................................25

19. Percentage Meeting DSM-IV Criteria for Past-Year Substance Abuse/ Dependence in Illinois, 2003.............................................................................................................................27

20. Population Estimates of Those Meeting DSM-IV Criteria for Past-Year Substance Abuse/Dependence in Illinois, by Demographic Characteristics, 2003...................................28

21. Percentage Meeting DSM-IV Criteria for Past-Year Substance Abuse in Illinois, by Demographic Characteristics, 2003 .........................................................................................30

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003 v

22. Population Estimates of Those Meeting DSM-IV Criteria for Past-Year Substance Abuse in Illinois, by Demographic Characteristics, 2003 ................................................................... 31

23. Percentage Meeting DSM-IV Criteria for Past-Year Substance Dependence in Illinois, by Demographic Characteristics, 2003 .........................................................................................32

24. Population Estimates of Those Meeting DSM-IV Criteria for Past-Year Substance Dependence in Illinois, by Demographic Characteristics, 2003 ..............................................33

25. Estimated Population in Need of Treatment and Unmet Treatment Need in Illinois, by Gender, Race/Ethnicity, and Age, 2003: Based on 2003 Illinois Household Survey ..............34

26. Estimated Population in Need of Treatment and Unmet Treatment Need in Illinois, by Gender, Race/Ethnicity, and Age, 2003: Based on DASA’s Automated Reporting and Tracking System........................................................................................................................35

27. Substance Use Prevalence and Abuse/Dependence Treatment Need in Illinois, by Medical Health Insurance Coverage, 2003...............................................................................37

28. Substance Use Prevalence and Abuse/Dependence Treatment Need in Illinois, by Presence of Physical Disability, 2003.......................................................................................38

29. Substance Use Prevalence and Abuse/Dependence Treatment Need in Illinois, by Presence of Mental Disorder, 2003 ......................................................................................... 40

30. Substance Use Prevalence and Abuse/Dependence Treatment Need in Illinois, by Lifetime Experience with Homelessness, 2003........................................................................41

31. Substance Use Prevalence and Abuse/Dependence Treatment Need in Illinois, by Past-Year Receipt of Public Financial Support, 2003...............................................................43

32. Substance Use Prevalence and Abuse/Dependence Treatment Need Among Those Age 18+ in Illinois, by Domestic Violence Victimization Status, 2003 .......................................... 44

33. Substance Use Prevalence and Abuse/Dependence Treatment Need in Illinois, by Presence of Gambling Problem, 2003 ......................................................................................46

34. Estimated Population in Need of Treatment Among Vulnerable Populations in Illinois, 2003 ..........................................................................................................................................47

vi Illinois Division of Alcoholism & Substance Abuse

¿¿¿ FIGURES

1. Treatment Need in Illinois, by Gender, Age, Race/Ethnicity, and Marital Status, 2003...........x

2. Treatment Need in Illinois, by Educational Attainment, Employment Status, and Income, 2003..............................................................................................................................x

3. Treatment Need Among Vulnerable Populations in Illinois, 2003...........................................xi

4. Substance Use Prevalence in Illinois, 2003..............................................................................xii

5. Study Sample Areas.....................................................................................................................3

6. Geographic Distribution of Completed Interviews.....................................................................4

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003 vii

¿¿¿ ACKNOWLEDGMENTS

We wish to acknowledge the following staff of the University of Illinois at Chicago Survey Research Laboratory who participated in the design, implementation, and processing of this survey: Marni Basic, Kathleen Chan, Antonio Cox, Douglas Hammer, Christine Orland, Geoffrey Parker, Vincent Parker, Karen Foote Retzer, David Schipani, and Lisa Kelly-Wilson.

This study was funded by the U.S. Center for Substance Abuse Treatment under the State Treatment Needs Assessment Program. We wish to acknowledge the support of our project officer, Debbie Fulcher.

viii Illinois Division of Alcoholism & Substance Abuse

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003 ix Executive Summary

¿¿¿ EXECUTIVE SUMMARY

Introduction

Nationwide, alcohol and other drug abuse and addiction cause approximately 120,000 deaths each year as well as afflicting millions of others. Their economic costs exceed $240 billion a year and are likely to increase significantly over the next several decades.

Research specific to Illinois estimates that alcohol, tobacco, and other drug dependence cost our state over $6 billion a year. With an estimated State Fiscal Year 2001 spending level of over $213 million, the Illinois Division of Alcoholism and Substance Abuse (DASA) is the largest funder of addiction intervention and treatment services in Illinois.

DASA licenses all non-hospital-based treatment programs, approves Medicaid services and payments, monitors service delivery, and coordinates substance abuse as well as gambling issues across all State departments. As part of these responsibilities, DASA continually assesses the needs of Illinois residents.

This document reports findings from a statewide survey of substance use patterns and treatment need among Illinois citizens.

Methods

Telephone interviews were conducted with 4,155 randomly sampled persons age 16 and older. Details of this survey include the following:

® Interviews were completed between January and August 2003.

® All survey estimates were weighted to represent the population of Illinois.

® The study was designed to provide estimated treatment need for a variety of vulnerable populations, including

§ Those lacking health insurance coverage,

§ Persons with physical disabilities, § Individuals with mental health

conditions, § Public financial support recipients, § Persons ever having been homeless, § Victims of domestic violence, and § Chronic gamblers.

® The study was reviewed and monitored by the University of Illinois at Chicago Institutional Review Board.

® All surveys are subject to limitations. Analyses suggest that although sampling, coverage, and nonresponse sources of error may not seriously bias the findings of this study, the presence of measurement error may contribute to underestimates of substance use behaviors and treatment need.

Need for Alcohol and Other Drug Treatment

The treatment need of Illinois citizens was assessed using Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria for substance abuse and dependence. Key findings include the following:

® Overall, 16.4% of all persons age 16 and older are estimated to have had substance abuse/dependence treatment needs during the past year. This

x Illinois Division of Alcoholism & Substance Abuse Executive Summary

represents approximately 1,560,000 persons with treatment need.

® Need for substance abuse/dependence treatment varies by gender, age, race/ethnicity, marital status,

educational attainment, employment status, and household income. These differences are summarized in Figures 1 and 2.

16.4

23.6

9.6

24.8

19.5

15.017.8

11.0

6.3

18.8

7.7

12.214.2

22.4

15.4

12.2

0

5

10

15

20

25

30

Per

cent

Mal

e

Fem

ale

16

-24

25

-34

35

-44

45

-54

55

-64

65

+

Whi

te

Afric

an A

mer

ican

His

pani

c

Oth

er

Nev

er m

arrie

d

Mar

ried

Oth

er

Figure 1. Treatment Need in Illinois, by Gender, Age, Race/Ethnicity, and Marital

Status, 2003

Age Race/ Ethnicity

Marital status

Gender

TOTAL

Figure 2. Treatment Need in Illinois, by

Educational Attainment, Employment Status, and

Income, 2003 16.4

9.6

11.7

19.120.8 19.4

11.2

7.4

10.712.4 12.6

18.3

26.8

0

5

10

15

20

25

30

Per

cent

< H

igh

scho

ol

Hig

h sc

hool

gra

duat

e

Som

e co

llege

Colle

ge g

rad

+

Wor

king

Not

wor

king

Inco

me

<$1

0,00

0

$1

0,0

00

-<$

20

,00

0

$2

0,0

00

-<$

30

,00

0

$3

0,0

00

-<$

50

,00

0

$5

0,0

00

-<$

70

,00

0

$7

0,0

00

+TOTAL

Employment status

Income Educational attainment

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003 xi Executive Summary

® Treatment need for alcohol abuse/dependence is about six times greater than is treatment need for drug abuse/dependence.

® It also is estimated that in excess of 90% of those with past-year substance abuse/dependence treatment need did not access treatment services during the past year.

® Substance abuse/dependence treatment need is greater among persons with mental disorders, those ever having experienced domestic violence, persons with gambling problems, and persons with current health insurance coverage. These findings are presented in Figure 3.

Figure 3. Treatment Need Among Vulnerable Populations in Illinois, 2003

Prevalence of Alcohol and Other Drug Use

Prevalence rates for lifetime, past-year, and past-month alcohol, tobacco, and other drug use are summarized in Figure 4.

Alcohol is by far the most common substance ever and currently being used by Illinois citizens. About 30% of all persons age 16 and older are estimated to have engaged in

binge drinking at least once during the past year.

Marijuana is the most common illicit substance ever and currently being used by Illinois citizens. During the past year, 6.3% are estimated to have used marijuana at least once.

Eight percent of respondents reported either using an illicit drug (including

16.4 17.3

11.0

14.2

16.9

29.2

14.8

11.0

17.0 17.716.1

24.9

12.9

33.0

15.4

0

5

10

15

20

25

30

35

Per

cent

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Physical disability

Mental disorder

Ever homeless

Public financial support

recipient

Domestic violence vict im

Problem gambler

Heal th insurance coverage

TOTAL

xii Illinois Division of Alcoholism & Substance Abuse Executive Summary

marijuana, cocaine, crack, heroin, hallucinogens, inhalants, and club drugs) or misusing a prescription medication (including stimulants, sedatives, tranquilizers, and analgesics) during the past year.

Approximately one in five persons (19.8%) age 16 and older in Illinois reported current cigarette use.

Persons ever having smoked cigarettes reported an average age of first use of 16.4 years; the average age of first use was 17.7 for alcohol, 18.1 for marijuana, and 20.4 for any other drug.

Figure 4. Substance Use Prevalence in Illinois, 2003

Alcoho

l

Cigare

ttes

Any ill

icit dru

gs

Marijuan

a

Psycho

therap

eutics

Cocai

ne

Halluc

inogen

s

Inhalan

ts

Club d

rugs

Heroin

Past-month use

Past-year use

Lifetime use

83.9

41.8

32.631.6

9.38.7

5.74.1

2.4 0.8

69.5

22.4

6.7 6.2 2.50.8

0.5 0.3 0.50.1

54.0

19.8

3.6 3.31.4

0.3 0 0 0.10

0

20

40

60

80

100

Per

cent

Substance

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003 1

¿¿¿ INTRODUCTION AND OVERVIEW

Alcohol and other drug abuse and addiction continue to harm the health and shorten the life expectancy of many Americans (Rehm, Gmel, Sempos, & Trevisan, 2003; U.S. Dept. of Health and Human Services, 2000). Misuse of these substances is responsible for approximately 120,000 deaths each year (McGinnis & Foege, 1993). Millions more are afflicted by alcoholism or drug addiction. One or both of these conditions have been linked with weakened immune (Cook, 1998) and cardiovascular systems (York & Hirsch, 1997), liver disease (Dufour et al., 1993), breast cancer (Smith-Warner et al., 1998), and increased susceptibility to accidents and fatal injuries (Felson, Kiel, Anderson, & Kannel, 1988; Lucas, 1987). The economic costs of alcohol and other drug abuse, including medical care and lost earnings, are estimated to exceed $240 billion per year (Harwood, Fountain, & Livermore, 1998). These costs are likely to increase significantly during the next several decades (Gfroerer & Epstein, 1999). Illinois-specific studies estimate that alcohol, tobacco, and other drug abuse and dependence cost our state approximately $6 billion each year (Office of Alcoholism and Substance Abuse, 1999).

The Illinois Division of Alcoholism and Substance Abuse (DASA) is the largest funder of addiction intervention and treatment services in Illinois. With an estimated State Fiscal Year 2001 spending level of over $213 million (over 70% of which are Illinois funds), DASA must plan for and manage the addiction treatment system in a way that insures accountability, accessibility, and outcomes.

Nationally, the substance abuse treatment need of an aging Baby Boomer population is expected to triple by 2020 (Gfroerer, Penne, Pemberton, & Folsom, 2003). Accurate estimates of those affected by alcohol and other drugs and in need of intervention or treatment are essential for planning for additional services and evaluation of current penetration rates, lengths of stay, and outcomes by substance, age grouping, geographic location, and population type.

As the single State authority for all substance abuse issues in Illinois, DASA licenses all non-hospital-based treatment programs, approves certificates of need for Medicaid services, approves Medicaid payments (appropriation/spending authority), monitors service delivery, and coordinates substance abuse and gambling issues across all other State departments and offices. DASA works closely with service providers, the communities they serve, the recovering community, and other advocacy groups to promote and address issues through an integrated service delivery system. DASA has responsibility for planning, development, coordination, and funding of a comprehensive culturally and gender appropriate system of addiction services. This system must respond to the needs of rural and small-town America, small cities, metropolitan suburbia, and inner-city Chicago. The needs of at-risk and addicted individuals and their families must be met through its network of community-based providers. DASA must take into account and respond to changes in the economic, social, and

2 Illinois Division of Alcoholism & Substance Abuse

political milieu that may profoundly influence the service system and those it serves. Issues such as “welfare reform,” mandated targeted populations, changes in benefit programs, and outcome and performance management must be balanced within plans for a geographically and culturally diverse population. To meet these challenges, DASA continuously has sought to have the most appropriate population

and geographically specific estimates of need for intervention and treatment services possible. As part of DASA’s continuing commitment to basing decisions on the most accurate and timely information, a needs assessment survey of Illinois adults age 16 and older was conducted in 2003. This report provides a summary of findings from that survey.

¿¿¿ METHODOLOGY SUMMARY ¿¿¿

This report presents primary analyses from the 2003 Illinois Household Treatment Need Assessment Survey. A total of 4,155 random telephone interviews were completed between January 15 and August 15, 2003, by the University of Illinois at Chicago Survey Research Laboratory. The sample was stratified in order to produce reliable estimates for each of eight geographic areas. The boundaries of each sample area are provided in Appendix C and summarized in Figure 5. Wherever possible, sampled households were sent an advance letter introducing the survey. An example of this advance letter is reproduced in Appendix D.

Within each sampled household, one person age 16 or older was randomly selected to be interviewed. Where persons age 16–17 were selected, parental consent was obtained in advance. Interviews averaged 29.8 minutes in length (standard deviation = 12.7) and were concerned primarily with issues of substance use involvement, severity, and treatment need. A listing of the specific substances covered during these interviews is provided in Appendix B, along with the definitions of each that were employed in questioning respondents. The survey’s response rate was 32.7%. The parental approval rate for participation by their underage children was approximately 67% (see Appendix A).

All estimates reported in this document are weighted to represent the population of Illinois. Details of the survey’s sample design, response rate calculations, and sample weighting are provided in Appendix A. Because many statistical tests are included in this report, only those attaining p<.001 are considered significant. In keeping with common practice, however, other p-values, including those p<.05 and p<.01, also are provided (but not discussed).

Standard errors (SE) for all population estimates, which can be used to estimate 95% confidence intervals for each estimate, are provided in each table. Details regarding how to do this are provided in Appendix E.

All surveys are subject to important limitations. Those relevant to this study are briefly reviewed in the Discussion section. Additional details regarding the main sources of survey error, including sampling, coverage, nonresponse, and measurement error, are further explored in Appendices E–H. Finally, this study was reviewed, approved, and monitored by the University of Illinois at Chicago Institutional Review Board (IRB).


¿¿¿ RESULTS

Sample Characteristics

Table 1 on page 5 provides a demographic profile of the study sample. Column 1 presents breakdowns of the number of unweighted survey respondents within each gender, age, racial/ethnic, education, employment, income, marital status, and regional subgroup. Columns 2 and 3 compare the percentages of the sample within each of these subgroups for the unweighted sample vs. the final weighted sample. As described in Appendix A, the final weighted sample has been statistically adjusted to more closely reflect the demographic composition of the Illinois population age 16 and older, as of the 2000 Census.

The unweighted sample was very similar to the final weighted sample (i.e., less than a three percentage point difference for any subgroup) for several demographic characteristics, including gender, educational attainment, and household income. Differences in composition were found for age, race/ethnicity, employment status, marital status, and region of residence. To summarize, the unweighted sample tended to overrepresent persons age 65 and older, Whites, persons not working, and persons who are divorced,

separated, or widowed. The unweighted differences in the distribution of completed interviews across sample areas was an anticipated consequence of the survey’s stratified sample design, which sought to complete roughly equal numbers of interviews within each of the state’s eight sample areas. Figure 5 presents a graphic representation of these sample areas.

Figure 5. Study Sample Areas


Figure 6 provides a summary of the geographic distribution of completed interviews. Interviews were completed in all 102 Illinois counties. The final weighted

sample adjusted for discrepancies between the unweighted sample and the 2000 Census. All other statistical data presented in this report are based on this final weighted sample.

Figure 6. Geographic Distribution of Completed Interviews

“Collar” Counties

As Figure 5 shows, “collar” counties are those that form a ring around Cook County. Thus, the “northern collar counties” form the northern half of the ring; the “southern collar counties” form the south half.


Table 1. Demographic Characteristics of Il l inois Household Survey Sample, 2003

Demographic Characterist ic Frequency

Unweighted Percent Unweighted Percent Weighted Standard Error

TOTAL 4,155 100.0% 100.0% —

Gender Male 1,912 46.0 48.3 (1.4) Female 2,243 54.0 51.7 (1.4)

Age (in years) 16–24 879 21.2 18.5 (0.8) 25–34 657 15.8 18.9 (1.0) 35–44 681 16.4 18.7 (1.1) 45–54 596 14.3 16.3 (1.3) 55–64 575 13.8 13.3 (0.9) 65+ 767 18.5 14.2 (1.1)

Race/Ethnici ty White 3,291 79.2 70.9 (1.4) African American 382 9.2 13.7 (1.2) Hispanic 316 7.6 10.6 (1.0) Other 166 4.0 4.7 (0.6)

Education <High school 603 15.4 16.0 (1.4) High school graduate 857 21.9 20.1 (1.0) Some college, no degree 1,281 32.7 33.4 (1.4) Bachelor’s degree+ 1,181 30.1 30.5 (1.2)

Employment status Working 2,564 64.7 68.4 (1.4) Not working 1,401 35.3 31.6 (1.4)

Household income <$10,000 159 4.6 3.9 (0.4) $10,000–<$20,000 326 9.3 8.8 (1.0) $20,000–<$30,000 505 14.5 15.0 (1.4) $30,000–<$50,000 814 23.3 21.0 (1.1) $50,000–<$70,000 651 18.6 18.9 (1.2) $70,000+ 1,037 29.7 32.5 (1.4)

Marital status Never married 1,150 28.8 26.2 (1.0) Married 1,977 49.5 58.5 (1.3) Divorced/Separated/Widowed 863 21.6 15.3 (0.9)

Sample area Chicago 555 13.4 23.2 (1.1) Suburban Cook County 493 11.9 20.1 (0.9) Northwest Illinois 530 12.8 5.2 (0.3) South collar counties 533 12.8 5.0 (0.3) North collar counties 484 11.6 17.7 (0.9) West central Illinois 518 12.5 12.2 (0.5) East central Illinois 537 12.9 6.7 (0.3) Southern Illinois 505 12.2 9.9 (0.5)

A number of population subgroups of special interest as being at high risk for substance abuse treatment need also were

examined in this survey. Table 2 provides prevalence estimates for each of these groups for persons age 16 and older in Illinois.


Fourteen percent are estimated to be currently without health insurance coverage, and 19% are estimated to have a physical disability. Just over 10% are estimated to have a current (i.e., within the past year) mental disorder (11.1%). Approximately 17% of the population is estimated to have received public financial support in the past year

(17.3%). One in twenty persons have experienced homelessness at some time during their lives (4.8%). About 5% are also problem gamblers (5.5%). Just under one-third of all persons age 18 and older report ever having been a victim of domestic abuse/violence (30.3%).

Table 2. Vulnerable Populations: Prevalence Estimates for Il l inois, 2003

Vulnerable Population Frequency Percent Standard Error

TOTAL 4,155 100.0% —

Medical health insurance coverage Yes 3,555 85.6 (1.3) No 583 14.0 (1.3)

Physically disabled Yes 789 19.0 (1.3) No 3,366 81.0 (1.3)

Mental disorder Yes 462 11.1 (0.7) No 3,693 88.9 (0.7)

Recipient of public financ ial support Yes 718 17.3 (1.0) No 3,437 82.7 (1.0)

Ever homeless Yes 200 4.8 (0.9) No 3,955 95.2 (0.9)

Domestic violence victim (among those age 18+) Yes 1,192 30.3 (1.4) No 2,746 69.7 (1.4)

Problem gambler Yes 229 5.5 (0.6) No 3,926 94.5 (0.6)


Substance Use Prevalence Estimates

Table 3 summarizes the estimated prevalence of alcohol, tobacco, and other drug (ATOD) use among persons 16 years of age and older in Illinois. Alcohol use is by far most common in terms of both recent (past month, 54.0%) and prior (lifetime and past year, 83.9% and 69.5%, respectively) consumption, followed by cigarette use (41.8% lifetime use, 22.4% past-year use, 19.8% past-month use). Of the illicit substances examined in this report, marijuana was the most commonly used, reported to have been used recently (within the past month) by 3.3% of

respondents. Past-year use of marijuana was reported by 6.3% and lifetime use by 31.6%. Among other illicit drugs, lifetime cocaine use was most common (8.7%), followed by hallucinogens (5.7%), club drugs (2.4%), crack cocaine (1.5%), methamphetamines (1.3%), and heroin (0.8%). Any lifetime use of inhalants was estimated to be 4.1%.

Of the psychotherapeutic substances that often are used without a prescription, the lifetime nonmedical use of stimulants was reported by 5.5% of respondents. Lifetime analgesic use was reported by 2.9%, sedative use by 2.5%, and tranquilizer use by 2.1%.

Table 3. Lifetime, Past-Year, and Past-Month ATOD Use in Il l inois, 2003

LIFETIME USE PAST-YEAR USE PAST-MONTH USE Substance Percent Standard Error Percent Standard Error Percent Standard Error

Cigarettes 41.8% (1.4) 22.4% (1.3) 19.8% (1.2)

Alcohol 83.9 (1.3) 69.5 (1.4) 54.0 (1.4)

Any ill icit drugs 32.6 (1.3) 6.7 (0.5) 3.6 (0.4) Marijuana 31.6 (1.3) 6.2 (0.5) 3.3 (0.4) Cocaine 8.7 (0.9) 0.8 (0.2) 0.3 (0.1) Crack 1.5 (0.3) 0.0 (0.0) 0.0 (0.0) Hallucinogens 5.7 (0.5) 0.5 (0.1) 0.0 (0.0) Club drugs 2.4 (0.4) 0.5 (0.1) 0.1 (0.1) Heroin 0.8 (0.2) 0.1 (0.0) 0.0 (0.0) Methamphetamines 1.3 (0.2) 0.2 (0.1) 0.1 (0.0) Other drugs 1.0 (0.2) 0.2 (0.1) 0.1 (0.0)

Nonmedical psychotherapeutics 9.3 (0.8) 2.5 (0.4) 1.4 (0.4) Analgesics 2.9 (0.4) 1.2 (0.3) 0.6 (0.2) Stimulants 5.5 (0.6) 0.2 (0.1) 0.0 (0.1) Tranquilizers 2.1 (0.4) 0.7 (0.3) 0.4 (0.3) Sedatives 2.5 (0.4) 1.0 (0.2) 0.6 (0.2)

Inhalants 4.1 (0.5) 0.3 (0.1) 0.0 (0.0)


¿ Cigarette use

Detailed prevalence estimates of cigarette

use in Illinois are presented in Table 4. About 20% of persons age 16 and older reported current cigarette use, and past-year use was reported by 22.4%. About two of every five persons interviewed indicated lifetime smoking (41.8%).

Age, race/ethnicity, household income, and marital status all were found to be associated with cigarette use. Specifically, older persons were more likely to report ever having smoked cigarettes. Whites and African Americans were more likely than Hispanics and persons from other racial/ethnic backgrounds to report lifetime cigarette use.

Household income also was associated with cigarette use, although not in a linear

manner. Persons reporting incomes between $20,000 and $30,000 per year were more likely to report lifetime cigarette use.

A majority of divorced, separated, or widowed persons reported ever having used cigarettes; never-married persons were least likely to ever have smoked. Currently married persons were least likely to have smoked cigarettes during the past year and during the past month.

When examining regional variation, lifetime cigarette use was found to be lowest in the northern collar counties and east central Illinois.

Table 5 provides estimates of the number of current and past cigarette users in Illinois. It is estimated that approximately 1,883,320 persons age 16 and older were recent users of cigarettes in 2003.


Table 4. Cigarette Use in Ill inois, by Demographic Characteristics, 2003

LIFETIME USE PAST-YEAR USE PAST-MONTH USE Demographic Characteristic Percent Standard Error Percent Standard Error Percent Standard Error

TOTAL 41.8% (1.4) 22.4% (1.3) 19.8% (1.2)

Gender ** * Male 45.6 (2.2) 25.4 (2.2) 21.9 (2.2) Female 38.3 (1.8) 19.5 (1.3) 17.7 (1.3)

Age (in years) *** 16–24 31.2 (2.0) 28.1 (2.0) 25.1 (1.9) 25–34 32.5 (2.4) 24.1 (2.2) 20.9 (2.1) 35–44 37.4 (3.3) 20.4 (2.5) 18.8 (2.4) 45–54 46.6 (4.4) 24.8 (3.3) 21.9 (3.2) 55–64 56.9 (3.8) 19.5 (2.6) 16.2 (2.3) 65+ 54.3 (4.1) 15.1 (6.1) 13.5 (6.2)

Race/Ethnici ty *** White 45.5 (1.5) 22.9 (1.2) 20.1 (1.1) African American 40.7 (5.3) 26.7 (6.0) 24.4 (6.1) Hispanic 26.0 (4.1) 16.3 (2.8) 14.7 (2.7) Other 25.4 (4.7) 14.7 (4.0) 13.7 (3.9)

Education <High school 35.1 (5.2) 25.1 (5.6) 22.7 (5.6) High school graduate 45.2 (2.8) 23.7 (2.1) 22.3 (2.1) Some college, no degree 40.6 (2.4) 23.7 (1.9) 21.2 (1.8) Bachelor’s degree+ 38.8 (2.3) 18.7 (1.9) 15.3 (1.8)

Employment status Working 40.7 (1.7) 23.5 (1.4) 20.8 (1.3) Not working 43.7 (2.8) 20.4 (3.0) 18.3 (3.0)

Household income *** * * <$10,000 45.6 (5.6) 30.9 (5.8) 28.9 (5.8) $10,000–<$20,000 36.8 (5.0) 19.9 (3.2) 17.5 (3.0) $20,000–<$30,000 51.3 (5.4) 32.8 (6.2) 30.1 (6.3) $30,000–<$50,000 46.1 (2.8) 25.1 (2.4) 23.1 (2.3) $50,000–<$70,000 40.3 (3.4) 22.2 (2.5) 19.8 (2.3) $70,000+ 40.7 (2.5) 18.8 (2.0) 15.7 (1.9)

Marital status *** ** ** Never married 33.2 (1.9) 28.5 (1.8) 25.4 (1.7) Married 42.1 (2.1) 19.4 (2.0) 16.9 (2.0) Divorced/Separated/Widowed 53.8 (3.0) 24.1 (2.7) 22.4 (2.7)

Sample area Chicago 41.3 (3.8) 26.2 (3.9) 23.4 (3.9) Suburban Cook County 43.6 (3.4) 20.6 (2.5) 17.1 (2.3) Northwest Illinois 41.5 (3.3) 20.7 (2.7) 17.1 (2.4) South collar counties 46.2 (3.6) 19.8 (2.0) 17.8 (1.9) North collar counties 35.3 (3.2) 18.7 (2.6) 16.7 (2.5) West central Illinois 47.1 (3.2) 21.8 (2.5) 20.2 (2.5) East central Illinois 36.0 (2.9) 20.9 (2.5) 18.7 (2.4) Southern Illinois 46.5 (3.5) 27.1 (3.1) 24.5 (3.0)

*p<.05; **p<.01; ***p<.001.


Table 5. Population Estimates of Cigarette Use in Il l inois, by Gender, Age, and Race/Ethnicity, 2003

LIFETIME USE PAST-YEAR USE PAST-MONTH USE Demographic Character ist ic Percent Population† Percent Population† Percent Population†

TOTAL 41.8% 3,986 22.4% 2,130 19.8% 1,883

Gender Male 45.6 2,098 25.4 1,170 21.9 1,009 Female 38.3 1,888 19.5 960 17.7 874

Age 16–24 31.2 552 28.1 496 25.1 442 25–34 32.5 584 24.1 433 20.9 376 35–44 37.4 667 20.4 364 18.8 336 45–54 46.6 724 24.8 385 21.9 340 55–64 56.9 723 19.5 248 16.2 206 65+ 54.3 736 15.1 204 13.5 183

Race/Ethnici ty White 45.5 3,078 22.9 1,550 20.1 1,355 African American 40.7 531 26.7 348 24.4 318 Hispanic 26.0 263 16.3 165 14.7 148 Other 25.4 114 14.7 68 13.7 61

†In thousands.

¿ Alcohol use

Compared to cigarette use, alcohol

consumption is far more common in Illinois (see Table 6). Over 83% of persons age 16 and older reported ever having used alcohol; approximately 70% reported past-year usage, and more than half drank alcohol in the past month (54.0%).

Compared to females, males were more likely to have consumed alcohol during the past year and the past month.

The lowest rates of lifetime alcohol use were reported by respondents under age 25 and over age 64. Respondents 65 and older

also were less likely to report past-year and past-month alcohol consumption.

Alcohol use was greater among Whites for each time interval examined.

In addition, lifetime, past-year, and past-month alcohol consumption consistently increased with education: respondents who did not graduate from high school were least likely to report using alcohol, while college graduates were most likely to be or to have ever been alcohol drinkers.

Employment status also was associated with alcohol use. For each time interval examined, persons currently employed full- or part-time were more likely to report alcohol use.


Table 6. Alcohol Use in Ill inois, by Demographic Characteristics, 2003

LIFETIME USE PAST-YEAR USE PAST-MONTH USE Demographic Characteristic Percent Standard Error Percent Standard Error Percent Standard Error

TOTAL 83.9% (1.3) 69.5% (1.4) 54.0% (1.4)

Gender ** *** *** Male 88.4 (2.0) 74.8 (2.1) 60.8 (2.2) Female 79.7 (1.6) 64.6 (1.8) 47.7 (1.9)

Age (in years) *** *** ** 16–24 75.6 (1.8) 67.4 (2.0) 49.7 (2.1) 25–34 87.9 (1.7) 80.4 (2.1) 59.8 (2.8) 35–44 88.8 (2.2) 74.6 (2.8) 59.6 (3.3) 45–54 88.1 (3.9) 74.2 (4.1) 60.2 (4.3) 55–64 90.6 (1.4) 72.7 (2.9) 57.2 (3.8) 65+ 72.0 (5.6) 43.1 (3.9) 34.7 (3.4)

Race/Ethnici ty *** *** *** White 90.4 (0.7) 76.0 (1.1) 61.8 (1.4) African American 66.2 (5.7) 48.1 (4.7) 37.4 (4.1) Hispanic 70.8 (5.6) 59.3 (5.4) 31.9 (4.4) Other 67.0 (6.0) 57.9 (6.1) 35.5 (5.6)

Education *** *** *** <High school 59.0 (5.4) 42.9 (4.6) 25.9 (3.1) High school graduate 83.2 (2.1) 63.3 (2.8) 48.3 (2.8) Some college 90.3 (1.1) 75.8 (1.9) 57.8 (2.4) Bachelor’s degree+ 90.9 (1.3) 82.4 (1.7) 69.6 (2.0)

Employment status *** *** *** Working 89.6 (1.0) 77.7 (1.3) 61.0 (1.7) Not working 73.0 (3.2) 54.2 (2.8) 41.0 (2.5)

Household income *** *** *** <$10,000 66.9 (5.7) 48.5 (5.5) 35.3 (5.1) $10,000–<$20,000 71.0 (7.1) 50.2 (6.0) 37.8 (5.2) $20,000–<$30,000 82.1 (6.8) 63.4 (6.0) 45.1 (5.2) $30,000–<$50,000 84.4 (2.3) 66.6 (2.8) 49.9 (2.7) $50,000–<$70,000 88.7 (1.8) 75.4 (2.7) 56.1 (3.6) $70,000+ 93.1 (1.1) 86.0 (1.6) 73.5 (2.0)

Marital status ** ** ** Never married 77.9 (1.8) 69.7 (1.9) 50.9 (2.0) Married 86.9 (2.1) 72.9 (2.1) 58.3 (2.2) Divorced/Separated/Widowed 85.0 (1.7) 60.4 (3.0) 46.5 (3.0)

Sample area ** ** * Chicago 75.9 (4.4) 64.4 (4.1) 51.0 (3.7) Suburban Cook County 84.8 (2.1) 71.1 (2.8) 57.6 (3.2) Northwest Illinois 87.6 (1.8) 74.2 (2.7) 53.9 (3.4) South collar counties 88.9 (1.5) 77.8 (2.6) 59.9 (3.2) North collar counties 86.4 (1.9) 76.3 (2.5) 61.2 (3.6) West central Illinois 86.1 (2.3) 65.3 (3.2) 47.1 (3.1) East central Illinois 87.8 (1.5) 73.0 (2.5) 53.2 (3.3) Southern Illinois 86.4 (1.7) 62.2 (3.4) 46.7 (3.5)

*p<.05; **p<.01; ***p<.001.


Table 7. Population Estimates of Alcohol Use in Il l inois, by Gender, Age, and Race/Ethnicity, 2003

LIFETIME USE PAST-YEAR USE PAST-MONTH USE Demographic Characteristic Percent Population† Percent Population† Percent Population†

TOTAL 83.9% 7,996 69.5% 6,627 54.0% 5,150

Gender Male 88.4 4,066 74.8 3,441 60.8 2,800 Female 79.7 3,930 64.6 3,186 47.7 2,350

Age (in years) 16–24 75.6 1,334 67.4 1,191 49.7 877 25–34 87.9 1,579 80.4 1,444 59.8 1,075 35–44 88.8 1,586 74.6 1,331 59.6 1,065 45–54 88.1 1,369 74.2 1,153 60.2 936 55–64 90.6 1,152 72.7 924 57.2 726 65+ 72.0 976 43.1 584 34.7 470

Race/Ethnici ty White 90.4 6,112 76.0 5,138 61.8 4,177 African American 66.2 862 48.1 627 37.4 487 Hispanic 70.8 719 59.3 601 31.9 324

†In thousands.

Household income was associated with

alcohol consumption in a linear manner. For lifetime, past-year, and past-month use, consumption increased with household income.

Persons who have never been married were least likely to have ever used alcohol. Those who were divorced, separated, or widowed were least likely to report current or past-year alcohol use.

With regard to regional variation, lifetime use of alcohol was lowest in Chicago. Past-year use was lowest in southern Illinois, west central Illinois, and Chicago. Recent alcohol use was lowest in southern and west central Illinois.

Estimates of the number of Illinois citizens who are current or past alcohol consumers are provided in Table 7.

Binge drinking is a commonly employed indicator of alcohol-related problems (Naimi et al., 2003). In these analyses, binge drinking is defined as the consumption of four or more drinks by a female or five or more drinks by a male on one occasion (within several hours) during the past year. Estimates of binge drinking during the past year are provided in Table 8. Statewide, it is estimated that 30.1% of all persons age 16 and older engaged in binge drinking at least once during the past year.

Binge drinking was more common among males, persons age 16–34, Whites, those with at least some college education, the currently employed, those with higher incomes, and never-married individuals.


Table 8. Past-Year Binge Drinking in Illinois, by Demographic Characteristics, 2003

Demographic Characterist ic Percent Standard Error

TOTAL 30.1% (1.2)

Gender *** Male 38.4 (2.0) Female 22.4 (1.3)

Age (in years) *** 16–24 44.9 (2.1) 25–34 45.7 (2.8) 35–44 29.1 (2.7) 45–54 26.5 (4.1) 55–64 18.6 (2.8) 65+ 6.6 (1.3)

Race/Ethnici ty *** White 34.3 (1.4) African American 15.1 (2.4) Hispanic 26.2 (4.7) Other 19.8 (4.8)

Education *** <High school 21.6 (3.3) High school graduate 24.4 (2.1) Some college, no degree 33.5 (2.3) Bachelor’s degree+ 34.6 (2.1)

Employment status *** Working 34.8 (1.6) Not working 20.2 (1.7)

Household income *** <$10,000 25.2 (4.7) $10,000–<$20,000 18.2 (3.1) $20,000–<$30,000 26.0 (4.6) $30,000–<$50,000 26.2 (2.1) $50,000–<$70,000 34.1 (3.0) $70,000+ 40.2 (2.4)

Marital status *** Never married 42.7 (2.0) Married 27.5 (1.8) Divorced/Separated/Widowed 18.1 (1.9)

Sample area Chicago 31.6 (3.0) Suburban Cook County 29.7 (3.3) Northwest Illinois 29.0 (2.8) South collar counties 31.9 (2.9) North collar counties 27.0 (2.9) West central Illinois 29.5 (2.8) East central Illinois 36.0 (3.3) Southern Illinois 29.6 (3.0)

***p<.001.

¿ Marijuana use

Estimates of marijuana use are presented

in Table 9. The lifetime prevalence of marijuana use in Illinois was 31.6% among persons 16 years of age and older. Past-year prevalence was 6.3%, while 3.3% of respondents reported past-month use.

For all time intervals, males were more likely to be marijuana users.

Past-year and past-month marijuana use were highest among the youngest respondents (those age 16–24) and lowest among the oldest respondents (age 65 and older).

Education was positively associated with lifetime use of this substance, with respondents with at least a bachelor’s degree most likely to report ever having used marijuana.

Persons currently employed full- or part-time were most likely to report lifetime marijuana use.

Lifetime marijuana use increased with household income. While less than 21% of respondents with a household income of less than $20,000 reported ever having used this substance, almost 45% of those in the highest income category ($70,000+) did so.

Marital status was associated with marijuana use during each time interval examined, with never-married respondents most likely to report marijuana use.

Table 10 provides estimates of the numbers of current and past marijuana users in Illinois. Although it is estimated that approximately 3 million persons have used marijuana at some time during their lives, the number of recent (i.e., past-month) users is estimated to be 317,743. Most recent users are male, between the ages of 16–24, and White.


Table 9. Marijuana Use in Il l inois, by Demographic Characteristics, 2003

LIFETIME USE PAST-YEAR USE PAST-MONTH USE Demographic Characterist ic Percent Standard Error Percent Standard Error Percent Standard Error

TOTAL 31.6% (1.3) 6.2% (0.5) 3.3% (0.4)

Gender *** *** *** Male 38.6 (2.1) 8.9 (1.0) 5.2 (0.8) Female 25.1 (1.5) 3.8 (0.5) 1.5 (0.3)

Age (in years) * *** *** 16–24 36.6 (2.1) 19.7 (1.8) 11.0 (1.5) 25–34 39.8 (2.7) 7.8 (1.6) 4.3 (1.4) 35–44 43.9 (3.4) 3.6 (0.8) 1.5 (0.5) 45–54 38.2 (4.4) 2.1 (0.6) 0.9 (0.4) 55–64 18.4 (2.5) 0.7 (0.4) 0.4 (0.3) 65+ 2.9 (0.8) 0.2 (0.2) 0.0 (0.0)

Race/Ethnici ty * White 33.7 (1.5) 6.6 (0.6) 3.4 (0.5) African American 31.0 (3.6) 5.3 (1.3) 3.5 (1.0) Hispanic 22.2 (4.5) 3.6 (1.0) 1.4 (0.5) Other 23.7 (5.0) 9.6 (3.8) 6.6 (3.4)

Education *** * <High school 19.5 (3.1) 8.5 (1.5) 5.1 (1.2) High school graduate 23.1 (2.2) 3.7 (0.9) 1.8 (0.5) Some college, no degree 35.1 (2.3) 6.7 (0.9) 3.7 (0.7) Bachelor’s degree+ 39.1 (2.3) 5.4 (1.1) 2.6 (0.9)

Employment status *** Working 37.0 (1.6) 6.4 (0.7) 3.1 (0.5) Not working 19.3 (1.6) 5.7 (0.9) 3.4 (0.7)

Household income *** ** <$10,000 20.6 (3.9) 5.2 (1.7) 3.7 (1.4) $10,000–<$20,000 20.5 (3.4) 4.9 (1.3) 2.6 (1.0) $20,000–<$30,000 24.8 (4.2) 4.0 (1.0) 2.3 (0.8) $30,000–<$50,000 27.3 (2.3) 3.5 (0.7) 1.8 (0.5) $50,000–<$70,000 36.0 (3.5) 8.1 (1.5) 4.1 (1.1) $70,000+ 44.8 (2.4) 8.6 (1.3) 4.2 (1.1)

Marital status *** *** *** Never married 37.4 (1.9) 15.6 (1.4) 8.5 (1.1) Married 30.4 (1.9) 3.0 (0.6) 1.5 (0.5) Divorced/Separated/Widowed 24.7 (2.6) 2.0 (0.6) 1.0 (0.4)

Sample area Chicago 31.7 (2.9) 8.3 (1.4) 4.3 (1.1) Suburban Cook County 36.5 (3.6) 5.6 (1.0) 3.2 (0.8) Northwest Illinois 30.4 (3.3) 4.6 (1.0) 1.4 (0.5) South collar counties 32.4 (3.2) 5.8 (1.0) 3.4 (0.8) North collar counties 34.2 (3.4) 7.5 (1.7) 3.9 (1.4) West central Illinois 28.3 (2.8) 5.6 (1.2) 3.0 (0.9) East central Illinois 30.9 (3.1) 4.7 (1.0) 2.9 (0.8) Southern Illinois 21.8 (2.5) 3.6 (0.9) 1.9 (0.7)

*p<.05; **p<.01; ***p<.001.


Table 10. Population Estimates of Marijuana Use in Illinois, by Gender, Age, and Race/Ethnicity, 2003

LIFETIME USE PAST-YEAR USE PAST-MONTH USE Demographic Characteristic Percent Population† Percent Population† Percent Population†

TOTAL 32.6% 3,014 6.7% 595 3.6% 318

Gender Male 40.1 1,776 9.5 409 5.6 242 Female 25.6 1,238 4.0 185 1.6 76

Age (in years) 16–24 37.9 647 20.8 348 11.3 194 25–34 42.1 716 7.9 140 4.6 77 35–44 44.5 785 4.2 63 1.8 27 45–54 39.1 593 2.5 33 1.3 15 55–64 18.4 234 0.7 8 0.4 4 65+ 3.0 39 0.2 2 0.0 —

Race/Ethnici ty White 34.1 2,278 7.0 445 3.7 230 African American 31.9 404 5.5 69 3.5 45 Hispanic 27.1 225 4.5 36 1.6 13 Other 25.4 107 9.6 595 6.6 30

†In thousands.

¿ Other illicit drug use

Lifetime prevalence estimates for several

other illicit drugs are presented in Table 11. The highest estimate was for cocaine and crack cocaine use (8.8%). Use of these substances was greatest among currently employed respondents, those age 35–44, and those reporting household incomes between $20,000 and $30,000.

Less than one percent (0.8%) of respondents indicated they ever had used heroin. Race/ethnicity was associated with lifetime use of this drug, however, with African-Americans most likely to report they ever had used it.

Lifetime use of hallucinogens was estimated to be 5.7%. Use was higher among the following subgroups: males, persons age 25–34, and Whites.

Just over 4% of those interviewed had ever used inhalants. This type of substance was more likely to have been tried by respondents age 25–34.

Table 11 also presents, in the final column, estimates of lifetime club drug use in Illinois. The overall rate among persons age 16 and older was 2.4%. Persons age 16–24 were more likely to report club drug use. College graduates also were more likely to ever have used club drugs, as were persons who never had been married.


Table 11. Lifetime Use of Crack/Cocaine, Heroin, Hallucinogens, Inhalants, and Club Drugs in Illinois, by Demographic Characteristics, 2003

Crack/Cocaine Heroin Hallucinogens Inhalants Club Drugs Demographic Character ist ic % SE % SE % SE % SE % SE

TOTAL 8.8% (0.9) 0 .8% (0.2) 5 .7% (0.5) 4 .1% (0.5) 2 .4% (0.4)

Gender ** *** * ** Male 11.6 (1.6) 1.0 (0.3) 8.1 (0.5) 5.6 (0.8) 3.3 (0.7) Female 6.2 (1.0) 0.6 (0.2) 3.4 (0.9) 2.8 (0.7) 1.5 (0.3)

Age ( in years) *** *** *** *** 16–24 5.3 (0.9) 0.3 (0.2) 6.9 (1.1) 5.1 (0.9) 6.4 (1.0) 25–34 10.5 (2.3) 1.2 (0.6) 11.0 (1.8) 7.3 (1.5) 5.2 (1.5) 35–44 17.0 (2.6) 0.7 (0.3) 7.2 (1.2) 6.3 (1.9) 0.8 (0.4) 45–54 14.5 (3.8) 1.8 (0.8) 4.7 (1.0) 2.7 (0.8) 0.6 (0.4) 55–64 2.0 (0.7) 0.5 (0.3) 1.6 (0.6) 1.4 (0.5) 0.0 (0.0) 65+ 0.2 (0.2) 0.0 (0.0) 0.2 (0.2) 0.3 (0.2) 0.0 (0.0)

Race/Ethnicity *** *** * White 9.5 (1.1) 0.5 (0.2) 7.0 (0.7) 4.8 (0.5) 2.6 (0.5) African American 5.6 (1.5) 2.7 (0.2) 1.8 (0.9) 0.7 (0.4) 1.4 (0.8) Hispanic 10.6 (4.5) 0.0 (0.0) 1.8 (0.7) 5.2 (3.1) 1.1 (0.6) Other 3.6 (1.8) 0.3 (0.2) 5.6 (2.2) 2.0 (1.0) 5.1 (2.4)

Educat ion * ** ** *** <High school 5.6 (2.6) 1.4 (0.8) 2.1 (0.6) 1.3 (0.4) 1.1 (0.4) High school graduate 5.7 (0.9) 0.3 (0.1) 4.6 (0.9) 2.7 (0.7) 0.6 (0.3) Some college, no degree 12.2 (2.0) 0.6 (0.3) 7.0 (0.9) 5.4 (1.2) 2.6 (0.6) Bachelor’s degree+ 8.1 (1.0) 0.6 (0.3) 7.4 (1.2) 5.8 (1.1) 4.3 (1.0)

Employment status *** ** * Working 10.2 (1.2) 0.8 (0.2) 6.7 (0.7) 4.9 (0.7) 2.9 (0.5) Not working 4.7 (0.7) 0.5 (0.2) 3.8 (0.7) 2.9 (0.6) 1.6 (0.4)

Household income ** * <$10,000 3.7 (1.7) 1.3 (1.3) 3.7 (1.4) 0.7 (0.7) 2.6 (1.3) $10,000–<$20,000 6.2 (1.8) 3.0 (1.6) 3.9 (1.2) 3.3 (1.1) 2.0 (1.0) $20,000–<$30,000 13.5 (4.6) 0.2 (0.1) 6.7 (1.7) 2.9 (1.0) 2.9 (0.8) $30,000–<$50,000 8.1 (1.7) 0.5 (0.3) 4.3 (0.8) 3.1 (0.6) 1.1 (0.4) $50,000–<$70,000 8.1 (1.3) 0.7 (0.3) 7.3 (1.3) 5.7 (1.1) 2.7 (0.9) $70,000+ 10.4 (1.6) 0.6 (0.3) 7.9 (1.3) 6.6 (1.6) 4.1 (1.1)

Mari ta l s tatus * ** *** Never married 7.1 (0.9) 1.3 (0.5) 8.6 (1.1) 5.6 (0.8) 6.3 (1.0) Married 8.6 (1.3) 0.4 (0.2) 4.5 (0.6) 4.0 (0.8) 1.3 (0.5) Divorced/Separated/ Widowed 10.2 (2.3) 0.6 (0.4) 5.8 (1.4) 3.1 (0.8) 0.3 (0.2)

Sample area * Chicago 10.1 (2.3) 1.1 (0.5) 4.6 (1.0) 6.0 (1.7) 3.1 (0.8) Suburban Cook County 12.2 (3.3) 0.9 (0.5) 4.6 (1.0) 2.6 (0.7) 1.9 (0.7) Northwest Illinois 7.8 (1.7) 0.8 (0.4) 6.9 (1.8) 4.1 (1.5) 0.9 (0.4) South collar counties 10.6 (2.4) 3.1 (2.0) 8.2 (2.3) 3.8 (0.8) 2.4 (0.6) North collar counties 7.6 (1.4) 0.3 (0.2) 7.9 (1.8) 4.5 (1.5) 3.4 (1.4) West central Illinois 6.3 (1.2) 0.1 (0.1) 5.3 (1.1) 3.6 (0.9) 1.8 (0.7) East central Illinois 5.0 (0.9) 0.2 (0.1) 5.4 (1.1) 3.6 (0.8) 2.5 (0.8) Southern Illinois 6.1 (1.6) 0.5 (0.3) 5.3 (1.1) 3.6 (0.9) 1.5 (0.7)

*p<.05; **p<.01; ***p<.001.


Table 12. Lifetime & Past-Year Illicit Drug† Use in Ill inois, by Demographic Characteristics, 2003

ILLICIT DRUGS INCLUDING MARIJUANA ILLICIT DRUGS EXCLUDING MARIJUANA Lifet ime Use Past -Year Use Lifetime Use Past -Year Use

Demographic Characteristic Percent Standard

Error Percent Standard



Error

TOTAL 32.6% (1.3) 6.7% (0.5) 12.2% (1.0) 1.5% (0.2)

Gender *** *** *** *** Male 40.1 (2.1) 9.5 (1.0) 16.3 (1.7) 2.3 (0.4) Female 25.6 (1.5) 4.0 (0.5) 8.4 (1.0) 0.8 (0.2)

Age (in years) *** *** *** *** 16–24 37.9 (2.1) 20.8 (1.8) 11.8 (1.4) 5.5 (0.9) 25–34 42.1 (2.8) 7.9 (1.6) 16.5 (2.6) 1.3 (0.5) 35–44 44.5 (3.4) 4.2 (0.8) 19.3 (2.6) 0.9 (0.3) 45–54 39.1 (4.4) 2.5 (0.7) 17.6 (3.8) 0.7 (0.3) 55–64 18.4 (2.5) 0.7 (0.4) 2.9 (0.8) 0.0 (0.0) 65+ 3.0 (0.8) 0.2 (0.2) 0.5 (0.3) 0.0 (0.0)

Race/Ethnici ty White 34.1 (1.5) 7.0 (0.6) 13.3 (1.2) 1.8 (0.3) African American 31.9 (3.7) 5.5 (1.3) 8.3 (1.9) 0.7 (0.5) Hispanic 27.1 (5.2) 4.5 (1.1) 12.2 (4.5) 1.4 (0.6) Other 25.4 (5.1) 9.6 (3.8) 7.6 (2.8) 1.0 (0.9)

Education *** * ** <High school 22.9 (3.8) 9.4 (1.6) 7.7 (2.7) 2.2 (0.7) High school graduate 23.6 (2.2) 4.0 (0.9) 7.5 (1.1) 1.1 (0.4) Some college, no degree 35.7 (2.3) 7.1 (0.9) 16.2 (2.1) 1.8 (0.4) Bachelor’s degree+ 39.7 (2.3) 5.7 (1.1) 12.9 (1.4) 1.3 (0.4)

Employment status *** *** Working 38.3 (1.7) 6.8 (0.7) 14.3 (1.3) 1.6 (0.3) Not working 19.5 (1.6) 6.0 (0.9) 7.0 (0.9) 1.4 (0.4)

Household income *** ** <$10,000 21.3 (4.0) 5.9 (1.9) 5.9 (2.1) 2.9 (1.4) $10,000–<$20,000 22.1 (3.6) 5.9 (1.5) 10.4 (2.4) 2.6 (0.9) $20,000–<$30,000 28.4 (4.8) 4.8 (1.1) 17.6 (4.6) 2.7 (0.9) $30,000–<$50,000 27.5 (2.3) 3.9 (0.7) 9.5 (1.7) 0.9 (0.4) $50,000–<$70,000 36.4 (3.5) 8.6 (1.5) 12.7 (1.8) 1.4 (0.5) $70,000+ 45.6 (2.4) 8.8 (1.3) 15.4 (1.8) 1.6 (0.5)

Marital status *** *** *** Never married 38.8 (1.9) 16.6 (1.5) 13.8 (1.4) 4.3 (0.7) Married 31.4 (2.0) 3.2 (0.6) 11.1 (1.4) 0.6 (0.2) Divorced/Separated/Widowed 25.0 (2.6) 2.1 (0.7) 11.9 (2.3) 0.5 (0.3)

Sample area Chicago 34.9 (3.2) 9.2 (1.5) 14.2 (2.5) 2.3 (0.6) Suburban Cook 36.7 (3.6) 5.8 (1.0) 14.6 (3.3) 0.6 (0.3) Northwest Illinois 31.3 (3.3) 4.6 (1.0) 11.1 (2.0) 0.9 (0.4) South collar counties 33.3 (3.2) 6.3 (1.1) 14.0 (2.5) 1.9 (0.7) North collar counties 34.2 (3.4) 7.6 (1.7) 12.1 (2.0) 1.1 (0.4) West central Illinois 28.5 (2.8) 5.9 (1.2) 8.5 (1.4) 2.0 (0.7) East central Illinois 31.3 (3.1) 5.4 (1.2) 10.1 (2.0) 2.5 (0.9) Southern Illinois 22.0 (2.5) 3.9 (1.0) 8.7 (1.7) 1.7 (0.7)

†Marijuana, cocaine, crack, heroin, methamphetamines, hallucinogens, club drugs, and other unspecified drugs. *p<.05; **p<.01; ***p<.001.


Lifetime and past-year prevalence estimates for the use of any illicit drugs are shown in the first two columns of Table 12. About a third of all persons age 16 years and older are estimated to ever have used an illicit drug (32.6%). The proportion estimated to have used an illicit drug in the past year is 6.7%. For both time frames, males were more likely to report any illicit drug use.

Lifetime use of any illicit drug was greatest among persons age 35–44, while past-year use was highest among those age 16–24.

Education and income were associated with lifetime illicit drug use. Those earning $70,000 or more were most likely to report lifetime use, as were respondents with at least some college education.

Never-married persons were more likely to be lifetime and past-year users of illicit drugs.

Differences in lifetime illicit drug use also were observed by employment status, with over 38% of the currently employed reporting such use. Less than 20% of the unemployed indicated they had ever used an illicit drug.

The final two columns of Table 9 examine the prevalence of any illicit drug use exclusive of marijuana, which is the most commonly used illicit substance. The lifetime prevalence estimate of use of any illicit drug other than marijuana is 12.2%. The past year estimate is 1.5%.

Males and respondents age 35–44 were more likely to report lifetime and past-year use of an illicit drug other than marijuana. With regard to recent use, those in the youngest age category (16–24) and never-married persons were most likely to have used such a substance in the past year.

Estimates provided in Table 13 indicate that approximately 1.1 million persons in Illinois age 16 and older ever have used an

illicit drug other than marijuana and that 148,205 are recent (i.e., past month) users. Most recent users were male, age 19–24, and White.

Table 13. Population Estimates of Illicit Drug† Use in Illinois, by Gender, Age, and Race/Ethnicity, 2003

LIFETIME USE PAST-YEAR USE Demographic Character is t ic Percent Population‡ Percent Population‡

TOTAL 12.2% 1,165 1.5% 148 Gender Male 16.3 749 2.3 107 Female 8.4 416 0.8 42

Age ( in years) 16–24 11.8 208 5.5 96 25–34 16.5 296 1.3 24 35–44 19.3 344 0.9 17 45–54 17.6 273 0.7 10 55–64 2.9 36 0.0 — 65+ 0.5 7 0.0 —

Race/Ethnicity White 13.3 900 1.8 121 African American 8.3 107 0.7 10 Hispanic 12.2 124 1.4 14 Other 7.6 35 1.0 4

†Includes cocaine, crack, heroin, methamphetamines, hallucinogens, club drugs, and other unspecified drugs; marijuana is not included. ‡In thousands. *p<.05; **p<.01; ***p<.001. ¿ Nonmedical use of prescription drugs

Table 14 summarizes estimates of the

lifetime prevalence of the nonmedical use of prescription drugs. For this survey, stimulants were defined as “amphetamines, Ritalin, speed, uppers, and go pills,” and 5.5% of respondents reported ever using such a substance. Lifetime stimulant use was most common among persons age 35–44 and least common among those age 65 and older. Also, lifetime use was more commonly reported by those currently employed full- or part-time.


The prevalence of lifetime nonmedical use of sedatives was estimated to be 2.5% in Illinois. Sedatives were defined as “downers, sleeping pills, bennies, and ludes.”

Tranquilizers were defined as “Ativan, Mandrax, Serax, Valium, and Xanax.” Two percent of respondents reported the nonmedical use of such drugs.

Just under 3% of the Illinois population age 16 and older were estimated to have ever used analgesics for nonmedical reasons. In this study, analgesics were defined to include “codeine products, Darvon, Demerol,

methadone, morphine, OxyContin, Percocet, and Vicodin.” Their use was greater among men and among persons age 16–24. In general, analgesic use declined with age.

The final column of Table 14 summarizes the prevalence of nonmedical use of any prescription medication. It was estimated that 9.3% of persons age 16 and older in Illinois ever have used one of these substances for nonmedical reasons. Use tended to be highest among persons age 35–44 and lowest among those older than 64.


Table 14. Lifetime Nonmedical Use of Prescription Drugs in Illinois, by Demographic Characteristics, 2003

STIMULANTS SEDATIVES TRANQUILIZERS ANALGESICS ANY PRESCRIPTION

DRUGS Demographic Character ist ic Percent

Standard Error Percent




Standard Error

TOTAL 5.5% (0.6) 2.5% (0.4) 2.0% (0.4) 2.9% (0.4) 9.3% (0.8)

Gender ** Male 6.6 (0.9) 2.9 (0.4) 2.4 (0.4) 3.9 (0.7) 10.9 (1.3) Female 4.4 (0.8) 2.1 (0.4) 1.7 (0.6) 1.9 (0.4) 7.8 (1.0)

Age (in ye ars) *** *** *** 16–24 3.1 (0.7) 3.9 (0.8) 2.7 (0.7) 6.1 (1.0) 9.5 (1.2) 25–34 8.4 (1.9) 2.3 (0.7) 1.8 (0.7) 4.1 (1.4) 13.1 (2.2) 35–44 10.5 (2.1) 3.3 (1.9) 3.2 (1.5) 2.7 (0.7) 15.6 (2.9) 45–54 6.6 (1.1) 2.4 (0.7) 2.7 (0.7) 2.2 (0.7) 9.3 (1.4) 55–64 1.4 (0.5) 1.3 (0.5) 0.9 (0.5) 0.5 (0.4) 2.9 (0.8) 65+ 0.3 (0.2) 1.0 (0.5) 0.3 (0.2) 0.2 (0.2) 1.5 (0.6)

Race/Ethnici ty * * White 6.5 (0.7) 2.0 (0.3) 2.4 (0.5) 3.4 (0.5) 9.9 (0.8) African American 2.3 (0.7) 1.6 (0.7) 1.4 (0.9) 1.4 (0.7) 6.0 (1.5) Hispanic 4.3 (3.0) 6.1 (0.3) 0.7 (0.3) 0.9 (0.4) 11.1 (4.4) Other 1.0 (0.7) 4.6 (2.0) 2.5 (1.2) 3.0 (1.8) 5.1 (2.0)

Education * <High school 2.4 (0.7) 4.9 (2.4) 2.1 (0.7) 3.2 (0.8) 9.2 (2.6) High school graduate 6.3 (1.6) 1.4 (0.5) 1.1 (0.4) 1.7 (0.5) 7.9 (1.6) Some college, no degree 7.7 (1.3) 2.7 (0.5) 3.0 (0.9) 2.8 (0.5) 12.0 (1.6) Bachelor’s degree+ 4.7 (1.0) 1.9 (0.4) 1.6 (0.4) 3.5 (1.0) 7.8 (1.1)

Employment status *** ** Working 6.8 (0.9) 2.7 (0.6) 2.2 (0.5) 3.3 (0.5) 10.8 (1.1) Not working 3.2 (0.5) 2.2 (0.5) 1.7 (0.4) 1.9 (0.5) 6.4 (0.9)

Household income * * <$10,000 2.2 (1.2) 2.2 (1.1) 2.2 (1.1) 2.9 (1.3) 5.2 (1.8) $10,000–<$20,000 7.8 (3.6) 3.3 (1.4) 2.9 (1.1) 2.9 (1.1) 13.4 (3.9) $20,000–<$30,000 4.8 (1.5) 2.5 (0.9) 1.7 (0.7) 1.9 (0.7) 6.9 (1.7) $30,000–<$50,000 5.7 (0.9) 1.4 (0.4) 2.7 (1.5) 2.2 (0.5) 9.9 (1.7) $50,000–<$70,000 5.4 (0.9) 5.4 (2.3) 2.4 (0.7) 3.8 (1.0) 11.9 (2.5) $70,000+ 7.3 (1.6) 1.4 (1.4) 1.3 (0.4) 3.6 (1.0) 9.9 (1.6)

Marital status ** Never married 4.3 (0.7) 3.9 (0.7) 3.0 (0.6) 4.5 (0.7) 9.9 (1.1) Married 6.0 (1.0) 2.1 (0.7) 1.2 (0.3) 2.5 (0.5) 8.9 (1.2) Divorced/Separated/ Widowed 6.3 (1.3) 1.8 (0.6) 3.5 (1.9) 1.2 (0.4) 10.2 (2.2)

Sample area * Chicago 5.7 (1.9) 2.4 (0.6) 1.6 (0.5) 2.6 (0.7) 9.0 (2.0) Suburban Cook County 3.5 (0.8) 2.9 (0.8) 3.1 (1.4) 3.1 (0.8) 8.4 (1.8) Northwest Illinois 5.1 (1.0) 1.4 (0.5) 1.4 (0.6) 2.3 (0.6) 7.4 (1.2) South collar counties 6.8 (1.4) 1.0 (0.4) 3.4 (2.0) 1.9 (0.6) 10.1 (2.3) North collar counties 6.6 (1.7) 4.1 (2.1) 0.9 (0.3) 3.4 (1.4) 12.2 (2.6) West central Illinois 6.9 (1.3) 1.0 (0.4) 1.6 (1.6) 2.6 (0.7) 8.9 (1.4) East central Illinois 6.5 (1.1) 2.9 (0.7) 1.8 (0.6) 2.5 (0.7) 8.6 (1.2) Southern Illinois 3.6 (0.9) 2.2 (0.7) 3.4 (1.0) 3.1 (0.9) 7.5 (1.4)

*p<.05; **p<.01; ***p<.001.


Table 15. Population Estimates of Lifetime Nonmedical Use of Prescription Drugs in Illinois, by Gender, Age, and Race/Ethnicity, 2003

LIFETIME USE PAST-YEAR USE Demographic Character ist ic % Population† % Population†

TOTAL 9.3% 884 2.5% 240

Gender Male 10.9 499 2.8 129 Female 7.8 385 2.3 111

Age (in years) 16–24 9.5 167 5.2 92 25–34 13.1 236 3.2 58 35–44 15.6 279 3.0 53 45–54 9.3 145 1.3 21 55–64 2.9 37 0.5 6 65+ 1.5 20 0.8 10

Race/Ethnici ty White 9.9 671 2.7 187 African American 6.0 78 1.8 23 Hispanic 11.1 112 1.6 17 Other 5.1 23 3.0 13

†In thousands.

Table 15 provides estimates of the numbers of persons in Illinois age 16 and above who have misused prescription medications. About 884,000 persons have ever taken prescription medications without appropriate medical supervision. It is estimated that 239,926 have done so during the past year.

¿ Any drug use

The lifetime, past-year, and past-month

use of any drug, including both illicit substances and the nonmedical use of prescription medications, is summarized in Table 16. It is estimated that just over a third of Illinois residents age 16 and older (34.3%) have ever used drugs illegally or without medical supervision. Eight percent and 4.3% are estimated to have done so during the past year and past month, respectively.

Males were more likely to report lifetime and past-year drug use. Adults age 25–44 were most likely to ever have used drugs, while younger persons (those age 16–24) were most likely to have used drugs during the past year and the past month.

College graduates reported the highest prevalence of lifetime drug use across all subgroups of educational achievement. Employment status was strongly associated with lifetime drug use as well. Those currently employed full- or part-time were nearly twice as likely to report ever having used drugs than were those not currently in the labor force.

Lifetime drug use also was strongly associated with household income, with the likelihood of ever having used drugs increasing with each increase in income level.

Finally, marital status was correlated with any drug use. Specifically, persons never having been married were most likely to report lifetime, past-year, and past-month use.


Table 16. Use of Any Drug in Il l inois, by Demographic Characteristics, 2003

LIFETIME USE PAST-YEAR USE PAST-MONTH USE

Demographic Characterist ic Percent Standard



Error

TOTAL 34.3% (1.3) 8.0% (0.6) 4.3% (0.5)

Gender *** *** ** Male 41.9 (2.1) 10.4 (1.0) 6.0 (0.8) Female 27.3 (1.5) 5.7 (0.7) 2.7 (0.6)

Age (in years) *** *** *** 16–24 40.0 (2.1) 22.2 (1.9) 11.8 (1.5) 25–34 45.2 (2.8) 9.3 (1.7) 5.2 (1.5) 35–44 45.6 (3.4) 6.7 (1.6) 3.9 (1.5) 45–54 40.7 (4.4) 3.5 (0.8) 1.8 (0.6) 55–64 19.5 (2.5) 1.3 (0.5) 0.5 (0.4) 65+ 4.4 (1.0) 0.8 (0.4) 0.7 (0.4)

Race/Ethnici ty White 35.9 (1.5) 8.3 (0.8) 4.4 (0.6) African American 33.0 (3.8) 7.2 (1.5) 4.6 (1.2) Hispanic 29.2 (5.2) 5.7 (1.3) 2.5 (0.8) Other 26.9 (5.2) 10.2 (3.8) 6.6 (3.4)

Education *** * <High school 25.4 (3.9) 10.6 (1.7) 5.6 (1.3) High school graduate 26.1 (2.4) 4.8 (1.0) 2.4 (0.6) Some college, no degree 37.2 (2.8) 9.3 (1.6) 5.4 (1.4) Bachelor’s degree+ 41.0 (2.0) 6.7 (1.0) 3.4 (0.8)

Employment status *** Working 40.2 (1.7) 8.4 (0.8) 4.4 (0.7) Not working 21.1 (1.7) 7.0 (1.0) 3.9 (0.7)

Household income *** <$10,000 23.0 (4.0) 8.1 (2.2) 4.4 (1.5) $10,000–<$20,000 28.0 (4.7) 8.5 (0.2) 4.9 (1.6) $20,000–<$30,000 29.0 (4.8) 5.3 (1.2) 2.9 (0.9) $30,000–<$50,000 29.5 (2.4) 6.4 (1.6) 3.8 (1.6) $50,000–<$70,000 38.7 (3.5) 9.7 (1.6) 5.0 (1.2) $70,000+ 46.3 (2.4) 9.7 (1.3) 4.9 (1.1)


Sample area * Chicago 37.1 (3.3) 10.3 (1.5) 5.6 (1.2) Suburban Cook County 39.0 (3.5) 8.2 (1.7) 5.5 (1.6) Northwest Illinois 33.2 (3.3) 6.0 (1.1) 1.8 (0.5) South collar counties 34.3 (3.2) 6.8 (1.2) 3.8 (0.9) North collar counties 35.2 (3.4) 8.7 (1.7) 4.6 (1.5) West central Illinois 29.5 (2.8) 6.7 (1.2) 3.4 (0.9) East central Illinois 33.1 (3.1) 6.5 (1.3) 3.6 (0.9) Southern Illinois 24.2 (2.6) 5.6 (1.2) 1.9 (0.7)

*p<.05; **p<.01; ***p<.001.


¿ Age of first use

The age of substance use initiation is

known to be a strongly associated with subsequent risk for substance abuse (Grant & Dawson, 1997, 1998; Yamaguchi & Kandel, 1984). Table 17 summarizes information regarding the reported age of first use of all substances included in this study. Overall, respondents who had ever smoked cigarettes were, on average, 16.4 years old at the time of their first use. The average age of first alcohol use was 17.7 years. The age of marijuana use initiation was 18.1 years; the age of first use for any other drug was 20.4 years.

Generally, males were younger than females when they first used cigarettes and alcohol. No gender differences in age of first use of marijuana or other drugs were found.

For all substances examined, the reported age of first use tended to increase with increasing age of the respondent. Similar relationships have been reported in other studies (Johnson & Mott, 2001), and there is some evidence that this phenomenon may be a consequence of retrospective reporting bias (Johnson, Gerstein, & Rasinski, 1998).

Racial/ethnic variations in the average age of first use were also found. Whites reported an earlier age of first use of cigarettes, alcohol, and drugs other than marijuana. African Americans reported the highest average age of first cigarette use, while Hispanics reported

the highest average age of first use of marijuana and other drugs.

Persons with at least a bachelor’s degree had the lowest average age of first reported use of alcohol. High school graduates, in contrast, had the lowest average age of first use of marijuana.

Further, average age of first use of two substances was associated with household income levels. Those with incomes greater than $70,000 reported the youngest average age for first experiencing alcohol and drugs other than marijuana. Persons with incomes of less than $10,000 reported the oldest average age for first alcohol use. Those with incomes between $20,000 and $30,000 reported the oldest average age for first use of drugs other than marijuana.

Age of substance use initiation varied by current marital status as well. Never-married respondents reported the earliest average age of use for each substance. Those currently divorced, separated, or widowed reported the oldest average age of initiation for each.

Finally, some variations were found in age of first use across geographic areas. The mean age of first cigarette use was lowest in the north collar counties and highest in Chicago. The average age of marijuana use initiation was lowest in the south collar counties and highest in the north collar counties.


Table 17. Average Age of Drug Use Initiation in Il l inois, by Demographic Characteristics, 2003

CIGARETTES ALCOHOL MARIJUANA OTHER DRUGS Demographic Characteristic Mean SE Mean SE Mean SE Mean SE

TOTAL 16.35 (.18) 17.68 (.12) 18.05 (.27) 20.41 (.55)

Gender *** *** Male 15.58 (.20) 16.82 (.16) 18.03 (.43) 20.65 (.81) Female 17.20 (.29) 18.56 (.16) 18.07 (.24) 20.02 (.50)

Age (in years) *** *** *** *** 16–24 14.62 (.23) 16.05 (.12) 15.73 (.18) 16.76 (.28) 25–34 15.68 (.22) 17.11 (.24) 17.41 (.29) 19.07 (.88) 35–44 15.45 (.26) 17.68 (.31) 17.70 (.88) 19.79 (.74) 45–54 17.35 (.66) 17.76 (.31) 18.74 (.35) 23.23 (1.55) 55–64 16.75 (.36) 18.54 (.38) 23.91 (.61) 28.11 (2.88) 65+ 17.63 (.53) 19.85 (.33) 29.95 (2.98) 35.66 (4.58)

Race/Ethnici ty *** *** *** *** White 15.99 (.17) 17.32 (.12) 17.85 (.19) 19.81 (.66) African American 18.13 (.88) 19.22 (.40) 18.15 (.41) 20.80 (.92) Hispanic 16.31 (.45) 18.24 (.46) 20.04 (2.69) 23.07 (1.26) Other 17.80 (.83) 19.30 (.82) 17.74 (.61) 22.82 (2.37)

Education *** *** ** <High school 15.95 (.46) 17.57 (.41) 18.49 (2.32) 21.48 (1.63) High school graduate 16.24 (.29) 18.48 (.24) 17.01 (.40) 17.81 (.48) Some college, no degree 16.25 (.22) 17.56 (.23) 17.80 (.31) 20.96 (1.31) Bachelor’s degree+ 16.64 (.35) 17.48 (.20) 18.71 (.23) 20.09 (.58)

Employment status *** Working 16.19 (.22) 17.39 (.14) 18.14 (.33) 20.50 (.68) Not working 16.51 (.25) 18.53 (.24) 17.91 (.37) 19.45 (.70)

Household income *** *** <$10,000 17.48 (.84) 19.17 (.63) 17.68 (.74) 20.41 (2.12) $10,000–<$20,000 16.26 (.51) 18.32 (.45) 18.62 (.97) 19.76 (1.04) $20,000–<$30,000 16.70 (.75) 18.87 (.44) 18.08 (.53) 23.29 (2.39) $30,000–<$50,000 16.20 (.27) 17.88 (.22) 17.90 (.36) 19.73 (.69) $50,000–<$70,000 15.85 (.33) 17.24 (.30) 18.93 (1.20) 20.76 (1.10) $70,000+ 16.17 (.23) 17.00 (.21) 17.87 (.25) 19.11 (.47)

Marital status *** *** *** *** Never married 15.49 (.21) 17.02 (.20) 16.64 (.21) 18.22 (.41) Married 16.21 (.17) 17.70 (.17) 18.60 (.45) 21.14 (.93) Divorced/Separated/Widowed 17.43 (.62) 18.82 (.31) 19.43 (.62) 21.29 (.81)

Sample area *** Chicago 17.43 (.60) 17.90 (.27) 17.63 (.29) 20.53 (.73) Suburban Cook County 16.21 (.34) 17.49 (.37) 17.87 (.41) 21.41 (1.75) Northwest Illinois 16.57 (.63) 17.68 (.44) 17.42 (.50) 20.19 (.87) South collar counties 15.84 (.33) 17.34 (.25) 17.07 (.53) 19.28 (1.03) North collar counties 15.71 (.30) 17.66 (.28) 19.30 (1.14) 20.90 (1.28) West central Illinois 16.08 (.40) 17.58 (.22) 17.73 (.42) 19.10 (.94) East central Illinois 15.75 (.34) 17.63 (.25) 18.04 (.48) 19.91 (.85) Southern Illinois 16.04 (.34) 17.95 (.24) 18.22 (.54) 18.53 (.74)

**p<.01; ***p<.001.


Substance Abuse Treatment Need Prevalence Estimates

Persons in need of substance abuse treatment were defined as those meeting Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria for substance abuse or substance dependence (American Psychiatric Association, 1994). Alcohol abuse and dependence questions were asked of all respondents who reported (1) having consumed alcohol at least once a week or more on average during any one year; (2) drinking three or more drinks per day during any one year, or drinking four or more drinks on one occasion if female or five or more drinks on one occasion if male; (3) ever thinking that they had a problem with alcohol or any drug; or (4) that friends or family members ever thought that they had a problem with alcohol

use. Drug abuse and dependence questions were asked of all respondents who reported (1) ever having used hallucinogens, inhalants, or club drugs; (2) having used crack/cocaine, heroin, or other unspecified drugs, or having used without a prescription analgesics, stimulants, or sedatives at least twice during any given year; or (3) having used marijuana at least ten times during any given year.

Persons were defined as meeting substance abuse criteria if they indicated having experienced one or more of four symptoms listed in the first section of Table 18 during the past year. Dependence criteria were met if respondents endorsed three or more of the seven symptoms listed in the second section of Table 18. Persons defined as meeting either substance abuse or dependence criteria were considered to be in need of treatment.

Table 18. Substance-Related DSM-IV Symptoms in Il l inois , 2003

DSM-IV Symptom Without

Diagnosis With

Diagnosis† Total

Sample

Alcohol/Drug Abuse

Role impairment: Failed work or home obligations due to substance use N/A 6.5% 1.1%

Hazardous use: Put self at physical risk by driving a car or riding a motorbike while intoxicated N/A 79.9% 13.1% Legal problems: Arrests for substance-related legal problems N/A 0.7% 0.1%

Social or interpersonal problems: Interpersonal problems caused by substance use N/A 18.8% 3.1% Alcohol/Drug Dependence

Increased tolerance: Increased amount of the substance to achieve desired effect 2.5% 36.0% 8.0% Withdrawal symptoms: Experienced substance withdrawal symptoms 1.0% 23.2% 4.7% Loss of control: Taking substance more than intended 3.0% 39.8% 9.0% Inability to cut down: Unsuccessful attempts to cut down on use 1.0% 20.3% 4.2% Increased time devoted to substance use: Spend much time getting, taking, or recovering 0.5% 18.8% 3.5% Impaired activities: Gave up important social, economic, or recreational activities 0.0% 7.8% 1.3% Use despite physical or psychological consequences 0.5% 14.4% 2.8%

†Respondents experiencing one or more of the abuse items or three or more of the dependence items were defined as being in need of treatment. N/A: Since persons experiencing one or more abuse symptoms were defined as DSM-diagnosed problem users, these cells are not applicable for those without diagnosis.


¿ Estimated treatment need

The proportion of all respondents age 16

and older reporting at least one of the symptoms listed in Table 18 was 21.8%. Among persons having any symptoms, the average number reported was 2.33. This proportion varied by both gender and race/ethnicity. Males were twice as likely as females to report at least one symptom (29.7% vs. 14.4%, respectively). Whites were most likely to report one or more symptoms (24.2%), compared to African Americans (12.7%), Hispanics (19.0%), and persons from other racial/ethnic backgrounds (18.7%).

Using the criteria described above, overall prevalence estimates of treatment need for alcohol, drug, and either form of abuse/ dependence were constructed. These are presented in Table 19. Over 16% of respondents were estimated to have substance abuse/dependence treatment need. Most of those with any treatment need also were identified as specifically in need of treatment for alcohol problems (15.5%). The proportion requiring drug abuse/dependence treatment was much smaller (2.4%).

Several demographic subgroups were found to be at increased risk for requiring substance abuse/dependence treatment. Males were more than twice as likely to require treatment for alcohol, drugs, and either type of substance. Whites were almost three times as likely as African Americans to be in need of treatment for alcohol. No racial/ethnic differences in treatment need for drug abuse/dependence were found, however. Overall, treatment need for alcohol, drug, and either form of abuse/dependence decreased with age. For each measure examined, persons

age 16–24 were at greatest risk for abuse/dependence. Never-married persons were more likely to need treatment for drug and for any substance abuse/dependence.

The prevalence of alcohol treatment need, as well as either alcohol or drug use treatment need, increased with level of education. Persons currently employed full- or part-time were more likely to require treatment for alcohol and any substance abuse or dependence. Alcohol and any treatment need also increased in a linear manner with household income. Those with household incomes of $70,000 and greater were far more likely to have alcohol and any substance abuse/dependence treatment need.

The proportions of all household respondents having experienced each DSM-IV abuse and dependence symptom during the past year are presented in Table 18 for the sample as a whole and separately for those classified as having and not having treatment need. The most commonly reported abuse or dependence symptom was hazardous use of alcohol or drugs, which was defined as placing oneself at physical risk by driving a car or riding a motorbike while intoxicated. This symptom was reported by 13.1% of the sample and by 79.9% of those classified as alcohol or drug abusers. Other common symptoms, endorsed by 9.0% and 8.0% of the sample, respectively, were loss of control (i.e., using a substance more than had intended) and increased tolerance (i.e., needing to increase the amount of the substance used in order to achieve the desired effect). These two symptoms also were each cited by more than a third of those classified as alcohol or drug use dependent.


Table 19. Percentage Meeting DSM-IV C riteria for Past-Year Substance Abuse/Dependence in Illinois, by Demographic Characteristics, 2003

ALCOHOL PROBLEM DRUG PROBLEM ALCOHOL OR DRUG PROBLEM

Demographic Characterist ic Percent Standard Error Percent Standard Error Percent Standard Error

TOTAL 15.5% (0.9) 2.4% (0.3) 16.4% (1.0)

Gender *** *** *** Male 22.4% (1.6) 3.8% (0.7) 23.6% (1.7) Female 9.1% (0.9) 1.2% (0.2) 9.6% (0.9)

Age (in years) *** *** *** 16–24 22.2% (1.9) 7.7% (1.1) 24.8% (1.9) 25–34 18.6% (2.1) 3.4% (1.3) 19.5% (2.1) 35–44 14.1% (2.2) 1.0% (0.3) 15.0% (2.2) 45–54 17.6% (3.4) 0.6% (0.3) 17.8% (3.4) 55–64 11.0% (1.8) 0.2% (0.1) 11.0% (1.8) 65+ 6.3% (1.2) — — 6.3% (1.2)

Race/Ethnici ty *** ** White 18.1% (1.1) 2.5% (0.4) 18.8% (1.1) African American 6.3% (1.3) 2.5% (0.8) 7.7% (1.5) Hispanic 10.4% (3.3) 2.0% (0.8) 12.2% (3.3) Other 14.2% (4.2) 2.0% (1.2) 14.2% (4.2)

Education *** *** <High school 7.7% (1.5) 4.3% (1.0) 9.6% (1.7) High school graduate 11.0% (1.4) 1.5% (0.5) 11.7% (1.4) Some college, no degree 17.8% (2.4) 2.7% (0.7) 18.6% (2.5) Bachelor’s degree+ 21.4% (1.6) 1.9% (0.7) 21.8% (1.6)

Employment status *** *** Working 18.5% (1.3) 2.7% (0.5) 19.4% (1.3) Not working 10.2% (1.1) 1.9% (0.4) 11.2% (1.1)

Household income *** *** <$10,000 5.9% (2.2) 3.7% (1.5) 7.4% (2.4) $10,000–<$20,000 8.8% (2.0) 2.9% (1.0) 10.7% (2.3) $20,000–<$30,000 11.8% (3.6) 2.7% (0.8) 12.4% (3.6) $30,000–<$50,000 12.1% (1.5) 1.4% (0.4) 12.6% (1.5) $50,000–<$70,000 17.1% (2.0) 2.0% (0.6) 18.3% (2.1) $70,000+ 26.3% (2.2) 3.2% (1.0) 26.8% (2.2)

Marital status ** *** *** Never married 20.4% (1.5) 6.5% (1.0) 22.4% (1.6) Married 14.9% (1.4) 1.0% (0.4) 15.4% (1.5) Divorced/Separated/Widowed 11.5% (1.5) 1.0% (0.4) 12.2% (1.6)

Sample area Chicago 16.1% (2.2) 2.9% (0.7) 17.6% (2.3) Suburban Cook County 12.9% (2.4) 1.4% (0.5) 13.2% (2.4) Northwest Illinois 16.6% (2.3) 1.4% (0.5) 17.1% (2.3) South collar counties 18.8% (2.6) 1.9% (0.6) 19.2% (2.6) North collar counties 16.4% (2.5) 3.1% (1.4) 17.3% (2.5) West central Illinois 16.2% (2.2) 3.2% (0.9) 17.8% (2.2) East central Illinois 16.5% (2.8) 2.9% (0.8) 16.9% (2.8) Southern Illinois 14.3% (2.2) 1.5% (0.6) 14.3% (2.2)

**p<.01; ***p<.001.


Table 20. Population Estimates of Those Meeting DSM-IV Criteria for Past-Year Substance Abuse/Dependence in Il l inois, by Demographic Characteristics, 2003

ALCOHOL PROBLEM

DRUG PROBLEM

ALCOHOL OR DRUG PROBLEM

Demographic Characteristic TOTAL

POPULATION† Percent N † Percent N † Percent N †

TOTAL 9,530 15.5% 1,477 2.4% 229 16.4% 1,562

Gender Male 4,601 22.4 1,030 3.7 173 23.6 1,089 Female 4,929 9.1 447 1.1 56 9.6 474

Age (in years) 16–24 1,766 22.2 392 7.7 136 24.8 439 25–34 1,798 18.6 335 3.4 62 19.5 352 35–44 1,786 14.1 253 1.0 19 15.0 269 45–54 1,555 17.6 273 0.6 10 17.8 277 55–64 1,271 11.0 141 0.2 2 11.0 141 65+ 1,355 6.3 85 — — 6.3 —

Race/Ethnici ty White 6,761 18.1 1,226 2.4 166 18.8 1,272 African American 1,304 6.3 83 2.5 33 7.7 101 Hispanic 1,014 10.4 105 2.3 22 12.2 125 Other 452 14.2 64 2.0 8 14.2 64

Education <High school 1,434 7.7 109 4.3 63 9.6 138 High school graduate 1,799 11.0 197 1.5 27 11.7 211 Some college, no degree 2,357 17.8 531 2.7 79 18.6 554 Bachelor’s degree+ 3,363 21.4 586 1.9 52 21.8 596

Employment status Working 6,205 18.5 1,150 2.7 169 19.4 1,205 Not working 2,866 10.2 294 1.9 55 11.2 321

Household income <$10,000 311 5.9 19 3.7 11 7.4 24 $10,000–<$20,000 704 8.8 62 2.9 20 10.7 76 $20,000–<$30,000 1,203 11.8 141 2.7 33 12.4 150 $30,000–<$50,000 1,691 12.1 205 1.4 23 12.6 215 $50,000–<$70,000 1,519 17.1 259 2.0 29 18.3 278 $70,000+ 2,613 26.3 688 3.2 83 26.8 699

Marital status Never married 2,386 20.4 487 6.5 156 22.4 536 Married 5,336 14.9 797 1.0 55 15.4 820 Divorced/Separated/ Widowed 1,394 11.5 161 1.0 14 12.2 170

Sample area Chicago 2,212 16.1 357 2.9 65 17.6 391 Suburban Cook County 1,913 12.9 247 1.4 28 13.2 252 Northwest Illinois 498 16.6 82 1.4 6 17.1 85 South collar counties 476 18.8 89 1.9 9 19.2 91 North collar counties 1,690 16.4 276 3.1 53 17.3 292 West central Illinois 1,158 16.2 188 3.2 37 17.8 207 East central Illinois 637 16.5 105 2.9 18 16.9 109 Southern Illinois 947 14.3 134 1.5 13 14.3 136

†In thousands.


Table 20 provides estimates of the numbers of persons in Illinois who met DSM-IV criteria for past-year substance abuse/dependence. Overall, it is estimated that 1,562,277 Illinois residents age 16 and older experienced substance abuse or dependence in 2003. The numbers estimated to have suffered specifically from alcohol and drug abuse/dependence were 1,477,480 and 228,791, respectively. It should be noted that the estimated numbers with alcohol vs. drug abuse/dependence do not sum to the total number having experienced any substance abuse/dependence because there is some overlap between these estimates. That is, some individuals have experienced both alcohol and drug abuse/dependence during the past year.

Of those experiencing any form of past-year substance abuse or dependence, more than one million were male, White, and currently employed. The region with the greatest numbers of persons with any form of substance abuse/dependence was the city of Chicago, where 390,923 were estimated to have met the DSM-IV criteria for either substance abuse or dependence during 2003.

¿ Abuse treatment need

Assessments of persons specifically meeting past-year substance abuse criteria are presented in Table 21. The overall estimate of persons in Illinois meeting past-year DSM-IV substance abuse criteria was 12.7 %. Again, alcohol abuse was far more prevalent (11.7%) than was drug abuse (0.7%). Alcohol abuse and alcohol or drug abuse each were associated with gender, race/ethnicity,

education, employment status, and household income in a manner similar to the combined abuse/dependence measures presented in Table 14. Drug abuse was associated with gender and age. As Table 22 indicates, a total of 1,157,223 Illinois residents were estimated to have met DSM-IV substance abuse criteria during 2003. Of these, approximately one million were estimated to have abused alcohol only; 42,219 were estimated to have abused drugs only; and 25,484 were estimated to have abused both drugs and alcohol.

¿ Dependence treatment need

Table 23 provides similar information regarding past-year substance dependence in Illinois. The proportion of all persons age 16 and older estimated to be substance dependent in 2003 was 4.7%. This represents 445,982 Illinois residents (see Table 24). Alcohol dependence (3.8%) was twice as common as drug dependence (1.7%).

As Table 24 indicates, approximately 285,000 persons were estimated to be alcohol dependent in 2003. Another 84,000 were estimated to be drug dependent; and 78,000 were estimated to be both alcohol and drug dependent in 2003. Alcohol dependence was greatest among males, young persons (age 16–24), persons representing other racial and ethnic groups, and the never married. Drug dependence was greatest among males, young persons, and the never married. Any substance dependence was greatest among males, young persons, never-married persons, and persons in other racial and ethnic groups.


Table 21. Percentage Meeting DSM-IV Criteria for Past-Year Substance Abuse in Illinois, by Demographic Characteristics, 2003

ALCOHOL ABUSE DRUG ABUSE ALCOHOL OR DRUG ABUSE

Demographic Characteristic Percent Standard Error Percent Standard Error Percent Standard Error

TOTAL 11.7% (0.8) 0.7% (0.2) 12.1% (0.8)


Age (in years) * ** * 16–24 11.7 (1.4) 2.2 (0.6) 13.0 (1.4) 25–34 13.6 (1.7) 0.5 (0.3) 13.9 (1.7) 35–44 12.0 (2.1) 0.8 (0.4) 12.6 (2.1) 45–54 15.8 (3.3) 0.3 (0.2) 16.0 (3.3) 55–64 9.8 (1.5) — — 9.7 (1.5) 65+ 5.9 (1.1) — — 5.9 (1.1)

Race/Ethnici ty *** *** White 14.2 (1.0) 0.7 (0.2) 14.6 (1.0) Black 4.1 (1.1) 0.4 (0.3) 4.6 (1.1) Hispanic 8.1 (3.2) 0.7 (0.4) 8.8 (3.2) Other 3.6 (1.7) 1.0 (0.9) 4.6 (1.9)

Education *** * *** <High school 3.0 (0.8) 1.6 (0.6) 4.0 (0.9) High school graduate 8.8 (1.3) 0.8 (0.3) 9.4 (1.3) Some college, no degree 13.4 (2.3) 0.6 (0.3) 13.8 (2.3) Bachelor’s degree+ 17.5 (1.4) 0.3 (0.2) 17.6 (1.4)

Employment status *** * *** Working 14.7 (1.2) 0.9 (0.2) 15.2 (1.2) Not working 6.5 (0.8) 0.3 (0.1) 6.8 (0.8)

Household income *** *** <$10,000 3.8 (1.9) 0.0 (0.4) 4.4 (1.9) $10,000–<$20,000 5.3 (1.5) 1.3 (0.6) 6.5 (1.7) $20,000–<$30,000 9.0 (3.5) 0.8 (0.4) 9.3 (3.5) $30,000–<$50,000 9.9 (1.4) 0.7 (0.3) 10.2 (1.4) $50,000–<$70,000 12.4 (1.6) 0.6 (0.4) 13.0 (1.6) $70,000+ 20.7 (2.0) 0.8 (0.3) 21.0 (2.0)

Marital status ** Never married 11.6 (1.2) 1.5 (0.4) 12.6 (1.2) Married 13.1 (1.4) 0.4 (0.1) 13.4 (1.4) Divorced/Separated/Widowed 8.8 (1.2) 0.7 (0.4) 9.2 (1.2)

Sample area Chicago 10.4 (1.9) 0.5 (0.2) 10.7 (1.9) Suburban Cook 10.8 (2.4) 1.0 (0.4) 11.3 (2.4) Northwest Illinois 14.1 (2.2) 0.5 (0.4) 14.3 (2.2) South collar counties 15.1 (2.5) 1.0 (0.4) 15.5 (2.5) North collar counties 13.4 (2.1) 0.4 (0.2) 13.8 (2.1) West central Illinois 12.4 (1.8) 1.2 (0.6) 13.3 (1.9) East central Illinois 12.2 (2.7) 0.7 (0.4) 12.9 (2.7) Southern Illinois 9.6 (1.4) 0.5 (0.3) 9.7 (1.4)

*p<.05; **p<.01; ***p<.001.


Table 22. Population Estimates of Those Meeting DSM-IV Criteria for Past-Year Substance Abuse in Illinois, by Demographic Characteristics, 2003 (Total Population in 2000=9,530,327)

ALCOHOL ABUSE ONLY

DRUG ABUSE ONLY

ALCOHOL & DRUG ABUSE

ALCOHOL OR DRUG ABUSE

Demographic Characteristic Percent N † Percent N † Percent N † Percent N †

TOTAL 11.4% 1,090 0.4% 42 0.3% 25 12.1% 1,157

Gender Male 16.3 750 0.7 34 0.5 21 17.5 806 Female 6.9 339 0.2 8 0.1 4 7.1 351

Age (in years) 16–24 10.9 192 1.3 23 0.9 15 13.0 230 25–34 13.3 239 0.3 5 0.3 5 13.9 250 35–44 11.8 211 0.6 12 0.1 2 12.6 225 45–54 15.6 243 0.2 3 0.2 3 16.0 248 55–64 9.8 124 — — — — 9.7 124 65+ 5.9 80 — — — — 5.9 80

Race/Ethnici ty White 13.9 938 0.4 26 0.4 25 14.6 989 African American 4.1 54 0.4 5 — — 4.6 59 Hispanic 8.1 82 0.7 7 — — 8.8 89 Other 3.5 16 0.9 4 0.1 — 4.6 20

Education <High school 2.3 33 0.8 12 0.8 11 3.9 57 High school graduate 8.7 156 0.7 13 0.1 2 9.5 170 Some college, no degree 13.2 394 0.4 12 0.2 7 13.8 413 Bachelor’s degree+ 17.3 472 0.1 4 0.2 6 17.6 481

Employment status Working 14.3 889 0.5 31 0.4 25 15.2 946 Not working 6.5 186 0.3 10 — — 6.8 196

Household income <$10,000 3.8 12 0.4 1 — — 4.4 13 $10,000–<$20,000 5.3 37 1.3 9 — — 6.5 47 $20,000–<$30,000 8.6 103 0.3 3 0.4 5 9.3 112 $30,000–<$50,000 9.5 161 0.3 6 0.3 6 10.2 173 $50,000–<$70,000 12.4 188 0.6 9 0.0 — 13.0 198 $70,000+ 20.2 528 0.4 10 0.4 12 21.0 549

Marital status Never married 11.0 262 1.0 23 0.6 15 12.6 300 Married 13.0 693 0.2 11 0.2 9 13.4 713 Divorced/Separated/ Widowed 8.7 121 0.5 6 0.2 2 9.2 129

Sample area Chicago 10.2 225 0.3 7 0.2 5 10.7 236 Suburban Cook County 10.3 196 0.5 9 0.5 10 11.3 215 Northwest Illinois 13.7 68 0.3 1 0.4 2 14.3 71 South collar counties 14.6 69 0.3 2 0.6 3 15.5 74 North collar counties 13.4 226 0.4 7 — — 13.8 233 West central Illinois 12.1 140 0.9 10 0.3 4 13.3 153 East central Illinois 12.1 77 0.8 5 0.1 1 12.9 83 Southern Illinois 9.3 88 0.1 1 0.3 3 9.7 92

†In thousands.


Table 23. Percentage Meeting DSM-IV Criteria for Past-Year Substance Dependence in Illinois, by Demographic Characteristics, 2003

ALCOHOL DEPENDENCE DRUG DEPENDENCE ALCOHOL OR DRUG DEPENDENCE

Demographic Characterist ic Percent Standard Error Percent Standard Error Percent Standard Error

TOTAL 3.8% (0.5) 1.7% (0.3) 4.7% (0.5)


Age (in years) *** *** *** 16–24 10.4 (1.4) 5.5 (1.0) 13.7 (1.6) 25–34 5.0 (1.4) 2.9 (1.3) 6.0 (1.5) 35–44 2.2 (0.7) 0.3 (0.1) 2.5 (0.7) 45–54 1.7 (1.1) 0.3 (0.2) 2.0 (1.1) 55–64 1.3 (1.0) 0.2 (0.1) 1.3 (1.0) 65+ 0.4 (0.4) — (0.0) 0.4 (0.4)

Race/Ethnici ty ** *

White 3.9 (0.6) 1.7 (0.4) 4.7 (0.6) African American 2.2 (0.7) 2.1 (0.7) 3.3 (0.9) Hispanic 2.2 (0.9) 1.5 (0.7) 3.6 (1.1) Other 10.5 (4.0) 0.9 (0.9) 10.5 (0.4)

Education <High school 4.5 (1.2) 2.7 (0.8) 6.2 (1.3) High school graduate 2.2 (0.6) 0.8 (0.3) 2.7 (0.7) Some college, no degree 4.4 (1.0) 1.9 (0.6) 5.2 (1.0) Bachelor’s degree+ 3.9 (0.9) 1.6 (0.7) 4.6 (1.0)

Employment status Working 3.8 (0.6) 1.8 (0.4) 4.8 (0.7) Not working 3.8 (0.8) 1.6 (0.4) 4.5 (0.8)

Household income <$10,000 2.2 (1.1) 3.3 (1.4) 3.8 (1.5) $10,000–<$20,000 3.5 (1.3) 1.5 (0.7) 4.4 (1.4) $20,000–<$30,000 2.7 (0.9) 2.1 (0.8) 4.0 (1.1) $30,000–<$50,000 2.3 (0.6) 0.7 (0.3) 2.9 (0.7) $50,000–<$70,000 4.7 (1.3) 1.3 (0.5) 5.3 (1.4) $70,000+ 5.7 (1.3) 2.4 (0.9) 6.5 (1.3)


Sample area * Chicago 5.8 (1.2) 2.4 (0.6) 7.4 (1.3) Suburban Cook County 2.1 (0.6) 0.5 (0.2) 2.3 (0.6) Northwest Illinois 2.4 (0.9) 0.6 (0.4) 2.9 (1.0) South collar counties 3.5 (0.9) 1.1 (0.5) 4.2 (0.9) North collar counties 3.0 (1.4) 2.7 (1.4) 4.0 (1.5) West central Illinois 3.9 (1.3) 2.0 (0.7) 5.2 (1.4) East central Illinois 4.2 (1.1) 1.9 (0.7) 4.9 (1.2) Southern Illinois 4.6 (1.9) 1.0 (0.5) 4.7 (1.9)

*p<.05; **p<.01; ***p<.001.


Table 24. Population Estimates of Those Meeting DSM-IV Criteria for Past-Year Substance Dependence in Illinois, by Demographic Characteristics, 2003 (Total Population in 2000=9,530,327)

ALCOHOL DEPENDENCE ONLY

DRUG DEPENDENCE ONLY

ALCOHOL & DRUG DEPENDENCE

ALCOHOL OR DRUG DEPENDENCE

Demographic Characteristic Percent N † Percent N

† Percent N † Percent N

†

TOTAL 3.0% 285 0.9% 84 0.8% 78 4.7% 446

Gender Male 4.3 199 1.3 58 1.3 60 6.9 316 Female 1.7 86 0.5 25 0.4 18 2.6 130

Age (in years) 16–24 8.1 144 3.2 57 2.3 41 13.7 241 25–34 3.1 56 1.0 18 1.9 34 6.0 108 35–44 2.2 39 0.2 4 — — 2.5 44 45–54 1.7 26 0.3 4 0.1 1 2.0 31 55–64 1.1 15 — — 0.2 2 1.3 17 65+ 0.4 5 — — — — 0.4 5

Race/Ethnic i ty White 3.0 204 0.8 56 0.9 59 4.7 319 African American 1.2 15 1.1 14 1.0 14 3.3 43 Hispanic 2.1 21 1.4 14 0.1 1 3.6 36 Other 9.7 44 — — 0.9 4 10.5 48

Education <High school 3.5 50 1.7 25 1.0 15 6.2 89 High school graduate 1.9 34 0.4 8 0.3 5 2.7 47 Some college, no degree 3.6 99 1.0 26 1.4 32 5.2 157 Bachelor’s degree+ 2.9 85 0.6 19 0.7 24 4.6 127

Employment status Working 3.0 184 1.0 61 0.8 51 4.8 296 Not working 2.9 84 0.7 20 0.9 25 4.5 129

Household income <$10,000 0.5 2 1.6 5 1.7 5 3.8 12 $10,000–<$20,000 2.9 20 0.9 6 0.6 4 4.4 31 $20,000–<$30,000 1.9 23 1.3 15 0.8 9 4.0 48 $30,000–<$50,000 2.2 37 0.6 11 0.1 1 2.9 49 $50,000–<$70,000 4.0 61 0.7 10 0.6 10 5.3 81 $70,000+ 4.2 109 0.8 22 1.5 40 6.5 171

Marital status Never married 6.6 158 2.7 65 2.2 53 11.5 276 Married 1.4 74 0.3 14 0.4 21 2.0 109 Divorced/Separated/ Widowed 2.6 36 0.2 3 0.2 2 2.9 41

Sample area Chicago 5.0 110 1.6 36 0.8 17 7.4 163 Suburban Cook County 1.9 35 0.2 4 0.3 6 2.3 45 Northwest Illinois 2.3 12 0.5 2 0.1 1 2.9 14 South collar Counties 3.1 15 0.6 3 0.4 2 4.2 20 North collar Counties 1.3 22 1.0 18 1.7 28 4.0 68 West central Illinois 3.2 37 1.3 16 0.7 8 5.2 60 East central Illinois 3.0 19 0.7 4 1.2 8 4.9 31 Southern Illinois 3.7 35 0.1 1 0.9 9 4.7 44

†In thousands.


¿ Unmet treatment need

For persons with substance use treatment

need, there often are personal, social, and/or structural barriers to obtaining needed help (Harris & McKellar, 2003; Horgan, Skwara, & Strickler, 2001). Consequently, unmet need for treatment is a serious problem. Measuring unmet need for substance abuse treatment is also difficult. In this study, it was measured using two alternative data sources. The first approach involved estimating the proportion of persons needing substance abuse treatment during the past year who also reported not receiving substance abuse treatment services during the past year. Overall, a large majority of persons with past-year treatment need (97.1%) indicated that they had not received treatment services during the past year. When examined separately by substance type, 97.2% of persons with alcohol treatment need reported not receiving services, and 91.4% of those in need of treatment for drug use indicated not having received treatment. Differences by gender and race/ethnicity were small. Respondent age, however, was

associated with the likelihood of having unmet treatment need. Specifically, persons age 16–24 were more likely to have come into contact with the treatment system during the past year than were those in any succeeding age cohort. These estimates are provided in Table 25.

A second approach to estimating unmet treatment need involved subtracting the actual numbers of persons in Illinois receiving 2003 DASA-supported treatment services from the 2003 survey estimates of the total numbers of persons in need of treatment. Overall, this method estimates that 94.1% of the adults age 16 and older living in Illinois who had substance use treatment need did not receive services during the previous year (see results in Table 26). The proportion estimated by this method to have unmet treatment need would have been somewhat lower if information regarding receipt of treatment from private facilities had been available. It is nonetheless notable that these varying methodologies produced estimates of unmet treatment need that ranged rather narrowly between 94.1% and 97.1%.

Table 25. Estimated Population in Need of Treatment and Unmet Treatment Need in Illinois, by Gender, Race/Ethnicity, and Age, 2003: Based on 2003 Ill inois Household Survey

Estimated Population in Need of Treatment †

% Unmet Treatment Need

% Received Treatment Past Year

Estimated Population with Unmet Treatment Need†

TOTAL 1,561 97.06% 2.94% 1,515

Gender Male 1,087 96.84 3.16 1,053 Female 474 97.57 2.43 462

Race/Ethnici ty White 1,272 96.94 3.06 1,233 African American 101 97.73 2.27 99 Hispanic 124 96.30 3.70 119 Other 64 100.00 — 64

Age (in years) 16–24 438 92.15 7.85 403 25–34 352 98.69 1.31 347 35–44 269 98.29 1.71 264 45+ 502 99.54 0.46 500

†In thousands.


Table 26. Estimated Population in Need of Treatment and Unmet Treatment Need in Illinois, by Gender, Race/Ethnicity, and Age, 2003: Based on DASA’s Automated Reporting and Tracking System

Estimated Population in Need

of Treatment † DASA-Treated

Population†

Estimated Treatment Unmet

Population† Estimated % Met Treatment Need

Estimated % Unmet Treatment

Need

TOTAL 1,561 93 1,468 5.95% 94.05%

Gender Male 1,087 30 1,058 2.72 97.28 Female 474 63 410 13.37 86.63

Race/Ethnici ty White 1,272 42 1,231 3.27 96.73 African American 101 40 61 39.71 60.29 Hispanic 124 9 115 7.12 92.88 Other 64 2 61 3.67 96.33

Age (in years) 16–24 438 27 410 6.25 93.75 25–34 352 23 329 6.61 93.39 35–44 269 27 242 10.01 89.99 45+ 502 15 487 3.06 96.94

†In thousands.


Prevalence and Treatment Need Estimates for Vulnerable Groups

In this section, the substance use prevalence and treatment need of several vulnerable populations are examined. The specific groups of interest include persons without health insurance coverage, persons with physical disabilities, persons with mental disorders, those ever having experienced homelessness, public financial support recipients, victims of domestic violence, and chronic gamblers. ¿ Persons without health insurance

coverage

Persons with substance abuse problems

are known to require more inpatient and outpatient services compared to those without such problems (Artalejo et al., 2001; Hunkeler, Hung, Rice, Wesiner, & Hu, 2001). The availability of health insurance coverage among persons likely to have substance use treatment need is thus an important policy question. Fourteen percent of the survey sample indicated that they did not have current health insurance coverage (Table 2). Table 27 examines the substance use prevalence and treatment need of those without coverage and compares them to persons with coverage. Those with health

insurance were more likely to have used alcohol during the past month. With this one exception, there were few differences in patterns of substance use and treatment need by insurance status. ¿ Persons with physical disabilities

The substance use prevalence and

treatment need of persons with and without physical disabilities are compared in Table 28. Persons with physical disabilities were defined as those reporting current vision impairment, hearing impairment, or any physical or mental condition that substantially limits daily activities. The specific items used to measure physical disability were adapted from the long form of the 2000 United States Census questionnaire. Overall, 19.0% of persons age 16 and older are estimated to have a physical disability, based on this definition. Those with physical disabilities were less likely to report alcohol use during the past year and during the past month. They also were less likely to report binge drinking during the past year.

In general, no differences were found in self-reported drug use between persons with and without physical disabilities. Need for alcohol and/or other drug treatment also were unrelated to physical disability status.


Table 27. Substance Use Prevalence and Abuse/Dependence Treatment Need in Illinois, by Medical Health Insurance Coverage, 2003

TOTAL WITH COVERAGE WITHOUT COVERAGE

Percent Standard



Error p

Use of…

Alcohol Lifetime 84.0% (1.3) 72.2% (6.6) 85.9% (0.9) ** Past year 69.6 (1.4) 56.1 (5.7) 71.9 (1.2) ** Past month 54.2 (1.4) 38.4 (4.4) 56.8 (1.4) *** Past-year binge 30.2 (1.2) 26.4 (3.8) 30.8 (1.3) n.s.

Marijuana Lifetime 31.6 (1.3) 28.6 (3.7) 32.1 (1.4) n.s. Past year 6.2 (0.5) 7.0 (1.5) 6.0 (0.6) n.s. Past month 3.2 (0.4) 3.8 (1.0) 3.2 (0.5) n.s.

Cocaine Lifetime 8.7 (1.0) 12.0 (3.2) 8.2 (1.0) n.s. Past year 0.8 (0.2) 1.5 (0.5) 0.7 (0.2) *

Other drugs (l i fet ime) Crack 1.5 (0.4) 3.4 (1.2) 1.2 (0.4) * Heroin 0.8 (0.2) 2.4 (1.1) 0.5 (0.2) ** Analgesics 2.9 (0.4) 3.6 (0.9) 2.8 (0.4) n.s. Methamphetamines 1.4 (0.2) 1.7 (0.7) 1.3 (0.2) n.s. Stimulants 5.5 (0.6) 5.7 (1.5) 5.4 (0.7) n.s. Hallucinogens 5.7 (0.5) 7.5 (1.8) 5.4 (0.5) * Tranquilizers 2.1 (0.4) 3.4 (1.1) 1.8 (0.4) * Sedatives 2.5 (0.4) 3.1 (1.0) 2.4 (0.5) n.s. Inhalants 4.1 (0.5) 3.3 (0.9) 4.3 (0.6) n.s. Club drugs 2.4 (0.8) 2.6 (0.4) 2.4 (0.4) n.s. Other drugs 1.0 (0.2) 1.7 (0.7) 0.9 (0.2) n.s.

Any nonmedical use of prescription drugs

Lifetime 9.3 (0.8) 11.5 (2.1) 8.9 (0.9) n.s.

Any illicit drugs excluding marijuana

Lifetime 12.3 (1.0) 16.8 (3.4) 11.5 (1.0) n.s. Past year 1.5 (0.2) 3.3 (0.8) 1.3 (0.2) **

Any drug Lifetime 34.4 (1.3) 36.0 (4.5) 34.1 (1.4) n.s. Past year 7.9 (0.6) 9.3 (1.7) 7.7 (0.7) n.s. Past month 4.3 (0.5) 4.5 (1.0) 4.2 (0.6) n.s.

Treatment need Alcohol 15.6 (0.9) 10.3 (1.7) 16.4 (1.0) ** Drug 2.4 (0.3) 1.9 (0.6) 2.5 (0.4) n.s. Alcohol or drug 16.4 (0.9) 11.0 (1.8) 17.3 (1.1) **

*p<.05; **p<.01; ***p<.001; n.s.: not significant at .05 level.


Table 28. Substance Use Prevalence and Abuse/Dependence Treatment Need in Illinois, by Presence of Physical Disability, 2003

TOTAL WITH PHYSICAL

DISABILITY WITHOUT PHYSICAL

DISABILITY

Percent Standard



Error p

Use of…

Alcohol Lifetime 83.9% (1.3) 80.8% (4.6) 84.6% (1.2) n.s. Past year 69.5 (1.4) 55.4 (3.9) 72.8 (1.3) *** Past month 54.0 (1.4) 37.8 (3.2) 57.8 (1.5) *** Past-year binge 30.2 (1.2) 21.8 (2.3) 32.1 (1.4) ***


Cocaine Lifetime 8.7 (0.9) 8.9 (1.4) 8.7 (1.1) n.s. Past year 0.8 (0.2) 0.6 (0.3) 0.8 (0.2) n.s.

Other drugs (l i fet ime) Crack 1.5 (0.4) 2.0 (0.6) 1.4 (0.4) n.s. Heroin 0.7 (0.2) 1.0 (0.4) 0.7 (0.2) n.s. Analgesics 2.9 (0.4) 3.8 (0.7) 2.6 (0.4) n.s. Methamphetamines 1.3 (0.2) 2.3 (0.6) 1.1 (0.2) * Stimulants 5.5 (0.6) 7.0 (1.2) 5.1 (0.7) n.s. Hallucinogens 5.7 (0.5) 7.2 (1.3) 5.3 (0.6) n.s. Tranquilizers 2.1 (0.4) 2.7 (0.7) 1.9 (0.4) n.s. Sedatives 2.5 (0.4) 3.8 (0.7) 2.2 (0.5) n.s. Inhalants 4.1 (0.5) 4.2 (0.9) 4.1 (0.6) n.s. Club drugs 2.4 (0.4) 1.8 (0.5) 2.6 (0.4) n.s. Other drugs 1.0 (0.2) 1.6 (0.5) 0.9 (0.2) n.s.


Lifetime 9.3 (0.8) 10.8 (1.5) 8.9 (0.9) n.s.


Lifetime 12.2 (1.0) 11.9 (1.6) 12.3 (1.2) n.s. Past year 1.5 (0.2) 1.5 (0.5) 1.5 (0.2) n.s.


Treatment need Alcohol 15.5 (0.9) 12.8 (1.7) 16.1 (1.1) n.s. Drug 2.4 (0.3) 2.9 (0.7) 2.3 (0.4) n.s. Alcohol or drug 16.4 (1.0) 14.2 (1.8) 16.9 (1.1) n.s.

*p<.05; ***p<.001; n.s.: not significant at .05 level.


¿ Persons with mental disorders

The associations between substance use

and mental health conditions are well documented (Kessler et al., 1997). Table 29 reveals similar relationships in Illinois. In this study, persons with mental disorders are defined as those with major depression, generalized anxiety disorder, social phobia, or panic attack based on the Composite International Diagnostic Interview Short Form (CIDI-SF; Kessler, Anidres, Mroczek, Ustun, & Wittchen, 1998). Using this definition, 11.1% of Illinois citizens age 16 and older have a mental disorder. These individuals were more likely to report binge drinking during the past year and to report lifetime, past-year, and past-month marijuana use. Persons with mental disorders also reported higher lifetime use of methamphetamines, hallucinogens, inhalants, and other drugs, as well as lifetime and past-year use of any illicit drug (other than marijuana). Further, they were more likely to report lifetime nonmedical use of psychotherapeutic substances, including stimulants, analgesics, tranquilizers, and sedatives, and lifetime, past-year, and past-month drug use of any kind was higher among these persons. Perhaps not surprisingly, then, all indicators of alcohol and/or drug treatment need were greater among this population.

In addition, it is estimated that 3.2% of all persons in Illinois age 16 and older can be classified as having comorbid mental health

and substance abuse/dependence conditions (SE = 0.4).

¿ Persons ever homeless

The association between substance use and

homelessness is well documented, if poorly understood (Johnson, Freels, Parsons, & VanGeest, 1997; Reardon, Burns, Preist, Sachs-Ericcson, & Lang, 2003). To examine the relationship between homeless experiences and substance use and treatment need, respondents were asked if they ever had considered themselves to be homeless. The specific questions used were adapted from a national study of lifetime homelessness (Link et al., 1994). Overall, it is estimated that 4.8% of all persons age 16 and older in Illinois have ever experienced homelessness. Table 30 provides comparisons of the ever-homeless vs. the never-homeless in Illinois. Persons indicating that they never had had a homeless experience were more likely to report alcohol consumption during the past month.

In contrast, persons with homeless experiences were more likely to report lifetime use of several illicit substances, including crack cocaine, heroin, and methamphetamines. Ever-homeless persons also were more likely to report past year-use of any illicit drug (other than marijuana). The alcohol and other drug treatment needs of the ever homeless and the never homeless, however, were not different.


Table 29. Substance Use Prevalence and Abuse/Dependence Treatment Need in Illinois, by Presence of Mental Disorder, 2003

TOTAL WITH MENTAL DISORDER

WITHOUT MENTAL DISORDER

Percent



Standard Error p

Use of…

Alcohol Lifetime 83.9% (1.3) 89.8% (1.8) 83.1% (1.4) ** Past year 69.5 (1.4) 77.1 (2.6) 68.6 (1.5) ** Past month 54.0 (1.4) 58.8 (2.9) 53.4 (1.6) Past-year binge 30.2 (1.2) 40.7 (3.0) 28.8 (1.3) ***

Marijuana Lifetime 31.6 (1.3) 48.3 (3.0) 29.5 (1.4) *** Past year 6.2 (0.5) 14.3 (2.0) 5.2 (0.5) *** Past month 3.3 (0.4) 9.1 (1.8) 2.6 (0.4) ***

Cocaine Lifetime 8.7 (0.9) 13.0 (2.0) 8.2 (1.0) * Past year 0.8 (0.2) 1.7 (0.6) 0.7 (0.2) *

Other drugs (l i fet ime) Crack 1.5 (0.4) 3.0 (0.9) 1.3 (0.4) * Heroin 0.7 (0.2) 1.3 (0.6) 0.7 (0.2) n.s. Analgesics 2.9 (0.4) 5.8 (1.1) 2.5 (0.4) *** Methamphetamines 1.3 (0.2) 3.5 (0.8) 1.1 (0.2) *** Stimulants 5.5 (0.6) 12.1 (1.8) 4.6 (0.6) *** Hallucinogens 5.7 (0.5) 11.7 (1.9) 4.9 (0.5) *** Tranquilizers 2.1 (0.4) 5.0 (1.1) 1.7 (0.4) *** Sedatives 2.5 (0.4) 7.2 (1.2) 1.9 (0.5) *** Inhalants 4.1 (0.5) 8.0 (1.4) 3.7 (0.6) *** Club drugs 2.4 (0.4) 4.8 (1.1) 2.1 (0.4) ** Other drugs 1.0 (0.2) 2.8 (0.8) 0.8 (0.2) ***


Lifetime 9.3 (0.8) 19.9 (2.2) 8.0 (0.9) ***


Lifetime 12.2 (0.1) 18.4 (2.2) 11.5 (1.1) ***

Any drug Lifetime 34.3 (1.3) 52.4 (3.0) 32.1 (1.4) *** Past year 8.0 (0.6) 18.0 (2.2) 6.7 (0.6) *** Past month 4.4 (0.5) 10.2 (1.8) 3.6 (0.5) ***

Treatment need Alcohol 15.5 (0.9) 26.8 (2.8) 14.1 (1.0) *** Drug 2.4 (0.3) 6.6 (1.3) 1.9 (0.4) *** Alcohol or drug 16.4 (1.0) 29.2 (2.8) 14.8 (1.0) ***



Table 30. Substance Use Prevalence and Abuse/Dependence Treatment Need in Illinois, by Lifetime Experience with Homelessness, 2003

TOTAL EVER HOMELESS NEVER HOMELESS

Percent Standard



Error p

Use of…

Alcohol Lifetime 84.3% (1.3) 72.4% (10.9) 84.9% (1.3) n.s. Past year 70.1 (1.4) 53.3 (9.6) 71.0 (1.4) * Past month 54.5 (1.5) 30.5 (6.9) 55.8 (1.4) *** Past-year binge 30.1 (1.2) 31.7 (8.0) 30.0 (1.2) n.s.

Marijuana Lifetime 31.3 (1.3) 35.2 (7.1) 31.1 (1.3) n.s. Past year 6.2 (0.5) 9.0 (2.7) 6.0 (0.5) n.s. Past month 3.3 (0.4) 6.5 (2.3) 3.1 (0.4) *

Cocaine

Lifetime 8.3 (0.9) 22.6 (7.7) 7.6 (0.8) ** Past year 0.9 (0.2) 3.0 (1.4) 0.7 (0.2) **

Other drugs (l i fet ime) Crack 1.4 (0.3) 8.5 (2.8) 1.0 (0.3) *** Heroin 0.7 (0.2) 4.5 (2.3) 0.5 (0.1) *** Analgesics 2.9 (0.4) 4.0 (1.4) 2.8 (0.4) n.s. Methamphetamines 1.4 (0.2) 6.0 (2.2) 1.1 (0.2) *** Stimulants 5.6 (0.6) 10.0 (2.7) 5.4 (0.6) * Hallucinogens 5.8 (0.5) 10.5 (3.0) 5.5 (0.5) * Tranquilizers 2.0 (0.4) 4.5 (1.7) 1.9 (0.4) * Sedatives 2.5 (0.5) 4.5 (1.8) 2.4 (0.5) n.s. Inhalants 4.3 (0.5) 6.5 (2.3) 4.1 (0.6) n.s. Club drugs 2.5 (0.4) 3.5 (1.4) 2.4 (0.4) n.s. Other drugs 1.1 (0.2) 3.0 (1.7) 1.0 (0.2) n.s.


Lifetime 9.3 (0.8) 15.5 (3.7) 9.0 (0.8) **

Any il l icit drugs excluding marijuana

Lifetime 11.9 (1.0) 29.5 (8.0) 11.0 (0.9) ** Past year 1.6 (0.2) 5.5 (1.9) 1.4 (0.2) ***

Any drug Lifetime 34.1 (1.3) 48.2 (9.2) 33.3 (1.3) n.s. Past year 8.0 (0.6) 14.0 (3.6) 7.6 (0.6) * Past month 4.3 (0.5) 9.5 (2.9) 4.1 (0.5) *

Treatment need Alcohol 15.9 (1.0) 8.5 (2.4) 16.3 (1.0) * Drug 2.4 (0.3) 6.3 (2.0) 2.5 (0.4) ** Alcohol or drug 16.7 (1.0) 11.0 (2.8) 17.0 (1.0) n.s.



¿ Recipients of public financial support

Historically, concern has been expressed

that public financial support recipients are at increased risk for substance abuse (Pollack, Danziger, Seefeldt, & Jayakody, 2002; Yacoubian, Peters, Urbach, & Johnson, 2002). In this study, public financial support recipients were defined as households in which survey respondents or immediate family members received income from unemployment compensation, worker’s compensation, Supplemental Security Income (SSI), Veteran’s Administration (VA), or Temporary Assistance to Needy Families (TANF) during the past 12 months. Based on this survey, 17.3% of all persons age 16 and older in Illinois are estimated to be receiving one or more of these forms of public assistance. As Table 31 indicates, no differences (significant at the p<.001 level) were found in the prevalence rates of alcohol and other drug use between those currently receiving and those not currently receiving public financial support. There also were no differences in the treatment need of persons who did and did not receive such support during the past year.

¿ Domestic violence victims

Victims of domestic violence have been

identified as a group at increased risk for

substance abuse treatment need (Testa, Livingston, & Leonard, 2003). Domestic violence victims were defined using eight items concerned with various aspects of physical and emotional abuse. Several of these items were adapted from the Conflict Tactics Scale (Straus & Gelles, 1990). Persons reporting any history of spousal or partner abuse such as being hit, pushed, shoved, insulted, or being physically isolated were classified as being a victim of domestic violence. The lifetime prevalence of domestic violence victimization among Illinois adults age 18 and older was 30.3%.

Several associations between domestic violence victimization and substance use were identified in the Illinois sample, as Table 32 shows. Persons ever the victim of domestic violence were more likely to report past-year binge drinking, lifetime use of marijuana, and lifetime and past-year use of cocaine. They reported more lifetime use of crack, methamphetamines, hallucinogens, inhalants, and nonmedical use of prescription analgesics, stimulants, and sedatives. They reported higher prevalence rates of lifetime and past-year illicit drug use (other than marijuana), and lifetime, past-year, and past-month of any drug use as well. Members of this population were significantly more likely to need treatment for alcohol use, drug use, and alcohol or drug use.


Table 31. Substance Use Prevalence and Abuse/Dependence Treatment Need in Illinois, by Past-Year Receipt of Public Financial Support, 2003

TOTAL RECEIVING PUBLIC SUPPORT

NOT RECEIVING PUBLIC SUPPORT

Percent



Standard Error p

Use of…

Alcohol Lifetime 83.9% (1.2) 85.8% (1.7) 83.5% (1.5) n.s. Past year 69.5 (1.4) 69.4 (2.7) 69.6 (1.6) n.s. Past month 54.0 (1.4) 51.2 (3.1) 54.6 (1.6) n.s. Past-year binge 30.2 (1.2) 29.8 (2.5) 30.2 (1.4) n.s.


Cocaine Lifetime 8.7 (0.9) 9.6 (1.5) 8.6 (1.1) n.s. Past year 0.8 (0.2) 1.3 (0.4) 0.7 (0.2) n.s.

Other drugs (l i fet ime) Crack 1.5 (0.4) 2.1 (0.6) 1.4 (0.4) n.s. Heroin 0.7 (0.2) 1.3 (0.6) 0.6 (0.2) n.s. Analgesics 2.9 (0.4) 4.2 (0.9) 2.6 (0.4) n.s. Methamphetamines 1.3 (0.2) 1.4 (0.4) 1.3 (0.2) n.s. Stimulants 5.5 (0.6) 7.1 (1.3) 5.1 (0.7) n.s. Hallucinogens 5.7 (0.5) 7.6 (1.3) 5.3 (0.5) n.s. Tranquilizers 2.1 (0.4) 2.2 (0.5) 2.0 (0.4) n.s. Sedatives 2.5 (0.4) 3.5 (0.8) 2.3 (0.5) n.s. Inhalants 4.1 (0.5) 3.6 (0.8) 4.2 (0.6) n.s. Club drugs 2.4 (0.4) 2.8 (0.8) 2.3 (0.4) n.s. Other drugs 1.0 (0.2) 1.8 (0.7) 0.9 (0.2) *


Lifetime 9.3 (0.8) 12.3 (1.6) 8.7 (0.9) *


Lifetime 12.2 (1.0) 14.8 (1.9) 11.7 (1.2) n.s. Past year 1.6 (0.2) 2.2 (0.6) 1.4 (0.2) n.s.


Treatment need Alcohol 15.5 (0.9) 16.4 (1.9) 15.3 (1.1) n.s. Drug 2.4 (0.3) 2.4 (0.6) 2.4 (0.4) n.s. Alcohol or drug 16.4 (1.0) 17.7 (1.9) 16.1 (1.1) n.s.

*p<.05; n.s.: not significant at .05 level.


Table 32. Substance Use Prevalence and Abuse/Dependence Treatment Need Among Those Age 18+ in Illinois, by Domestic Violence Victimization Status, 2003

TOTAL REPORTING DOMESTIC VIOLENCE

NOT REPORTING DOMESTIC VIOLENCE

Percent



Standard Error p

Use of…

Alcohol Lifetime 85.4% (1.4) 91.7% (2.3) 82.7% (1.7) ** Past year 70.7 (1.4) 77.8 (2.6) 67.7 (1.7) ** Past month 55.4 (1.5) 62.9 (2.8) 52.2 (1.8) ** Past-year binge 30.6 (1.3) 37.5 (2.5) 27.6 (1.5) ***

Marijuana Lifetime 32.0 (1.3) 47.7 (2.8) 25.2 (1.4) *** Past year 5.7 (0.5) 8.1 (1.2) 4.6 (0.5) ** Past month 3.0 (0.4) 5.1 (1.1) 2.1 (0.3) **

Cocaine Lifetime 9.1 (1.0) 14.4 (1.8) 6.8 (1.2) *** Past year 0.8 (0.2) 1.7 (0.5) 0.5 (0.1) ***

Other drugs (l i fet ime) Crack 1.6 (0.4) 3.4 (0.1) 0.8 (0.2) *** Heroin 0.8 (0.2) 0.8 (0.3) 0.8 (0.3) n.s. Analgesics 2.8 (0.4) 4.9 (1.0) 1.9 (0.3) *** Methamphetamines 1.4 (0.2) 2.9 (0.6) 0.7 (0.2) *** Stimulants 5.7 (0.6) 11.0 (1.7) 3.4 (0.5) *** Hallucinogens 5.8 (0.5) 10.5 (1.4) 3.8 (0.5) *** Tranquilizers 2.1 (0.4) 3.9 (1.1) 1.3 (0.3) ** Sedatives 2.5 (0.5) 5.4 (1.4) 1.2 (0.3) *** Inhalants 4.3 (0.5) 8.4 (1.6) 2.5 (0.4) *** Club drugs 2.4 (0.4) 4.2 (1.0) 1.6 (0.3) ** Other drugs 1.1 (0.2) 2.2 (0.6) 0.6 (0.2) ***


Lifetime 9.5 (0.8) 18.4 (2.2) 5.6 (0.6) ***


Lifetime 12.7 (1.1) 20.2 (2.1) 9.4 (1.2) *** Past year 1.5 (0.2) 2.7 (0.6) 1.0 (0.2) ***


Treatment need Alcohol 15.8 (1.0) 23.6 (2.2) 12.4 (1.0) *** Drug 2.4 (0.4) 4.5 (1.0) 1.2 (0.3) *** Alcohol or drug 16.5 (1.0) 24.9 (2.2) 12.9 (1.0) ***

**p<.01; ***p<.001; n.s.: not significant at .05 level.


¿ Chronic gamblers

Gambling addictions have been associated

with substance use and substance use treatment need (Spunt, Dupont, Lesieur, Liberty, & Hunt, 1998). In this study, persons were defined as chronic gamblers if they were diagnosed as “at-risk,” “problem,” or “pathological” gamblers based on 17-item DSM-IV-based diagnostics (Gerstein et al., 1999). It is estimated from this survey that 5.5% of all persons age 16 and older in Illinois are lifetime chronic gamblers. Table 33 provides comparisons of persons classified as chronic gamblers with all other survey respondents. These individuals were more likely to report lifetime, past-year, and past-month marijuana use. Chronic gamblers also were more likely to report lifetime use of

hallucinogens and inhalants, as well as the lifetime use of any illicit drug other than marijuana. Further, the lifetime, past-year, and past-month use of any drug was greater among chronic gamblers. Treatment need for alcohol and for any substance use were higher among these persons.

¿ Estimated treatment need for vulnerable groups

Using the expansion weights described in

Appendix A, the estimated number of persons within each of these vulnerable groups in need of treatment for alcohol use, drug use, and either alcohol or drug use are presented in Table 34.


Table 33. Substance Use Prevalence and Abuse/Dependence Treatment Need in Illinois, by Presence of Gambling Problem, 2003

TOTAL

WITH GAMBLING PROBLEM

WITHOUT GAMBLING PROBLEM

Percent



Standard Error p

Use of…

Alcohol Lifetime 83.9% (1.3) 90.0% (2.6) 83.5% (1.4) n.s. Past year 69.5 (1.4) 78.7 (3.6) 69.0 (1.4) * Past month 54.0 (1.4) 64.2 (5.2) 53.4 (1.5) n.s. Past-year binge 30.1 (1.2) 45.4 (5.2) 29.2 (1.3) **

Marijuana Lifetime 31.6 (1.3) 50.2 (5.3) 30.5 (1.3) *** Past year 6.2 (0.5) 17.9 (4.5) 5.6 (0.5) *** Past month 3.3 (0.4) 13.1 (4.3) 2.8 (0.3) ***

Cocaine Lifetime 8.7 (0.9) 14.3 (3.0) 8.4 (1.0) * Past year 0.8 (0.2) 0.1 (0.0) 0.9 (0.2) n.s.

Other drugs (l i fet ime) Crack 1.5 (0.4) 2.2 (1.1) 1.5 (0.4) n.s. Heroin 0.7 (0.2) 3.1 (1.9) 0.6 (0.2) ** Analgesics 2.9 (0.4) 8.7 (4.0) 2.5 (0.3) ** Methamphetamines 1.3 (0.2) 3.5 (1.3) 1.2 (0.2) * Stimulants 5.5 (0.6) 13.9 (4.2) 5.0 (0.6) ** Hallucinogens 5.7 (0.5) 18.3 (4.4) 5.0 (0.5) *** Tranquilizers 2.0 (0.4) 3.5 (1.4) 2.0 (0.4) n.s. Sedatives 2.5 (0.4) 5.7 (1.9) 2.3 (0.5) * Inhalants 4.1 (0.5) 12.2 (4.1) 3.7 (0.5) *** Club drugs 2.4 (0.4) 7.4 (4.0) 2.1 (0.3) * Other drugs 1.0 (0.2) 0.9 (0.5) 1.0 (0.2) n.s.


Lifetime 9.3 (0.8) 19.1 (4.4) 8.7 (0.8) **


Lifetime 12.2 (1.0) 27.1 (4.8) 11.4 (1.0) *** Past year 1.6 (0.2) 1.7 (0.8) 1.6 (0.2) n.s.


Treatment need Alcohol 15.5 (0.9) 32.3 (4.8) 14.5 (0.9) *** Drug 2.4 (0.3) 8.6 (4.0) 2.0 (0.3) ** Alcohol or drug 16.4 (1.0) 33.0 (5.0) 15.4 (1.0) ***



Table 34. Estimated Population in Need of Treatment Among Vulnerable Populations in Illinois, 2003

TREATMENT NEED FOR… Alcohol Use Drug Use Alcohol or Drug Use

Total population† Percent Population† Percent Population† Percent Population†

TOTAL 9,530 15.5% 1,477 2.4% 228 16.4% 1,561

Medical health insurance coverage

Yes 8,154 16.4 1,339 2.5 202 17.3 1,412 No 1,337 10.3 137 1.9 26 11.0 147

Physically disabled Yes 1,810 12.8 231 2.9 53 14.2 256 No 7,720 16.1 1,246 2.3 176 16.9 1,305

Mental disorder Yes 1,059 26.8 283 6.6 70 29.2 309 No 8,472 14.1 1,194 1.9 159 14.8 1,252

Ever homeless Yes 458 8.5 38 2.4 29 11.0 49 No 8,704 16.3 1,415 2.4 197 17.0 1,484

Public financial support recipient

Yes 1,647 16.4 270 6.3 39 17.7 291 No 7,883 15.3 1,208 2.4 189 16.1 1,269

Domestic violence victim Yes 2,733 23.6 645 4.5 123 24.9 682 No 6,299 12.4 780 1.2 76 12.9 812

Problem gambler Yes 526 32.3 170 8.6 45 33.0 173 No 9,004 14.5 1,307 2.0 184 15.4 1,388

†In thousands.


¿¿¿ DISCUSSION

Substance use and abuse continues to afflict the lives of many Illinois citizens. This need assessment reflects DASA’s ongoing commitment to monitor substance use and treatment need patterns in Illinois as part of its mission to serve the citizens of our State. Copies of earlier need assessment reports can be found at www.srl.uic.edu/publist/DASA.htm.

Most important are the findings that 16.4% of all persons age 16 and older were identified as having treatment need for alcohol and/or drug abuse/dependence during the past year. This represents approximately 1.5 million persons estimated to currently need treatment in Illinois. Given that the prevalence of alcohol-related treatment need (15.5%) is six times greater than the treatment need associated with other drug use (2.4%), this serves as an important reminder of the relative impact of these various substances on the lives of our citizens.

Other findings include the relationships between several sociodemographic indicators and need for treatment. Alcohol or other drug treatment need is greater among households in the highest income group ($70,000/year and above), persons currently working, college graduates, Whites, younger persons, males, and those never having been married. One particularly notable finding is that treatment need for alcohol abuse/dependence is greatest among college graduates, while treatment need for other drug abuse/ dependence is greatest among persons with less than a high school

degree. In other words, alcohol treatment need increases with education and other drug treatment need decreases with education.

Among the vulnerable populations examined, alcohol or other drug treatment need was greater among those currently having health insurance than those without health insurance. Illinois does not have a parity law; however, that would require equal insurance coverage for substance abuse intervention and treatment services. Consequently, it is likely that many health insurance policies do not provide adequate coverage for the treatment of substance abuse conditions.

Persons classified as having mental disorders, those ever having been a victim of domestic violence, and persons with gambling problems also were found to be at increased risk for having alcohol and/or other drug treatment need. Clearly, these groups deserve additional attention as substance abuse or dependence is believed to covary with and contribute to each of these conditions.

Findings that more than 90% of all persons identified with alcohol and/or other drug abuse/dependence conditions have unmet treatment need is also of critical importance. Clearly, many persons in need of assistance for substance use conditions are not obtaining help. Explanations as to why so many persons with treatment need do not receive it are unfortunately beyond the scope of this need assessment. We would speculate, however, that structural, social, economic, and other personal factors would all need to be considered in order


to account for such a large gap between treatment need and receipt of treatment.

Finally, several limitations of this research should be acknowledged. Most importantly, and as several of this report’s technical appendices indicate, there are numerous potential sources of error that may bias the estimates from any survey. In exploring potential sources of survey error specific to this study, measurement error associated with the

underreporting of substance use behaviors is the most likely source of bias (see Appendix H). Consequently, the findings presented in this report are likely to be conservative, with the true prevalence estimates being somewhat higher. In addition, this study was designed to produce independent estimates for eight sample areas within the State. Estimates cannot be made for smaller geographic units, such as counties or cities (other than Cook County and the city of Chicago).


¿¿¿ REFERENCES

American Psychiatric Association. (1994).

Diagnostic and statistical manual of mental disorders (4th ed., DSM-IV). Washington, DC: Author.

Artalejo, F. R., de Andres Manzano, B., Guallar-Castillon, P., Mendizabal, M. T. P., Enriquez, J. G., & del Rey Calero, J. (2000). The association of tobacco and alcohol consumption with the use of health care services in Spain. Preventive Medicine, 31, 554–561.

Cook, R. T. (1998). Alcohol abuse, alcoholism, and damage to the immune system—A review. Alcoholism: Clinical and Experimental Research, 22, 1927–1942.

Felson, D. T., Kiel, D. P., Anderson, J. J., & Kannel, W. B. (1988). Alcohol consumption and hip fractures: The Framingham Study. American Journal of Epidemiology, 128, 1102–1110.

Gerstein, D., Murphy, S., Toce, M., Hoffmann, J., Palmer, A., Johnson, R. et al. (1999). Gambling impact and behavior study: Report to the National Gambling Impact Study Commission. Chicago: National Opinion Research Center.

Gfroerer, J. C., & Epstein, J. F. (1999). Marijuana initiates and their impact on future drug abuse treatment need. Drug and Alcohol Dependence, 54, 229–237.

Gfroerer, J., Penne, M., Pemberton, M., & Folsom, M. (2003). Substance abuse treatment need among older adults in 2020: The impact of the aging Baby-Boom cohort. Drug and Alcohol Dependence, 69, 127–135.

Grant, B. F., & Dawson, D. A. (1997). Age at onset of alcohol use and its association with DSM-IV alcohol abuse and dependence: Results from the National Longitudinal Alcohol Epidemiologic Survey. Journal of Substance Abuse, 9, 103–110.

Grant, B. F., & Dawson, D.A. (1998). Age at onset of drug use and its association with DSM-IV drug abuse and dependence: Results from the National Longitudinal Alcohol Epidemiologic Survey. Journal of Substance Abuse, 10, 163–173.

Harris, K. M., & McKellar, J. D. (2003). Demand for alcohol treatment. Frontlines: Linking Alcohol Services Research & Practice, June, 1–2.

Harwood, H., Fountain, D., & Livermore, G. (1998). The economic costs of alcohol and drug abuse in the United States, 1992. Publ. No. 98-4327. Rockville, MD: National Institutes of Health.

Horgan, C., Skwara, K. C., & Strickler, G. (2001). Substance abuse: The nation’s number one health problem. Princeton, NJ: The Robert Wood Johnson Foundation.

Hunkeler, E. M., Hung, Y., Rice, D. P., Wesiner, C., & Hu, T. (2001). Alcohol consumption patterns and health care costs in an HMO. Drug and Alcohol Dependence, 64, 181–190.

Johnson, R. A., Gerstein, D. R., & Rasinski, K. A. (1998). Adjusting survey estimates for response bias: An application to trends in alcohol and marijuana use. Public Opinion Quarterly, 62, 354–377.

Johnson, T. P., Freels, S. A., Parsons, J. A., & VanGeest, J. B. (1997). Substance abuse and homelessness: Social selection or social adaptation? Addiction, 92, 437–445.

Johnson, T. P., & Mott, J. A. (2001). The reliability of self-reported age of onset of tobacco, alcohol and illicit drug use. Addiction, 96, 1187–1198.

Kessler, R. C., Anidres, G., Mroczek, D., Ustun, B., & Wittchen, H. U. (1998). The World Health Organization Composite International Diagnostic Interview Short Form (CIDI-SF).


International Journal of Methods in Psychiatric Research, 7, 171–185.

Kessler, R. C., Crum, R. M., Warner, L. A., Nelson, C. B., Schulenberg, J., & Anthony, J. C. (1997). Lifetime co-occurrence of DSM-III-R alcohol abuse and dependence with other psychiatric disorders in the National Comorbidity Survey. Archives of General Psychiatry, 54, 313–321.

Lahaut, V. M. H. C. J., Jansen, H. A. M., van de Mheen, D., & Garretsen, H. F. L. (2002). Non-response in a sample survey on alcohol consumption. Alcohol & Alcoholism, 37, 256–260.

Link, B. G., Susser, E., Stueve, A., Phelan, J., Moore, R. E., & Struening, E. (1994). Lifetime and five-year prevalence of homelessness in the United States. American Journal of Public Health, 84, 1907–1912.

Lucas, E. G. (1987). Alcohol in industry. British Medical Journal, 291, 460–461.

McGinnis, J. M., & Foege, W. (1993). Actual causes of death in the United States. JAMA, 270, 2207–2212.

Naimi, T. S., Brewer, R. D., Mokdad, A., Denny, C., Serdula, M. K., & Marks, J. S. (2003). Binge drinking among US adults. JAMA, 289, 70–75.

Office of Alcoholism and Substance Abuse. (1999, April). Illinois Department of Human Services: A substance abuse plan for the new millennium. Chicago: Illinois Department of Human Services, Office of Alcoholism and Substance Abuse.

Pollack, H. A., Danziger, S., Seefeldt, K. S., & Jayakody, R. (2002). Substance abuse among welfare recipients: Trends and policy responses. Social Science Review, 76, 256–274.

Reardon, M. L., Burns, A. B., Preist, R., Sachs-Ericsson, N., & Lang, A. R. (2003). Alcohol use and other psychiatric disorders in the formerly homeless and never homeless: Prevalence, age of onset, comorbidity,

temporal sequencing, and service utilization. Substance Use & Misuse, 38, 601–644.

Rehm, J., Gmel, G., Sempos, C. T., & Trevisan, M. (2003). Alcohol-related morbidity and mortality. Alcohol Research & Health, 27, 39–51.

Smith-Warner, S., Spiegelman, D., Yaun, S., van den Brandt, P., Folsom, A., Goldbohm, A. et al. (1998). Alcohol and breast cancer. JAMA, 279, 535–540.

Straus, M. A., & Gelles, R. J. (1990). Physical violence in American families: Risk factors and adaptations to violence in 8,145 families. New Brunswick, NJ: Transaction Books.

Spunt, B., Dupont, I., Lesieur, H., Liberty, H. J., & Hunt, D. (1998). Pathological gambling and substance misuse: A review of the literature. Substance Use & Misuse, 33, 2535–2560.

Testa, M., Livingston, J. A., & Leonard, K. E. (2003). Women’s substance use and experiences of intimate partner violence: A longitudinal investigation among a community sample. Addictive Behaviors, 28, 1649–1664.

U.S. Dept. of Health and Human Services. (2000). 10th Special Report to the U.S. Congress on alcohol and health. Washington, DC: National Institute on Alcohol Abuse and Alcoholism.

Yacoubian, G. S., Peters, R. J., Urbach, B. J., & Johnson, R. J. (2002). Comparing drug use between welfare-receiving arrestees and non-welfare-receiving arrestees. Journal of Drug Education, 32, 139–147.

Yamaguchi, K., & Kandel, D. B. (1984). Patterns of drug use from adolescence to young adulthood: III. Predictors of progression. American Journal of Public Health, 74, 673–681.

York, J. L., & Hirsch, J. A. (1997). Association between blood pressure and lifetime drinking patterns in moderate drinkers. Journal of Studies on Alcohol, 58, 480–485.

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003 A-1

¿¿¿ APPENDIX A

Study Methodological Report Karen Retzer and David Schipani

OVERVIEW

The Survey Research Laboratory (SRL) at the University of Illinois conducted this study on behalf of the Illinois Division of Alcohol and Substance Abuse (DASA). The purpose was to measure the prevalence of drug and alcohol abuse among individuals age 16 and older in Illinois and also to estimate the substance abuse treatment need of this population. This report discusses the pretest, describes the main study sample design, presents the disposition of sample and sample rates for the entire sample, and includes a description of the weights.

PRETEST

The pretest sample for this telephone study was obtained from Survey Sampling, Inc. on September 26, 2002. The pretest sample was not stratified by region, as the main study sample was, but was a simple random sample of the state of Illinois. The pretest consisted of 100 cases, and data collection started on December 11 and ended on December 17, 2002. Tables A-1 and A-2 show the disposition of sample and the rates for the pretest, respectively. Exhibit A contains definitions of the sample disposition codes for the pretest, and Exhibit B defines the sample rates. A description of how the rates are calculated can be found in the section on Final Disposition of Sample and Sample Rates for the main study.

Table A-1. Disposition of Sample: Pretest

Code Disposition Number Percent

(1) Completed interview, English 17 17.0% (30) No answer 11 11.0% (31) Answering service/machine 6 6.0% (32) Eligible R not available 6 6.0% (33) Unscreened R not available 6 6.0% (40) Final refusal to screener 13 13.0% (41) Final refusal interview 4 4.0% (55) Not able to interview during survey period 2 2.0% (56) Never able to interview 3 3.0% (86) Nonworking 18 18.0% (87) Nonresidential 12 12.0% (88) Ineligible foreign language 2 2.0%

TOTAL 100 100.0%

A-2 Illinois Division of Alcoholism & Substance Abuse

Exhibit A. Disposition Codes: Pretest

Code Description

(01) Completed interview—English

(30) No answer Telephone numbers that were never answered or always rang busy.

(31) Answering machine/answering service Answering devices or answering services.

(32) Eligible respondent not available Screened respondent was not at home or was busy and could not be interviewed at time of contact. No appointment could be made.

(33) Unscreened R not available Someone answered the phone but was not available for screening. No appointment could be made.

(40) Final refusal before screener completed Refusal to screener or to completing screener.

(41) Final refusal interview Eligible respondent refused to be interviewed or to complete the interview.

(55) Not able to interview during survey period There was clear indication that respondent would be unavailable to participate within time confines of survey period. Respondent might have been away on vacation, hospitalized, etc.

(56) Never able to interview Eligible respondent was too hard of hearing, permanently ill, incapacitated, or for some other reason would never be able to be interviewed. Not related to the timeframe of the data collection effort.

(86) Nonworking Phone number was disconnected or otherwise not working.

(87) Nonresidential Phone number was for business or other nonresidential.

(88) Ineligible, foreign language Contact person was unable to complete interview in English.

Table A-2 . Sample Numbers and Rates: Pretest

Number Rate

Total sample 100 Nonduplicates 100 100.0% Working numbers 82 82.0% Residential numbers 70 85.4% Contact to screener 53 75.7% Cooperation to screener 29 54.7% Eligible 27 93.1% Contact to final 21 77.8% Cooperation to final 17 81.0% Response rate 27.4% Refusal rate 25.9% Cooperation rate 51.4%

SAMPLE DESIGN: MAIN STUDY

The sample design was a stratified random-digit-dial telephone survey within Illinois. The sample consisted of eight strata: Chicago, suburban Cook County, northwest Illinois, south collar counties, north collar counties, west central Illinois, east central Illinois, and

southern Illinois. Appendix C lists the tracts that made up the Chicago and suburban Cook County strata and the counties of the other six sample areas.

The population of interest was individuals 16 years of age and older. Because we wanted to oversample the younger ages, the selection process was as follows: if a household had individuals age 16–24 and individuals age 25 and older, an individual age 16–24 was selected 90% of the time and an individual age 25 or older was selected 10% of the time. Furthermore, if there was more than one individual in the age group (16–24 or 25+), one person was randomly selected using the Troldahl-Carter-Bryant method of selection (Bryant, 1975). If the household only had individuals in one of the two age groups, the respondent was selected using Troldahl-Carter-Bryant method of selection. No incentives were used in this study.


The goal was to achieve 500 completed interviews in each of the eight sample areas, for a total of 4,000 completed interviews. Because all study designs are different, there is never a

way to predict precisely the starting sample size necessary to achieve a targeted number of completed interviews. In lieu of a precise model, SRL makes estimates. SRL based the anticipated completion rates for this study (and consequently, the starting sample size necessary) on actual completion rates from studies conducted by SRL in the past. We forecast the following sampling rates for this study: 76% of the numbers we started with would be working numbers, 87% of those would be residential numbers, 81% of the residential numbers would be contacted to conduct a screener, 67% of these would cooperate to the screener, 98% of those who cooperated would be eligible, 93% of the eligible would be contacted for the final interview, and 70% of those would complete the interview. Based on these rates, we originally estimated a total starting sample size of 17,473.

In January 2003, SRL obtained 18,847 pieces of sample from Survey Sampling, Inc. Sample was randomized and divided into 10 replicates of 1,600 pieces each. These first 10 replicates were released between January and April 2003. Interviewing began on January 15, 2003. In March, it became clear that we were not going to have enough sample to attain 4,000 completes. On June 6, we obtained an additional 2,162 pieces of sample. We calculated rates in each of the sample areas to make projections as to the amount of sample that would be needed to reach our goal. Replicates 11–16 were released in an effort to achieve 500 completes in each of the sample areas. Table A-3 shows the date that and location at which each replicate was released. As can be seen, some of the replicates were divided with a portion of the replicate released on different dates and in different calling centers. This happened for two reasons. First, if we were releasing a replicate sooner than had

Exhibit B. Sample Rate Definitions Total sample: the total number of phone numbers called for this study.

Working numbers: the number of phone numbers that were working phone numbers (includes all final disposition codes with the exception of code 86).

Residential numbers: the number of phone numbers that were households, not businesses (the number of “working numbers” above minus the number of cases with final disposition code 87).

Contact to screener: the total number of households that were contacted for the screener (the number of “residential numbers” above minus the number of cases with final disposition codes 30 or 31).

Cooperation to screener: the total number of households that completed the screener (the number of “contact to screener” above minus the number of cases with final disposition codes 33, 40, 47, 55, or 56—in other words, the number of cases with final dispositions 1 –8, 32, 41, 42, 48, 60, 70, 85, 88, or 90).

Eligible: the number of respondents selected (the number of “cooperation to screener” above minus the number of cases with final disposition codes 70, 85, 88 or 90—in other words, the number of cases with final dispositions 1 –8, 32, 41, 42, 48 or 60).

Contact to final: the total number of respondents who were contacted for an interview (the number of “eligible” above minus the number of cases with final disposition codes 32 and 60—in other words, the number of cases with final dispositions 1 –8, 41, 42 or 48).

Cooperation to final: the total number of respondents who completed an interview (the number of “contact to final” above minus the number of cases with final disposition codes 41, 42 or 48—in other words, the number of cases with final dispositions 1 –8).


been scheduled, we needed to allow time for the advance letters to reach the households, thus sample released later for some of the replicates was the sample for which we had mailing addresses. Second, with two calling centers, we released sample where staff were optimally available to make the needed phone calls.

Table A-3. Sample Replicate Release Dates

Replicate Size Date Location 1 1,600 1/16/03 Urbana-Champaign 2 1,600 1/16/03 Chicago 3 1,600 1/20/03 Urbana-Champaign 4 1,600 1/29/03 Chicago 5 1,600 2/6/03 Urbana-Champaign 6 1,600 2/10/03 Chicago 7 1,600 2/12/03 Urbana-Champaign 8 1,600 2/14/03 Chicago 9 1,033

567 3/21/03 4/7/03

Urbana-Champaign Urbana-Champaign

10 1,027 573

4/18/03 4/24/03


11 350 568 237 300

6/1/03 6/1/03 6/6/03 6/6/03

Chicago Urbana-Champaign Urbana-Champaign Chicago

12 318 177 476 264

6/13/03 6/25/03 6/13/03 6/25/03

Chicago Chicago Urbana-Champaign Urbana-Champaign

13 626 374

7/8/03 7/11/03


14 786 7/16/03 Urbana-Champaign 15 148

100 7/18/03 7/23/03


16 50 7/28/03 Urbana-Champaign

TOTAL 20,774

We release sample in randomized replicates over time for two reasons. First, if we need to conduct analysis or end data collection before all of the sample has been released, the sample we will have worked at that point will

be a random selection of the entire sample. Second, as we work through the sample and finalize replicates, we can use the information from earlier replicates to improve the projections about needed sample for upcoming sample releases.

Because telephone exchange boundaries do not correspond to census tract boundaries, sampling telephone numbers for geographic areas defined by census tract and county usually involves a tradeoff between coverage of the geographic area and incidence (the percent of the sample that actually lives in the area of interest). The coverage rate, incidence, and sample size for each stratum are listed in Table A-4. Table A -4. Coverage Rate, Incidence, & Sample Size, by Stratum

Stratum Coverage Rate (%)

Incidence (%)

Sample Size

Chicago 98.8 98.8 3,214

Suburban Cook 96.0 96.0 2,816

Northwest Illinois 99.2 99.7 2,570

South collar counties 94.4 97.4 2,578

North collar counties 98.8 95.5 2,803

West central Illinois 99.7 99.3 2,182

East central Illinois 99.4 99.6 2,169

Southern Illinois 99.7 99.8 2,442

Final Disposition of Sample & Sample Rates

The final disposition of sample and rates for the overall sample are contained in Tables A-5 and A-6, respectively. Exhibit C contains a description of the main study sample dispositions, while Exhibit B defines the sample rates (including the sample disposition codes that go into each rate).


Table A-5. Final Dis position of Sample: Total Sample

Code Disposition Number Percent

(01) Completed interview—English, age 16–17 237 1.14% (02) Completed interview—Spanish, age 16–17 4 0.02% (03) Completed interview—English, age 18–24 541 2.60% (04) Completed interview—Spanish, age 18–24 32 0.15% (05) Completed interview—English, age 25+ 3,082 14.84% (06) Completed interview—Spanish, age 25+ 83 0.40% (07) Completed interview—age 18+ 1 0.00% (08) Partial completed interview 175 0.84% (30) No answer 2,278 10.97% (31) Answering machine/answering service 1,024 4.93% (32) Eligible R not available 940 4.52% (33) Unscreened R not available 632 3.04% (40) Final refusal before screener completed 2,304 11.09% (41) Final refusal English interview 1,855 8.93% (42) Final refusal Spanish interview 34 0.16% (47) Unscreened refusal—privacy manager 98 0.47% (48) English screened refusal—privacy manager 49 0.24% (55) Not able to interview during survey period 44 0.21% (56) Never able to interview 204 0.98% (60) Other eligible 2 0.01% (70) Ineligible, age 18 0.09% (85) Deceased 3 0.01% (86) Nonworking 4,161 20.03% (87) Nonresidential 2,795 13.45% (88) Ineligible, foreign language 177 0.85% (90) Other ineligible 1 0.00%

TOTAL 20,774 100.00%

Table A -6 . Sample Numbers & Rates: Total Sample

Number Rate

Total sample 20,774 Nonduplicates 20,774 100.0% Working numbers 16,613 80.0% Residential numbers 13,818 83.2% Contact to screener 10,516 76.1% Cooperation to screener 7,234 68.8% Eligible 7,035 97.2% Contact to final 6,093 86.6% Cooperation to final 4,155 68.2% Response rate 32.7% Refusal rate 33.7% Cooperation rate 49.3%

The response rate is the proportion of the eligible respondents who completed the interview. The American Association of Public Opinion Research’s (AAPOR) Standard Definitions includes six different methods for calculating response rates (pp. 36–37). This

sampling report describes the calculation of response rate number 3 (RR3) in detail.1 In RR3, the numerator includes completed interviews, while the denominator includes interviews, refusals, noncontact of eligible respondents, and a proportion of households whose eligibility status is unknown.

In the overall sample, there were 4,306 cases for which a screening questionnaire 1 Response rate 1 (RR1) and response rate 5 (RR5) are included here to show the variation in what researchers might report as a response rate for this study. RR1 and RR5 have the same numerator as that for RR3, i.e., the number of completed interviews. RR1 and RR5 differ from RR3 because they make different assumptions about what is in the denominator of the ratio. RR1 is the most conservative (i.e., lowest) response rate; RR5 is the highest. While RR3 includes a portion of the cases for which eligibility is unknown, RR1 includes all of the cases for which eligibility is unknown and RR5 includes none of the unknown eligible cases. RR1 for this study is 30.5%; RR5 is 59.1%.


could not be administered (Table A-6’s contact to screener minus cooperation to screener, plus Table A-5’s code 31 answering machines). Based on the eligibility of those cases where we actually did get cooperation to the screener (see Table A-6), we assumed that 97.2% of these 4,306 cases would have been eligible. In another 2,278 cases (with final disposition code 30 in Table A-5), the phone rang continuously at each contact attempt and was never answered. Based on the numbers in Table A-6, we assumed 80.0% of those 2,278 cases were working numbers, 83.2% were household numbers, and 97.2% were eligible. Consequently, the total number of cases with assumed eligibility is estimated as 5,659. This is the sum of 97.2% of 4,306 (4,185) plus 64.7% (80.0% * 83.2% * 97.2%) of 2,278 (1,474 cases). Thus, the response rate is computed as the ratio of 4,155 (from Table A-6) completed and partial (see definition in Exhibit B) interviews to the sum of the cases known to be eligible (7,035), from Table A-6, and the estimated number of eligible cases among the cases of unknown eligibility (5,659). The resulting response rate is 32.7%.

The refusal rate is the proportion of the eligible respondents who either refused to complete an interview or broke off an interview. AAPOR’s Standard Definitions includes three different methods for calculating refusal rates. This report uses refusal rate 2 (p. 39). In this rate, the numerator includes refusals (actual refusals of eligible respondents plus a proportion of refusals of households whose eligibility is unknown); the denominator is the same as that of response rate 3 described above (5,659 + 7,035 = 12,694). Therefore, the total number

of refusals is those who refused after screening, 1,938 (disposition 41, 42, and 48), plus 97.2% of the 2,402 (2,335) who refused prior to screening (disposition 40 and 47), for a total of 4,273 refusals. The refusal rate is 33.7%.

The cooperation rate is the number of completed interviews (and partially completed interviews) divided by the number of completed interviews (including partials) plus the number of refusals. AAPOR’s Standard Definitions includes four different methods for calculating cooperation rates. This report uses cooperation rate 4 (p. 38). Thus, the cooperation rate is computed as the ratio of 4,155 completed and partial cases to the sum of these completed and partial cases (4,155) and the number of refusal cases in the numerator of the refusal rate calculated above (4,273). The cooperation rate is 49.3%.

Refusal Conversion & Quality Control

SRL makes a substantial effort to contact

respondents during the data collection period. In a phone survey, interviewers contact a household up to 20 times before assigning a final disposition. If a respondent refuses to complete either a screener or an interview, interviewers continue calling up to the 20 attempts or until they get a second refusal, learn that the household is ineligible, or attain a complete interview, at which point the case is finalized. Table A-7 shows some refusal conversion rates to measure the refusal conversion effort.


Table A-7 . Refusal Conversion Rates: Total Sample

Rate Number Percent

Rate A. Percent of completed interviews that ever refused 783/3,980 19.7% Rate B. Percent of completed or partial interviews that ever refused 829/4,155 20.0% Rate C. Percent of refusals that turned into complete/partial or ineligible 875/5,562 15.7% Rate D. Percent of refusals that turned into a completed interview 783/5,562 14.1% Rate E. Percent of refusals that turned into a completed or partially completed interview 829/5,562 14.9%

Parental Approval Rate & Several Refusal Rates for Underage Respondents

This study included respondents age 16 and 17, and before interviewing such underage respondents, we first obtained approval from a parent or guardian. The process of selecting and obtaining consent to interview a 16- or 17-year-old was as follows:

(1) If the cell for 16–24 year olds in a household was chosen (the other cell was for individuals age 25 and older), our regular Troldahl-Carter-Bryant method of selection was used to pick the 16–24 year old respondent.

(2) We then asked whether the selected respondent was under the age of 18.

(3) If yes, we asked to speak to a parent or guardian and obtained approval to interview the selected minor. The script we used to get parental approval read:

We are seeking your permission to interview your teenager, the (youngest/oldest) (male/female) age 16 to 17, about substance use. It will take about 25 minutes to complete. Participation is completely voluntary and all answers will be kept confidential. (He/She) is free to refuse to answer any questions (he/she) does not want to answer. May we have your permission to speak with this teenager?

(4) After obtaining parental approval, we then proceeded as we did for adult respondents,

to try to contact the respondent so as to complete the interview with him or her.

(5) If at any point a parent or guardian said that they did not want us to interview the minor respondent (even if this, or another, parent or guardian had previously given us permission to do the survey), we immediately dispositioned the case as a final refusal; no refusal conversion was done for such cases.

Table A-8 shows that we selected 1,519

respondents in the 16–24 cell, of which 33% were confirmed to be minors age 16 or 17. (Note that in 22 cases we could not get an answer to the question of whether the selected respondent was under the age of 18; in none of those 22 cases did we proceed to do the interview.)

Table A-8. Respondents Age 16 to 24

Age Number Percent

16 to 17 501 33.0% 18 to 24 996 65.6% Just know that 16 to 24 22 1.4%

TOTAL 1,519 100.0%

Table A-9 shows the results of our efforts

to get parental approval for and then complete interviews with the 499 eligible 16- and 17-year-old respondents (2 of the 501 respondents shown in Table A-8 turned out to speak neither English nor Spanish, which made them ineligible).


Table A-9 . Final Refusal Status * Initial Parental Approval for Underage Respondents†

INITIAL PARENTAL APPROVAL Final Refusal Status Yes No TOTAL

No one in household refused 316 (a) 34 (e) 350 Respondent refused 17 (b) 0 (f) 17 Unknown household member refused 15 (c) 8 (g) 23 Parent/guardian refused 9 (d) 100 (h) 109

TOTAL 357 142 499

†The final dispositions (see Exhibit C) of the cases in Table A-9’s eight cells (a-h) are as follows: Cell a: 237 FNL=1; 4 FNL=2; 8 FNL=8; 61 FNL=32; 3 FNL=55; 2 FNL=56; 1 FNL=90 Cell b: 16 FNL=41; 1 FNL=48 Cell c: 12 FNL=41; 3 FNL=48 Cell d: 9 FNL=41 Cell e: 28 FNL=32; 2 FNL=55; 3 FNL=56; 1 FNL=90 Cell f: (no cases in this cell) Cell g: 8 FNL=41 Cell h: 100 FNL=41

The parental approval rate is somewhere between 66.7% and 69.7%, depending upon how one interprets the refusals from an “unknown household member” after an initial parental approval (the 15 cases in Table A-9’s cell c). The low estimate for the parental approval rate assumes that all 15 of these cases were refusals from a parent or guardian following the initial approval, so that only 333 (cells a + b) out of 499 cases (66.7%) ended up with parental approval for the study. The high estimate for the parental approval rate assumes that none of these 15 refusals was a refusal from a parent or guardian, so that 348 (333 + 15) out of 499 cases (or 69.7%) ended up with parental approval for the study. The truth is somewhere in between: some but not all of those 15 cases were probably refusals from a parent or guardian, revoking the parental approval we had obtained previously.2

2 The range for the parental approval rate would edge down somewhat if we assume a few more respondents age 16 or 17 by including in the denominator some of the 22 respondents from Table A-8 for whom we could not get any more information than that they were age 16 to 24 (i.e., for whom we could not get an answer to the question of

Several other rates also can be calculated from Table A-9.

(1) (All refusals)/(all eligible 16–17

respondents) = 29.9%

Calculated as b+c+d+ f+g+h/499=149/499.

(2) (Parental refusals)/(all eligible 16–17

respondents) = somewhere between 21.8%

and 26.5%.

Low range calculated as d+h/499=109/499. High range calculated as c+d+g+h/499=132/499.

Note that this is another way of approaching

what the parental approval rate shows.

whether the selected respondent was under the age of 18). For example, we could assume that the percentage of 16- or 17-year-olds in this group of 22 respondents was the same as the percentage of 16- or 17-year-olds in the group of 1,497 respondents for whom we obtained an answer to this question. And since we learned that 501 of 1,497 respondents in this latter group were age 16 or 17 (33.5%), this would mean assuming that 33.5%, or around 7, of the 22 respondents in the mystery group were also age 16 or 17. Adding 7 respondents to the denominator of the parental approval rate would make the range 65.8% (333/506) to 68.8% (348/506), instead of the slightly higher current range of 66.7% to 69.7%.


(3) (Respondent refusals)/(all eligible 16–17

respondents) = somewhere between 3.4%

and 8.0%.

Low range calculated as b/499=17/499. High range calculated as b+c+g/499=40/499; note that we include cell g because even though we had not yet obtained parental refusal, the unknown HH member who refused us may have been the selected respondent.

This is lower than the comparable rate #2

above.

(4) (Parental refusals)/(all refusals for

eligible 16–17 respondents) = somewhere

between 73.2% and 88.6%.

Low range calculated as d+h/b+c+d+ f+g+h= 109/149. High range calculated as c+d+g+h/b+c+d+ f+ g+h=132/149.

This is a way of comparing rates #2 and #3

above and shows that most of the refusals, by a

very large margin, came from parents.

(5) (Respondent refusals)/(all refusals for

eligible 16–17 respondents) = somewhere

between 11.4% and 26.8%.

Low range calculated as b/b+c+d+f+g+h=17/149. High range calculated as b+c+g/b+c+d+f+g+h= 40/149.

This is the converse point from the one made

for rate #4 above. (The assumption is that all

of the refusals could be coming from just

parents and respondents, with no refusals from

other members of the household, which means

that the high parent and low respondent

estimates add up to 100%, and ditto for the low

parent and high respondent estimates.)

(6) (Respondent refusals after parental

approval)/(all cases with parental

approval) = somewhere between 4.9% and

9.2%.

Low range calculated as b/a+b+c=17/348. High range calculated as b+c/a+b+c= 32/348.

So once we obtained parental approval, the

respondent was unlikely to refuse.

(7) (Respondent completions & partials)/

(all cases with parental approval) =

somewhere between 71.6% and 74.8%.

Low range calculated as 237+4+8/a+b+c= 249/348. High range calculated as 237+4+8/a+b= 249/333.

Note that the numerator for both the low and

high range comes from footnote #2, which

notes that cell a contains 237 underage

respondents with an FNL=1, 4 with FNL=2,

and 8 with FNL=8. So it’s not just that

respondents were unlikely to refuse after

parental approval (rate #6) but that they were

very likely to complete (as opposed to being

unreachable, for example).

(8) (Parental refusals after approval)/(all

cases with initial parental approval) =

somewhere between 2.5% and 6.7%.

Low range calculated as d/a+b+c+d=9/357. High range calculated as c+d/a+b+c+d=24/357.

This rate shows that very few parents revoked

their approval after having given it.

Sample Weights Selection Weights

In the first stage of the sample selection, within each stratum, telephone numbers were sampled with equal probabilities of selection. However, given our primary interest in the probability of selection of the household, we need to translate the chance of a telephone number being sampled into the chance of a household being sampled. Because households have varying numbers of phone lines,3 they have different probabilities of being sampled. 3 By “phone line,” we mean lines with separate telephone numbers that are used to receive outside calls. Lines designated solely for computers, fax machines, etc. are not included in the calculations.


For example, all else being equal, a household that has two telephone lines has twice the chance of being sampled as a household with one line. Therefore, we calculated a phone line weight that is simply the inverse of the number of phone lines in the household.

Once we made contact with a household, we sampled one respondent from among all eligible respondents. The sampling was designed to oversample persons age 16–24 where available. As discussed earlier, if there was more than one individual eligible, one person was randomly selected using the Troldahl-Carter-Bryant method of selection (Bryant, 1975). Sample weights were calculated to equal the inverse of the probability of selecting the age group multiplied by 1 over the number of people in that selected age group.

The overall selection weight for each case at this stage is the product of the phone line weight and the eligible respondent weight.

Poststratification Weights

Poststratification weights were calculated for the data set to ensure that the marginal distribution of several demographic variables conformed to the marginal distribution of the same variables from the 2000 U.S. census for the eight sample areas. The variables used in the weighting procedure were age group (15–20; 21–39; 40–59; 60+), gender, and race (White, non-Hispanic; Black, non-Hispanic; Hispanic; and other). In addition, we adjusted the sample for stratum size; this was necessary because we drew the sample disproportionately from each of the eight sample areas. Calculating the poststratification weights involved several steps outlined below.

After applying the selection weight, we compared the age distribution in the sample data to that of the census data. The initial age weights were the ratio of the census percentages to the study sample percentages (e.g., % individuals age 21–39 in the census data divided by % age 21–39 in the sample data).

We weighted the study data with the initial age-group weight and compared the weighted frequencies of the race groups to the race groups in the census. We computed the initial race weights as the ratio of the census percentages to the study percentages. This weight then was multiplied by the age weight from the first step to create a second-stage weight.

We weighted cases with the second-stage race weight and compared the study gender distribution with that of the census. Gender weights were computed as the ratio of census percentages to study percentages. The gender weights were multiplied by the race weights from the second stage for a third stage of weights.

Weighting was completed in this manner until the study sample percentages for each region had been adjusted to the demographic variables in the census for each region. This took one more iteration of age weights for six of the eight sample areas and one more iteration of the race weights for four of the eight sample areas. At that time the sample had been adjusted to match the census data on the demographic variables.

Because of the disproportionate sampling, we calculated a weight to adjust for the size of each region. We compared the percent of the completed interviews in each of the eight strata to the census estimates of the percent of population age 15 and over in those sample areas. The weight was simply equal to the ratio


of the census percent to the survey data percent. This weight was multiplied by the final race weight for the final weight in the data file (fnlwgt). This final weight is employed for all analyses reported in this document.

Expansion Weights

Expansion weights are used to estimate the size of the target populations of interest. These weights are equal to the final weight multiplied by the inverse of sampling ratio. The equation for the expansion weight is

Expansion weight= final weight * (total population in

Illinois 16 or older/total sample

size)

= final weight * (9,530,327/4,155).

REFERENCES American Association for Public Opinion

Research. (2000). Standard definitions: Final dispositions of case codes and outcome rates for surveys. Ann Arbor, MI: Author.

Bryant, B. E. (1975). Respondent selection in a time of changing household composition. Journal of Marketing Research, 12, 129–135.


Exhibit C. Disposition Codes: Main Study

Code Description (01) Completed interview—English, age 16–17

(02) Completed interview—Spanish, age 16–17

(03) Completed interview—English, age 18–24

(04) Completed interview—Spanish, age 18–24

(05) Completed interview—English, age 25+

(06) Completed interview—Spanish, age 25+

(07) Completed interview—English, age 18+

(08) Partially completed interview Partials were defined as incomplete interviews where the respondent had finished answering questions at least through the end of the G section (which asked about alcohol/drug treatment experiences). Since the G section was skipped by respondents who didn't indicate use of alcohol (on C1) or drugs (on D1a–D1l), in practice this meant that partials were defined as all incomplete cases with a non-missing value for the "etime" variable, which is filled with the current time immediately after the G section.

(30) No answer Telephone number was never answered or always rang busy.

(31) Answering machine/answering service Answering devices or answering services.

(32) Eligible respondent not available Screened respondent was not at home or was busy and could not be interviewed at time of contact. No appointment could be made.

(33) Unscreened R not available Someone answered the phone but was not available for screening. No appointment could be made.

(40) Final refusal before screener completed Refusal to screener or to completing screener.

(41) Final refusal English interview English-speaking eligible respondent refused to be interviewed or to complete the interview.

(42) Final refusal Spanish interview Spanish-speaking eligible respondent refused to be interviewed or to complete the interview.

(47) Unscreened refusal—privacy manager Household has privacy manager phone service that prevented completion of screener.

(48) English screened refusal—privacy manager After screener was conducted in English and respondent selected, privacy manager phone service prevented completion of an interview.

(55) Not able to interview during survey period There was clear indication that respondent would be unavailable to participate within time confines of survey period (away on vacation, hospitalized, etc).

(56) Never able to interview Eligible respondent was too hard of hearing, permanently ill, incapacitated, or for some other reason would never be able to be interviewed. Not related to the timeframe of the data collection effort.

(70) Ineligible, age No one age 16 or older lived in household.

(85) Deceased Selected respondent died before we could conduct interview.

(86) Nonworking Phone number was disconnected or otherwise not working.

(87) Nonresidential Phone number was for business or other nonresidential.

(88) Ineligible, foreign language Contact person spoke a language other than English or Spanish; was unable to complete interview in English.

(90) Other ineligible Situation did not seem to be covered by any other category and the case was ineligible.

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003

¿¿¿ APPENDIX B

Substance Use Definitions

Alcoholic beverage. Beer, wine, and liquor; a drink of alcohol is measured as (a) a can or bottle of beer, (b) a four-ounce glass of wine, (c) a one-ounce shot of liquor, or (d) a mixed drink. A 40-ounce bottle of beer equals four drinks.

Analgesics. Painkillers, including codeine products, Darvon, Demerol, methadone, morphine, OxyContin, Percocet, and Vicodin, used for nonmedical reasons.

Any drug use. Use of marijuana, cocaine, crack, heroin, pain relievers, methamphetamines, other stimulants, hallucinogens, tranquilizers, sedatives, inhalants, club drugs, and/or other drugs.

Any other drugs. Any drugs besides those mentioned that were not prescribed or that were used only for the feeling or experience they caused.

Cigarettes. Cigarettes (lifetime cigarette use was measured as having smoked more than 100 cigarettes).

Club drugs. MDMA or ecstasy, ghb, ketamine, special k, or k, rohypnol or roofies, Xanax, e, liquid ecstasy, soap, easy lay, liquid g, Georgia home boy, forget-me-not, roach, vitamin k.

Cocaine. Powder cocaine, used for nonmedical reasons; coke, coca-cola, lines, pine, snow, white, or china.

Crack cocaine. Including rock, stone, or redirock.

Binge drinking. Four or more drinks by females or five or more drinks by males on one occasion (within several hours) during the past year.

Hallucinogens. Any hallucinogenic drug, including LSD, PCP, and psilocybin/mushrooms; acid, dosing, tripping, angel, angel dust, horst, ozone, wack, rocket fuel, zoot, killer joints, wicky sticks, “shrooms,” crystal supergrass.

Heroin. Heroin and other heroin-like drugs; Brown, boy, dope, “H,” harry, hop, horse, it, junk, smack.

Inhalants. Nitrous oxide, “whippets,” rush/poppers (amyl nitrite), spray paint, cleaning fluids, glue, snappers, paint thinner, room fresheners, and gasoline.

Marijuana. Marijuana, pot, grass, reefer, mary jane, dope, weed, blunt, joint, bong.

Methamphetamines. Any kind of nonprescription “speed,” such as “crystal meth,” “crank,” “ice,” “glass,” or “pink.”

Other stimulants. Amphetamines or speed-like stimulants, including Ritalin if not prescribed, uppers, go pills, used for nonmedical reasons.

Sedatives. Sedatives and sleeping pills, such as, amytal, barbiturates, Quaaludes, Tuinal, Seconal, Nembutal, cough syrup, Benzodiazepines, Dalmane, Halcion, Librium, downers, “bennies,” used for nonmedical reasons.

Tranquilizers. Valium, Xanax, Ativan, Mandrax, or Serax, used for nonmedical reasons.

Illinois Division of Alcoholism & Substance Abuse


¿¿¿ APPENDIX C

Definitions of Illinois Sample Areas

Sample Area Counties/tracts

1 Chicago All Census tracts less than 7702 and 7705–7709, 8104, 8106, 8116, 8117.01, 8209.02, 8233.04

2 Suburban Cook County Census tracts greater or equal to 7702 (with the exception of those mentioned above).

3 Northwest Illinois Counties: Boone, Carroll, DeKalb, Jo Daviess, Lee, Ogle, Stephenson, Whiteside, Winnebago.

4 South collar counties Counties: Grundy, Kankakee, Will.

5 North collar counties Counties: DuPage, Kane, Kendall, Lake, McHenry.

6 West central Illinois Counties: Adams, Brown, Bureau, Calhoun, Cass, Christian, Fulton, Greene, Hancock, Henderson, Henry, Jersey, Knox, LaSalle, Logan, Macoupin, Marshall, Mason, McDonough, Menard, Mercer, Montgomery, Morgan, Peoria, Pike, Putnam, Rock Island, Sangamon, Schuyler, Scott, Stark, Tazewell, Warren, Woodford.

7 East central Illinois Counties: Champaign, Clark, Coles, Cumberland, DeWitt, Douglas, Edgar, Ford, Iroquois, Livingston, Macon, McLean, Moultrie, Piatt, Shelby Vermillion.

8 Southern Illinois Counties: Alexander, Bond, Clay, Clinton, Crawford, Edwards, Effingham, Fayette, Franklin, Gallatin, Hamilton, Hardin, Jackson, Jasper, Jefferson, Johnson, Lawrence, Madison, Marion, Massac, Monroe, Perry, Pope, Pulaski, Randolph, Richland, Saline, St. Clair, Union, Wabash, Washington, Wayne, White, Williamson.

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003 D-1

¿¿¿ APPENDIX D

Advance Letter Sent to Sampled Households

D-2 Illinois Division of Alcoholism & Substance Abuse

SURVEY RESEARCH LABORATORY

Dear Resident, We are writing to ask your help in a study of Illinois residents regarding alcohol and other drug use. This study, sponsored by the Illinois Department of Human Services and the University of Illinois at Chicago, is part of an effort to better understand the alcohol and drug treatment needs in the state of Illinois. For this study, we will be conducting telephone interviews with Illinois residents age 16 and older to ask about experiences with alcohol and other substances. Your household has been randomly selected to take part in this project. You should know that the University of Illinois at Chicago Survey Research Laboratory is conducting this study for research purposes only. We do not want to sell you anything. In the next couple of weeks, one of our professional interviewers will call your household to select one person who is eligible to participate and request an interview. He or she will ask about that person’s experiences with alcohol and other drugs, and will also be prepared to answer any questions you may have about the project. The interview is voluntary and completely confidential. Your household’s participation in this study is important. By completing the interview, the selected person will be helping the Department of Human Services better understand these important issues. Your help would be greatly appreciated. If you have any questions concerning this study, you may contact the Project Coordinator, David Schipani at (312) 996-6129. Thank you in advance for your assistance. Sincerely, Lillian Pickup Timothy Johnson, Ph.D. Administrator of Planning and Program Analysis Director Illinois Department of Human Services Survey Research Laboratory

CHICAGO OFFICE 412 South Peoria, Sixth Floor, M/C 336

Chicago, Illinois 60607-7069 (312) 996-5300 Fax: (312) 996-3358

E-mail: [email protected] http://www.srl.uic.edu

URBANA-CHAMPAIGN OFFICE 505 East Green, Suite 3, MC-442 Champaign, Illinois 61820-5723

(217) 333-4273 Fax: (217) 244-4408 E-mail: [email protected] http://www.srl.uic.edu


¿¿¿ APPENDIX E

Survey Limitations I: Sampling Errors Linda Owens

Sampling error is a result of calculating a

statistic based on a sample of the population rather the entire population. It is not an indicator of flaws in the sample design but is a measure of variation. The final sample of 4,155 Illinois households in this study is only one of an astronomical number of samples that could have been drawn from a population of approximately 4.5 million households. Each sample of 4,155 households that could be drawn would provide an estimate of a statistic (e.g., percent of Illinois residents who used marijuana in the last 12 months). The sampling error (or standard error) is the amount of variation in the statistic among all potential samples. In its simplest form, it is equal to the sample standard deviation divided by the square root of the sample size. The standard error varies directly with the sample size—the larger the sample size, the smaller the standard error. In complex sample designs that use clustering or disproportionate stratification, the standard error will increase relative to that of a simple random sample.

Many basic statistical software packages assume simple random sampling when calculating statistics. As a consequence, the estimates of standard errors used in significance tests are too low, resulting in researchers overstating the significance of their results. Analyses of data collected with a

complex sample design should be conducted with software packages that can incorporate the sample design into the analysis (e.g., STATA, SUDAAN).

This study utilizes a disproportionate stratified sample. Therefore, a software package that adjusts for sample design was used to analyze the data. STATA adjusts for the impact of sample design as well as the use of sample and poststratification weights.

There is no one single standard error for a survey. It varies from item to item. For example, if one variable is measured as a proportion with 50% of the sample answering yes and 50% answering no, it will have a different standard error from a variable in which 80% of the sample answers yes and 20% answer no. Many tables in this report provide standard errors for the specific items being reported. These standard errors can be used to compute 95% confidence intervals for each estimate provided in this report. To do so, a standard error should be multiplied by 1.96 to approximate the confidence interval. For example, in Table 3, it is estimated that 19.8% of all persons in Illinois age 16 and older have smoked cigarettes during the past month. The margin of error for this estimate is +/- 2.4, suggesting that the actual proportion having smoked cigarettes during the past month was between 17.4% and 22.2%.

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003 F-1

¿¿¿ APPENDIX F

Survey Limitations II: Coverage Errors Linda Owens

Undercoverage refers to the use of a sampling frame that does not include all members of the population of interest. In this survey, we are interested in the substance use and treatment need of Illinois residents age 16 and older. However, because we conducted the study by telephone, residents who live in households without phones are not included in the sampling frame and therefore are not represented by the sample. As a result of this undercoverage, it is possible that substance use and treatment need estimates are biased, most likely in the direction of underreporting.

According to the November 2003 Current Population Survey (CPS), approximately 8.3% of Illinois households had no telephone service (Belinfante, 2004). The conventional wisdom regarding such households is that the vast majority of them have intermittent telephone service rather than no service at all (Brick, Waksberg, & Keeter, 1996). In other words, they have telephone service when they can afford it, but it is shut off when the bills become unmanageable. Unfortunately, no data exist regarding the amount of time households spend without telephone service.

Although the CPS data suggest that 8.3% of Illinois households had no phone service in 2000, 6.9% of respondents to this survey had interrupted service at some point during the 12-month period prior to the interview. This suggests that the survey missed some of the nontelephone households. These are likely the households that never had phone service. However, because the data are from two

different years, it is difficult to determine exactly the degree of undercoverage of this survey.

If undercoverage is random—meaning population members omitted from the sampling frame are no different from those included—then it is not a problem. However, when those omitted differ from those included in ways related to the primary outcome variables, then undercoverage leads to bias in the outcome estimates. In this survey, our primary interest is in the degree to which households without phone service differ from households with uninterrupted service in regard to substance use and related treatment need.

Given that households without phone service are similar to those with interrupted service (Keeter, 1995), one way to assess the degree of bias in telephone surveys is to look at differences between households with interruptions and those with continuous service. Table F-1 shows the relationship between several substance use and treatment need estimates and phone service interruption, both before and after weighting to census data.

Prior to weighting, interrupted telephone service is significantly associated with all outcome variables except past-year alcohol treatment need. In all instances, a higher proportion of respondents with interrupted telephone service reported substance use and treatment need compared to those with continuous service. This suggests that those households that the survey missed also have

F-2 Illinois Division of Alcoholism & Substance Abuse

higher rates of substance use and treatment need.

Weighting the data to census totals of certain demographic characteristics can correct for some undercoverage because the census data contain all households, not just those with telephones. However, it does not completely solve the problem because the demographic data may or may not be related to telephone service. The weights may overadjust some variables and underadjust others (Brick, Burke, & West, 1992; Massey & Botman, 1988).

In this survey, the poststratification weights are based on region, race, age, and gender. Table F-1 shows that after weighting for these characteristics, only one significant association between phone service and the outcome measures remains—past-year cigarette use. Because nonphone households are similar to households with interrupted service, the remaining significant associations point to areas where bias remains particularly problematic after weighting. The general patterns revealed in the weighted estimates presented in Table F-1 suggest that the undercoverage of nontelephone households in Illinois results in small but mostly nonsignificant underestimates of substance use and treatment need indicators in this survey. Investigations of the effects of coverage error on national estimates of substance use behaviors have reached similar conclusions (McAuliffe, LaBrie, Woodworth, & Zhang, 2002).

REFERENCES

Belinfante, A. (2004). Telephone subscribership in the United States. Retrieved June 29, 2004, from the Federal Communications Commission Web site: http://www.fcc.gov/Bureaus/ Common_Carrier/Reports/FCC-State_Link/ IAD/subs0703.pdf.

Brick, J., Burke, J., & West, J. (1992). Telephone undercoverage bias of 14- to 21-year-olds and 3- to 5-year–olds (National Household Education Survey Technical Report No. 2, NCES 92-101). Washington, DC: U.S. Department of Education, Office of Educational Research and Improvement.

Brick, J. M., Waksberg, J. & Keeter, S. (1996). Using data on interruptions in telephone service as coverage adjustments. Survey Methodology, 22, 185–197.

Keeter, S. (1995). Estimating noncoverage bias from a phone survey. Public Opinion Quarterly, 59, 196–217.

Massey, J., & Botman, S. (1988). Weighting adjustments for random digit dialed surveys. In R. M. Groves, P. P. Biemer, L. E. Lyberg, J. T. Massey, W. L. Nicholls, II, & J. Waksberg (Eds.), Telephone survey methodology (pp. 143–160). New York: Wiley.

McAuliffe, W. E., LaBrie, R., Woodworth, R., & Zhang, C. (2002). Estimates of potential bias in telephone substance abuse surveys due to exclusion of households without telephones. Journal of Drug Issues, 32, 1139–1154.

.

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003 F-3

Table F -1. Past-Year Treatment Need and Substance Use, by Interrupted Telephone Service in the Past Year, Unweighted and Weighted

UNWEIGHTED WEIGHTED Interrupted service in past year Interrupted service in past year

TOTAL Yes No TOTAL Yes No

TOTAL N 3,974 230 3,744 3,961 275 3,686

Past Year-Treatment Need Alcohol 16.2% 20.4% 15.9% 15.9% 17.8% 15.7% Drug 3.1% 6.1%* 2.9% 3.1% 3.3% 3.1% Alcohol or drug 17.1% 23.9%* 16.7% 16.7% 19.6% 16.5%

Past -Year Substance Use Cigarettes 22.4% 35.7%* 21.6% 22.5% 30.4%* 21.9% Alcohol 69.6% 75.7%* 69.2% 70.2% 74.6% 69.8% Any illicit drug 6.6% 12.2%* 6.3% 6.6% 6.5% 6.6% Any psychoactive drug 2.3% 5.2%* 2.2% 2.5% 4.0% 2.4%

*p<05.

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003 G-1

¿¿¿ APPENDIX G

Survey Limitations III: Nonresponse Errors Young Ik Cho and Timothy P. Johnson

Background

Potential survey errors arising from failure to interview all sampled persons is known as nonresponse error. Response rates to social and behavioral surveys have been declining for several decades (de Leeuw & de Heer, 2002; Groves & Couper, 1998; Steeh, 1981), increasing the likelihood that differences between respondents and nonrespondents may be sufficient to bias survey estimates (Caetano, 2001). A small body of research has investigated the potential effects of nonresponse on the reporting of alcohol and other drug use. Several methods have been applied, including (1) the use of nonresponse surveys that attempt to conduct follow-up interviews with persons not responding to the initial survey, (2) panel attrition studies that compare the baseline characteristics of respondents who are and are not lost to follow-up waves of interviewing, (3) comparisons of early vs. late responders to survey requests (under the assumption that late responders are more similar to nonresponders), and (4) records match analyses that compare the personal and/or community characteristics of households that do and do not respond to survey requests. Findings from the available research are summarized below.

Follow-up surveys of nonrespondents have produced mixed findings, with some indicating less substance use among nonrespondents (Hill & Roberts, 1997; Lemmens, Tan, & Knibbe, 1988; Mulford &

Miller, 1959; Lahaut, Jansen, van de Mheen, & Garretsen, 2002) and some indicating no relationship between substance use and survey nonresponse (Casper, 1992; Crawford, 1986; Gmel, 2000). These findings can be interpreted as evidence that persons with fewer substance use experiences, for whom the topic of the survey may be of less interest or saliency, may be less likely or willing to participate. There are several important limitations of nonresponse surveys, however, that may call into question the quality of these findings and account for their empirical discrepancies. For example, nonresponse surveys typically have low response rates; thus it is unclear whether nonresponse survey respondents are actually representative of nonresponders. Available evidence from panel attrition studies (reviewed below) strongly suggests that substance users may in general be less available. If substance users are less available for participation in both surveys and nonresponse follow-ups, the studies reviewed here may be vulnerable to correlation bias limitations. That is, substance users may be more likely to be excluded from both surveys and follow-up surveys, thus calling into question estimates obtained from each. In addition, differences in the mode of data collection between main and follow-up interviews are common (Gmel, 2000; Hill & Roberts, 1997; Lemmens et al., 1988; Lahaut et al., 2002), and this may confound nonresponse with substance use measurement quality.

Panel attrition studies are less vulnerable to these limitations, as common baseline data

G-2 Illinois Division of Alcoholism & Substance Abuse

are available for all respondents and nonrespondents to follow-up interview requests. In contrast to the findings of nonresponse surveys reported above, eight of nine of the available studies find evidence that substance use is positively associated with nonresponse to follow-up interviews (Bucholz et al., 1996; Caetano, Ramisetty-Mikler, & McGrath, 2003; Cottler, Zipp, Robins, & Spitznagel, 1987; Johnson, 1988; McAlister & Gordon, 1987; Plant, Chick, & Kreitman, 1980; Snow, Tebes, & Arthur, 1992; Wild, Cunningham, & Adlaf, 2001). One other study, though, found no differences in substance use behavior of those who did and those who did not complete a follow-up survey (Iversen & Klausen, 1986). An important limitation of these studies, however, is that they mostly fail to differentiate between respondents who refuse and those who cannot be contacted, making it difficult to determine if substance use is associated with availability or with willingness to participate. One exception is a study reported by Eaton and colleagues (1992), which found drug and alcohol abuse/dependence to be associated with failure to contact for follow-up interview but not refusal to participate once contacted.

Some nonresponse research has been conducted on the assumption that difficult-to-reach respondents are most similar to nonrespondents. Two studies that have examined these difficult-to-reach respondents have found them to consume more alcohol than the less difficult-to-reach (Crawford, 1987; Wilson, 1981). Cottler et al. (1987) found similar evidence. They reported that persons with current alcohol disorders required the greatest number of contact attempts, on average, in the St. Louis Epidemiologic Catchment Area study. Two other studies have found no differences in alcohol consumption among early vs. late respondents (Etter &

Perneger, 1997; Voigt, Koepsell, & Daling, 2003).

Finally, records match analyses are another approach to investigating survey nonresponse (Groves & Couper, 1998). There are two examples available of applications of this technique to assessment of nonresponse bias in substance use research. Needle, McCubbin, and Lorence (1985) examined the service utilization records of a health maintenance organization and found no differences between households that did and households that did not respond to a survey in terms of alcohol and drug treatment service utilization. More recently, Gfroerer, Lessler, and Parsley (1997) analyzed the nonresponse patterns of 5,030 households sampled to participate in the 1990 National Household Survey on Drug Abuse (NHSDA). Household and community-level information from the 1990 decennial census for each sampled household was appended to the data set. Overall, they found that males and Hispanics were less likely to participate in the NHSDA. At the household level, owner-occupied households, those with greater value, and single-person households consisting of persons age 65 and older were less likely to respond. At the block level, urbanicity was associated with lower participation. Several interviewer characteristics were also predictive of survey nonresponse. However, the degree to which these measures were associated with the substance use indictors being estimated by the survey was not reported. The key advantage of this approach is that information regarding all sampled households is available for analysis, providing researchers with the opportunity to make better generalizations from these data. We note, however, that this approach has not been employed to evaluate nonresponse bias in telephone surveys of substance use.


Analysis

To examine the potential effects of nonresponse error in this study, indicators of survey response, refusal, and noncontact w ere first constructed using the final sample dispositions assigned to each of the 20,774 telephone numbers included in the survey’s sample frame. The final dispositions for each of these numbers are presented in Appendix A, Table A-5. Several nonresponse measures were constructed from these data. Two general indicators of response were used to contrast those households completing telephone interviews with (1) all households identified as eligible but that did not participate, regardless of reason, and (2) all households identified as eligible plus all unscreened telephone numbers that likely included some eligible households. Additional measures were developed to contrast (1) households that refused with those that participated, and (2) households in which the randomly selected respondent could be directly contacted and those in which the respondent could not.

The specific disposition codes (from Exhibit C in Appendix A) employed to develop each of these measures are as follows. For the first general nonresponse indicator, households completing interviews included disposition codes 01–08, and identified households not completing interviews included codes 31–60. The second general nonresponse indicator was identical to the first with the exception of also including among the nonresponding cases those telephone numbers coded as having never been answered (disposition code 30). For the refusal indicator, households classified as refusals included those with disposition codes 40–48. These were contrasted with nonrefusal households (codes 01–08, 31–33). The noncontact indicator used disposition codes

31–33, 40, 47, 55, and 56 to identify households where the selected respondent was never directly reached. Codes 01-08, 41, 42, 48, and 60 represented households in which the selected respondent was directly contacted. Households not speaking English or Spanish (disposition code 88) were excluded from these analyses.

Data from several other sources were next merged with this sample frame file for analysis. First, survey responses for key outcome measures of interest were appended to those phone numbers that yielded completed interviews. Three dependent variables of interest were examined: current alcohol use treatment need, current drug use treatment need, and current treatment need for either alcohol or drug use.

Second, each telephone exchange/area code combination included in the survey’s sample frame was linked with the predominant area code that it served. Next, a set of approximately 70 measures from the 2000 Census (American Fact Finder [http://factfinder.census.gov/home/saff/ main.html?_lang=en] were aggregated at the ZIP code level and merged with the sample frame information. The census variables included a variety of housing, income, occupation, and sociodemographic dimensions that collectively represented broad social conditions within each local geographic area.

Using this merged data file, an exploratory ecological factor analysis was conducted to determine if a smaller set of underlying dimensions could be identified to represent these census data. Results presented in Table G-1 provide a summary of this factor analysis. Specifically, this table provides loadings for ten orthogonal (i.e., uncorrelated) factors identified from these data, along with


summary descriptive labels that were assigned to each.

Hierarchical linear modeling (HLM) next was used to examine potential associations between each of the ten ecological factors and survey nonresponse, survey refusal, and survey noncontact (Raudenbush, Bryk, Cheong, & Congdon, 2000). Since the outcomes are all binary (coded 0 for “no” and 1 for “yes”), the first-level equations for these outcomes use a logit link function (Hedeker & Gibbons, 1994) to estimate the log-odds of occurring “yes.” State geographic region was included as a covariate in each of these analyses. In Table G-2, two equations are presented, one examining the ZIP-code-level predictors of survey nonresponse excluding phone numbers with unknown eligibility (i.e., the “ring, no answer” numbers; equation 1 in Table G-2), and the second examining the same predictors of survey nonresponse including phone numbers with unknown eligibility (equation 2). Three factors were found to be associated with likelihood of survey response (excluding unknown eligibility cases): community income, community poverty, and average housing size. As the first column of Table G-2 indicates, survey response was lower in areas with higher incomes, with greater levels of poverty, and with more crowded housing conditions. The second equation examined in Table G-2, which includes unknown eligibility cases, also identified high-income areas and areas with more crowded housing conditions as having lower survey response rates. Those in areas with more white-collar employees also were less likely to participate in the survey. We also note that the average log-odds (estimated probability) of outcomes (=1) did not vary significantly between ZIP code areas, as indicated at the bottom part of this and subsequent tables.

Analyses of ecological predictors of refusals and noncontacts are presented in Table G-3. The first equation presented in this table examines the predictors of survey refusal among households determined to be eligible. None of the ten ecological measures examined were found to be associated with likelihood of refusing to participate. Equation 2 in Table G -3 examines the predictors of survey noncontact. Three of the factors examined were found to be independently associated with likelihood of noncontact. Specifically, households in areas with more immigrants, more crowded housing conditions, and those in more urban areas were all less likely to have been successfully contacted as part of the survey.

Next, a series of HLM models were estimated to examine potential associations between the ecological measures and the alcohol and other drug use treatment need of survey respondents, controlling for geographic region of the state. Table G-4 presents these findings. The first equation in Table G-4 indicates that respondents living in higher income areas were more likely to report alcohol treatment need. In contrast, persons living in high poverty areas were less likely to report alcohol treatment need. The second equation in Table G-4 indicates that only one ecological measure was associated with drug use treatment need. Respondents in areas with higher proportions of government workers were more likely to report drug use treatment need. Finally, equation 3 indicates that the likelihood of reporting treatment need for either alcohol or drug use was greater among respondents living in higher income areas and lower among respondents living in higher poverty areas.

The findings reported in Tables G-2 through G-4 suggest that two community measures are associated with both likelihood of survey nonresponse and respondent


treatment need for alcohol and either alcohol or drug use. The direction of these relationships indicates that persons in higher income areas are less likely to be included in the survey and more likely to have treatment need. In addition, persons in high poverty areas appear less likely to have responded to the survey but to be less likely to have alcohol and alcohol or drug treatment need. The findings regarding community income levels suggests that survey treatment need estimates may be underestimated as a consequence of nonresponse bias. With regards to community poverty levels, an opposite conclusion may be drawn: Because households in higher poverty communities are less likely to respond and because they also have lower rates of alcohol and alcohol/drug treatment need, treatment need estimates may be somewhat overestimated.

There are a number of strong and well known sociodemographic correlates of substance use that were included in the survey’s poststratification weights that have yet to be examined, however, including gender, age, and race/ethnicity. We next re-estimated the models presented in Table G-4 to determine if community income level and poverty level remain associated with treatment need once these measures are introduced as controls. Indeed, findings presented in Table G-5 indicate that community income level is not predictive of alcohol and alcohol or other drug treatment need once these sociodemographic characteristics are introduced into the models. For both of these treatment need measures, gender, age, and race/ethnicity each were found to be significantly associated with treatment need.

Based on this evidence, we conclude that nonresponse bias is unlikely to adversely affect the quality of the survey data presented

in this report. This conclusion is based on the findings that the one variable predictive of both nonresponse and substance use treatment need—community income levels—does not remain predictive of treatment need once demographic measures included in the study’s poststratification weights are added to the analyses presented in this appendix.

References Bucholz, K. K., Shayka, J. J., Marion, S. L.,

Lewis, C. E., Pribor, E. F., & Rubio, D. M. (1996). Is a history of alcohol problems or of psychiatric disorder associated with attrition at 11-year follow-up? Annals of Epidemiology, 6, 228–234.

Casper, R. A. (1992). Follow-up of nonrespondents in 1990. In C. F. Turner, J. T. Lessler, & J. C. Gfroerer (Eds.), Survey measurement of drug use: Methodological studies (pp. 155–173). Rockville, MD: National Institute on Drug Abuse.

Caetano, R. (2001). Non-response in alcohol and drug surveys: A research topic in need of further attention. Addiction, 96, 1541–1545.

Caetano, R., Ramisetty-Mikler, S., & McGrath, C. (2003). Characteristics of non-respondents in a U.S. national longitudinal survey on drinking and intimate partner violence. Addiction, 98, 791–797.

Cottler, L. B., Zipp, J. F., Robins, L. N., & Spitznagel, E. L. (1987). Difficult-to-recruit respondents and their effect on prevalence estimates in an epidemiological survey. American Journal of Epidemiology, 125, 3–36.

Crawford, A. (1986). A comparison of participants and non-participants from a British general population survey of alcohol drinking practices. Journal of the Market Research Society, 28, 291–297.

Crawford, A. (1987). Bias in a survey of drinking habits. Alcohol and Alcoholism, 22, 167–179.

de Leeuw, E., & de Heer, W. (2002). Trends in household survey nonresponse: A longitudinal and international comparison. In R. M. Groves, D. A. Dillman, J. L. Eltinge,


& R. J. A. Little (Eds.), Survey nonresponse (pp. 41–54). New York: Wiley.

Eaton, W. W., Anthony, J. C., Tepper, S., & Dryman, A. (1992). Psychopathology and attrition in the Epidemiologic Catchment Area Surveys. American Journal of Epidemiology, 135, 1051–1059.

Etter, J. F., & Perneger, T. V. (1997). Analysis of non-response bias in a mailed health survey. Journal of Clinical Epidemiology, 50, 1123–1128.

Gfroerer, J., Lessler, J., & Parsley, T. (1997). Studies of nonresponse and measurement error in the National Household Survey on Drug Abuse. In L. Harrison & A. Hughes (Eds.), The validity of self-reported drug use: Improving the accuracy of survey estimates (pp. 273–295). Rockville, MD: National Institute on Drug Abuse.

Gmel, G. (2000). The effect of mode of data collection and of non-response on reported alcohol consumption: A split-sample study in Switzerland. Addiction, 95, 123–134.

Groves, R. M., & Couper, M. P. (1998). Nonresponse in household interview surveys. New York: Wiley.

Hedeker, D., & Gibbons, R. (1994) A random-effects ordinal regression model for multilevel analysis. Biometrics, 50, 993–994.

Hill, A., & Roberts, J. (1997). Non-response bias in a lifestyle survey. Journal of Public Health in Medicine, 19, 203–207.

Iversen, L., & Klausen, H. (1986). Alcohol consumption among laid-off workers before and after closure of a Danish ship-yard: A 2-year follow-up study. Social Science and Medicine, 22, 107–109.

Johnson, T. P. (1988). The social environment and health. Unpublished Ph.D. dissertation. Lexington: University of Kentucky.

Lahaut, V. M. H. C. J., Jansen, H. A. M., van de Mheen, H., & Garretsen H. F. L. (2002). Non-response bias in a sample survey on alcohol consumption. Alcohol and Alcoholism, 37, 256–260.

Lemmens, P. H. H. M., Tan, E. S., & Knibbe, R. A. (1988). Bias due to non-response in a

Dutch survey on alcohol consumption. British Journal of Addiction, 83, 1069–1077.

McAlister, A., & Gordon, N. P. (1987). Attrition bias in a cohort study of substance abuse onset and prevention. Evaluation Review, 10, 853–859.

Mulford, H. A., & Miller, D. E. (1959). Drinking in Iowa, I. Socio-cultural distribution of drinkers: With a methodological model for sampling, evaluation, and interpretation of findings. Quarterly Journal of Studies on Alcohol, 20, 634–726.

Needle, R., McCubbin, H., & Lorence, J. (1985). A test of non-respondent bias in a family-based study: A research note. International Journal of the Addictions, 20, 763–769.

Plant, M. A., Chick, J., & Kreitman, N. (1980). The effects of response rates on level of self-reported alcohol consumption and alcohol-related problems: Conclusions from a Scottish study. British Journal on Alcohol and Alcoholism, 15, 158–163.

Raudenbush, S. W., Bryk, A. S., Cheong, Y. F., & Congdon, R. T., Jr. (2000). HLM 5: Hierarchical linear and nonlinear modeling (Statistical software manual). Lincolnwood, IL: Scientific Software International.

Snow, D. L., Tebes, J. K., & Arthur, M. W. (1992). Panel attrition and external validity in adolescent substance use research. Journal of Consulting and Clinical Psychology, 60, 804–807.

Steeh, C. G. (1981). Trends in nonresponse rates 1952–1979. Public Opinion Quarterly, 45, 40–57.

Voigt, L. F., Koepsell, T. D., & Daling, J.R. (2003). Characteristics of telephone survey respondents according to willingness to participate. American Journal of Epidemiology, 157, 66–73.

Wild, T. C., Cunningham, J., & Adlaf, E. (2001). Nonresponse in a follow-up to a representative telephone survey of adult drinkers. Journal of Studies on Alcohol, 62, 257–261.

Wilson, P. (1981). Improving the methodology of drinking surveys. The Statistician, 30, 159–167.


Table G-1. Ecological Factor Analysis of Census Indicators at ZIP Code Level

Loadings Factor 1: High Income Areas

Median income in 1999 (dollars): households 0.854 Median income in 1999 (dollars): families 0.901 Per capita income in 1999 (dollars) 0.917 Median earnings in 1999 of full-time, year-round workers (dollars): male 0.891 Median earnings in 1999 of full-time, year-round workers (dollars): female 0.802

Factor 2: High Poverty Areas Percent with no vehicles available 0.596 Percent with no telephone service available 0.573 Income in 1999 below poverty level: percent of population for whom poverty status is determined, all ages 0.883 Income in 1999 below poverty level: percent of population for whom poverty status is determined, related children under 18 years

0.849

Income in 1999 below poverty level: percent of population for whom poverty status is determined, age 65+ 0.660 Income in 1999 below poverty level: percent of families 0.884

Factor 3: Immigrant Populations Total population: percent foreign born 0.900 Foreign-born population: number 0.785 Population 5 years and over: percent who speak a language other than English at home, total 0.918 Population 5 years and over: percent who speak a language other than English at home and speak English less than "very well" 0.940

Factor 4: Smaller Housing Unit s Median rooms -0.755 Percent 1 unit, detached or attached -0.716 Percent in buildings with 10 or more units -0.759 Percent with 3 or more bedrooms -0.796

Factor 5: Urbanicity Total housing units 0.833 Occupied housing units 0.830 Total population, number 0.822

Factor 6: Working Families Population 16 years and over: percent in labor force, female, with own children under 6 years 0.943 Own children: percent with all parents in family in labor force, under 6 years 0.941 Own children: percent with all parents in family in labor force, 6 to 17 years 0.526

Factor 7: Poor Housing Percent lacking complete plumbing facilities 0.944 Percent lacking complete kitchen facilities 0.928

Factor 8: Government Employees Percent distribution by occupation: production, transportation, and material moving occupations -0.695 Percent in selected industries: manufacturing -0.876 Percent government workers (local, state, or federal) 0.739

Factor 9: White Collar Employees Percent distribution by occupation: sales and office occupations 0.503 Percent distribution by occupation: farming, fishing, and forestry occupations -0.853 Percent in selected industries: agriculture, forestry, fishing, and hunting -0.786

Factor 10: Newer Housing Units Percent: structure built 1990 to March 2000 0.734 Percent in which householder moved into unit 1999 to March 2000 0.593 Population 5 years and over: percent who lived in different house in 1995 and lived in same state 0.774


Table G-2. Hierarchical Model of the Regional Effects on the Likelihood of Survey Response

SURVEY RESPONSE 1 (1=yes) SURVEY RESPONSE 2 (1=yes)

Level 1 N: 10,997 13,102 Level 2 N: 779 796

Fixed Effects Coefficient (S.E.) T -ratio Coefficient (S.E.) T -ratio

Intercept -0.441 (0.064) -6.911*** -0.640 (0.062) -10.267***

ZIP code level factors Factor 1: High income areas -0.059 (0.027) -2.157* -0.095 (0.027) -3.568** Factor 2: High poverty areas -0.047 (0.020) -2.347* -0.029 (0.020) -1.429 Factor 3: Immigrant populations -0.026 (0.022) -1.191 -0.039 (0.021) -1.804 Factor 4: Smaller housing units -0.042 (0.020) -2.155* -0.076 (0.019) -3.937*** Factor 5: Urbanicity -0.032 (0.019) -1.703 -0.018 (0.018) -0.984 Factor 6: Working families 0.083 (0.044) 1.882 0.075 (0.043) 1.739 Factor 7: Poor housing 0.086 (0.069) 1.246 0.060 (0.064) 0.940 Factor 8: Government employees 0.011 (0.029) 0.393 0.016 (0.028) 0.583 Factor 9: White collar employees -0.066 (0.046) -1.426 -0.110 (0.045) -2.447* Factor 10: New housing -0.023 (0.021) -1.079 -0.032 (0.021) -1.544

Individual-level covariates Sample areas (Ref: southern Illinois)

Chicago -0.126 (0.129) -0.977 -0.173 (0.126) -1.371 Suburban Cook County -0.125 (0.109) -1.145 -0.188 (0.107) -1.751 Northwest Illinois 0.201 (0.095) 2.113* 0.160 (0.093) 1.720 South collar counties -0.144 (0.097) -1.482 -0.152 (0.096) -1.579 North collar counties -0.103 (0.109) -0.948 -0.202 (0.107) -1.893 West central Illinois 0.269 (0.088) 3.040** 0.218 (0.086) 2.540* East central Illinois 0.367 (0.090) 4.085*** 0.349 (0.088) 3.977***

Random Effects Variance Component

(df) χ2 Variance Component

(df) χ2

Average log-odds of dependent variable between ZIP code sample areas 0.008 (768) 771.352 0.013 (785) 807.859

*p<.05; **p<.01; ***p<.001. Note: Survey Response 1 excludes unknown eligibility cases; Survey Response 2 includes unknown eligibility cases.


Table G -3. Hierarchical Model of the Regional Effects on the Likelihood of Survey Refusal and Non-contactable

REFUSALS (1=yes) NONCONTACTS (1=yes)

Level 1 N: 10,753 10,997 Level 2 N: 771 779

Fixed Effects Coefficient (S.E.) T -ratio Coefficient (S.E.) T -ratio

Intercept -0.319 (0.063) -5.094*** -0.413 (0.065) -6.402***

ZIP code level factors Factor 1: High income areas 0.022 (0.026) 0.846 0.077 (0.027) 2.887 Factor 2: High poverty areas 0.023 (0.019) 1.202 0.014 (0.020) 0.694 Factor 3: Immigrant populations -0.024 (0.021) -1.151 0.063 (0.022) 2.895** Factor 4: Smaller housing units -0.003 (0.018) -0.148 0.065 (0.020) 3.275** Factor 5: Urbanicity 0.003 (0.018) 0.143 0.039 (0.019) 2.067* Factor 6: Working families -0.077 (0.043) -1.786 -0.066 (0.045) -1.483 Factor 7: Poor housing -0.089 (0.069) -1.283 -0.127 (0.071) -1.784 Factor 8: Government employees 0.005 (0.028) 0.166 -0.005 (0.030) -0.162 Factor 9: White collar employees 0.020 (0.046) 0.438 0.091 (0.048) 1.890

Factor 10: Newer housing units -0.014 (0.020) -0.704 0.031 (0.021) 1.431

Individual-level covariates Sample area (Ref: southern Illinois)

Chicago -0.210 (0.122) -1.724 0.304 (0.128) 2.381* Suburban Cook County -0.136 (0.103) -1.320 0.196 (0.108) 1.823 Northwest Illinois -0.131 (0.092) -1.417 -0.095 (0.097) -0.978 South collar counties -0.031 (0.091) -0.338 0.228 (0.097) 2.345* North collar counties -0.092 (0.103) -0.895 0.153 (0.107) 1.428 West central Illinois -0.272 (0.089) -3.048** -0.018 (0.090) -0.198 East central Illinois -0.276 (0.089) -3.088** -0.274 (0.094) -2.933**

Random Effects Variance Component


(df) χ2

Average log-odds of dependent variable between ZIP code sample areas 0.000 (760) 724.190 0.014 (768) 800.967

*p<.05; **p<.01; ***p<.001.


Table G -4. Hierarchical Model of the Regional Effects on the Likelihood of Being Diagnosed for Substance Abuse Treatment (Level 1 N=4,038; Level 2 N=627)

ALCOHOL PROBLEM (1=yes) DRUG PROBLEM (1=yes) ALCOHOL OR DRUG PROBLEM (1=yes) FIXED EFFECTS Coefficient (S.E.) T -ratio Coefficient (S.E.) T -ratio Coefficient (S.E.) T -ratio

Intercept -1.902 (0.148) -12.866*** -3.942 (0.330) -11.935*** -1.911 0.146 -13.085***

ZIP Code Level Factors Factor 1: High income areas 0.127 (0.063) 2.028* -0.038 (0.138) -0.276 0.119 (0.061) 1.957† Factor 2: High poverty areas -0.100 (0.049) -2.056* 0.027 (0.096) 0.282 -0.089 (0.047) -1.899† Factor 3: Immigrant populations 0.031 (0.050) 0.611 0.035 (0.099) 0.356 0.041 (0.048) 0.862 Factor 4: Smaller housing units 0.033 (0.044) 0.749 0.054 (0.089) 0.604 0.043 (0.042) 1.012 Factor 5: Urbanicity 0.046 (0.041) 1.108 0.145 (0.087) 1.664 0.066 (0.040) 1.654 Factor 6: Working families 0.108 (0.098) 1.093 0.080 (0.211) 0.379 0.117 (0.096) 1.220 Factor 7: Poor housing -0.059 (0.162) -0.366 0.203 (0.292) 0.694 0.001 (0.154) 0.005 Factor 8: Government employees 0.085 (0.063) 1.360 0.334 (0.125) 2.675** 0.099 (0.061) 1.635 Factor 9: White collar employees 0.080 (0.107) 0.754 -0.134 (0.215) -0.620 0.119 (0.106) 1.127 Factor 10: Newer housing units 0.082 (0.046) 1.788 0.070 (0.098) 0.711 0.079 (0.044) 1.785

Individual-Level Covariates Sample area (Ref: southern Illinois)

Chicago 0.245 (0.291) 0.841 0.090 (0.613) 0.148 0.251 (0.281) 0.894 Suburban Cook County -0.097 (0.255) -0.382 0.332 (0.536) 0.620 -0.069 (0.247) -0.279 Northwest Illinois 0.214 (0.210) 1.021 -0.042 (0.521) -0.080 0.270 (0.205) 1.315 South collar counties 0.101 (0.222) 0.454 0.225 (0.502) 0.449 0.154 (0.216) 0.712 North collar counties 0.110 (0.248) 0.443 0.569 (0.530) 1.073 0.200 (0.240) 0.833 West central Illinois 0.104 (0.194) 0.534 0.447 (0.412) 1.085 0.230 (0.188) 1.219 East central Illinois -0.065 (0.201) -0.323 0.289 (0.428) 0.675 -0.019 (0.197) -0.098

RANDOM EFFECTS

Variance Component (df) χ2



Average log-odds of dependent variable between ZIP code sample areas 0.058 (616) 639.406 0.180 (616) 517.829 0.042 (616) 636.637

†p<.06; *p<.05; **p<.01; ***p<.001.


Table G -5. Hierarchical Model of the Regional Effects on the Likelihood of Being Diagnosed for Substance Abuse Treatment (Level 1 N=4,038; Level 2 N=627)

ALCOHOL TREATMENT NEED (1=yes)

DRUG TREATMENT NEED (1=yes)

ALCOHOL OR DRUG TREATMENT NEED (1=yes)

FIXED EFFECTS Coefficient (S.E.) T -ratio Coefficient (S.E.) T -ratio Coefficient (S.E.) T -ratio

Intercept -1.573 (0.178) -8.858*** -2.130 (0.374) -5.695*** -1.489 (0.174) -8.539***

ZIP Code Level Factors

High income areas 0.046 (0.050) 0.925 -0.075 (0.100) -0.745 0.037 (0.049) 0.764

High poverty areas -0.012 (0.049) -0.241 0.050 (0.090) 0.556 -0.010 (0.047) -0.206

Individual Demographic Factors Gender (male) 0.949 (0.093) 10.249*** 0.860 (0.199) 4.321*** 0.961 (0.090) 10.621***

Age -0.214 (0.027) -8.056*** -0.824 (0.089) -9.210*** -0.239 (0.026) -9.126***

Race/Ethnicity African American -1.107 (0.219) -5.046*** -0.187 (0.348) -0.537 -0.917 (0.203) -4.516*** Hispanic -1.021 (0.208) -4.911*** -0.882 (0.392) -2.249* -0.852 (0.193) -4.416*** Other -1.007 (0.285) -3.530** -0.262 (0.489) -0.537 -1.069 (0.285) -3.749***

Sample area (Ref: southern Illinois) Chicago 0.741 (0.215) 3.452** 0.704 (0.419) 1.681 0.773 (0.208) 3.709*** Suburban Cook County 0.162 (0.216) 0.752 0.180 (0.448) 0.402 0.185 (0.211) 0.874 Northwest Illinois 0.210 (0.197) 1.061 -0.429 (0.487) -0.883 0.235 (0.194) 1.211 South collar counties 0.283 (0.205) 1.382 0.092 (0.443) 0.208 0.302 (0.201) 1.506 North collar counties 0.237 (0.217) 1.093 0.331 (0.445) 0.743 0.316 (0.211) 1.499 West central Illinois 0.138 (0.198) 0.696 0.418 (0.410) 1.020 0.257 (0.193) 1.333 East central Illinois -0.002 (0.200) -0.010 0.345 (0.403) 0.856 0.030 (0.197) 0.154

RANDOM EFFECTS Variance Component



(df) χ2

Average log-odds of treatment need between ZIP code sample areas

0.094 (624) 593.356

0.168 (624) 399.089

0.008 (625) 582.377

*p<.05; **p<.01; ***p<.001.

Alcohol, Tobacco, & Other Drug Use in Illinois: Prevalence & Treatment Need, 2003 H-1

¿¿¿ APPENDIX H

Survey Limitations IV: Measurement Error Timothy P. Johnson

A variety of sources of measurement error may influence survey results. For example, question wording, the ordering of questions within the instrument, respondent characteristics, interviewer characteristics, and the mode of data collection may each affect data quality and should be considered when interpreting survey findings. None of these forms of potential measurement error can be definitively eliminated. However, careful pretesting of survey instruments, as was done in this study, can help identify and correct many serious forms of measurement error. In addition, the Survey Research Laboratory requires that survey instruments be reviewed by a Questionnaire Review Committee (QRC) that is specifically responsible for identifying and addressing potential sources of measurement error in survey questionnaires.

Measurement error specific to the collection of survey data concerned with substance use behaviors are of particular concern due to the social desirability of these behaviors and the fact that use of some of these substances is illegal. Available information comes from studies that have investigated the effects of data collection mode and from research that has attempted to validate self-reported substance use behaviors. Although the survey described in this report was not designed to directly evaluate measurement errors, we briefly review here available research that is relevant.

The available evidence suggests that illicit drug use is underreported in telephone surveys, compared to interviews that employ face-to-face and self-administered methodologies. Several studies have reported

lower prevalence estimates of recent drug use in telephone surveys vs. face-to-face interviews with samples of the general population (Gfroerer & Hughes, 1991; Johnson et al., 1989). However, other research has found no differences in estimates of alcohol use between telephone and face-to-face modalities (Greenfield et al., 2000). Comparisons of telephone and self-administered surveys suggest that drug use behaviors also are underreported in telephone interviews (Aquilino, 1994; Aquilino & LoSciuto, 1990; Kraus & Augustin, 2001; Schober, Caces, Pergamit, & Branden, 1992). Although telephone surveys have many advantages (McAuliffe, Geller, LaBrie, Paletz, & Fournier, 1998), available evidence suggests that some underreporting of drug use behaviors can be expected.

A small body of research also has attempted to validate the accuracy of substance use behaviors reported in community surveys. These studies have compared self-reported substance use to toxicological findings from hair, urine, and/or saliva assays. The validity of tobacco use reports has been well established (Fendrich, Johnson, Wislar, Hubbell, & Spiehler, 2004; Patrick et al., 1994). However, drug use reporting is not reported with equal accuracy. In examining results from a high-risk community sample of adults in Chicago, Fendrich and colleagues (1999) reported that estimates of recent cocaine and heroin use derived from hair testing were considerably higher than self-reports obtained from the same respondents. A follow-up study found that higher rates of cocaine and heroin were obtained from drug testing (using hair, saliva and urine), compared to self-reports from a

H-2 Illinois Division of Alcoholism & Substance Abuse

community sample of adults age 18–40 in Chicago (Fendrich et al., 2004). A higher estimate of marijuana use, however, was obtained from self-reports in comparison to drug test assays, a finding attributed to weaknesses in the ability of hair testing to detect marijuana use. Underreporting of cocaine and heroin also have been found in general population surveys conducted by Colón and colleagues (2001; 2002) in Puerto Rico.

Collectively, this research suggests the presence of measurement error in the substance use prevalence estimates presented in this report. Although no direct evidence is available, the consistency and direction of the findings from this literature indicate that it is highly likely that these behaviors have been underreported in the current study. The reported estimates might thus be best interpreted as lower bound estimations of substance use behaviors in Illinois. These errors of measurement are to some extent inevitable, and there are unfortunately no known examples of epidemiologic surveys of substance use that are not subject to these limitations.

REFERENCES

Aquilino, J. S. (1994). Interview mode effects in

surveys of drug and alcohol use. Public Opinion Quarterly, 58, 210–240.

Aquilino, J. S., & LoScuito, L. A. (1990). Effect of interview mode on self-reported drug use. Public Opinion Quarterly, 54, 362–395.

Colón, H. M, Robles, R. R. & Sahai, H. (2001). The validity of drug use responses in a household survey in Puerto Rico: Comparison of survey responses of cocaine and heroin use with hair tests. International Journal of Epidemiology, 30, 1042–1049.

Colón, H. M., Robles, R. R., & Sahai, H. (2002). The validity of drug use reports among hard core drug users in a household survey in

Puerto Rico: Comparison of survey responses of cocaine and heroin use with hair tests. Drug and Alcohol Dependence, 67, 269–279.

Fendrich, M., Johnson, T. P., Sudman, S., Wislar, J., & Spiehler, V. (1999). Validity of drug use reporting in a high-risk community sample: A comparison of cocaine and heroin survey reports with hair tests. American Journal of Epidemiology, 149, 955–962.

Fendrich, M., Johnson, T. P., Wislar, J. S., Hubbell, A., & Spiehler, V. (2004). The utility of drug testing in epidemiological research: Results from a general population survey. Addiction, 99, 197–208.

Gfroerer, J. C., & Hughes, A. L. (1991). The feasibility of collecting drug abuse data by telephone. Public Health Reports, 106, 284–393.

Greenfield, T. K., Midanik, L. T., & Rogers, J. D. (2000). Effects of telephone versus face-to-face interview modes on reports of alcohol consumption. Addiction, 95, 277–284.

Johnson, T. P., Hougland, J., & Clayton, R. (1989). Obtaining reports of sensitive behaviors: A comparison of substance use reports from telephone and face-to-face interviews. Social Science Quarterly, 70, 174–183.

Kraus, L., & Augustin, R. (2001). Measuring alcohol consumption and alcohol-related problems: Comparison of responses from self-administered questionnaires and telephone interviews. Addiction, 96, 459–471.

McAuliffe, W. E., Geller, S., LaBrie, R., Paletz, S., & Fournier, E. (1998). Are telephone surveys suitable for studying substance abuse? Cost, administration, coverage and response rate issues. Journal of Drug Issues, 28, 455–482.

Patrick, D., Cheadle, A., Thompson, D. C., Diehr, P., Koepsell, T., & Kinne, S. (1994). The validity of self-reported smoking: A review and meta-analysis. American Journal of Public Health, 84, 1086–1093.

Schober, S., Caces, M. F., Pergamit, M., & Branden, L. (1992). Effects of mode of administration on reporting of drug use in the National Longitudinal Survey. In C. Turner, J. Lessler, & J. Gfroerer (Eds.), Survey measurement of drug use: Methodological studies (pp. 267–276). Rockville, MD: National Institute on Drug Abuse.

Alcohol, Tobacco, and Other Drug Use in Illinois ... Tobacco, and Other Drug Use in Illinois:...

Documents

Transcript of Alcohol, Tobacco, and Other Drug Use in Illinois ... Tobacco, and Other Drug Use in Illinois:...