In the United States, between 2010 and 2014 the rate of people living with HIV steadily increased from 275.7/100,000 population to 299.5/100,000 population, respectively. In 2010, the rate of HIV diagnoses was 14.2/100,000 population and the rate declined to 12.3/100,000 population in 2015 [1]. The highest rates of new diagnoses in 2015 were among blacks/African Americans (hereafter referred to as blacks) (44.3/100,000 population), more specifically black males (84.8/100,000 population), blacks aged 20-24 years (111.2/100,000 population) and blacks 25-29 years (112.2/100,000 population) [1]. Hispanics/Latinos also experienced elevated rates of HIV diagnoses over whites although the rates remained at approximately one-third to one-half the rates of blacks with the same sex and age characteristics. The vast majority of males with HIV infection diagnosed in 2015 had infection attributed to male-to-male sexual contact or male-to-male sexual contact in conjunction with injection drug use (86.2%); females had infection attributed primarily to heterosexual contact (86.3%) [1]. For these and other risk groups, prevention services are essential to avoid further spread of HIV infection and, for those living with HIV infection, coordinated supportive services and medical care are crucial [2].

Goals of the National HIV/AIDS strategy for the United States: Updated to 2020 [3] include reducing new HIV infections, increasing access to care and improving health outcomes for people living with HIV infection, reducing HIV-related disparities and health inequities, and achieving a more coordinated national response to the HIV epidemic. Progress toward these goals by 2020 includes: increasing to 85% the number of newly diagnosed persons who are linked to HIV medical care within one month of diagnosis, increasing to 90% the number of persons living with diagnosed HIV who are retained in HIV medical care, increasing to 80% the number of persons diagnosed with HIV who have suppressed viral load, and reducing disparities by focusing on high risk groups comprised of young black gay and bisexual males and black females [4] for whom disparities in HIV diagnoses and treatment outcomes have been demonstrated [5-8].

Data from the National HIV Surveillance System, described elsewhere [9, 10], show that for people with HIV infection diagnosed in 2015 in 37 states and the District of Columbia, 75.0% were linked to care within one month of diagnosis and for those with HIV diagnosed by year-end 2013 and alive at year-end 2014, 72.5% were retained in care and 57.9% had suppressed viral load [11]. All three indicators fall short of national goals. In 2014, for 32 states and the District of Columbia, outcomes for these three indicators were even lower for black males (69.9%, 67.0%, and 47.9% respectively), males aged 18-24 years (66.2%, 71.4%, and 44.4% respectively), and people who inject drugs (74.3%, 64.3%, and 47.1% respectively) [12].

Assessing changes in linkage to care, received care, and viral suppression over the last decade has been hindered by a lack of comprehensive HIV laboratory reporting laws that require reporting of all CD4 results regardless of value and all viral load results including detectable and undetectable; and by incomplete laboratory reporting within states with comprehensive reporting laws [13]. This study examined annual changes for 2012 to 2015 in linkage to care, receipt of care, and viral suppression among adults and adolescents using data reported for 17 jurisdictions to the National HIV Surveillance System. These jurisdictions represented approximately 46% of the U.S. population. We aimed to explore these changes by patient characteristics, stage of HIV infection at diagnosis and transmission category.

Data reported to the National HIV Surveillance System by December 2016 were used to identify cases of HIV infection that met the CDC HIV infection case definition [14] among persons aged 13 years and older at time of diagnosis and whose residence at diagnosis (for linkage to care) and most current residence (for receipt of care and viral suppression) were within one of 17 identified jurisdictions with complete laboratory reporting. To be eligible for inclusion jurisdictions were required to meet the following three criteria for each year between 2012 and 2015:

• The jurisdiction’s laws/regulations required the reporting of all CD4 and viral load results to the state/city health department,
• Laboratories that perform HIV-related testing for the areas must have reported a minimum of 95% of HIV-related test results to the state/city health department, and
• By December 31, 2016, the area had reported to CDC at least 95% of all CD4 and viral load test results received from January 2012 through December 2015.

The jurisdictions that met inclusion criteria were California, District of Columbia, Hawaii, Iowa, Illinois, Indiana, Louisiana, Maryland, Michigan, Missouri, New Hampshire, New York, North Dakota, South Carolina, Texas, Utah, and West Virginia.

Linkage-to-care analysis included people with HIV infection diagnosed between January 1^st and December 31^st of the outcome year (e.g., the outcome for 2012 includes all diagnoses between January 1, 2012 and December 31, 2012). Linkage to care was defined as ≥1 CD4 or viral load test performed within 1 month of diagnosis.

Receipt of care and viral load suppression analyses included people with HIV diagnosed before January 1^st of the outcome year and not known to be deceased on December 31^st of the outcome year (e.g., the result for 2012 includes all persons with an HIV diagnosis before January 1, 2012 and not known to be deceased on December 31, 2012). Receipt of care was defined as ≥1 CD4 or viral load test performed during the outcome year. Viral suppression was defined as a viral load result of <200 copies/mL or, if the quantitative value was missing, a test interpretation value of “undetected”, at the time of the most recent viral load test during the outcome year.

Laboratory results with missing month or year of specimen collection were excluded from the analysis (<0.36% for linkage-to-care and <0.16% for receipt of care and viral suppression). All duration times were calculated using the month and year for both HIV infection diagnosis and laboratory results. If a patient had two tests in the same month with different viral suppression results, we applied a conservative approach and used the test result that indicated a higher viral load. In addition, laboratory tests with a missing result and tests with specimen collection dates prior to the date of HIV infection diagnosis were excluded.

The twelve-month interval between the last observation year (2015) and dataset used (reporting through December 2016) allowed time for reporting of diagnoses, laboratory results, and deaths. Data were adjusted for unknown or missing transmission category [15]. Results are presented by age group (13-24, 25-34, 35-44, 45-54, ≥55), sex, race/ethnicity (black, Hispanic/Latino, white, and other) (those with missing race/ethnicity were excluded from the analysis), and transmission category (male-to-male sexual contact, people who inject drugs, male-to-male sexual contact and injection drug use, heterosexual contact, and other). Stage of disease at diagnosis was categorized as diagnosis of HIV-infection Stage 3 [AIDS] within 3 months of an HIV diagnosis, or the absence of HIV-infection Stage 3 [AIDS] diagnosis within 3 months of an HIV diagnosis. The estimated annual percent change (EAPC) was calculated for each person characteristic and considered statistically significant at P-value <.05. All analyses were conducted using SAS 9.3 (SAS Institute, Inc., Cary, NC). This project was approved by CDC as a retrospective, secondary data analysis using HIV surveillance data. This analysis did not constitute research involving identifiable human subjects requiring IRB review.

Linkage to care among the 81,174 (2012=20,876; 2013=19,887; 2014=20,467; and 2015=19,944) people with HIV infection diagnosed between January 1, 2012 and December 31, 2015 increased each year from 2012-2015 (79.9%, 81.4%, 82.9%, and 83.7% respectively, EAPC = 1.6, 95% CI 0.9-2.3, P<.001) (Table 1). In 2015, linkage to care was higher among whites compared with all other races/ethnicities, older age groups compared with younger age groups, and those with HIV-infection Stage 3 [AIDS] compared with those without stage 3 [AIDS] diagnosis. Linkage was also higher among women and those with infection attributed to heterosexual contact. Linkage to care increased significantly from 2012-2015 among males (EAPC=1.6, 95% CI 0.9-2.4, P<.001), blacks (EAPC=2.2, 95% CI 1.1-3.3, P<.001), those with HIV diagnosed earlier than stage 3 (AIDS) (EAPC=2.4, 95% CI 1.6-3.2, P<.001), males with infection attributed to male-to-male sexual contact (EAPC=1.6, 95% CI 0.7-2.4, P<.001), those aged 13-24, 25-34, and 35-44 years at diagnosis (EAPC=3.2, 95% CI 1.7-4.7, P<.001, EAPC=1.3, 95% CI 0.1-2.5, P=.039, and EAPC=1.7, 95% CI 0.2-3.2, P=.031 respectively), and those with a transmission category of heterosexual contact (EAPC=1.8, 95% CI 0.4-3.3, P=.014).

By race/ethnicity and sex, linkage to care increased among black males overall (EAPC=2.5, 95% CI 1.2-3.8, P<0.001), and specifically among black males aged 13-24 years (EAPC=3.4, 95% CI 1.1-5.8, P=0.004), those in earlier than stage 3 of the disease at diagnosis (EAPC=3.8, 95% CI 2.2-5.3, P<0.001) and those with infection attributed to male-to-male sexual contact (EAPC=2.4, 95% CI 1.0-3.9, P=0.001) (Table 2). Black females showed no significant improvements in linkage to care from 2012-2015 with the exception of those with HIV-infection diagnosed without stage 3 [AIDS] (P=0.045). Similarly, linkage to care among Hispanic/Latino males aged 13-24 years (EAPC=3.2, 95% CI 0.3-6.2, P=0.033) and those with HIV infection without a diagnosis of stage 3 [AIDS] (EAPC=2.1, 95% CI 0.5-3.8, P=0.010) increased. There were no significant increases in linkage to care from 2012-2015 among Hispanic/Latino females or white males or females.

Overall, the proportion of people with diagnosed HIV infection who received care increased from 2012 to 2015 (EAPC=0.6, 95% CI 0.4-0.8, P<0.001) (Table 3). Receipt of care increased significantly for all individual categories of person characteristics assessed from 2012 to 2015, with the exception of those identifying as other race/ethnicity and other transmission category, those aged 45-54 years, and people who inject drugs. In each of the four years, the proportion of people who received care was consistently higher among whites and those classified as other race/ethnicity compared to blacks and Hispanics/Latinos and consistently lowest among those with a transmission category of injection drug use. There was little variation among age groups within years.

By race/ethnicity and sex, significant increases in the proportion of those who received care annually from 2012-2015 were found among black males of all ages except ≤55 years, Hispanic/Latino males aged 13-24 and 25-34 years, and all racial/ethnic groups diagnosed without Stage 3 [AIDS] except Hispanic/Latino and white females (Table 4). Receipt of care increased from 2012-2015 among all racial/ethnic groups of men who have male to male sexual contact (blacks EAPC=1.0, 95% CI 0.6-1.4, P<.001; Hispanics/Latinos EAPC=0.8, 95% CI 0.4-1.2, P<.001; whites EAPC=0.5, 95% CI 0.1-0.8, P=.004). Receipt of care also increased among black females aged 35-44 years (EAPC=1.0, 95% CI 0.2-1.8, P=.020) and transmission category of heterosexual contact (EAPC=0.7, 95% CI 0.2-1.2, P=.008).

Viral suppression among people with diagnosed HIV infection increased annually from 2012 to 2015 overall (53.4%, 56.4%, 58.8%, and 59.2% respectively) (EAPC=3.5, 95% CI 3.3-3.7, P<.001) (Table 5) and for all categories of person characteristics assessed. Within years, the proportion of people achieving viral suppression was consistently greater among males, whites, and people with HIV infection Stage 3 [AIDS], and consistently lowest among those with a transmission category of injection drug use. The proportion of those achieving viral suppression increased with age each year.

By race/ethnicity and sex, viral suppression increased annually from 2012-2015 for all characteristics except transmission category of other for all racial/ethnic groups, and Hispanic/Latino and white females aged 13-24 years (Table 6).

This analysis explored progress in linkage to care, receipt of care, and viral suppression among people with HIV infection for outcome years 2012-2015 in 17 U.S. jurisdictions meeting the criteria for complete laboratory reporting for each of the four years. Linkage-to-care increases were found among some high-risk populations, particularly young people, blacks, males, and those engaging in male-to-male sexual contact. However, an increase was found in only one category of black women, those with a diagnosis of HIV Stage 3 [AIDS], despite black women being a priority population for HIV prevention. Notably, Hispanic/Latino and white females and white males showed no improvement in linkage to care from 2012-2015; however, the proportions at which they were linked to care were close to or higher than the national standard and exceeded linkage for black males by approximately five percentage points.

Generally, receipt of care improved among the same populations as linkage to care and also showed improvements for white men who have male to male sexual contact. Once again, no improvements were found among Hispanic/Latino or white women. Despite limited improvement in linkage to and receipt of care, and similar to another study [16], increases in viral suppression were seen in nearly every category of person characteristics assessed. This is potentially due to a number of factors including improvements in linkage to care and receipt of care that were found and increased prescribing of or compliance with antiretroviral therapy.

The distinction in linkage to and receipt of care and viral suppression between those with HIV diagnosed at stage 3 [AIDS] and not at stage 3 [AIDS] is prominent. Those with stage 3 [AIDS] are linked to care in greater proportions than those not in stage 3 [AIDS]. However, the differences between the two groups disappear for receipt of care. Then, once again, those with HIV diagnosed at Stage 3 [AIDS] achieve viral suppression more often than those not at stage 3 [AIDS]. These variations suggest the need to better understand why those with stage 3 [AIDS] are linked but did not receive care in any greater proportion than those not at stage 3 [AIDS].

A similar scenario is seen for linkage to and receipt of care and viral suppression by age group. While linkage to care and viral suppression both increase as age increases, this is not the case for receipt of care where the proportions vary within age categories. A better understanding of the impact of age on receipt of care could be important in increasing the proportion of those virally suppressed for all age groups.

Studies documenting the impact of poverty, poor education, substance use, mental health challenges, domestic violence, transportation to medical care, and lack of social support and employment on HIV care and treatment are abundant, and demonstrate disproportional impact on black men and women and Hispanics/Latinos, and on high-risk populations such as people who inject drugs and young men having sexual contact with men [7, 17-23]. Despite these barriers a number of evidence-based programs targeting these populations, including pre-exposure prophylaxis, demonstrate effective outcomes [24-26]. However, within years substantial differences remain in linkage to and receipt of care and viral suppression with whites exceeding blacks and Hispanics/Latinos, suggesting the effectiveness of programs to reduce the health disparities are limited [27].

The analysis was subject to several limitations. First, the outcomes assessed included only cases identified through 2015 and were under the guidance of the National HIV/AIDS Strategy for the United States: July 2010 [28]. The 2010 guidance established a linkage to care goal of 65% which was met for virtually every person characteristic group assessed and a retention in care goal of 80% for which our more liberal definition showed no person characteristic group met the goal. The 2010 guidance also set a goal of a 20% increase in the proportion of HIV diagnosed gay and bisexual men as well as Blacks with undetectable viral load which we cannot assess as we used 2012-2015 data. Improvements in testing and care and treatment since 2015 make it critical to continue to assess these outcomes under the new 2020 strategy released in July 2015. Second, the 17 jurisdictions may not be representative of all people with HIV infection in the United States. To mitigate the lack of representation we looked back no further than 2012. Including earlier years would have further reduced the number of states eligible for the analysis. Third, the EAPC for linkage to care is based on only four years of data due to the lack of complete laboratory reporting in other jurisdictions. Fourth, documentation of the most recent viral load may not be indicative of consistent viral suppression in this population over time [29] and further studies are needed to understand factors contributing to long term viral suppression. Fifth, exclusion of laboratory results with missing month or year of specimen collection date may underestimate linkage-to-care, receipt-of-care, and viral suppression. To address the majority of the limitation above, states continue to work with their legislatures to enact mandatory HIV-related laboratory test result reporting laws and all but six (Idaho, Kansas, New Jersey, Pennsylvania, Vermont, and the Virgin Islands) have now done so with the latest being Arizona in 2018. Opportunities to expand analyses will occur as states collect this data. Additional studies are needed as more jurisdictions begin to meet the laboratory reporting requirements.

Improving care and treatment for people with HIV infection and reducing HIV-related disparities across the three indicators studied show some promising results; however, linkage-to-care and viral suppression indicators fall short of National HIV/AIDS Strategy for the United States: Updated to 2020 goals. While not a national indicator, the more inclusive definition used in this study for receipt of care (one HIV-related medical visit per year) still falls short of the retention in care (two or more HIV-related medical visits more than three months apart in a year) national goal and thus there also remain opportunities for improvement in receipt of care.

Prevention programs that are scientifically proven, cost-effective, scalable and targeted to high-risk populations are the foundation for stopping transmission of HIV infection. Frequent testing to support early diagnosis and prompt linkage to medical care, particularly among young men who have male to male sexual contact and black and Hispanic/Latino populations, are key to reducing transmission at all stages of disease, and specifically among those with acute infection when transmission risk is high. The substantial disparity in receipt of care and viral suppression among racial/ethnic groups suggests the need for improved targeting of interventions based on social determinants of health. Resources are still needed to monitor and improve outcomes across the HIV continuum of care. The transition from the July 2010 national strategy to the 2020 updated national strategy presents an opportunity to reassess local, state, and national programs to reach these care continuum goals.