To the Editor:
Young adults, aged from 18 to 24 years, with HIV (YHIV) in the United States are less likely to be aware of their HIV status, linked to or retained in care, and virally suppressed than older adults with HIV.1 Health insurance is an important prerequisite to high-quality, continuous health care services,2 especially for young adults,3 and Medicaid is the primary source of insurance for YHIV.4
Insurance discontinuity—defined as disenrollment with or without reenrollment—is common among Medicaid enrollees5 and is associated with poor health outcomes,2 sustained periods without insurance,6 and higher costs due to increased utilization and uncompensated care.7 Lack of insurance is a barrier to adherence to combination antiretroviral therapy (ART) in persons with HIV.8
Twenty-six percent of adults with HIV enrolled in Medicaid experience discontinuity,9 but little is known about the rates of discontinuity among YHIV with Medicaid and the factors that may be associated with discontinuity. The objective of this study was to characterize discontinuity in Medicaid coverage among YHIV in the United States before the Affordable Care Act (ACA).
We evaluated YHIV enrolled in Medicaid at 6 sites, with the largest proportion of Medicaid-insured individuals, out of the 21 sites of the HIV Research Network (HIVRN), a federally funded clinical trial network. We merged clinical and demographic data from sites with coverage and eligibility data from the Medicaid Analytic eXtract files using patient identification numbers, as described elsewhere.9
Data from 336 participants who enrolled from January 1, 2006 to December 31, 2010 were included. Exposure time (person years) began at HIVRN enrollment or 18th birthday and ended with death, transfer, loss to follow-up, or 25th birthday. Data were excluded if age was ≤17 or ≥25 on January 1 of given year, total days of Medicaid enrollment were ≤30, gender was transgender (due to low numbers, n = 5), or CD4 count was missing in given year.
The primary outcome was incidence of discontinuity, defined as number of days without Medicaid coverage per person year. Age on January 1 was categorized as 18–21 or 22–24. Self-reported race/ethnicity was categorized as non-Hispanic black, non-Hispanic white, or other. Self-reported HIV transmission risk factor was categorized as perinatally HIV infected (PHIV) or nonperinatally HIV infected (nPHIV).10 First CD4 count in given year was categorized as ≤200, 201–499, or ≥500 cells/mm3. On ART was defined as prescription of combination ART at any time in given year. Calendar year was included to assess for secular trends during the study period. Medicaid eligibility type was categorized as disability, low-income, or other.
We calculated proportions for participant characteristics. The outcome had excess zeros because most individuals had no discontinuities; therefore, we applied a zero-inflated Poisson regression model. This two-part model estimates odds ratios (OR) for the binary part of the model (zero discontinuity versus having discontinuity) and incidence rate ratios (IRRs) for the number of discontinuities >0. We used robust standard errors to account for potential correlation among observations from the same person, and a complete case analysis approach because missing data were likely not missing at random. We conducted sensitivity analysis to determine if categorizing risk differently (e.g., men who have sex with men and IV drug use) changed the results. Variable inflation factors were used and no covariates were collinear.
We used Stata 13.1 (College Station, TX). IRBs at each site and Johns Hopkins University School of Medicine approved the study.
Our overall sample included 593 person years contributed by 336 YHIV. Most individuals were male (62%), black (66%), nPHIV (61%), 18–21 years (65%), and low-income (56%). Thirty-six percent were disabled. Forty-two percent had CD4 count 201–499, and 63% were on ART (Table 1).
Table 1.
Characteristics of Young Adults with HIV at Study Entry (n = 336)
| Characteristics | Individuals, n (col %)a |
|---|---|
| Year | |
| 2006 | 115 (34) |
| 2007 | 48 (14) |
| 2008 | 62 (18) |
| 2009 | 55 (16) |
| 2010 | 56 (17) |
| Gender | |
| Female | 128 (38) |
| Male | 208 (62) |
| Age (years) | |
| 18–21 | 217 (65) |
| 22–24 | 119 (35) |
| Race/ethnicityb | |
| Black | 223 (66) |
| White | 53 (16) |
| Other | 60 (18) |
| Medicaid eligibility typec | |
| Low income | 189 (56) |
| Disability | 120 (36) |
| Otherd | 27 (8) |
| Risk categoryd | |
| nPHIV | 205 (61) |
| PHIV | 131 (39) |
| CD4 count (cells/mm3) | |
| ≤200 | 70 (21) |
| 201–499 | 142 (42) |
| ≥500 | 124 (37) |
| On ARTe | |
| No | 126 (38) |
| Yes | 210 (63) |
Totals may not add to 100% due to rounding.
White includes non-Hispanic; black includes African American and Caribbean; and Other includes Hispanic/Latino, Asian, Pacific Islander, American Indian, Unknown, or coded as “Other”
Other eligibility type includes foster care, disabled and low income, and those coded “Other”
nPHIV includes male-to-male sexual contact, IV drug use, heterosexual contact, transfusion, female sexual contact with females, other, and unknown. PHIV includes those infected at birth.
Combination ART.
ART, antiretroviral therapy; nPHIV, nonperinatally HIV infected; PHIV, perinatally HIV infected.
Overall, 120 of the 336 individuals (36%) had a discontinuity. Multivariable analysis of Medicaid discontinuity is shown in Table 2. Incidence of discontinuity was higher among white YHIV [adjusted incidence rate ratio (aIRR) 1.51 (95% CI 1.10–2.06)] and those with higher CD4 counts [CD4 201–499: aIRR 1.54 (1.05–2.28); CD4 ≥ 500: aIRR 1.65 (1.14–2.39)]. Compared to 2006, incidence was lower in 2009 [aIRR 0.62 (0.45–0.86)]. Disabled YHIV had higher odds of having complete coverage, compared with low-income YHIV [aOR 2.18 (1.26–3.78)]. Transmission risk factor was not significantly associated with discontinuity, regardless of dichotomous or categorical definition.
Table 2.
Multivariable Analysis of Factors Associated with Medicaid Discontinuity (n = 336)
| Variable | Incidence of discontinuity | Odds of zero discontinuity | ||
|---|---|---|---|---|
| IRR (95% CI) | aIRRa(95% CI) | OR (95% CI) | AORa(95% CI) | |
| Year | ||||
| 2006 | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| 2007 | 0.94 (0.61,1.43) | 0.97 (0.63,1.48) | 1.19 (0.68,2.10) | 1.14 (0.63,2.08) |
| 2008 | 0.89 (0.65,1.22) | 0.90 (0.68,1.21) | 0.93 (0.53,1.63) | 0.93 (0.51,1.69) |
| 2009 | 0.64 (0.44,0.92) | 0.62 (0.45,0.86) | 0.95 (0.55,1.66) | 0.92 (0.50,1.71) |
| 2010 | 0.78 (0.56,1.08) | 0.74 (0.53,1.03) | 1.07 (0.63,1.83) | 1.07 (0.58,1.97) |
| Gender | ||||
| Female | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| Male | 0.99 (0.77,1.27) | 1.20 (0.95,1.50) | 0.66 (0.43,1.03) | 0.71 (0.44,1.13) |
| Age | ||||
| 18–21 | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| 22–24 | 0.95 (0.75,1.19) | 1.02 (0.80,1.31) | 0.89 (0.61,1.30) | 1.07 (0.66,1.75) |
| Race/ethnicityb | ||||
| Black | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| White | 1.19 (0.80, 1.78) | 1.51 (1.10, 2.06) | 1.16 (0.64, 2.11) | 1.83 (0.93, 3.61) |
| Other | 0.93 (0.61, 1.42) | 0.99 (0.67, 1.47) | 1.55 (0.86, 2.81) | 1.44 (0.73, 2.85) |
| Medicaid eligibility type | ||||
| Low income | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| Disability | 0.98 (0.76,1.27) | 0.96 (0.75,1.24) | 2.25 (1.43,3.57) | 2.18 (1.26,3.78) |
| Otherc | 0.73 (0.50,1.06) | 0.91 (0.60,1.40) | 0.59 (0.34,1.05) | 0.63 (0.30,1.31) |
| Risk categoryd | ||||
| nPHIV | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| PHIV | 0.98 (0.77,1.23) | 1.22 (0.92,1.60) | 1.46 (0.94,2.25) | 1.55 (0.70,3.43) |
| CD4 count | ||||
| ≤200 | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| 201–499 | 1.33 (0.95,1.87) | 1.54 (1.05,2.28) | 0.70 (0.40,1.21) | 0.87 (0.49,1.56) |
| ≥500 | 1.47 (1.04,2.07) | 1.65 (1.14,2.39) | 0.78 (0.42,1.43) | 1.03 (0.53,2.00) |
| On ARTe | ||||
| No | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) | 1.0 (ref) |
| Yes | 0.95 (0.76,1.19) | 1.01 (0.82,1.25) | 1.20 (0.78,1.84) | 1.16 (0.75,1.80) |
Results in bold are statistically significant (p < 0.05).
Adjusted models include site, in addition to variables listed in table.
Race/ethnicity: white includes non-Hispanic; black includes African American and Caribbean; and other includes Hispanic/Latino, Asian, Pacific Islander, American Indian, Unknown, or coded as “Other”
Other eligibility type includes foster care and those coded “other”
nPHIV includes male-to-male sexual contact, IV drug use, heterosexual contact, transfusion, female sexual contact with females, other, and unknown. PHIV includes those infected at birth.
Combination ART.
aIRR, adjusted incidence rate ratio; AOR, adjusted odds ratio; ART, antiretroviral therapy; CI, confidence interval; IRR, incidence rate ratio; nPHIV, nonperinatally HIV infected; OR, odds ratio; PHIV, perinatally HIV infected.
Over one-third (36%) of YHIV experienced a Medicaid discontinuity. Higher CD4 count, low-income eligibility type, and white race were associated with discontinuity.
Young adults with HIV are at risk for poor health outcomes compared with older adults with HIV,1 and an important driver of this difference may be disparity in discontinuity of Medicaid coverage faced by young adults versus older adults—36% versus 26% in Fleishman et al.9
This pre-ACA baseline for comparison can inform future changes to the Medicaid program that account for the dynamics of coverage for YHIV. Post-ACA research is needed to describe the impact of other strategies on discontinuities, such as the Ryan White Program (and its coordination with Medicaid), Medicaid expansion, dependent coverage expansion, and work requirements, especially post-ACA. Medicaid discontinuity can place an already vulnerable population at risk for further barriers to achieving the goal of sustained viral suppression.10 Unfortunately, private insurance is unlikely to fill the gap for young adults with Medicaid discontinuity, given young adults often work in low wage, entry-level jobs without employer sponsored health insurance.3
This study also identifies possible predictors of discontinuity for YHIV. Compared to disabled eligibility type, low-income was associated with discontinuity; however, states vary with respect to Medicaid eligibility requirements for people with HIV. State-specific analyses are needed to determine how YHIV—who may not have experienced opportunistic infections that many states use as a marker of disability—are affected differently than older adults by eligibility requirements. In addition, those with higher CD4 counts may have less contact with the health care system and may be more susceptible to administrative churning—the frequent exit and reentry of beneficiaries.5 Even short periods without insurance coverage may impact access to care for YHIV. Race is another important factor to explore further given its association with income and eligibility type, which impact participation in Medicaid.11
Future research should explore etiologies of discontinuity—such as administrative churning, loss of eligibility, lack of reenrollment, and other factors that may simultaneously impact Medicaid eligibility and discontinuity, such as race, income, and disability. Support from clinic staff may decrease the frequency and impact of discontinuities, but must be delieverd in an empowering and patient-centered manner.12
Limitations of our study include complete case restriction, exclusion of transgender individuals due to low numbers, imprecise measures of immune status, ART utilization, few measures of social capital such as income, and inability to determine the insurance status (i.e., private or uninsured) of participants during a Medicaid discontinuity. Despite these limitations, the inclusion of multiple sites in multiple states and use of both HIVRN and Medicaid datasets are significant strengths of this work. By drawing from multiple sites caring for YHIV with confirmed Medicaid coverage, we avoided relying on clinic-reported or self-reported insurance status common in clinical databases.
Our results demonstrate that many YHIV experienced discontinuity in their Medicaid coverage before the ACA. Those caring for YHIV should anticipate and inquire about discontinuity, and policy makers should consider the impact of discontinuity on YHIV when considering changes to the Medicaid program.
Acknowledgments
Dr. Rusley received support from the Adolescent Health Promotion Research Training Program (T32HD052459-07, NIH) and the Leadership Education in Adolescent Health Training Grant (T71MC08054, HRSA). The HIV Research Network is supported by the Agency for Healthcare Research and Quality (HHSA290201100007C), the Health Resources and Services Administration (HHSH250201200008C), the National Institutes for Health (U01 DA036935), and the Clinical Investigation and Biostatistics Core of the UC San Diego Center for AIDS Research (AI036214). This publication was also made possible with help from the Johns Hopkins University Center for AIDS Research, an NIH-funded program (P30AI094189), which is supported by the following NIH Co-Funding and Participating Institutes and Centers: NIAID, NCI, NICHD, NHLBI, NIDA, NIMH, NIA, FIC, NIGMS, NIDDK, and OAR. Dr. Matson was supported on NIDA funded career development award (K01DA035387). The views expressed in this article are those of the authors. No official endorsement by the Department of Health and Human Services, the National Institutes of Health, or the Agency for Healthcare Research and Quality is intended or should be inferred. The authors acknowledge Hoover Adger, Maria Trent, and Krishna Upadhya from the Division of General Pediatrics and Adolescent Medicine at Johns Hopkins University. We gratefully acknowledge the extensive work of the HIV Research Network sites and principal investigators: Alameda County Medical Center, Oakland, California (Howard Edelstein, MD); Children's Hospital of Philadelphia, Philadelphia, Pennsylvania (Richard Rutstein, MD); Drexel University, Philadelphia, Pennsylvania (Amy Baranoski, MD, Sara Allen, CRNP); Fenway Health, Boston, Massachusetts (Stephen Boswell, MD); Johns Hopkins University, Baltimore, Maryland (Kelly Gebo, MD, Richard Moore, MD, Allison Agwu, MD); Montefiore Medical Group, Bronx, New York (Robert Beil, MD); Montefiore Medical Center, Bronx, New York (Uriel Felsen, MD); Mount Sinai St. Luke's and Mount Sinai West, New York, New York (Judith Aberg, MD, Antonio Urbina, MD); Oregon Health and Science University, Portland, Oregon (P. Todd Korthuis, MD); Parkland Health and Hospital System, Dallas, Texas (Ank Nijhawan, MD, Muhammad Akbar, MD); St. Jude's Children's Research Hospital and University of Tennessee, Memphis, Tennessee (Aditya Gaur, MD); Tampa General Health Care, Tampa, Florida (Charurut Somboonwit, MD); Trillium Health, Rochester, New York (William Valenti, MD); University of California, San Diego, California (W. Christopher Mathews, MD). We acknowledge the sponsoring agencies of the HIV Research Network: Agency for Healthcare Research and Quality, Rockville, Maryland (Fred Hellinger, PhD, John Fleishman, PhD, Irene Fraser, PhD); and Health Resources and Services Administration, Rockville, Maryland (Robert Mills, PhD, Faye Malitz, MS). This study was made possible through the extensive support of the staff of the HIV Research Network Data Coordinating Center at Johns Hopkins University (Richard Moore, MD, Jeanne Keruly, CRNP, Kelly Gebo, MD, Cindy Voss, MA, Charles Collins, MPH, Rebeca Diaz-Reyes, MSPH). Preliminary findings of this study were presented at the 2017 Society of Adolescent Health and Medicine Annual Meeting.
Contributor Information
Collaborators: for the HIV Research Network, Howard Edelstein, Amy Baranoski, Sara Allen, Robert Beil, Uriel Felsen, Antonio Urbina, P. Todd Korthuis, Muhammad Akbar, Aditya Gaur, Charurut Somboonwit, William Valenti, W. Christopher Mathews, Fred Hellinger, John Fleishman, Irene Fraser, Robert Mills, Faye Malitz, Jeanne Keruly, Kelly Gebo, Cindy Voss, Charles Collins, and Rebeca Diaz-Reyes
Author Disclosure Statement
No competing financial interests exist.
References
- 1. Lally MA, van den Berg JJ, Westfall AO, et al. . HIV continuum of care for youth in the United States. JAIDS J Acquir Immune Defic Syndr 2018;77:110–117 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Sommers BD. From Medicaid to uninsured: Drop-out among children in public insurance programs. Health Serv Res 2005;40:59–78 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Han X, Zhu S, Jemal A. Characteristics of young adults enrolled through the Affordable Care Act-Dependent coverage expansion. J Adolesc Heal 2016;59:648–653 [DOI] [PubMed] [Google Scholar]
- 4. Yehia BR, Fleishman JA, Agwu AL, et al. . Health insurance coverage for persons in HIV care, 2006–2012. J Acquir Immune Defic Syndr 2014;67:102–106 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Czajka JL. Medicaid Enrollment Gaps, 2005 to 2007, Vol 6. Princeton, NJ: Mathematica Policy Research, 2012. [Google Scholar]
- 6. Sommers BD. Loss of health insurance among non-elderly adults in Medicaid. J Gen Intern Med 2009;24:1–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Rimsza ME, Johnson WG. The effects of access to pediatric care and insurance coverage on emergency department utilization. Pediatrics 2004;113(3 Pt 1):483–487 [DOI] [PubMed] [Google Scholar]
- 8. Rudy BJ, Murphy DA, Harris DR, Muenz L, Ellen J; Adolescent Trials Network for HIV/AIDS Interventions (ATN). Patient-related risks for nonadherence to antiretroviral therapy among HIV-infected youth in the United States: A study of prevalence and interactions. AIDS Patient Care STDS 2009;23:185–194 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Fleishman JA, Monroe AK, Voss CC, Moore RD, Gebo KA. Expenditures for persons living with HIV enrolled in Medicaid, 2006–2010. J Acquir Immune Defic Syndr 2016;72:408–415 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Agwu AL, Lee L, Fleishman JA, Voss C, Yehia BR, Althoff KN. Aging and loss to follow-up among youth living with human immunodeficiency virus in the HIV Research Network. J Adolesc Heal 2015;56:345–351 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Duckett P, Artiga S. Health Coverage for the Black Population Today and Under the Affordable Care Act | The Henry J. Kaiser Family Foundation. Kaiser Family Foundation. Published 2013. Available at: https://www.kff.org/disparities-policy/fact-sheet/health-coverage-for-the-black-population-today-and-under-the-affordable-care-act (Last accessed June14, 2018).
- 12. Grieb SM, Kerrigan D, Tepper V, Ellen J, Sibinga E. The clinic environment as a form of social support for adolescents and young adults living with HIV. AIDS Patient Care STDS 2018;32:208–213 [DOI] [PMC free article] [PubMed] [Google Scholar]