Transcript The Research Challenges of Caring for HIV+ Substance Users

Practicing Out of the Box:
The Research Challenges of Caring
for HIV+ Substance Users
Chinazo Cunningham, MD
Division of General Internal Medicine
Dept of Family and Social Medicine
MMC/AECOM
A community perspective
• Background & Programs
• Research/Evaluation
– Evaluation of medical outreach
– Self report vs. medical records
• Partnership with the community & harm
reduction
– Benefits and challenges
– Lessons learned
Background
• Montefiore Medical Center
• CitiWide Harm Reduction
– CBO serving HIV-infected drug users living in
SRO hotels in NYC
– Started as needle exchange program in 1994,
now offers numerous programs
– Collaboration with Montefiore since 1998
• Medical care/outreach
• Research/evaluation
CitiWide’s programs
•
•
•
•
•
•
•
Outreach in SRO hotels
Needle exchange
Case management
Health services
Mental wellness
Support groups
Peer education
•
•
•
•
Housing
Holistic health
Transportation
Clothing, showers,
meals
• Research/Evaluation
Health Services Program
Providers
Locations
2-3 MDs (internists) CitiWide’s drop-in center
Services
HIV primary care
1 NP
Patient’s SRO hotel room Acute care
1 PA
CHCC
Gyn care
1 program admin
Vaccinations
1 admin assistant
Hep C eval & tx
2 outreach workers
Referrals to medical
specialties
Referrals to nonmedical services
SRO Hotel Residents
Characteristic
Age (mean)
Male
Race:
%
45 yrs
72
Black
59
Hispanic
30
White/other
11
Heterosexual
79
< High school education
49
Income < $8000/yr
90
Medicaid
83
Drugs:
Cocaine/crack
49
Heroin/opioids
21
Have regular HIV provider
85
Research & Evaluation
Objectives / Model
Patient characteristics
•Trust
•Health beliefs
•Social support
Pt-Provider dynamic
•Cultural concordance
•Relationship with provider
Engagement in HIV Primary Care
•HIV ambulatory visits
•HIV tests (VL, CD4)
•HIV meds (ART, PCP & MAC prophylaxis)
Community outreach
Health-related characteristics
•HIV disease severity
•Mental health
•Substance use
Evaluation of a Medical
Outreach Program
• Background
– Marginalized HIV+ populations have less access to
care, poor health outcomes
– Outreach programs aim to improve access
– Few programs evaluated
• Objectives
– To evaluated a medical outreach program that targets
HIV+ SRO hotel residents in NYC
– To examine patient- and program-related factors
associated with keeping medical appointments
Methods
• Examined 2781 medical appt records (2003-2005)
– CitiWide and Montefiore databases
• Patient-related factors
– sociodemographic info
• Program-related factors
– Appt type: same day/walk-in vs. future appt
– Appt location: CitiWide vs. SRO hotel vs. CHCC
– Provider making appt: medical vs. non-medical
• Analysis
– Chi-square, regression analysis
Future appts
kept
Same day
appts kept
357 (23.3)*
309 (41.9)
245 (28.1)*
309 (41.9)
SRO hotel
32 (10.6)
--
CHCC
80 (22.1)
--
Medical provider
143 (18.1)*
21 (30.4)*
Non-medical provider
214 (28.7)
288 (43.1)
Total
Location of appt
CitiWide
Person making appt
Summary
• Overall 29% of appts were kept
• Program characteristics, NOT patient
characteristics assoc with kept appts
• Appts kept more often when:
– At CitiWide’s drop-in center
– Same day / walk-in
– Made by non-medical provider
Implications
• Changed Health Services Program to
provide more appts at CitiWide and same
day/walk-in
• Medical community must examine
program-related factors (not just pt-related
factors) in delivery of care to marginalized
HIV+ populations
– Same day access
– “One stop shopping”
SRO Hotel Residents
Characteristic
Age (mean)
Male
Race:
%
45 yrs
72
Black
59
Hispanic
30
White/other
11
Heterosexual
79
< High school education
49
Income < $8000/yr
90
Medicaid
83
Drugs:
Cocaine/crack
49
Heroin/opioids
21
Have regular HIV provider
85
Comparison of self report vs. medical
records HIV utilization measures
• Background
– Numerous studies examine HIV health services
using self-reported outcomes
– Few studies examined validity of these
outcomes in marginalized populations
– Crucial to understand validity of outcome
measures for program evaluation
• Objective: To examine agreement between
self-report and medical record HIV health
services utilization measures
Methods
• Cross-sectional study design
• Sample
– 522 HIV+ individuals living in 14 SRO hotels in NYC
• Data
– Self report from ACASI
– Medical record extraction by MD
• Variables
– HIV-related ambulatory care visits (0, 1, >2 visits)
– HIV lab markers (CD4, VL)
– HIV-related medications (ART, PCP, MAC)
• Analysis
– percent agreement & Kappa statistic
Results
Selfreport
(%)
Medical
records
(%)
Agreement Kappa
> 2 outpt visits/6 mos
84.7
56.9
54.9
0.09
Taking ART
69.8
64.7
75.0
0.43
Taking PCP prophylaxis
50.4
34.5
69.0
0.38
Taking MAC prophylaxis
22.5
14.5
75.8
0.23
(%)
Results
Selfreport
(%)
Medical
records
(%)
Agreement
Kappa
Had CD4 count
performed
81.7
73.2
64.8
0.06
Had VL performed
81.9
63.9
61.3
0.06
CD4 count value
(mean, cells/mm3)
357
329
81.3
0.71
Undetectable VL
(undetectable)
42.2
34.9
75.9
0.49
(%)
Conclusions
• Agreement between self-report and medical
records was:
–
–
–
–
Poor for ambulatory visits ( = 0.09)
Poor to fair for medication use ( = 0.23-0.43)
Poor for lab tests performed ( = 0.06)
Good for CD4 count value ( = 0.71)
• Most disagreement was from patient over-reporting
• When examining health services utilization in
marginalized populations, the use of self-reported
measures as outcomes raises concerns.
Partnership with the community
& harm reduction
Benefits to working with a CBO
• Large number of community members in
one place
• “Special population” not in clinical settings
• Facilitate trust
• Direct access to community (SRO hotels)
• Attractive to funders (community-based
participatory research)
Challenges to working with a CBO
• Different priorities
– research vs. service
• Philosophical clash
– traditional medical system vs. harm reduction*
• Power, money, resources
– large academic medical center vs. small CBO
• Supervision / oversight
– Two different geographic locations
• Structural issues
– Computers, heating, supplies, payroll, etc
• Staffing
– professionals vs. para-professionals
Harm Reduction vs. Medical Model
Harm Reduction
Medicine
Structure
Inclusive, community
decisions
Hierarchical chain
of command
System
Low threshold
High threshold
Provider role
Provides info, collaborative Prescribes
decision making
treatment
Client role
Makes informed decisions
Complies with
treatment
Locus of
control
Client-centered
Physician-centered
Lesson Learned - Redefine Health
• Health is NOT the absence of disease
• Biopsychosocial model including…
–
–
–
–
–
–
drugs
housing
support system
finances
violence
criminal justice issues…
• Life priorities of HIV+ IDUs
– Only 37% ranked HIV as most important
– Top priorities: housing, money, safety from violence
Mizuno
Lessons Learned –
Redefine Goals and Success
• Success is NOT just:
– Undetectable viral load
– Abstinence from drug use
• Success also includes:
– Making it to appointments
– Preventative care (PCP/MAC prophylaxis,
vaccinations, PAP smears, PPD)
– Less, safer, more controlled drug use
– Improvement in non-medical areas (housing,
support system, criminal activity, etc.)
Lessons Learned
•
•
Working with a CBO is an investment
Integration into CBO
–
–
–
•
•
•
legitimizes the research
notice problems before they become too big
“buy-in” from other CBO staff
Difficult to conduct research in setting where it
is not a priority
Need face-to-face time and close oversight of
research staff
Communication and transparency