Prescription Drug Abuse and the Emergence of Hepatitis C in Rural Appalachia

Jennifer R. Havens, PhD, MPHAssociate ProfessorDepartment of Behavioral Science

Center on Drug and Alcohol Research

University of Kentucky College of Medicine

Percent of Respondents Using Pain Relievers Nonmedically: 2001 – 2011

Source: National Survey on Drug Use and Health 2001 – 2011

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 20110

1

2

3

4

5

6

Past Year Illicit Drug Abuse/Dependence in Millions

Source: National Survey on Drug Use and Health 2002, 2011

Marijuana

Pain Relievers

Cocaine

Tranquilizers

Stimulants

Hallucinogens

Heroin

0 1 2 3 4 5

2011 2001

Source: MMWR, May 6, 2011; 60(17): 537-541

Injection Drug Use among Rural Drug Users

Previous research: IDU rare among rural drug users in Appalachian Kentucky (Leukefeld et al., Substance Use and Misuse, 1997)

Recent research suggested a much higher prevalence of IDU among Appalachian drug users (>40%)

(Havens et al., Drug and Alcohol Dependence, 2007).

IDU among Appalachian Drug Users

Majority of rural IDUs reported injecting OxyContin® and other prescription drugs NOT designed for injection

Fewer than 10% had ever injected heroin and/or cocaine

Self-reported hepatitis C infection significantly higher among the IDUs versus non-IDUs (p<0.001)

(Havens et al., Drug and Alcohol Dependence, 2007)

Rural Substance Abuse – Mid-2000’s

Clear need for researchLittle known about trends in drug use in

rural Appalachia in particularEven fewer empirical reports around

injection drug use and other medical consequences of drug use

Specifically, HIV, HCV and other STI’s

Study Rationale

Gaps in the literature in rural populations: Injection drug use Infectious disease prevalence and incidenceSocial network and geospatial factors in

disease transmission

Social Networks among Appalachian People (SNAP) study

Purpose: determine prevalence and incidence of HCV, HIV and HSV-2 and other risk behaviors in relation to social network characteristics among rural prescription drug users

Follow-up at 6-, 12-,18-, 24-, 30- and 36-months

Participants

500 rural out-of-treatment injection and non-injection drug users recruited and followed at 6-, 12-, 18-, 24-, 30-, and 36-months post-baseline

Recruitment began in November 2008 and was completed in August 2010

Storefront location in rural townParticipants recruited via Respondent

Driven Sampling (RDS)

Hazard, KY

Participant Characteristics N=503

n %

Male 286 56.7

Age, median (IQR) 31 (26 ,38)

Caucasian 474 94.2

Employed Full-Time 173 34.4

Lifetime Injection Drug Use 394 78.3

Effect of Change in OxyContin Formulation on Abuse

Baseline 6-Month 12-Month 18-Month 24-Month0

50

100

150

200

250

300

350

400

450

500

OxyContin (ER Oxycodone) Roxicodone (IR Oxycodone)

New Formulation Released – August

2010

Emerging Trends in Prescription Drug Abuse

Neurontin (gabapentin)165% increase in abuse between 2013 and

20142950% increase in abuse between 2008

and 2014Participants reporting a mean of 25 days of

use in past 30More likely (p<0.05) to also be abusing IR

oxycodone, buprenorphine and benzodiazepines

Smith, Lofwall and Havens, Am J Psychiatry, 2015

Proportion of Participants Accessing Substance Abuse Treatment

6-Month 12-Month 18-Month 24-Month 30-Month0

2

4

6

8

10

12

14

OST Uptake

6-Month 12-Month 18-Month 24-Month 30-Month0

1

2

3

4

5

6

7

8

9

10

MMT Buprenorphine

Per

cen

t o

f P

arti

cip

ants

En

roll

ing

Proportion of Participants Injecting Drugs

Baseline 6-Month 12-Month 18-Month 24-Month0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

IDU HCV+

Baseline Data – Drug Type for Injecting (n=394 Lifetime IDUs)

n %

Rx Opiates 348 88.3

Rx Stimulants 27 6.8

Rx Benzodiazepines 23 5.8

Methamphetamine 39 9.9

Cocaine 271 68.8

Heroin 95 24.1

Speedball (cocaine/OxyContin) 60 15.2

Initiation to Injection by Age

<=25

26-30

31-40

41+

65 70 75 80 85 90

Final Baseline

Baseline Prevalence – HIV, HCV, HSV-2

HIV HCV HSV-2 HCV/HSV-20

5

10

15

20

25

30

35

40

45

50

0 Cases

222 Cases

59 Cases

33 Cases

HCV Prevalence by Age

<=25 26-30 31-40 41+0

10

20

30

40

50

60

70

Baseline Final

Hepatitis C Seroconversion

Baseline

6-Months

12-Months

18-Months

24-Months

0 50 100 150 200 250 300 350

Cumulative Cases Incident Cases

N=45

N=10

N=8

N=19

N=222

HCV Incidence for PWID

<25 25-29 30-34 35-44 45+0

5

10

15

20

25

Age Group

HC

V I

nci

den

ce p

er 1

00 P

Y

22.1 (15.7 – 30.4)

13.0(8.17 – 19.7)

16.5 (10.2 – 25.3)

8.76(4.98 – 14.3)

1.24(0.06 – 6.11)

Drug Network

Probability of Heroin Initiation

0 6 12 18 24 300

0.05

0.1

0.15

0.2

0.25

0.04

0.020.03

0.00

0.02

p(Hazard) 95% CI

Months in Study

Heroin Availability and Potential for Abuse

One in ten participants noted an increase in the availability of heroin since 2013

However, when asked if they would abuse it if regularly available, less than 4% indicated they would

Conclusions

HCV highly prevalentAdditional incident cases at each visitYoung PWID particularly at risk for

seroconversionHeroin use not prevalent

Future Directions

Interventions Network based?

HCV treatment engagementHarm reduction

Syringe exchangeSubstance abuse treatment

HIV prevention

Acknowledgements

NIH/NIDA (R01-DA024598 and R01-DA033862)

Drs. Carl Leukefeld, Carrie Oser, Rick Crosby, Michelle Lofwall, Sharon Walsh (Co-Investigators)

Study Staff – Hazard and LexingtonStudy Participants