HIV Testing and Care Among Publicly Insured Enrollees With Schizophrenia: Is There a Benefit to Dual Eligibility? Authors: Karly Murphy, Marilyn D. Thomas, Paola Alonso‐Fraire, Alexander R. Bazazi, Priya Dahiya, Mark Olfson, Francine Cournos, Emily Arnold, James Dilley, Paul Wesson, Richard Hermida, James Walkup, Stephen Crystal, Christina Mangurian Published: 2026 ABSTRACT BackgroundPatients with schizophrenia experience gaps in the human immunodeficiency virus (HIV) care continuum. We examined the impact of dual enrollment in Medicare upon HIV testing and retention among Medicaid‐enrolled patients with schizophrenia.MethodsUsing Medicaid Analytic eXtract and Medicare files from 2008 to 2012, we examined adult participants with schizophrenia enrolled in Medicaid or dually enrolled in Medicare and Medicaid. Our outcomes were (a) HIV testing probability among participants without HIV and (b) retention in HIV care among people living with HIV (PLWH) (two HIV viral loads or CD4+ counts >90 days apart and within 1 year).ResultsThe average annual HIV testing probability was 4.9% among dually enrolled and 7.4% among Medicaid‐enrolled patients (p < 0.001). Compared to Medicaid‐only, HIV testing for dually enrolled patients were highest among those with opioid use disorder or sexually transmitted infection (STI); and lowest among study sample aged 50–59 years, Asian or White race. Among PLWH, the average retention probability in care was 54.7% for dually enrolled and 44.7% for Medicaid‐enrolled patients (p < 0.001). Compared to Medicaid‐only, retention probability for dually enrolled patients were highest among those with an STI or dyslipidemia, and lowest among younger sample.Conclusions and RelevanceDual Medicare enrollment for Medicaid‐enrolled patients benefitted HIV retention efforts more than HIV testing efforts. Differences in public insurance coverage between HIV care retention and testing among patients with schizophrenia, emphasizes the need for ongoing evaluation of the HIV care continuum as payor coverage and policies continue to evolve. The human immunodeficiency virus (HIV) care continuum (the HIV Continuum) is a public health model that guides the primary prevention of HIV and treatment of people living with HIV infection (PLWH) (1). HIV testing determines the appropriate treatment pathway: primary prevention for those who are HIV‐negative or treatment to achieve viral suppression for PLWH (1). Guided by the U.S. National HIV/AIDS Strategy (NHAS), the federal plan released by the White House Office of National AIDS Policy, the Centers for Disease Control and Prevention (CDC) prioritizes outreach among historically prioritized, high‐risk and marginalized populations (2). The NHAS identified five priority populations: gay, bisexual, and other men who have sex with men with a focus on Black, Latinx, and American Indian men; Black women, transgender women, youth aged 13–24 years, and people who inject drugs (3). In support of these priorities, the CDC directs funding and other resources and partners with state and local health departments and community‐based organizations to deliver HIV prevention, linkage, and treatment services (3). Consistent with NHAS, the CDC also tracks progress along the HIV Continuum, including reducing new HIV infections by 90% by 2030, increasing knowledge of HIV status to 95% and increasing linkage to care to 95%. In the U.S., the HIV epidemic has disproportionately affected populations with severe mental illness (e.g., schizophrenia, bipolar disorder), with the HIV prevalence being 6% in populations with severe mental illness (4), a rate more than 10‐fold higher than the general population (0.4%) (5). Populations with severe mental illness are at risk of acquiring HIV due to traditional risk behaviors, such as injection drug use or condomless sex, and social and structural risks from sexual violence, housing insecurity, and poverty (6, 7, 8, 9). Alongside increased risk of acquiring HIV, individuals with schizophrenia may face challenges accessing HIV prevention and HIV care continuity. They are more likely to experience bias when accessing health services (10, 11), which erodes trust, reduces uptake of preventive services, and is associated with reduced treatment adherence (12, 13). Populations with schizophrenia are also more likely to experience low socioeconomic status, housing instability, disability, and low health literacy, which create additional challenges accessing health services (14, 15). Finally, behavioral health services, where populations with schizophrenia receive longitudinal care, have historically been delivered separately from physical health services, where HIV care has typically been delivered (16). This separation contributes to fragmented care for an already marginalized and vulnerable population. Prior studies examining care quality along the HIV Continuum for people with schizophrenia have limitations. First, many studies focused on single institutions or individual U.S. state‐level data (17, 18, 19, 20, 21). Second, Medicaid and Medicare are the payers for nearly 90% of individuals with schizophrenia (22). Medicare is a federal health insurance program that primarily covers adults aged 65 years or older and younger individuals with permanent disabilities; Medicaid is a jointly funded federal‐state program that provides coverage for individuals based on income and other factors (23). Individuals may qualify for both programs are considered dual‐eligible beneficiaries and received insurance coverage through both Medicare and Medicaid. Beneficiaries who are dually enrolled in Medicare and Medicaid are more likely to be of younger age, non‐white race, and have lower incomes (24). Yet prior Medicaid‐based studies excluded individuals dually enrolled in Medicare due to unavailability of data (25, 26, 27). This exclusion left out an estimated 28% of U.S. dually enrolled adults with schizophrenia (22). Non‐elderly adults with schizophrenia dually eligible for Medicare and Medicaid based on disability from impairment from mental illness, may face differential challenges in accessing services. Dually enrolled patients receive primary coverage through Medicare and additional benefits through state Medicaid. Medicare also offers higher reimbursement than Medicaid for many healthcare services. Our study fills this knowledge gap by examining differences along the HIV Continuum for Medicaid‐enrolled patients with schizophrenia in the U.S. We examined whether non‐elderly Medicaid‐enrolled patients with schizophrenia differed in HIV testing and care retention compared with those dually enrolled in Medicare. METHODS Study Sample Cohorts were constructed with Medicaid Analytic eXtract and Medicare enrollment and claims files from Centers for Medicare and Medicaid Services as previously described (25, 26, 27). Data was constructed at the person‐year level. The dataset underwent comprehensive cleaning and validation to ensure accuracy and completeness. Due to delays in Medicare data availability, this analysis used data from January 1, 2008 to January 31, 2012. This study was deemed exempt due to use of de‐identified data by the University of California, San Francisco Institutional Review Board (17‐21998). Medicaid‐Enrolled Patients Living With Schizophrenia Study participants were Medicaid beneficiaries ages 20–64 years, living in one of 45 states, and at least 11 months of eligibility during the calendar year. Schizophrenia diagnoses were identified by having (a) one or more inpatient claims for schizophrenia (ICD‐9 295.x) or (b) two or more uniquely dated outpatient claims for schizophrenia within a 6‐month period (25, 26, 27). Individuals were excluded if they had a documented HIV diagnosis (ICD‐9 042, 079.53, V08, 795.71) in the 12 months prior to the index year. Individuals diagnosed with HIV were excluded in all subsequent years. Those without HIV were included in subsequent years' analyses. PLWH and Schizophrenia We identified a separate cohort of Medicaid beneficiaries living with HIV and schizophrenia. PLWH were identified if they had at least one documented HIV diagnosis (ICD‐9 042, 079.53, V08, 795.71) in the prior 12 months of the index year. Exposure Our exposure was dual enrollment in Medicare (28, 29). Medicare enrollment was measured by months of coverage, with individuals assigned dual eligibility if they were enrolled in Medicare for at least one calendar month (28, 29). Outcome Measures HIV Testing For individuals eligible for HIV testing, HIV testing was assessed as a binary outcome (yes/no) for each calendar year using CPT codes (26). Retention in HIV Care For PLWH, retention in HIV care was a yearly binary outcome (yes/no) for each calendar year, defined as ≥2 HIV viral load or CD4+ T‐cell count (CPT codes 87536, 86360, 8631) ≥90 days apart, based on the CDC definition of retention in HIV care (30). In a sensitivity analysis, we examined two less stringent measures for each calendar year: (a) ≥1 HIV viral load or CD4+ T‐cell test completed or (b) ≥2 outpatient visits at least ≥90 days apart (30). Covariates We identified a priori covariates that may confound the relationship between insurance and outcomes of interest. Selected covariates were coded categorically and included age (20–29, 30–39, 40–49, 50–59, 60–64 years); sex (male, female), and race/ethnicity (Non‐Hispanic Black, American Indian/Alaska Native, Asian, Hispanic/Latino, Native Hawaiian/Pacific Islander, Multiracial, Unknown, Non‐Hispanic White), and calendar year. We included substance use disorder (alcohol, opioids, cocaine, amphetamine, cannabis, or other drug use disorder) and sexually transmitted infections (STIs; herpes simplex virus, gonorrhea, chlamydia, syphilis) diagnoses. We also included chronic infectious disease of hepatitis B (HBV) and hepatitis C (HCV) and chronic medical conditions of diabetes, dyslipidemia, and hypertension associated with healthcare utilization (31, 32, 33). All used at least one ICD‐9 billing codes (Appendix A) (25, 26, 27). Statistical Analyses We used frequencies to describe the demographics and comorbid conditions by insurance status. At the person‐year level, we estimated associations between insurance status and the likelihood of (a) receiving HIV testing among people without HIV and (b) retention in HIV care among PLWH. We calculated average annual predicted probabilities using multivariable logistic regression adjusted for patient‐level demographics (age, sex, race/ethnicity) and comorbidities (substance use disorder, STIs, chronic infectious diseases, chronic medical conditions). We estimated two‐sided associations with robust standard errors to account for individuals enrolled in multiple years. Covariate categories were collapsed for race and ethnicity (5‐level) and any STI (yes/no). In a secondary analysis, we included state as a fixed effect in the adjusted multivariable logistic regression because Medicare enrollment is based on a national policy. Statistical analyses were performed using STATA 17 (College Station, TX). RESULTS Sample Characteristics HIV Testing for People Living With Schizophrenia but Without HIV The study sample eligible for HIV testing included 2,707,205 unique individuals, of which 1,261,564 (46.6%) were dually enrolled and 1,445,641 (53.4%) were Medicaid‐only (Table 1). Of those dually enrolled, 95.3% were dually enrolled all 12‐months of the calendar year (Appendix A). Dually enrolled patients were older and more likely to be male than Medicaid‐only patients (60.9% vs. 52.5%). Dually enrolled patients were less likely to be Black (25.1% vs. 36.3%) or Hispanic (5.7% vs. 10.6%) compared with Medicaid‐only patients. More dually enrolled patients had hypertension (45.3% vs. 37.3%), dyslipidemia (42.4% vs. 26.5%), or diabetes (29.2% vs. 22.9%); but fewer had opioid (2.2% vs. 3.7%), cocaine (3.6% vs. 5.4%), cannabis (4.6% vs. 5.6%), or other substance use disorder (7.9% vs. 8.7%), or chlamydia (0.1% vs. 0.2%) and syphilis (0.9% vs. 0.16%) compared with Medicaid‐only patients. TABLE 1Characteristics of unique individuals with schizophrenia in the analytic sample, stratified by insurance enrollment (dually enrolled in Medicare and Medicaid vs. exclusively in Medicaid). a Persons without HIVPersons living with HIVDual (N = 1,261,564)Medicaid (N = 1,445,641)Dual (N = 10,698)Medicaid (N = 32,049) n % n % n % n %DemographicsAge20–29103,8368.23238,14617.174614.3120986.5530–39229,29518.18264,58418.30161815.12489315.2740–59376,69929.86371,60125.70434040.5712,75039.7850–59429,70234.06429,78229.73368434.4410.6080.0360–65122,0329.67131,5289.675955.5617005.30SexFemale492,73039.06686,82647.51344432.1914,96346.69Male768,83460.94758,81552.49725467.8117,08653.31Race/ethnicityBlack316,69425.10524,79336.30596455.7519,81661.83Amer. Indian/Alaskan Native83770.6611,9690.83600.561270.40Asian15,4181.2229,1532.02400.371370.43Hispanic or Latino71,7965.69152,55410.5510609.91462114.42Native Hawaiian/Pacific Isl.98860.7823,2351.61390.361000.31Multiracial28920.2343060.30520.491570.49Unknown95,0107.53123,9168.575765.3819366.04White741,49158.78575,71539.82290727.17515516.08ComorbiditiesSubstance use disorderAlcohol126,52610.03153,30910.60223420.88822625.67Opioid28,0942.2353,9473.739138.53528816.50Cocaine44,9903.5778,6785.44197018.41831725.95Amphetamine12,7341.0116,4001.132472.315881.83Cannabis58,2684.6280,7195.58107910.09377311.77Other drug100,0097.93125,9428.71243922.80826725.79General medicalHepatitis B66410.5390510.636936.4818565.79Hepatitis C39,4033.1259,8504.14240022.43807425.19Diabetes mellitus368,68029.22330,85622.89314529.40794824.80Hypertension571,39245.29538,50437.25545450.9814,57645.48Dyslipidemia534,22042.35383,21226.51436340.78765823.89STIHerpes simplex12,6791.0114,1370.985725.3517975.61Chlamydia15630.1229460.20580.541590.50Gonorrhea32960.2643420.30500.471510.47Syphilis11330.0922470.164003.749022.81 a PLWH = persons living with HIV; STI = sexually transmitted infection. Characteristics of unique individuals with schizophrenia in the analytic sample, stratified by insurance enrollment (dually enrolled in Medicare and Medicaid vs. exclusively in Medicaid). a PLWH = persons living with HIV; STI = sexually transmitted infection. HIV Retention for PLWH and Schizophrenia Our analytic sample of PLWH and schizophrenia included 42,747 individuals, of which 10,698 (25.0%) were dually enrolled and 32,049 (75.0%) enrolled in Medicaid‐only (Table 1). Dually enrolled patients were older and more likely to be male compared with Medicaid‐only patients (67.8% vs. 53.1%). Fewer dually enrolled patients were Black (55.8% vs. 61.8%) or Hispanic (9.9% vs. 14.4%) compared with Medicaid‐only patients. More dually enrolled patients had diabetes (29.4% vs. 24.8%), hypertension (51.0% vs. 45.5%), or dyslipidemia (40.8% vs. 23.9%), and fewer had alcohol (20.9% vs. 25.7%), opioid (8.5% vs. 16.5%), and cocaine use disorder (18.4% vs. 27.0%), when compared to Medicaid‐only patients. HIV Testing The average annual HIV testing predicted probabilities were 7.40% (95% CI: 7.36–7.44) among Medicaid‐only patients and 4.90% (95% CI: 4.86–4.94) among dually enrolled patients (p < 0.001). On average, the adjusted percent point difference was 1.40 (95% CI: 1.35–1.53) higher for Medicaid‐only versus dually enrolled sample. Annual testing probabilities remained higher among Medicaid‐only enrolled patients (vs. dually enrolled) across years (Figure 1A). Adjustment for state did not change observed trends (Appendix A). A. Trends in HIV testing rates among adults with schizophrenia dually enrolled Medicare and Medicaid individuals compared with adults with schizophrenia enrolled in Medicaid alone. B. Trends in retention in HIV care (at least 2 CD4 or viral load tests, at least 90 days apart) among adults with schizophrenia dually enrolled Medicare and Medicaid compared with adults with schizophrenia enrolled in Medicaid alone. HIV = human immunodeficiency virus. This differential pattern of HIV testing by insurance type persisted across subgroups (Figure 2, Appendix A). Individuals diagnosed with a STI in a calendar year had the highest HIV annual testing probability (Medicaid‐only: 31.7% [95% CI: 31.10–32.37] vs. Dual: 23.3% [95% CI: 22.61–23.95]). Similarly, annual HIV testing probability were highest among individuals enrolled in Medicaid‐only with alcohol (12.3% [95% CI: 12.14–12.49] vs. 8.8% [95% CI: 8.60–8.94]); opioid (16.6% [95% CI: 16.25–16.93] vs. 12.7% [95% CI: 12.27–13.10]), cocaine (16.9% [95% CI: 16.62–17.19] vs. 13.5% [95% CI: 13.13–13.82]); amphetamine (13.9% [95% CI: 12.04–13.30] vs. 12.7% [95% CI: 13.36–14.47]), cannabis use disorder (13.4% [95% CI: 13.15–13.66] vs. 10.4% [95% CI: 10.09–10.63]), or other drugs (15.2% [95% CI: 15.02–15.45] vs. 11.3% [95% CI: 11.08–11.51]) than for those dually eligible. Within demographic subgroups, across Medicaid‐only and dually enrolled patients, annual HIV testing probability were higher among younger study sample (ages 20–29 years 10.3% [95% CI: 10.20–10.47] vs. 8.3% [95% CI: 8.12–8.48]; ages 30–39 years: 8.8% [95% CI: 8.70–8.95] vs. 6.8% [95% CI: 6.72–6.96]), females (8.8% [95% CI: 8.75–8.91] vs. 5.9% [95% CI: 5.85–6.01]), and non‐White Black (9.7% [95% CI: 9.65–9.84] vs. 7.3% [95% CI: 7.18–7.39]), Hispanic or Latino (9.6% [95% CI: 9.40–9.74] vs. 7.4% [95% CI: 7.14–7.59]), and Multi‐racial (10.1% [95% CI: 9.10–11.15] vs. 6.6% [95% CI: 5.54–7.67]) study sample. HIV testing rates across covariates among dual Medicare‐Medicaid enrolled patients with schizophrenia and Medicaid‐only enrolled patients with schizophrenia. AMP = amphetamine; Any STI = herpes simplex, chlamydia, gonorrhea, syphilis; DLD = dyslipidemia; DM = diabetes mellitus; Hep B = hepatitis B; Hep C = hepatitis C; HIV = human immunodeficiency virus; HS or Latino = Hispanic or Latino; HTN = hypertension; Other = Other Drug. Retention in HIV Care The average probability of retention in HIV care was 54.7% (95% CI: 53.5–56.0) among dually enrolled PLWH and 44.7% (95% CI: 44.0–45.4) among Medicaid‐only enrolled PLWH (p < 0.001). The average adjusted percent point difference was 8.1 (95% CI: 6.7–9.6) higher for the dually enrolled sample than Medicaid‐only sample. Retention remained higher for dually enrolled patients (vs. Medicaid‐only) across years (Figure 1B). Similar patterns were observed for secondary outcomes with less stringent measures of HIV retention in care and adjustment by state (Appendix A). Differential probabilities of retention by insurance type were observed across most PLWH subgroups (Figure 3, Appendix A). Retention in HIV care was lowest among groups ages 20–29 years (Dual: 39.0% [95% CI: 33.91–44.18] vs. Medicaid: 32.2% [95% CI: 29.86–34.59]) and 30–39 years (48.3% [95% CI: 45.34–51.20] vs. 40.1% [95% CI: 38.42–41.73]). The greatest percent point differences in retention rates between dual enrolled and Medicaid‐only patients were observed among those with substance use disorders: alcohol (51.2% [95% CI: 48.70–53.62] vs. 43.6% [95% CI: 37.29–39.73]); opioid (53.8% [95% CI: 49.97–57.59] vs. 38.8% [95% CI: 37.27–40.38]); amphetamine (55.1% [95% CI: 48.30–61.82] vs. 41.8% [95% CI: 37.42–46.26]), cannabis (49.2% [95% CI: 47.62–52.79] vs. 36.5% [95% CI: 39.50–41.97]), and other drug use (53.1% [95% CI: 50.71–55.40] vs. 40.6% [95% CI: 39.40–41.86]). Similarly larger differences in retention were observed among those with HBV (59.2% [95% CI: 54.77–63.56] vs. 45.6% [95% CI: 43.02–48.26]), HCV (61.8% [95% CI: 59.40–64.10] vs. 46.9% [95% CI: 45.61–48.27]), diabetes (55.3% [95% CI: 53.10–57.49] vs. 42.8% [95% CI: 41.39–44.12]), hypertension (53.9% [95% CI: 52.21–55.50] vs. 42.9% [95% CI: 41.94–43.90]), dyslipidemia (61.4% [95% CI: 59.67–63.22] vs. 48.5% [95% CI: 47.17–49.85]), or STI (60.9% [95% CI: 57.51–64.33] vs. 45.3% [95% CI: 43.33–47.32]). Rates of retention in HIV care (at least two CD4 or viral load tests, at least 90 days apart) among dual Medicare‐Medicaid enrolled patients with schizophrenia and Medicaid‐only enrolled patients with schizophrenia. AMP = amphetamine; Any STI = herpes simplex, chlamydia, gonorrhea, syphilis; DLD = dyslipidemia; DM = diabetes mellitus; Hep B = hepatitis B; Hep C = hepatitis C; HIV = human immunodeficiency virus; HS or Latino = Hispanic or Latino; HTN = hypertension; Other = Other Drug. DISCUSSION In this study of the HIV continuum, public insurance status was differentially associated with HIV testing and care retention among Medicaid enrollees with schizophrenia. First, we observed higher probabilities of HIV testing among Medicaid‐only enrolled study sample compared to dually enrolled study sample (Medicaid and Medicare). Second, we observed that dually enrolled PLWH with schizophrenia had higher retention in HIV care probabilities than Medicaid‐only enrolled PLWH with schizophrenia. Notably, differences in outcomes along the HIV continuum by insurance status persisted across individual subgroups, adding robustness to our findings. Our study extends prior work focused on HIV care quality among Medicaid beneficiaries (25, 26, 27, 34) to include populations with schizophrenia dually enrolled in Medicare and Medicaid. In 2012, HIV testing rates were 5.8% among Medicaid beneficiaries without schizophrenia and 7.2% among Medicaid beneficiaries with schizophrenia (26). In our study, the 2012 HIV testing rate for dual enrollees with schizophrenia 5.9%. Similarly in 2012, retention in HIV care was 39%–51% among Medicaid beneficiaries without schizophrenia and 30% among Medicaid beneficiaries with schizophrenia (25, 34). In our study, retention in 2012 among dual enrollees with schizophrenia was 52.4%. Our findings also confirm studies demonstrating lower retention in care among younger adults, females, and non‐Hispanic Black populations (35, 36). Our findings have important financial and public health implications for HIV testing and continuity of care among individuals with schizophrenia. Medicaid‐enrolled patients with schizophrenia (with and without HIV) have nearly double the annual costs of care compared to those without mental illness, with even greater costs for those dually enrolled in Medicare (37). Given the substantial healthcare expenditures associated with schizophrenia, improving early detection of HIV through testing and HIV continuity of care may represent an opportunity to reduce downstream costs associated with delayed diagnosis and untreated HIV infection, and HIV‐related complications (38). We also observed Medicaid‐only enrolled patients were younger, of non‐White race or ethnicity, with higher probabilities of substance use disorders and lower rates of chronic conditions compared with dual enrollees. The observed differences may reflect underlying factors associated with insurance status. Dual enrollment may reflect age, disability, psychiatric and medical complexity, and other unmeasured social and structural barriers. The variation across insurance groups highlights the importance of considering insurance status when developing and testing future interventions, as reimbursement structures and care delivery processes may affect their scalability and sustainability. While differential lower testing rates do not necessarily indicate that a group is receiving a deficit of testing, our study raises concern that HIV early detection efforts may miss high‐risk patients with schizophrenia. Although the overall difference across groups was small (1.4 percent points), disparities were larger (3–8 percentage points) among those with HIV‐related risk factors (e.g., substance use disorder, STIs) in the dually enrolled sample. Moreover, HIV testing remained suboptimal among high‐risk individuals with schizophrenia across both insurance groups. Among those diagnosed with an STI, less than one in four dually enrolled patients received HIV testing; and only one in three for Medicaid‐only enrolled patients. It is not clear why individuals were not also tested for HIV, despite a similar mode of transmission. Differential engagement may be driven by individual‐level factors, such as disability, impaired executive function, or perceived stigma (39, 40). Lower HIV testing rates among dually enrolled patients may be due to misconceptions concerning patient risk, not including HIV testing as part of routine care, and competing clinical priorities (41, 42). Structural barriers, such as transportation or access to care and logistical challenges, such as obtaining a blood‐based HIV screening test (vs. urine test for STI) may contribute. Efforts to improve HIV testing will require systematic and community‐based approaches to reach patients with schizophrenia where they are receiving chronic disease or mental health services. Given the above challenges, the higher HIV care retention probabilities observed among dually enrolled sample may be counter‐intuitive. One possibility is that dual enrollees with schizophrenia and HIV were preferentially enrolled in Medicaid managed care plans, such as the Dual Eligible Special Needs Plans (D‐SNPs) (43, 44). Beginning in 2006, D‐SNPs sought to integrate Medicare and Medicaid prescription drug benefits and behavioral health and coordination services (44). While we are unable to determine the study sample enrolled in D‐SNPs, older individuals, those with disability as the reason for Medicare enrollment, and socioeconomic disadvantaged groups were more likely to be enrolled in these programs (44). Dual enrollment status may reflect these care management efforts or more extensive coverage and reduction of pocket expenditures not covered by Medicaid (45). Another possibility is that dual enrollees with HIV and schizophrenia had obtained permanent disability status through schizophrenia or HIV, with assistance from family and/or social services. This same support translated to attendance at medical appointments and retention in HIV care (46, 47). Similarly, this group may have had greater care seeking behavior and a higher likelihood of maintaining dual eligibility with re‐enrollment in Medicaid (48, 49). Our study has limitations. First, our study data are from more than a decade ago due to the lengthy processing time for data availability from both Medicare and Medicaid. While Medicaid policy has evolved since 2012, our study is the first to examine HIV testing and care continuity among dually enrolled adults with schizophrenia. Given that approximately 90% of individuals with schizophrenia have public insurance, of whom 26% are dually enrolled in Medicare and Medicaid (22), our study, spanning 45 states, provides initial insight as to how to move forward for HIV testing and continuity of care. Although ART treatment regimens and Medicaid policies have evolved since the study period, our findings establish an important baseline for understanding where gaps in HIV testing and care continuity existed and for future evaluations of whether these gaps persist today. Our findings should only be interpreted within the policy and clinical context of the study period, and contemporary analyses are needed to assess whether these associations have evolved. Second, we restricted the study population to individuals with 11 months of eligibility during the calendar year (25, 26, 27). This approach may bias the Medicaid‐only group towards individuals who have a more stable engagement with healthcare services. Our study's observations may not extend to individuals without stable insurance. Our results may be susceptible to survival bias. However, we attempted to minimize this potential bias by conducting the analysis as a multiple cross‐sectional design. Individuals who died were not included in subsequent years. Similarly, we defined the exposure as at least 1 month of Medicare enrollment in a calendar year, which could lead to misattribution of events and person‐time. However as >95% of individuals were enrolled in Medicare for the entirety of the calendar year, we expect any residual bias to be minimal. Third, Medicaid and Medicare data do not provide information on health literacy, housing, or family support, which may contribute to the ability to navigate the healthcare system. Fourth, we did not track antiretroviral medications for primary HIV prevention. Fifth, our study is observational in nature. Residual confounding likely exists, which may influence both the insurance status and the outcome. Finally, Medicaid and Medicare claims data do not provide information on CD4 counts, viral loads, or whether viral suppression was achieved. While, HIV care retention is associated with viral suppression, futures studies should examine receipt of antiretroviral prescriptions or levels of viral suppression. This study examined care quality along the HIV care continuum using a national cohort of publicly insured individuals with schizophrenia from Medicaid and Medicare. Major differences were found in HIV testing and retention based on type of public insurance, suggesting that clinical, structural, and social factors contribute to challenges in HIV care for this population. Given the evolving landscape of Medicaid expansion and recent unwinding, changing eligibility requirements and regional differences in D‐SNP programs (50, 51), future observational studies should assess how policy and programmatic changes affect HIV care quality, including testing and continuity of care, for adults with schizophrenia. Simultaneously, public health outreach must focus on population‐based efforts to improve HIV testing within traditional clinical settings. Future implementation work is needed to support systemic and community‐based interventions to improve retention in care for PLWH.