Transcript HIV/STD Risk Behaviors in Methamphetamine User Networks

Three-Stage Sampling Plan
• Stage 1: Randomly selecting
particular sites (venues) within strata
• Stage 2: Randomly selecting sessions
(half-days of operation within sites
• Stage 3: Randomly selecting
respondents within sessions at
selected sites
Stage 1: Sampling Sites
• Two strata
– STD clinics operated in L.A. County, Long
Beach City (13 sites; 23,000 patients)
– High-risk sexual environments
• “Sites” are physical establishment
(clinic) or physical entity (public park)
– Locations that are co-located or proximal
considered one “site”
• Enumeration of sites in each stratum,
with number of locations in each site
Stage 2: Sampling Scheduled Sessions
• Each site sampled in Stage 1 will
determine when MSM, MSM/W are most
likely to congregate
• Time blocks will be 4 hours (am, pm, eve)
• Obtain reasonable estimate of how size is
distributed across sessions
• We will sample sessions with PPES
• We will sample 90 sessions from each
stratum (~ 7 sessions for STD sites; 4
sessions for sexual locations; Total
sessions = 180
Stage 3: Sampling Participants
within Scheduled Sessions
• We will deploy 2 interviewers to
each sampled session and location
to target MSM and MSM/W for
sampling and interviewing
• Sample drawn sequentially
Branding and Visibility
• High-profile study adopted by L.A.
County and Long Beach City STD
Programs
• Press releases and information about the
study through county materials
• Announce study in gay magazines, local
newspapers, websites
• Flyers/postcards to local gay-oriented
establishments
• CAB to advise the study and visibility
aspects
Index Participants:
Screening Procedures
• On specified sites and times, teams first
capture number of MSM and MSM/W at
location
• Identify potential cohort member and explain
this is a study of the sexual health of
L.A./Long Beach area men and their sexual
partners, both drug using and non-drug using
• All cohort members will be asked to provide
information about joining the study to at least
1 of their current sexual partners, especially
female partners
Enrollment of Index Participants
• Men screened will be given
appointments at the field offices in
Hollywood and Long Beach
• Men who decline are thanked and
visible information about the individual
recorded
• Consent procedures completed and
data collection begins
Enrollment of Sexual Partners
• Enrollment occurs indirectly
• Materials describe the study and
provide toll-free telephone number
• Appointment is made from phone call
• Partners asked to bring invitation
materials to appointment for
verification of the ask
• Partners asked to provide consent for
linking of data to index participant
Follow-up Assessments
• Pre-paid calling cards provided to
contact staff members
• $50 compensation for self-report and
biological samples (6- and 12-months)
• Participants who terminate early will be
asked to complete 5 to 10-minute
discussions to solicit reasons for
termination
– Probes for positive and negative
experiences
Interviewer Training
• Approximately 50 hours training
– Interviewing substance using
individuals
– CASI
– Data collection routines using ACCESS
– Behavioral practice on scheduling
calls, mock interviews, follow-up
appointment calls
– Standard operating procedures
– Ethics
– Managing difficult persons
Quality Assurance/Supervision
• Each interview audiotaped; 15%
randomly reviewed by Project
Director
• Tapes reviewed within 2 weeks and
feedback delivered
• QA checklist developed for each
assessment
• Interviewers attend bi-monthly
supervision
Assessment Plan
• Baseline measures require 1.5 h
• Self-report data for drug use
and sexual risk behaviors cover
a 3-month recall period
• Additional resources may
provide for re-contact of
participants at 3-, 9-months
Demographics/Acculturation
• Demographics (Baseline only)
– Age, gender, race/ethnicity, education,
employment status, income, relationship status,
residential stability
• Acculturation (Baseline; termination)
– Internalized Homonegativity Inventory –
personal homonegativity, gay affirmation,
morality of homosexuality (Mayfield 2001)
– Latino, African Americans use measure to
assess language, media and ethnic social
relations (Caetano, 1987); parallel measure for
Asians (Suinn et al., 1995)
Drug Use Measures
Baseline, 6-, 12-Months
• Addiction Severity Index
– Scales tapping 7 areas
– 30 day recall for recent drug use
• Time Line Follow-Back
– Recall of drug use for 3 months prior
to assessment point
– Periods longer than 3 months
unreliable in active drug users
Sexual Behavior Measures
Baseline, 6-, 12-Months
• NIMH Multisite Trial Cooperative Agreement
(for heterosexuals), Gorbach’s measures,
Behavioral Questionnaire for last 3 months on:
– Number and gender of partners
– Type of partners (primary, regular,
nonexclusive, new)
– Knowledge of serostatus of partner
– Type and frequency of sexual acts
– Frequency and type of condom use in each act
– Use of drugs/alcohol during sexual encounters
– Risk factors among partners
Sexual Behaviors (cont’d)
• Series of questions will investigate role of
drug use in exaggerating risk
– Facilitate recall of specific, detailed
information on type and amount of drug
use
– Estimate of duration of sexual episode(s)
– Abrasions, ulcerations, bleeding
– Rougher sex than usual
– Whether subject had STD at the time
• Partnership calendar will capture
specific behaviors in a partnership
during the last week
Partnership Assessments
Baseline, 6-, 12-Months
• Partnership Assessment Scale
• Concurrency Typology Index
• Characteristics of Partnerships
– Duration, emotional intensity, resource
sharing, cohabitation, sexual frequency,
frequency of social interaction,
concurrent partnership, level of social
integration, place of sexual encounter,
source of meeting
– Frequency of social interaction (times,
modes)
Measuring Partnership Types
UNKNOWN PERSON- someone that you had
never met before you had sexual contact and
never plan to see again
ONETIME PARTNER- someone you knew of or
had seen before, but you had sexual contact
with only one time
ACQUAINTANCE- someone you have had
sexual contact with more than once, but not on
a regular basis, and who you don't socialize
with
FRIEND- someone you have had sexual contact
with more than once, but not on a regular
basis, and you normally socialize with
Types of Concurrent Partnerships
Primary
Partner
Reciprocal
Reactive
Separational
Transitional
Compensatory
No Primary
Partner
“Fuckbuddy”
Survival Sex=
(Sex for $ & drugs)
“1 Night Stands”=
“Flings”
Co-parent=
“Baby’s Mom/Dad”
Experimental
Group sex
“Bootycall”
Environments
• Chart the geographic density of risk
establishments within neighborhoods
– Types of establishments
– Density of MSM and MSM/W at the establishmen
– Facilitation in providing condoms; prevention
– Extent of drug use
– Types of sexual risk behaviors
– Proportion of risk behaviors under the influence
– Frequency of behaviors that resulted in prior
HIV/STD infections
– Social, sexual, and drug-using contacts that
extend from the venue
Mental Health
Baseline, 6-, 12-Months
• Symptom Check List – 90 – R (Derogatis)
– Somatization, OCD, interpersonal
sensitivity, depression, anxiety, hostility,
phobic anxiety, paranoid ideation,
psychoticism
• Quality and Well-being (RAND Mental
Health Inventory)
– General positive affect, emotional ties,
anxiety, depression, loss of behavioral or
emotional control
STD Biomarkers
Baseline, 12-Months
• HIV testing – rapid test,
confirmatory (Elisa) for positives,
detuned assay
• STD testing – gonorrhea, chlamydia
– urine and anal swab using nucleicacid amplification; syphilis – blood
Ethnography Baseline, 12-Month
• 2-hour, semi-structured, in-depth, openended interviews with random sample of 40
participants and their sexual partners
• Understanding of sexual identity, drug use,
HIV risk behaviors, partnerships, social and
cultural networks
– Probes designed to explore social meanings
attached to active drug use and risk behaviors
within the relationship, concurrent or
monogamous to understand infectious disease
status and risk for disease transmission in the
context of the participants’ lives
Ethnography Procedure
• Pseudonyms used in reporting
findings
• Interviews analyzed for content; first
level coding scheme of re-emerging
themes
• Themes identified using interview
segments that have same labels
• Coded by 2 individuals;
discrepancies resolved
• Text analyzed in Atlas.ti
Data Analysis
• Main variables are disease status of
HIV, STDs, confirmed at baseline and
12-months
– Disease movement slow
• Self-report of drug use and
psychological status collected at 6month intervals
• Disease status modeled in MSM,
MSM/W and their sexual partners at
individual level, partnership level, and
community/environmental level
Hypothesis 1: Drug Involvement
and Infectious Disease
• Higher prevalence of HIV/STDs among
methamphetamine-using MSM and MSM/W
at baseline and 12-months
• Methamphetamine users more likely than
others to engage in IDU and drug-associated
sexual-risk behaviors during the follow-up
period
• Incidence of HIV infection is higher among
methamphetamine users than users of other
substances or non-drug using MSM and
MSM/W
Hypothesis 2: Sexual Partnerships
and Disease Transmission
• Methamphetamine users will have more
concurrent partnerships than comparison
peers
• Drug-using MSM, MSM/W will practice highrisk behaviors with non-drug using partners,
i.e., a bridge
• Partnership dynamics will associate with risky
behaviors (no condoms, concurrency)
• Pattern and distribution of different types of
concurrent partnerships will vary between
drug-using MSM and non-drug using
MSM/W, with MSM/W at higher concurrency
Hypothesis 3: Modeling the Diffusion
of Sexual Risk among Drug Users
• Evaluate how the incidence of HIV and
STDs reflect the size and connectedness
of the sexual networks of each group
studied
• To determine the impact of sexual
network structure in the future
transmission of HIV in Los Angeles
within and beyond MSM and
heterosexual groups
Modeling Approach
Schematic of the individual-based modeling software (SimPy) that
extends the software to implement the rules by which individuals choose
partners, and their behaviors within each partnership
Two Approaches to Modeling
• Classical population-level
epidemiological models based on
differential equations
– Pair approximations will model the rate of
change of partnerships of different types
solved by XPPAUT, which assumes a very
large population size
• Contrasting approach involves
individual-based models of HIV/STD
transmission that considers
heterogeneity in partnerships
Uses of the Models
• Infer the contact network structure from data
on partnerships across individuals and the
correlation in numbers and types of
partnerships and small sub-networks
• Implications of the inferred contact network
and risk behavior differences between
partnership types on disease spread will be
simulated
• Exploration of the implications of potential
biases in our sampling design on our ability to
describe the network of contacts, which allows
evaluation of the robustness of our solutions
Project Time Line
• Months 0-6
– Set-up
• Months 6-44
– Recruit cohort; enroll sexual partners;
begin ethnography; build databases; ethno
papers
• Months 18-55
– Conduct follow-up evaluations; data
queries and resolutions; Follow-ups end
month 55
• Months 48-60
– Data analysis and manuscript publication