NYC RYAN WHITE PART A CARE COORDINATION & THE CHORDS STUDY * CHORDS is a research partnership of: Stephanie Chamberlin, MIA, MPH Care and Treatment Unit Bureau of HIV/AIDS Prevention and Control New York City Department of Health and Mental Hygiene New York City Department of Health and Mental Hygiene, New York, NY * CHORDS is supported through a grant from NIMH (1R01MH101028). & CUNY School of Public Health, New York, NY 1
BACKGROUND: NYC RYAN WHITE PART A CCP CCP Goal: Ensure that HIV+ Ryan White clients at risk for suboptimal health outcomes receive support to achieve full engagement in care and treatment through coordinated care strategies 2
CCP INTERVENTION DESCRIPTION CCP model provides: case management patient navigation, including accompaniment adherence support, including directly observed therapy (DOT) health promotion in home visits assistance with medical/social services See CDC Compendium of Evidence-based Interventions: http://www.cdc.gov/hiv/pdf/prevention/research/compendium/cdc-hiv- HIVCCP_EI_Retention.pdf 3
CARE COORDINATION PROGRAM (CCP) POPULATION Eligibility based on indicators of need: newly diagnosed; previously lost to care/never in care; irregularly in care; initiating a new regimen; and/or showing incomplete medication adherence or response to treatment. 4
28 CCP AGENCIES IN NYC 5
STUDY AIMS Aim 1: To assess short and long-term CCP effectiveness by comparing care engagement and VL suppression among CCP participants with those of similar PLWH in HIV care who do not receive the CCP intervention.
Aim 1: Pre- Post Study METHODS 7
DATA SOURCES Matched CCP programmatic data with NYC HIV Registry data Programmatic Data: Ryan White Service Provider Reporting (eshare=electronic System for HIV/AIDS Reporting and Evaluation) Merge HIV Surveillance Data: Registry of NYC HIV cases (laboratory VL and CD4 tests, HIV diagnostic events) 8
STUDY ELIGIBILITY 7,337 Clients enrolled in CCP on or before March 31, 2013.1% clients excluded: did not match to the Registry 279 (3.8%) clients excluded: died within 12 months of CCP enrollment 7,058 (96.2%) Clients living 12 months post- CCP enrollment STUDY POPULATION 1117 (15.8%) Newly diagnosed (past 12 mos) at CCP enrollment 4,827 (68.4%) Current to Care at CCP enrollment (6 mos) Previously Diagnosed 1114 (15.8%) Out of Care at CCP enrollment (6 mos) 9
STATISTICAL MEASURES Outcome Measures: Engagement in Care (EiC): 2 CD4 or VL tests 90 days apart, with 1 in each half of 12-month period Viral Load Suppression (VLS): VL 200 copies/ml on most recent test in second half of 12-month period* Estimated post- vs. pre- CCP enrollment relative risks (RRs) for EiC and VLS using GEE *Missing VL in 2nd half of 12-month period considered equivalent to unsuppressed VL. 10
PSYCHOSOCIAL BARRIERS Psychosocial Barriers (Definitions)* Unstable housing: Homelessness or residence in temporary/transitional housing Lower mental health functioning: Mental component summary (MCS) score 37.0 on the SF-12(v2) functional health assessment Recent hard drug use: Self-report of using heroin, cocaine, methamphetamines, or Rx drugs to get high (past 3 months) * Based on CCP Assessment: Baseline = Intake Assessment; Post-baseline=Reassessment 11
PSYCHOSOCIAL BARRIER REDUCTION Reduction of Psychosocial Barriers (Definitions)* Housing barrier reduction: If unstable housing present at baseline, evidence of stable housing post-baseline Mental health barrier reduction: If MCS 37.0 at baseline, a post-baseline MCS score increase 3.5 points Drug-related barrier reduction: If recent hard drug use present at baseline, no use of these drugs post-baseline * Based on latest CCP Assessment during the year of follow-up 12
RESULTS Aim 1: Pre-Post Study 13
ENGAGEMENT IN CARE, PRE & POST 100% RR=1.30 (95% CI 1.28-1.33) RR=1.35 ( 95% CI 1.32-1.38) % with EiC 80% 60% 40% 90% 91% 70% 83% 86% 92% 20% 0% N/A Newly diagnosed ALL previously diagnosed 0% Out of care Current to care Among previously diagnosed 12 months prior to CCP enrollment 12 months post CCP enrollment 14
VL SUPPRESSION, PRE & POST 100% 80% RR=1.80 (95% CI 1.73-1.87) RR=1.46 (95% CI 1.41-1.51) % with VLS 60% 40% 20% 70% 30% 54% 55% 54% 37% 0% N/A Newly diagnosed ALL previously diagnosed 0% Out of care Current to care Among previously diagnosed 12 months prior to CCP enrollment 12 months post CCP enrollment 15
Length of CCP Enrollment (previously dx d): Post- vs. Pre- Enrollment Change, Relative POST- VS. PRE- ENROLLMENT CHANGE, RELATIVE RISK Risk BY LENGTH OF CCP ENROLLMENT (PREVIOUSLY DX D) 2.2 2 1.8 1.6 1.4 EiC RR VLS RR 1.2 1 0.8 6 months 6 months - 1 year > 1 year 16 16
PSYCHOSOCIAL BARRIER PREVALENCE: PREVIOUSLY DX D BASELINE % 51.8 23 30.2 16.2 UNSTABLE HOUSING LOWER MENTAL HEALTH HARD DRUG USE AT LEAST 1 BARRIER 17
ENGAGEMENT IN CARE: POST- VS. PRE-ENROLLMENT (RR, 95% CI) Enrollment/baseline Characteristics Housing Status Mental Health Hard Drug Use Psychosocial Barrier Unstably Housed Stably Housed Lower SF-12 MCS Higher SF-12 MCS Recent Use No Recent Use At least 1 Barrier No Barrier 1 1.1 1.2 1.3 1.4 1.5 Overall RR 18
VIRAL LOAD SUPPRESSION: POST- VS. PRE-ENROLLMENT (RR, 95% CI) Enrollment/baseline Characteristics Housing Status Mental Health Hard Drug Use Psychosocial Barrier Unstably Housed Stably Housed Lower SF-12 MCS Higher SF-12 MCS Recent Use No Recent Use At least 1 Barrier No Barrier Overall RR 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 19
PROPORTION OF THOSE WITH BARRIER AT BASELINE WHO SUBSEQUENTLY EXPERIENCED BARRIER REDUCTION POST-BASELINE BARRIER REDUCTION % 48.1 36.1 15.3 Housing (Unstable Stable) Mental Health ( ) Hard Drug Use (Yes No) 20
EIC BY BARRIER REDUCTION: POST- VS. PRE-ENROLLMENT (RR, 95% CI) Reduced Not Reduced Unstable Housing Reduced Not Reduced Mental Health Barrier Reduced Not Reduced Recent Hard Drug Use 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2 Overall RR 21 Post-enrollment Characteristics
VLS BY BARRIER REDUCTION: POST- VS. PRE-ENROLLMENT (RR, 95% CI) Reduced Not Reduced Mental Health Barrier Unstable Housing Reduced Not Reduced Reduced Not Reduced Recent Hard Drug Use 1.0 1.3 1.6 1.9 2.2 2.5 2.8 3.1 3.4 3.7 4.0 4.3 Overall RR 22 Post-enrollment Characteristics
PRELIMINARY COMPARISON GROUP WORK Aim 1: Comparison Group 23
CHORDS COMPARISON GROUP ANALYSIS: CCP EFFECTIVENESS Comparison Group Aim: To compare care engagement and VL suppression among CCP participants with those of similar PLWH in HIV care who do not receive the CCP intervention. Why?: To control for patterns (e.g. upward trends) in care engagement and VL suppression that may occur independently of the CCP intervention. EiC and VLS have been steadily improving in NYC overall, so some amount of improvement in any one program should not surprise us. 24
METHODS: COMPARISON GROUP Step 1: Select base comparison group of similar non-ccp enrollees from all PLWH in NYC, to include those PLWH who: Had at least 1 CD4/VL reported to surveillance December 2007 March 2013 Met CCP eligibility anytime between December 2009 and March 2013 Were not enrolled in the CCP before March 2014 Step 2: Assign each non-ccp PLWH a pseudo enrollment date (aka anchor date ) that: Results in a set of anchor dates fitting the CCP enrollment date distribution (by enrollment month) Step 3: Perform propensity score matching to identify those PLWH were: Most similar to CCP participants with regard to many measured factors 25
REVISED STATISTICAL MEASURES Outcome Measures: Engagement in Care (EiC): 2 CD4 or VL tests 90 days apart in the 12-month period Viral Load Suppression (VLS): VL 200 copies/ml on most recent test in the 12-month period* Estimated CCP vs. non-ccp relative risks (RRs) for EiC/VLS (at 12-month follow-up) using GEE Stratified outcomes to compare groups with same baseline EiC/VLS (or newly Dx d) status *Missing VL in the 12-month period considered equivalent to unsuppressed VL. 26
MIMICKING THE ENROLLMENT DATE DISTRIBUTION BETWEEN CCP AND NON-CCP WHO MEET ELIGIBILITY CRITERIA PLWH CCP 27 27
PRELIMINARY DATA: CARE ENGAGEMENT AT 12-MONTH FOLLOW-UP (%) - CCP VERSUS NON-CCP, BY BASELINE STATUS 100 91 95 96 86 88 84 CCP 88 Non-CCP 80 75 % Engaged in Care 60 40 20 49 51 0 Newly Diagnosed (N=1,105) EIC + No VLS (N=2,539) EIC + VLS (N=1,680) No EIC + No VLS (N=1,345) No EIC + VLS (N=361) EIC Engagement in Care status at baseline VLS Viral Load Suppression status at baseline 28
PRELIMINARY DATA: VIRAL SUPPRESSION AT 12-MONTH FOLLOW-UP (%) - CCP VERSUS NON-CCP, BY BASELINE STATUS 100 CCP Non-CCP % Virally Suppressed 80 60 40 20 73 60 40 36 82 79 52 27 79 65 0 Newly Diagnosed (N=1,105) EIC + No VLS (N=2,539) EIC + VLS (N=1,680) No EIC + No VLS (N=1,345) No EIC + VLS (N=361) EIC Engagement in Care status at baseline VLS Viral Load Suppression status at baseline 29
DISCUSSION 30
CONCLUSIONS (PRE-POST ANALYSES) Significant EiC/VLS increases occurred in all subgroups examined, including those with key barriers to HIV care and treatment adherence Findings suggest a link between support to reduce psychosocial barriers and greater improvement on 12-month EiC/VLS outcomes CCP shows promise for increasing health/survival opportunities among those at highest risk for suboptimal HIV health outcomes 31
CONCLUSIONS (COMPARISON GROUP ANALYSES) In short-term (12-mo.) comparisons of CCP clients post-enrollment EiC and VLS to the EiC and VLS achieved among other, similar PLWH: The CCP demonstrates effectiveness at increasing EiC and VLS overall (above and beyond the increases among similar PLWH in usual care over the same period) Benefits of the CCP appear less substantial for individuals who were already engaged in care at the time of CCP enrollment. Evidence of CCP effectiveness over usual care suggests the public health value of intervention scale-up, focusing on PLWH with the greatest need, lowest engagement in care 32
STUDY AIMS Aim 1: To assess short and long-term CCP effectiveness by comparing care engagement and VL suppression among CCP participants with those of similar PLWH in HIV care who do not receive the CCP intervention. Aim 2: Among those who enroll in the CCP, identify individual and CCP site-level determinants of care engagement and VL suppression up to 36 months following CCP enrollment. Aim 3: To assess the cost-effectiveness (cost per qualityadjusted life year [QALY]) of the CCP relative to usual care outside the CCP, considering downstream cost-savings and individual and public health benefits due to improved VL suppression and HIV infections averted.