Brian Montague, DO MS MPH Assistant Professor of Medicine

1. Linkage to HIV Care on Release from Incarceration: Data from the LINCS Project 2010-2012 in RI and NC Brian Montague, DO MS MPH Assistant Professor of Medicine Alpert School of Medicine at Brown University Michael Costa, MPH Senior Associate Abt Associates Funding: NIH, NIDA 1R01DA030778

2. Presentation Outline • Study Overview and Rationale • Data Access – Systems-Level Challenges and Solutions • Study Findings from Rhode Island and North Carolina Sites • Conclusions and Implications

3. Study Overview and Rationale

4. LINCS Study Overview • Goal: to develop scalable metrics to assess quality of linkage to for persons with HIV on release from corrections using existing corrections and clinical data sources • Framework: • Uses eUCI from HRSA/HAB as means of confidentially linking records between corrections release data and Ryan White HIV/AIDS Program clinical data sources • Key metrics assessed include time to linkage for persons linking to care and virologic status at time of first community assessment

5. Participating LINCS Study Sites Rhode Island North Carolina Massachusetts Georgia Dallas, Texas

6. Why Is This Important? • HIV treatment is critical to maintaining the health of HIV+ individuals • Sicker individuals are harder and more expensive to treat • Treated, individuals are less much less likely to pass HIV on to uninfected partners

7. HIV in Corrections • Since the early years of the HIV epidemic, HIV has disproportionately impacted prisoners. • In 2008, the prevalence of HIV was 1.6% among US state prisoners, representing 20,449 people.1 • Approximately 150,000 HIV-infected persons, 14% of all Americans with HIV, pass through corrections each year.2, 3 • The prevalence of HIV within correctional settings ranges from 2.5 to more than 3 times that of the general population, with prevalence in high prevalence communities such as Baltimore and Washington D.C. as high as 6.6%.1, 3, 4 • Minority disparities in HIV care are amplified in corrections. Maruschak LM. December 2009, revised 1/28/10. Bureau of Justice Statistics Bulletin: HIV in Prisons 2007-2008. Washington, DC: US Department of Justice. Spaulding AC, Seals RM, Page MJ, Brzozowski AK, Rhodes W, et al. (2009) HIV/AIDS among Inmates of and Releasees from US Correctional Facilities, 2006: Declining Share of Epidemic but Persistent Public Health Opportunity. PLoS ONE 4(11): e7558. Boutwell A, Rich JD. HIV infection behind bars. Clin Infect Dis. 2004 Jun 15;38(12):1761-3. Solomon L, Flynn C, Muck K, Vertefeuille J. Prevalence of HIV, syphilis, hepatitis B, and hepatitis C among entrants to Maryland correctional facilities. J Urban Health. 2004 Mar;81(1):25-37.

8. Disproportionate Impact on Minorities • African Americans are incarcerated at 6 times the rate of whites.1 • HIV disproportionately impacts African Americans. • 7 times the rate of HIV infection • Constitute 45% of new HIV infections nationwide2 • Nearly twice as likely to lack health insurance3 • Nearly 50% of Ryan White program clients are African American.4 http://irishgreeneyes-welcometomyworld.blogspot.com/2011/06/infographic-not-guilty-program-seeks-to.html Sabol WJ, West HC, Cooper M. Prisoners in 2008. Bureau of Justice Statistics. December 2009, NCJ 228417. Revised 6/30/2010 . Available at: http://bjs.ojp.usdoj.gov/content/pub/pdf/p08.pdf. Hall HI, Song R, Rhodes P, Prejean J, An Q, Lee LM, Karon J, Brookmeyer R, Kaplan EH, McKenna MT, Janssen RS; HIV Incidence Surveillance Group. Estimation of HIV incidence in the United States. JAMA. 2008 Aug 6;300(5):520-9. Schwartz K, Howard J, Tolbert J, Lawton E, Chen V. The Uninsured: A Primer. October 2010.. Publication #7451-06. Washington, DC: The Kaiser Commission on Medicaid and the Uninsured, Kaiser Family Foundation. Available at: http://www.kff.org/uninsured/upload/7451-06.pdf Heath Resources and Services Administration, HIV/AIDS Bureau. Going the distance: The Ryan White HIV/AIDS Program, 20 years of leadership, a legacy of care. August 2010. Rockville, MD: Health Resources and Services Administration. Available at: http://hab.hrsa.gov/data/files/2010progressrpt.pdf

9. HIV in Corrections and on Reentry • Incarceration is often the only time these individuals access HIV testing, education, counseling, and treatment services. • Limited data regarding the experience of HIV-infected persons on reentry. • Often marginalized in their communities due to addiction, mental health disorders, unemployment, and racial disparities • Decreased access to health care • High rates of relapse to substance abuse and other transmission risk behaviors • High mortality

10. Risks on Reentry • Health gains during the stay in corrections are often lost at the time of reentry • CD4 declines, viral load increases at reincarceration and return to care1,2 • Texas experience: 5.4% of prison inmates receiving ART while incarcerated fill ARV prescription in time to avoid gap in treatment3 http://www.mlive.com/news/muskegon/index.ssf/2010/01/mentoring_program_focused_on_c_1.html • Springer SA, Pesanti E, Hodges J, Macura T, Doros G, Altice FL. Effectiveness of antiretroviral therapy among HIV-infected prisoners: reincarceration and the lack of sustained benefit after release to the community. Clin Infect Dis. 2004;38 (12):1754–60.. • Stephenson BL, Wohl DA, Golin CE, Tien HC, Stewart P, Kaplan AH. Effect of release from prison and re-incarceration on the viral loads of HIV-infected individuals. Public Health Rep. 2005 Jan-Feb;120(1):84-8.. • Baillargeon J, Giordano TP, Rich JD, Wu ZH, Wells K, Pollock BH, Paar DP. Accessing antiretroviral therapy following release from prison. JAMA. 2009 Feb 25;301(8):848-57.

11. Key Assumptions • Ryan White providers, as the safety net providers, will be the first service provider for persons on release from corrections • Sentenced persons who are incarcerated will be offered antiretroviral therapy and in most cases achieve virologic suppression prior to release • Given limited supplies of medicine on release and social instability in the post release period, early engagement in community HIV care is critical to treatment success

12. Data Access – Systems-Level Challenges and Solutions

13. The Problem of Data Linkage Linkage Corrections Data Clinical Data Incarceration Release HIV Care Clinical Services Prescription HIV Status & Care Barriers Procedural Confidentiality/HIPAA Legal Motivational

14. Assessing Linkage to Care • Lack of scalable metrics to assess linkage to care • How were they at the time of release? • How long did it take on reentry to return to care? • What was their condition at the time of the first visit? • Ryan White client level data reporting (RSR) offers a unique opportunity to assess linkage and retention • National Corrections Reporting Program (NCRP) provides case-level data on released prisoners

15. National Corrections Reporting Program (NCRP) • Bureau of Justice Statistics (BJS) funds and directs data collection from state depts of correction and the Federal Bureau of Prisons • Centralized incarceration/release data • Contains data on sentenced persons in 41 states

16. HRSA, HAB Ryan White HIV/AIDS Program • Grant-based payor of last resort. • Ryan White Program - only care option for most releasees. • Ryan White Reporting System (RSR): • Demographics • Dates of service • Key clinical variables (CD4, VL) • Other services provided

17. Confidentiality in Ryan White CLD File • Client-level data records identified by electronic Unique Client Identifier (eUCI) • Encrypted identifier derived from 1st & 3rd letters of first and last name, DOB, and gender • Hash algorithm SHA-1 prevents reverse engineering client identifiers

18. eUCI Source: Coombs E, O’Brien-Strain P. “UCI and You.” Webcast. SPHERE Institute. November 10, 2008.

19. Matching RSR & NCRP Data HIPAA De-identified Matched Records Data File Mask Dates (PHI)

20. Data Access Challenges • The RSR data for this study is provided directly by participating RW grantees • HAB does not provide any data • Grantees own and determine the use of their data for such studies

21. Data Challenges - Permissions • For just Rhode Island and North Carolina sites we have initiated: • 22 Data Use Agreements • 8 IRB Reviews • 3 Letters of Authorization with OHRP

22. Data Challenges - Quality • NCRP and RSR have undergone many years of QI/QC • State HIV surveillance data is a possible source • Uneven completeness and quality across states • State laws vary regarding the ability share surveillance data with outside institutions • State DPHs have severe capacity constraints to conduct such analyses in-house

23. Data Access Solutions • We have taken the door to door approach out of necessity. • Once a partnership is established, data transfer can become routinized • There are still grantees that did not participate due to burden or caution – we respect both • Clear implications for State and Federal Leadership to institutionalize ongoing assessment

24. Study Findings from Rhode Island and North Carolina Sites

25. Study Phases • Validation • Rhode Island • Quantitative data • Rhode Island • North Carolina

26. Correctional & Clinical Data Sources • Rhode Island • Corrections: NCRP, single correctional institution • Clinical: Ryan White Data from Miriam Hospital (primary provider serving patients following release) • High level of ascertainment of HIV status in incarceration • Small release cohort so fewer false positive matches • North Carolina • Corrections: NCRP, multiple correctional institutions • Clinical: Ryan White data from 37 of 71* Ryan White providers • Larger release cohort with more potential for false positives • Not all Ryan White providers provided study data, so only partial coverage of clinical care sites serving patients post release *Source for total number of providers: 2011 HAB State Profile (number of providers can change year to year)

27. Correctional & Clinical Data Sources • Rhode Island • Corrections: NCRP, single correctional institution • Clinical: Ryan White Data from Miriam Hospital (primary provider serving patients following release) • High level of ascertainment of HIV status in incarceration • Small release cohort so fewer false positive matches • North Carolina • Corrections: NCRP, multiple correctional institutions • Clinical: Ryan White data from 37 of 71* Ryan White providers • Larger release cohort with more potential for false positives • Not all Ryan White providers provided study data, so only partial coverage of clinical care sites serving patients post release *Source for total number of providers: 2011 HAB State Profile (number of providers can change year to year)

28. Validation of the eUCI in RI • Gold standard linkage assessment performed in RI • eUCIs generated for names and for known aliases. Match for any eUCI was treated as a match for the individual • Linkage as assessed by eUCI compared to gold standard

29. Key Observations: • eUCI matching performed comparably to probabilistic matching • Inclusion of aliases significant improves matching performance • False positives and false negatives occurred but bias in estimates of time to linkage and rates of virologic suppression was small • Density plots demonstrated clustering of false positives at earlier linkage times

30. Validation of eUCI Linkage time distribution comparison False Positives False positives 10-30 days

31. Handling False Positive Matches • Based on RI validation sample, false positives tended to be clustered at earlier linkage times. • Similar clustering seen in North Carolina sample • Statistically, models created in which estimates with earlier linkage times are weighted less than those with later linkage time for the purposes of analyses of linkage to care • Clustering can may change over time as practice patterns change (spacing out routine care visits) and programs to impact linkage shorten time to linkage to care

32. Metrics • Time to linkage: time from release date to first ambulatory care service in the community • Clinical status at linkage: viral load at first assessment following release (on treatment/off treatment) • Retention: remaining in care based on visit frequency or persistent virologic suppression following linkage

33. Benchmarks and Comparisons • In+Care (2014): nationwide, cross-section HIV providers • Virologic suppression (<200): 75% • Texas (Baillargeon et al) TX reentry prisons • Althoff (2013): Multisite, jail reentry

34. Demographics of Sample

35. Linkage

36. Linkage

37. Time to Linkage To Care Rhode Island Linkage time assessed among those who ultimately link to care North Carolina: Known HIV+

38. Site Variability • 14 sites with > 5 release events • Range meeting retention metrics 27% to 55% • Considerations: • Low volume sites <=10 (what happens if you deal with this rarely) • range 29%-78% • High volume sites (greater experience, potentially more complex patients) • range 27% - 55% • Gives ability to identify sites with demonstrated success and potentially provide mechansim for identifying and sharing best practices

39. Key Findings • Despite significant differences between correctional and care systems between RI and NC, linkage experience was comparable • Delays in engagement in care are frequent with median time to first service in range of 40-50 days • 37% have recurrent viremia at first visit indicating significant lapses in treatment • Of those retained in care, about 30% have persistent viremia one year post release • RI and NC experience better than prior reports but significant gaps remain

40. Additional Analytic Questions • Given known barriers to linkage, to what extent can these be identified using additional data sets • Mental health: service data in corrections • Substance abuse: ? charge data, ? Ryan White service data • Unstable housing: homeless service data from community • Access to insurance: program policy, ? Ryan White data • Access to case management or other social supports post release: care providers accessed (if designated CM providers), service data • Distance to care providers: site of release, site of care

41. Conclusions and implications

42. Conclusions • This method provides a valuable framework for assessing the quality of linkage to care • Additional support is needed at all levels to promote engagement and retention following release • Release planning in corrections • Active community based case management peri-release • Active programs at community care sites to engage these high risk patients • ACA initiated Medicaid payment for care may present an opportunity to strengthen urgency for linkage Funding: NIDA 1R01DA030778

43. Acknowledgements North Carolina DPH: Jacquelyn Clymore Brian Berte All 17 Participating RW grantees DPS: David Edwards Co-Investigators Jacques Baillargeon David Wohl David Rosen Carmen Albizu Garcia Brown University Josiah Rich (PI) Cara Sammartino Roee Gutman Fizza Gillani Nick Zaller RI Dept of Corrections Jeff Renzi Pauline Marcussen Erin Boyar Abt Associates Liza Solomon (PI) Michael Costa Lisa LeRoy David Izrael Sara Donohue Tom Rich Michael Shively Jennifer Davis Alyssa Kogan Massachusetts DPH: Kevin Cranston Noelle Cocoros Betsey John DOC: Rihanna Kohl Texas Ank Nijhawan Princess Iroh BJS/NCRP William Sabol E. Ann Carson Georgia Jane Kelly, DPH Ron Henry, DOC Christine Helms, DOC Funding: NIH, NIDA 1R01DA030778

44. Viral Load Suppression: NC & RI Rhode Island North Carolina (HIV+ only)

45. Viral Load Suppression: Rhode Island

46. Virologic Suppression: North Carolina