Transcript Update on Alcohol and Health

Update on
Alcohol, Other Drugs,
and Health
January–February 2012
www.aodhealth.org
1
Studies on
Interventions &
Assessments
www.aodhealth.org
2
Lack of Efficacy of a
3-Medication Treatment
Protocol for Methamphetamine
Dependence
Ling W, et al. Addiction. 2012;107(2):361–369.
Summary by Kevin L. Kraemer, MD, MSc
www.aodhealth.org
3
Objectives/Methods



The PROMETA™ protocol (including the
benzodiazepine antagonist flumazenil, the
GABA-agonist gabapentin, and the H1-histamine
antagonist hydroxyzine) is unproven for the
treatment of methamphetamine dependence.
This double-blind trial randomized 120 patients
with methamphetamine abuse or dependence to
PROMETA* or placebo (hydroxyzine only).
*Intravenous flumazenil 2 mg on days 1, 2, 3, 22, and 23; oral gabapentin
titrated to 1200 mg per day by day 4 and continued to day 40; and oral
hydroxyzine 50 mg prior to flumazenil infusion and as take-home medication
through day 10.
www.aodhealth.org
4
Objectives/Methods (cont’d)


Each group received 14 weekly sessions of
cognitive behavioral therapy.
Outcomes were methamphetamine and other
drug use documented by urine testing, selfreported methamphetamine craving,
treatment retention, and adverse events.
www.aodhealth.org
5
Results



Fifty percent of the experimental group and 70%
of the placebo group remained in the study
through the medication phase (day 40).
Only 30% of the experimental group and 43% of
the placebo group remained until the end of the
study (day 108).
During follow-up at days 23, 40, and 108, the
experimental and placebo groups did not differ
significantly with regard to the following:
www.aodhealth.org
6
Results (cont’d)




proportion of methamphetamine-negative urine tests
(0.5, 0.4, and 0.3 versus 0.5, 0.5, and 0.4,
respectively);
mean methamphetamine craving score (2.8, 3.1, and
2.2 versus 2.9, 2.9, and 2.4, respectively); or
percentage with 3 consecutive negative urine tests
(36% and 39% versus 46% and 51% at days 40 and
108, respectively).
Mild, moderate, and severe adverse events were
reported by 73%, 23%, and 4% of the
experimental group and 59%, 40%, and 1% of
the placebo group, respectively.
www.aodhealth.org
7
Comments


The PROMETATM protocol has been promoted
and used as a treatment for methamphetamine
dependence despite lack of a strong scientific
rationale and proven efficacy.
This randomized placebo-controlled trial found
no benefit of PROMETA over placebo for
treatment of methamphetamine abuse or
dependence and raises serious questions about
the use of PROMETA for methamphetamine
dependence.
www.aodhealth.org
8
Retention in Opioid Agonist
Treatment after Prison Release
Reduces Re-incarceration
Larney S, et al. Addiction. 2012;107(2):372–380.
Summary by Peter D. Friedmann, MD, MPH
www.aodhealth.org
9
Objectives/Methods


Opioid agonist treatment (OAT) in prison and
after release might influence the risk of reincarceration.
This prospective cohort study linked data on
OAT and incarceration among 375 men with
heroin use originally recruited in 1996–1997 for
a randomized controlled trial of OAT in prison in
New South Wales, Australia. Participants were
followed through 2006.
www.aodhealth.org
10
Results




During 9+ years of observation, 331 participants
engaged in OAT 1081 times, with a median of 2
episodes per participant (mean length of
engagement, 156 days); 58% started OAT in prison.
Ninety percent of participants were re-incarcerated
after the first incarceration.
Engagement in OAT at the time of release had no
effect on re-incarceration.
Post-release retention in OAT was associated with a
one-fifth reduction in the number of reincarcerations.
www.aodhealth.org
11
Comments


This study affirms that retention in OAT following
release is associated with reduced reincarceration among former prisoners with opioid
dependence.
Although other investigations have shown that
initiating OAT prior to release maximizes postrelease treatment retention, the current study
suggests active linkage to ongoing treatment is
an essential component.
www.aodhealth.org
12
Comments (cont’d)


Continuing or initiating OAT during incarceration
is necessary but not sufficient to optimize postrelease outcomes among opioid-dependent
inmates.
Correctional systems and treatment providers
must also provide transitional assistance to
ensure that former inmates reach OAT programs
after release.
www.aodhealth.org
13
Adding N-Acetylcysteine to
Glucocorticoids May Improve
Outcomes in Patients with
Severe Alcoholic Hepatitis
Nguyen-Khac E, et al. N Engl J Med. 2011;365(19):1781–1789.
Summary by Darius A. Rastegar, MD
www.aodhealth.org
14
Objectives/Methods


This randomized controlled trial conducted in 11
French university hospitals assessed whether Nacetylcysteine [NAC], an antioxidant used to treat
acetaminophen-induced hepatitis, further reduced
mortality in patients treated with prednisolone for
severe alcoholic hepatitis.
Subjects (N=174) were:




age 18 or older.
had consumed, on average, >50 g alcohol per day in the
past 3 months.
had a Maddrey’s discriminant function* of 32 or more.
had histologic findings consistent with alcoholic hepatitis.
*Calculated as 4.6 x [patient’s prothrombin time - control prothrombin time (in seconds)] +
serum bilirubin level (mg/dL).
15
Objectives/Methods (cont’d)



The long list of exclusion criteria included other
possible causes of liver disease (e.g., hepatitis B
or C), HIV infection, and “serious cardiac,
respiratory or neurologic disease.”
All patients received oral prednisolone for 28
days.
Eighty-five also received intravenous NAC on
days 1–5.
www.aodhealth.org
16
Results

Mortality rates and hazard ratios (HR) for patients
in the prednisolone-only (PRED) and prednisolone
plus NAC (PRED-NAC) arms were as follows:
Outcome
PRED
(n=89)
PRED-NAC
(n=85)
HR
95% CI
1-month mortality
24%
8%
0.58
0.14 - 0.76
3-month mortality
34%
22%
0.33
0.33* - 1.04
6-month mortality
38%
27%
0.62
0.37 - 1.06
*As reported in the paper.

Adverse events were no higher in the PRED-NAC
arm, and 2 major adverse events (hepatorenal
syndrome and infection) were significantly lower.
www.aodhealth.org
17
Comments


Although this study failed to show significantly
reduced mortality at 6 months (the primary
endpoint), a significant short-term benefit was
seen in the PRED-NAC arm with no serious
adverse events.
The failure may simply have been due to lack
of sufficient power. It is also possible that
longer term outcomes would be better with
longer courses of treatment with NAC.
www.aodhealth.org
18
People Receiving Buprenorphine
and Methadone Are More Likely to
Be Responsible for Traffic Crashes
Corsenac P, et al. Drug Alcohol Depend. November 18, 2011
[Epub ahead of print].
Summary by Tae Woo Park, MD, & Alexander Y. Walley, MD, MSc
www.aodhealth.org
19
Objectives/Methods



Opioid agonist treatment (OAT) does not result in
substantial driving impairment in driving-simulation
studies; however, the association between traffic
crashes and OAT has not been studied since
buprenorphine became available.
This French study of 72,685 drivers involved in
traffic accidents between 2005–2008 investigated
the association between risk of being responsible for
a crash and having a prescription for buprenorphine
or methadone on the day of the crash.
Responsibility was determined by police reports
matched with a national crash database and linked
to national pharmacy data.
www.aodhealth.org
20
Results


After adjusting for age, gender, road/vehicle
conditions, and prescriptions for other medications
known to impair driving ability, drivers prescribed
buprenorphine or methadone had a 2-fold risk of
being responsible for a road traffic crash
compared with those who were not (odds ratio
[OR], 2.02).
The 0.3% of drivers prescribed buprenorphine or
methadone were more likely to be men, to be
younger, and to have been prescribed other
medications that impair driving ability (such as
anxiolytics) than the 99.7% who were not.
www.aodhealth.org
21
Comments



Although OAT patients may not be involved in
more crashes, when they are involved, this
study suggests that they are more likely to be
responsible.
However, we do not know if the increased risk
is because of OAT, because such persons tend
to be riskier drivers, or if there is some other
reason.
Furthermore, we do not know how these risks
stack up against the risk among patients with
opioid addiction who are not receiving OAT.
www.aodhealth.org
22
Comments (cont’d)

In any case, when discussing driving risk with
patients receiving OAT, it seems reasonable to
reassure them that driving is not impaired under
experimental conditions, while at the same time
recognizing that, under real-world conditions,
such patients may be more likely to be
responsible for a traffic crash.
www.aodhealth.org
23
Recent Notable Developments
in Addiction
Pharmacotherapies
Colfax GN, et al. Arch Gen Psychiatry. 2011;68(11):1168–1175.
Plebani JG, et al. Drug Alcohol Depend. 2012;121(1):163–166.
Stein MD, et al. Drug Alcohol Depend. 2012;120(1):65–73.
Summary by Richard Saitz, MD, MPH
www.aodhealth.org
24
Objectives/Methods

Several recent studies have potentially
important implications for practice.
25
Results

In one small randomized trial (Colfax et al.), 60
methamphetamine-dependent men who have
sex with men were assigned to mirtazepine or
placebo, both added to counseling. Although
medication adherence was modest,
methamphetamine-positive urine tests were
half as frequent among men in the mirtazepine
group, who also reported fewer risky sexual
behaviors than men in the placebo group.
www.aodhealth.org
26
Results (cont’d)


In another small randomized trial (Plebani et al.),
37 people with cocaine dependence were
assigned to varenicline or placebo. All participants
received counseling. Those on varenicline were
twice as likely to have negative urine tests for
cocaine, and they also reported less reward from
cocaine use.
In a larger (N=137) randomized placebocontrolled trial (Stein et al.), trazodone had no
effect on sleep or drug use among methadonemaintained opioid-dependent patients with sleep
complaints.
www.aodhealth.org
27
Comments


Despite being early reports and thus
inconclusive, the 2 studies among stimulantdependent patients are important, because
pharmacological interventions targeting this
group have not been particularly successful.
These results hold promise if they can be
replicated.
Trazodone is often recommended for insomnia
in patients with substance dependence because
it has little addiction risk, but the study among
methadone-maintained patients suggests it may
not work. Thus, the search for safe and
effective alternatives should continue.
www.aodhealth.org
28
Studies on
Health Outcomes
www.aodhealth.org
29
Modest Marijuana Exposure
Does Not Adversely Affect
Pulmonary Function, but High
Cumulative Exposure Does
Pletcher MJ, et al. JAMA. 2012;307(2):173–181.
Summary by Hillary Kunins, MD, MPH, MS
www.aodhealth.org
30
Objectives/Methods


Previous investigations have not shown a
consistent impact of marijuana smoking on
pulmonary function.
Investigators conducted a longitudinal analysis
of 5016 adults* enrolled in the CARDIA† study to
assess the impact of marijuana exposure on
pulmonary function (FEV1 and FVC).
*Black/non-Hispanic and white/non-Hispanic men and women aged 18–30 years and
healthy at enrollment in 1985.
†CARDIA=Coronary Artery Risk Development in Young Adults.
www.aodhealth.org
31
Objectives/Methods (cont’d)


Marijuana exposure was assessed at 6 follow-up
exams, permitting calculation of “joint-years”
(where 365 joints=1 joint-year) and tobacco
pack-years.
Fifty-six percent of participants (n=2807)
attended the 20-year follow-up examination
without differential attrition by marijuana use.
www.aodhealth.org
32
Results

In adjusted analyses, the association between
marijuana use and pulmonary function was
nonlinear:




at low lifetime exposure, marijuana use was associated
with increased FEV1 and FVC.
at >7 joint-years, the positive association between
marijuana use and pulmonary function leveled off.
at >10 joint-years, there was a nonsignificant decline in
FEV1.
at >20 episodes of marijuana use per month, the
decline in FEV1 was significant.
www.aodhealth.org
33
Comments


The study design and analysis of linearity allowed
investigators to elucidate the complex impact of
marijuana on pulmonary function: the positive effect
of marijuana smoking on FVC (hypothesized to be a
“stretching and training” effect) dominated
marijuana’s impact on pulmonary function at lower
exposures, whereas its negative effect on FEV1
seemed to dominate at higher exposures.
Estimating the impact of heavy marijuana use was
imprecise because of the low number of heavy users
and few non-tobacco users among the heaviest
marijuana users. Nevertheless, this study may help
define thresholds for riskier marijuana use similar to
those defined for alcohol.
www.aodhealth.org
34
Wine-specific Mortality Benefits
Disappear When Studied
Appropriately
Holahan CJ, et al. J Stud Alcohol Drugs. 2012;73(1):80–88.
Summary by Richard Saitz, MD, MPH
www.aodhealth.org
35
Objectives/Methods



Some have suggested that wine might be more
beneficial to health than other types of alcoholic
beverages because of compounds in wine besides
alcohol.
To test this hypothesis, investigators studied 802
past-month abstainers and drinkers of low-risk
(“moderate”) amounts of alcohol (1 to <3 [14 g]
drinks per day) age 55–65 years at enrollment.
They assessed mortality over 20 years.
www.aodhealth.org
36
Results


The mortality rate was 69% for abstainers, 50%
for low-wine-consumption drinkers (<one-third of
alcohol from wine) (adjusted* odds ratio [AOR],
0.67), and 32% for high-wine-consumption
drinkers (AOR, 0.59).
Although an unadjusted analysis showed a large
reduction in mortality associated with high-wineconsumption drinking compared with low-wineconsumption drinking, the difference disappeared
after controlling for overall alcohol consumption,
socioeconomic and health factors, and health
behaviors.
*Adjusted for age, gender, income and education, medical diagnoses, smoking, and
physical activity.
www.aodhealth.org
37
Comments


If there is a mortality benefit from drinking,
these results suggest beverage type doesn’t
matter. But, as is very common in studies
examining the potential benefits of drinking, this
study tested associations between typical alcohol
consumption during 1 month and mortality over
20 years without updating consumption.
It also compared drinkers with those who had
stopped drinking—a group well known to be less
healthy (“sick quitters”).
www.aodhealth.org
38
Comments (cont’d)


“Moderate” drinkers are well known to be
healthier than others—not because of drinking,
but because health-conscious people often drink
low-risk amounts.
The fact that the large benefit of wine
consumption disappeared with appropriate
statistical adjustment should serve as yet another
reminder to use caution concluding that observed
associations between drinking and health benefits
are causal. It would not be the first time
consistent results from numerous observational
studies were consistently wrong.
www.aodhealth.org
39
Thirty Years of Observational
Studies of Alcohol’s
Cardioprotective Effects:
Uncertainty Remains
Roerecke M, Rehm J. Addiction. January 9, 2012
[Epub ahead of print]. doi: 10.1111/j.1360-0443.2012.03780.x.
Summary by Peter D. Friedmann, MD, MPH
www.aodhealth.org
40
Objectives/Methods

This meta-analysis combined 44 observational
studies from 1980–2010 that reported a relative
risk for ischemic heart disease (IHD) in relation
to average alcohol intake.
www.aodhealth.org
41
Results



The well-known J-shaped curve was confirmed
(i.e., compared with abstainers, IHD risk is lower
in people with low-level alcohol consumption but
rises with increasing consumption).
The maximum cardioprotective effects for
mortality occurred between 32–63 g alcohol* per
day for men and between 11–31 g per day for
women.
The effects were heterogeneous, even at low
levels of intake.
*One standard drink averaged to 12 g pure alcohol in this study.
www.aodhealth.org
42
Comments


Although this study reaffirms the populationlevel association between low-level alcohol
intake and reduced cardiovascular morbidity
and mortality, the high level of heterogeneity
means it is very difficult for clinicians to make
inferences about individual patients.
Some people benefit from low-level drinking
while others are harmed, and we cannot
differentiate these groups.
www.aodhealth.org
43
Comments (cont’d)


Furthermore, meta-analysis of even a large
number of studies does not eliminate the
possibility that residual confounding could explain
the findings (e.g., IHD was reduced because of
better risk-factor profiles among those who drink
low-level amounts).
Although advising patients about lower risk
drinking limits is the current standard of practice,
considerable uncertainty remains about what
constitutes a safe level of consumption, and for
whom.
www.aodhealth.org
44
Lifestyle and Environmental
Factors, Including Tobacco and
Alcohol Use, and Risk of Cancer
Parkin DM, et al. Br J Cancer. 2011;105(Suppl 2):S77–S81.
Summary by R. Curtis Ellison, MD
www.aodhealth.org
45
Objectives/Methods

Researchers estimated the fraction of cancers
occurring in the UK in 2010 that could be
attributed to exposure to 14 lifestyle and
environmental risk factors:

tobacco; alcohol; diet (meat, fruit, vegetable,
fiber, and salt consumption); overweight; lack of
exercise; occupation; infection; radiation (ionizing
and solar); hormone use; and breastfeeding.
www.aodhealth.org
46
Results


Exposure to less-than-optimum levels of the 14
risk factors was responsible for 45.3% of
cancers in men and 40.1% of cancers in women
(a total of about 134,000 cases).
Of the lifestyle factors studied, tobacco smoking
had the largest effect on the risk of cancer,
responsible for 19.4% of all new cases.
www.aodhealth.org
47
Results (cont’d)


In men, deficient intake of fruits and vegetables
(6.1%), occupational exposures (4.9%), and
alcohol consumption (4.6%) had the next
largest effects on risk.
In women, overweight or obesity (because of
the association with breast cancer) (6.9%) and
infectious-agent exposure (3.7%) had the next
largest effects on risk.
www.aodhealth.org
48
Comments


In this well-done analysis, smoking had, by far,
the largest effect on cancer risk, while 4 percent
of all cancer cases were attributable to alcohol
intake (compared with abstaining).
Although the use in this analysis of abstinence
as the optimal level of alcohol consumption
related to cancer risk may be reasonable, it
would not be the case for cardiovascular
diseases or total mortality.
www.aodhealth.org
49
Comments (cont’d)



Furthermore, the analysis did not examine levels
of alcohol use, an important point since alcoholrelated cancer risk may relate to consumption
amounts.
The effects related to diet may have been
overestimated as well, as the values used in this
analysis were much greater than those seen in
most studies.
Nevertheless, the results highlight the importance
of targeting certain behaviors to reduce cancer
risk.
www.aodhealth.org
50
Studies on
HIV and HCV
www.aodhealth.org
51
Baclofen: New Hope for Alcohol
Abstinence in Patients with
Alcohol- and HCV-related
Cirrhosis?
Leggio L, et al. Addict Behav. 2012;37(4):561–564.
Summary by Jeanette M. Tetrault, MD
www.aodhealth.org
52
Objectives/Methods


Alcohol consumption accelerates HCV-related
liver fibrosis; thus, no safe level of drinking
exists in patients with HCV.
Baclofen, a GABAB receptor agonist, is a
potential therapeutic agent for alcohol
dependence.
www.aodhealth.org
53
Objectives/Methods (cont’d)


This post-hoc analysis of a clinical trial of
baclofen for the treatment of alcohol
dependence in patients with cirrhosis explored
whether the safety and efficacy of baclofen
persisted in a subset of alcohol-dependent
patients with cirrhosis and HCV.
Of 84 patients enrolled in the main trial, 24 had
alcohol dependence, HCV infection, and
cirrhosis. Of these, 12 patients received baclofen
(10 mg orally 3 times per day) and 12 received
placebo for 12 weeks.
www.aodhealth.org
54
Results



Ten patients receiving baclofen, compared with
3 receiving placebo, achieved total alcohol
abstinence (p=0.01).
In the baclofen group, compared with placebo,
albumin values increased, and a trend toward
reduction in international normalized ratio (INR)
levels was demonstrated.
No patients discontinued baclofen due to sideeffects.
www.aodhealth.org
55
Comments


In this post-hoc analysis of data from a larger
randomized controlled trial, baclofen showed
promise at improving alcohol abstinence in a
subgroup of alcohol-dependent HCV-infected
patients with cirrhosis.
If these results are replicated in larger clinical
trials, baclofen may add to existing alcohol
pharmacotherapies for those with HCV and
cirrhosis.
www.aodhealth.org
56
Needle and Syringe Provision and
Opioid Agonist Treatment May
Reduce the Spread of HCV in
People with Injection Drug Use
Turner KM, et al. Addiction. 2011;106(11):1978-1988.
Summary by Judith Tsui, MD, MPH
www.aodhealth.org
57
Objectives/Methods


Interventions that impact injection drug use
behaviors, such as needle and syringe provision
(NSP) and opioid agonist treatment (OAT), may
decrease the spread of HCV.
This meta-analysis pooled data from 6 studies to
assess whether NSP* (alone or in combination)
and OAT were associated with reduced
incidence of HCV.
*Needle and syringe provision was defined as "high" or "low" (high meaning 1 or more
sterile needles were obtained from a provider for each injection reported).
www.aodhealth.org
58
Objectives/Methods (cont’d)


Studies were included if they were published in
the UK after 2000, if they contained individual
level data on NSP and/or OAT, and if they
measured newly acquired cases HCV infection.
Six observational studies were identified:


4 cross-sectional studies.
2 cohort studies.
www.aodhealth.org
59
Results



Participants receiving OAT had 59% lower odds
of new HCV infection (adjusted odds ratio
[AOR], 0.41).
Participants engaged in high NSP had 52%
lower odds of new HCV infection (AOR, 0.48).
The combined effect of the interventions was
stronger than each alone (AOR, 0.21).
www.aodhealth.org
60
Comments



This meta-analysis suggests that NSP and OAT
may reduce the incidence of HCV.
Limitations include the design of the studies (i.e.,
observational versus randomized controlled trials)
and moderate heterogeneity among studies
included for OAT.
These results support those of a prior metaanalysis* that found evidence supporting the
effectiveness of needle exchange and OAT as
components of interventions to reduce HCV
seroconversion in people with IDU.
*Tetrault J. Alcohol, Other Drugs, and Health, September-October 2011. Available at
www.bu.edu/aodhealth/issues/issue_sept11/tetrault_hagan.html.
61
Mental-Health and Substance
Use Disorders Impact the
Development of AIDS-defining
Illness and Death in
HIV-infected Veterans
Nurutdinova D, et al. AIDS. 2012;26(2):229–234.
Summary by Jeanette M. Tetrault, MD
www.aodhealth.org
62
Objectives/Methods



Depression is associated with HIV medication
nonadherence, poor HIV outcomes, and mortality,
however, the effect of other mental-health or
substance use disorders on HIV disease progression
is unclear.
This retrospective cohort study assessed the impact
of mental-health and substance use disorders on
HIV disease progression and death in 27,574 HIVinfected veterans who initiated combined antiretroviral treatment (cART) between 2000–2006.
Sixty-nine percent of the sample had at least 1
mental-health or substance use disorder.
www.aodhealth.org
63
Results

After adjusting for age, race, baseline CD4 cell
count, baseline comorbidity, cART adherence, and
health-care utilization,



schizophrenia (hazard ratio [HR], 1.40), bipolar disorder
(HR, 1.32), and substance use disorders (HR, 1.23)
were associated with all-cause mortality, while anxiety
disorders were inversely associated with all-cause
mortality (HR, 0.80).
depressive disorders were not associated with all-cause
mortality.
only substance use disorders were associated with
development of AIDS-defining illness (HR, 1.19).
www.aodhealth.org
64
Comments


Although this study was limited by reliance on
administrative data, the findings suggest that,
despite adequate cART therapy, specific mentalhealth and substance use disorders impact both
the development of AIDS-defining illness and allcause mortality.
Lack of association between depressive disorders
and the primary outcomes is notable. Attempts to
replicate this finding should be made in further
studies.
www.aodhealth.org
65
Political Engagement Is
Associated with Reduced HIV
Risk Behaviors
Mino M, et al. Am J Drug Alcohol Abuse. 2011;37(6):520–524.
Summary by Darius A. Rastegar, MD
www.aodhealth.org
66
Objectives/Methods



This study examined the impact of civic and
political engagement on HIV risk behaviors in
people with injection drug use (IDU).
Subjects (N=162) were recruited from 6
methadone maintenance programs in New York
City/New Jersey. All subjects were of Puerto Rican
origin and reported current use of heroin or cocaine
and IDU in the past 30 days.
Subjects were asked about IDU risk behaviors and
political engagement (i.e., being registered to vote,
identifying as part of an organized political party,
and paying attention to politics as measured on a
Likert scale).
www.aodhealth.org
67
Results


In the past 30 days, 38% of participants
had shared injection equipment, and 15%
had used “shooting galleries” (location
where people gather to inject illegal drugs).
With regard to political engagement,



63% reported being registered to vote.
62% identified with a political party.
30% reported paying a “fair amount” or “a lot”
of attention to politics.
www.aodhealth.org
68
Results (cont’d)


On bivariable analysis, higher levels of political
engagement were significantly associated with
reduced HIV risk behaviors on 3 measures: those
registered to vote and those identifying with a
political party were less likely to share injection
equipment, and those who paid attention to politics
were less likely to use shooting galleries.
On multivariable analysis, after adjusting for age,
gender, educational level, and homelessness,
sharing injection equipment was significantly less
likely among those who identified with a political
party, and using a shooting gallery was less likely
among those who paid more attention to politics.
www.aodhealth.org
69
Comments


Although the results are not impressive, one
would expect persons who are politically
engaged to be more mindful of the impact of
their behaviors on others. Therefore, it is not
surprising that individuals with higher levels of
political engagement were less likely to engage
in HIV risk behaviors.
It would be interesting to see if political
engagement can be increased among people
with IDU, and if this engagement can be used to
reduce HIV (and other health) risk behaviors.
www.aodhealth.org
70