1 Sampson & Laub (2004). Criminology, 41 (3): 555-592.

1Sampson & Laub (2004). Criminology, 41 (3): 555-592.

Drug Abuse Epidemiology Datasets

Wilson M. Compton, M.D., M.P.E.Director, Division of Epidemiology, Services and Prevention Research

National Institute on Drug Abuse

16 May 2007

3

Primary ConceptsPrimary Concepts

1.Do the major datasets provide good secondary analysis scientific opportunities?

2.How can NIDA promote use of these resources?

Ultimate Goal is to Enhance the Process of Scientific Discovery

4

1. Cross-Sectional1. Cross-Sectional

5

Changes in Current Smoking Among US Students Before/After Launch of the "truth" Campaign in 2000Grade 1997 2002 Change 1997-1999 2000-2002

All 28.0 18.0 –35.7 –3.2 (–3.8, –2.6) –6.8 (–7.5, –6.1)

8th 19.4 10.7 –44.8 –3.4 (–4.6, –2.1) –9.0 (–10.4, –7.6)

10th 29.8 17.7 –40.6 –4.6 (–5.6, –3.6) –8.7 (–9.8, –7.5)

12th 36.5 26.7 –26.8 –1.8 (–2.7, –1.0) –5.1 (–6.1, –3.9)

Farrelly et al, AJPH 2005;95:425-431

Odds of smoking

Media market exposure (48 states shown) of the national "truth" campaign, 2000–2002

Ave. Cumulative GRP x 10,000

6

0

1

2

3

4

5

Odd Ratio for Reporting Chance to Buy Drugs

Aggression, Delinquency, and Opportunity to Buy Drugs: NHSDA

Rosenberg & Anthony, Drug Alc Dependence 2001;63:245-252

Delinquency

Aggression

Low

Low

Low

High

High

Low

High

High

* P < .01 Compared to Low/Low

Reference Group

*

*

7

10

20

30

40

50

60

70

80

90

Cigarettes Alcohol Marijuana Cocaine Rx

White

Afr-Am

Lifetime Prevalence of Use Among Whites and African Americans 12+ Years Old (NSDUH 2004)

Denise Kandel, NIDA Trajectories Conference 11/13/2006]

Per

cent

age

of L

ifetim

e U

se

***

***

***

*** ***

8

10

20

30

40

50

60

70

80

12-17 18-25 26-34 35-49

White 2004Afr-Am 2004

Current Smoking by Age Among White and African American Lifetime Smokers in 2004 and in the 1973-1978 Birth Cohorts in 1990, 1996 and 2004 (NHSDA/NSDUH )

Per

cent

age

of L

ast M

onth

Use

Age CategoryDenise Kandel, NIDA Trajectories Conference 11/13/2006

9

Current Smoking by Age Among White and African American Lifetime Smokers in 2004 and in the 1973-1978 Birth Cohorts in 1990, 1996 and 2004 (NHSDA/NSDUH )

10

20

30

40

50

60

70

80

12-17 18-25 26-34 35-49

Per

cent

age

of L

ast M

onth

Use

Age Category

White 1973-78 birth cohortAfr-Am 1973-78 birth cohort

1990

1990

1996

1996

20042004

2004

2004

Denise Kandel, NIDA Trajectories Conference 11/13/2006

10

Current Smoking by Age Among White and African American Lifetime Smokers in 2004 and in the 1973-1978 Birth Cohorts in 1990, 1996 and 2004 (NHSDA/NSDUH )

10

20

30

40

50

60

70

80

12-17 18-25 26-34 35-49

Per

cent

age

of L

ast M

onth

Use

Age Category

White 2004Afr-Am 2004

White 1973-78 birth cohortAfr-Am 1973-78 birth cohort

1990

1990

2004

2004

1996

1996

2004

20042004

2004

2004

2004

Denise Kandel, NIDA Trajectories Conference 11/13/2006

11

1011121314151617181920212223

Cigarettes Alcohol Marijuana Cocaine Rx

White

Afr-Amer

Mean Ages of Onset of Substances Among Whites and African Americans 18-49 Years Old (NSDUH 2004)

Mea

n A

ge o

f O

nset

***

*** ***

***

*

Denise Kandel, NIDA Trajectories Conference 11/13/2006

12

Last Year Dependence Among Lifetime Users Age 12+ Among Whites and African Americans (NSDUH 2004)

5

10

15

20

25

Nicotine Alcohol Marijuana Cocaine Non-RX

White

Afr-Am

Per

cen

t D

epen

den

t

***

***

**

Denise Kandel, NIDA Trajectories Conference 11/13/2006

13

Last Year Dependence Among Lifetime MJ Users by Age Among Whites & African Americans (NSDUH 2004)

Per

cent

age

of L

ast

Yea

r D

epen

denc

e

Age Category

WhiteAfr-Am

Denise Kandel, NIDA Trajectories Conference 11/13/2006

5

10

15

20

25

30

12-17 18-25 26-34 35-49

14

5

10

15

20

25

30

12-17 18-25 26-34 35-49

Last Year Dependence Among Lifetime Cocaine Users by Age Among Whites and African Americans (NSDUH 2004)

Per

cent

age

of L

ast

Yea

r D

epen

denc

e

Age Category

WhiteAfr-Am

Denise Kandel, NIDA Trajectories Conference 11/13/2006]

15

1.5%

0.5%0.3%

2.3%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

Misuse Disorder

1991-1992 NLAES 2001-2002 NESARC

Prescription Drug Misuse and Disorders in 1991-1992 (NLAES) and 2001-2002 (NESARC)

Blanco C et al., in press

* P < 0.01

*

*

16

1. Examples of National U.S. Cross-Sectional Data

1. Examples of National U.S. Cross-Sectional Data

1) Monitoring the Future (MTF)

2) National Survey on Drug Use and Health (NSDUH)

3) National Epidemiologic Survey on Alcohol and Related Conditions (NESARC)

4) National Comorbidity Survey (NCS) and related

5) Youth Risk Behavior Survey (YRBS)

6) Treatment Episode Data Set (TEDS)

7) Drug Abuse Warning Network (DAWN)

17

Select Examples of Other Large Data SetsSelect Examples of Other Large Data Sets

1) Adolescent Treatment Studies

2) SPF-SIG data

3) Criminal Justice Datasets

4) Medicaid and Medicare Administrative Data

5) V.A. Health Care Data

6) Large Managed Care Systems (e.g. Group Health Plan, Kaiser)

7) NIDA’s Clinical Trials Network (CTN) Database: www.ctndatashare.org

18

2. Longitudinal2. Longitudinal

19Jackson et al. J Abnormal Psychology 2005;114:612-626

Smoking Intensity Decreasing Across Birth Cohorts But Stable Across Time (MTF Data)Smoking Intensity Decreasing Across Birth

Cohorts But Stable Across Time (MTF Data)

20

21

NSPY Study: Distinguishing Vulnerable/Resolute Non-Using Adolescents

NSPY Study: Distinguishing Vulnerable/Resolute Non-Using Adolescents

0.0%

37.5%

57.8%66.2%

0.0%9.6%

19.3%26.0%

0%

20%

40%

60%

80%

100%

Round 1 Round 2 Round 3 Round 4

Vulnerable Resolute

Crano WD, NSPY Users Group Meeting May 2007

Sample with complete data at all rounds, Ages 12-18 (N = 2111)

Predictors:

Demographic

Social

Academic

Parental

Peer

Other substances

Marijuana Use

22

Benefit Cost Markedly Greater for Life Course View of Heroin Addiction (DATOS)

05

10152025303540

1 Tx Episode Life CourseZarkin et al. Health Economics 005;14:1133-1150

23

24

1) Monitoring the Future (MTF)

2) National Survey of Parents and Youth (NSPY)

3) Drug Abuse Treatment Outcomes Study (DATOS)

4) Adolescent Health (AddHealth)

5) National Longitudinal Survey of Youth (NLSY)

6) National Epidemiologic Survey on Alcohol and Related Conditions (NESARC)

2. Examples of National U.S. Longitudinal Data

2. Examples of National U.S. Longitudinal Data

25

Select Examples of Other Large Data SetsSelect Examples of Other Large Data Sets

1) Adolescent Treatment Studies

2) Multi-wave locally representative data

Great Smoky Mountain Study

and more than 50 others….

26

3. Using Data Systems: Advantages and Problems

3. Using Data Systems: Advantages and Problems

27

Access to DataAccess to Data

1) SAMHDA Website http://www.icpsr.umich.edu/SAMHDA/

28

Access to DataAccess to Data

1) SAMHDA Website http://www.icpsr.umich.edu/SAMHDA/

2) UCLA Data Program Project (Bentler, PI) and Penn State Prevention Methods Center (Collins, PI)

3) Confidentiality concerns

29

Knowledge of data setsKnowledge of data sets

1) Common problem in using data collected by someone else

2) Solutions: Seminars, publications, user-groups

1) NSPY User Group

2) NESARC User Group

30

3. Improving Use of Data3. Improving Use of Data

1) Support for secondary work• Gene Environment Development

Initiative (GEDI)

2) Support for release and access to data?

3) Developing common metrics for newly proposed studies that are not primarily concerned with drug abuse/addiction?

31

ConclusionsConclusions

Multiple rich sources of data exist and provide opportunities for developing and testing important hypotheses.

Promoting the effective and efficient use of these data requires management and strategic planning.

32