The HIV treatment cascade and care continuum: updates, goals, and recommendations for the future

Entry into the HIV treatment cascade begins with diagnosis. Current CDC guidelines recommend that healthcare professionals offer every patient between the ages of 13 and 64 an HIV test at least once, as an initiative to make HIV screening a routine rather than risk-based practice in healthcare settings [12]. HIV testing has never been more efficient or patient-friendly, as rapid test results can be available in under half an hour [13]; thus, integrating testing into routine health care visits is not a difficult task. Moreover, home testing is now available, providing individuals with enhanced access and control in learning their HIV serostatus. The US Preventative Services Task Force (USPSTF) has guidelines similar to the CDC’s for routine HIV testing [14], which has important implications for insurance coverage and payment for this screening test. However, it is estimated that approximately one in eight individuals who are HIV-positive are not aware of their serostatus, leaving ample room for improvements in domestic HIV testing and diagnosis [11]. This is not only a public health risk as it increases the probability that PLWH will unknowingly transmit HIV to others, but it is also detrimental to individual medical outcomes, as late diagnosis is associated with increased morbidity and mortality [15]. Further, Skarbinski and colleagues estimate that 91.5% of all new HIV infections are attributable to PLWH who are either undiagnosed or who are diagnosed but not retained in care [16]. In order to address these weaknesses in the continuum, one of the key goals set forth by the 2020 NHAS is to increase the number of PLWH who are aware of their status to 90% [6]. The Joint United Nations Programme on HIV/AIDS (UNAIDS) recently set international goals for 2020 termed the “90–90–90 target,” which aim to increase the number of PLWH who know their serostatus to 90%, and, among those who are diagnosed, to increase the percentage who are prescribed ART to 90% [17]. Likewise, among PLWH who are both diagnosed and on ART, the 90–90–90 target aims for 90% of PLWH to achieve and maintain viral suppression [17]. In response to the 90–90–90 target initiative, five cities shared their continuum of care data at the 21st International AIDS Conference, held in Durban, South Africa on July 18–22, 2016. [18]. These cities included San Francisco, California; Denver, Colorado; Amsterdam, Netherlands; Paris, France, and Kyiv, Ukraine, all of which have joined the global Fast-Track Cities Initiative that seeks to fulfill UNAIDS targets in urban areas with high HIV burdens [18]. In these fast-track cities, the percentage of PLWH who know their serostatus ranges from 51% in Kyiv to 93% in Amsterdam and San Francisco [18]. At the country and province levels, the US falls somewhere in the middle range, above British Columbia but below Denmark and Australia [19]. These numbers underscore the variability that exists even at the entry point of the continuum.

For low-income individuals in the US, Medicaid is their most common source of health coverage. As originally written, the ACA would have made routine HIV testing a universally-covered preventative screening measure; however, unequal Medicaid expansion has led to state-level variation in reimbursable routine HIV testing [20]. Currently, just 34 states cover routine HIV testing, while 16 cover only “medically necessary” HIV testing, which is defined by the state [20]. Notably, the states that do not offer reimbursable routine HIV testing are predominantly the ones with the highest HIV infection rates, and they are overwhelmingly found in the South [6, 20, 21]. This coverage disparity may bar individuals from getting tested, and it ultimately prevents PLWH who are unaware of their serostatus from receiving treatment.

Linkage to care (LTC) is the second step in the HIV care continuum. The CDC and Institute of Medicine (IOM) both define LTC as a period of three months or less between documentation of diagnosis and initiation of medical treatment with an HIV care provider/prescriber [1, 22]. The virtue of using a standard indicator is that it enables coordination among data tracking systems and facilitates policy analysis [22]; however, state reporting practices for reporting HIV diagnosis in the US vary, making it difficult to establish a standard baseline for measurement purposes [23, 24]. It is even more difficult to compare LTC rates among countries because there are multiple statistical “back calculation” methods used to estimate the number of undiagnosed PLWH. [25]. According to a recent systematic review of the literature, outside of the US, only the countries of Georgia, Denmark, and Australia and the Canadian province of British Columbia have population data at all steps of the continuum. [25].

Although viral suppression is achieved more quickly if ART is started within three months of diagnosis [26], estimates of the proportion of PLWH who are linked to care within 3 months in the US are between 59% [2] and 80% [1]. In light of low LTC estimates, the updated NHAS has set a 2020 goal of 85% of PLWH who are linked to care within 1 month of diagnosis, rather than 3 months, an ambitious target [6]. In other high-income countries and provinces, LTC proportions are higher, although none reach the targeted 85%. In British Columbia, between 61 and 67% of PLWH who are aware of their serostatus are linked to care, while estimates of LTC percentages in Denmark and Australia are higher, at 78% of PLWH in Australia and 81% in Denmark [19, 27]. This exemplifies the fact that even in high-income regions with universal health care systems, like British Columbia and Denmark, many PLWH drop out of the continuum before they have even received care. In low-income countries LTC proportions are appreciably lower, although estimates are derived from targeted intervention data rather than population-based data. In these countries, it is common for PLWH to delay treatment for several years until the disease has rapidly progressed [28]. For example, the proportion of PLWH linked to care in Kenya is 42% [29] and in South Africa, 37% [30].

Several multi-site studies have developed strategies to facilitate LTC. For example, although created for the purpose of testing and linking drug users to HIV care, the seek, test, treat, and retain (STTR) data collection and harmonization initiative is an HIV care model that is broadly applicable to many vulnerable populations [31]. As the authors discuss, what makes this model successful for HIV care is that it is built around a specific subgroup (e.g., drug users) and involves the collaboration of an interdisciplinary, multi-site team. Further, data interoperability is ensured through the use of standardized surveys and questionnaires which collect data on a number of measures, such as drug and alcohol use, mental health, and HIV testing history [31]. By streamlining the strategy’s approach and standardizing data collection and measurement techniques, the STTR Initiative facilitates coordinated research that, in turn, helps identify at-risk populations and link them to care [31]. Similarly, the multi-site access to care (A2C) initiative was developed for a specific subgroup—those living in poverty—and it expands on the HIV care continuum, primarily LTC [32]. In addition to using a set of quantifiable indicators, A2C also collects qualitative case studies to fully capture the experiences of staff members at each program site [32]. Further, A2C includes a cost-analysis component, which is especially useful for sites who may not have many resources. The STTR and A2C programs exemplify how LTC strategies may be developed in such a way that they address the needs of the target population, while also maximizing the resources available to the provider site(s).

There is a lack of consensus about how to best measure retention, or continuity, in HIV care. While the IOM, NHAS, and CDC define HIV retention in care (RiC) as the proportion of PLWH who have “two or more visits for routine HIV medical care in the preceding 12 months at least three months apart” [22], these measures only account for adherence to scheduled medical appointments and not for missed or cancelled appointments [3]. Thus, it is recommended that any measure of RiC include at least two indicators—one for kept appointments and one for missed appointments—as they appear to provide complementary information [3, 33]. In fact, the number of missed appointments is a significant predictor in measuring clinical outcomes. Several studies have found that the number of missed medical appointments is significantly associated with higher viral loads and lower CD4 counts [4, 34]. Moreover, national racial disparities in HIV health outcomes [24, 35] may actually belie the presence of other factors. For example, in a multi-site trial, Zinski and colleagues found that the lower proportion of viral suppression found among African Americans in comparison to other racial groups loses statistical significance when accounting for the number of missed visits [36].

Compared to other high-income areas of the world, the US has the lowest proportion of PLWH who are retained in care. In British Columbia, 57% of diagnosed PLWH are retained in care [19, 27], and 75 and 76% are retained in Denmark and Australia, respectively [19]. In the US, the CDC (2014) approximates that just 40% of diagnosed PLWH are retained in care [1]. However, Yehia and colleagues suggest that this estimate is too low because it only measures the number of kept clinic visits over a single year, which neither captures longitudinal trends in accessing HIV care nor measures missed clinic appointments [37]. In reality, PLWH may transition in and out of care over time, referred to as “churning,” increasing the error margin of a point estimate [3, 4, 37]. Moreover, PLWH may be inaccurately designated as “out of care” due to undocumented changes in their health provider, incarceration, or death [3]. For example, the results of a San Francisco-based HIV surveillance project found that more than half of the patients who qualified as disengaged from care in local health records had merely changed providers [9]. This further illustrates the need for coordination among service providers so that data reporting can be timely and accurate. This is particularly challenging and pertinent for measuring the longitudinal estimate of HIV RiC, which is iteratively more complex than the dichotomous steps of testing and LTC.

Although an International Association of Physicians in AIDS Care (IAPAC) expert committee recommends that health providers use patient self-report as a measurement of ART adherence, there is no standard indicator for confirming that patients consistently stay on ART medication [7, 8, 38]. Thus, while estimates for the number of all PLWH (undiagnosed and diagnosed, regardless of care engagement) prescribed ART in the US range from 24 [2] to 37% [1], it is unknown how many PLWH are consistently adherent to prescribed regimens. Internationally, the US again performs poorly on this measure: adherence estimates in other high-income areas range from 44% in British Columbia [27] to 66% in Australia [19].

One difficulty with monitoring adherence is that it is challenging to measure, as it relies largely on patient self-report. To illustrate, the US Medical Monitoring Project (MMP) found that 86% of PLWH in medical care reported taking ART as prescribed over a period of 3 days [39]. However, this estimate offers just a snapshot of patients’ self-reported adherence patterns, rather than a longitudinal record, and must be interpreted with caution. ART adherence is vital, as it is the primary determinant of viral suppression, which reduces the chance that PLWH will transmit the virus by 96% and helps prevent the development of ART-resistant strains of the virus [8, 22, 38, 40].

While the standard HIV care model places ART adherence after the retention stage, Hallett and Eaton suggest that a unidirectional model fails to represent other entry points into a “leaky” continuum, wherein PLWH are lost at each step [41]. In fact, their model predicts that the number of PLWH who are at the ART Adherence stage is greater than the standard linear model would suggest [41]. To explain this contradiction, they suggest that the traditional continuum model does not reflect the experience of PLWH who may reengage in care after dropping out of for a period of time, or who may even wait until their symptoms have become unmanageable to begin ART [41]. Other researchers have similarly proposed that PLWH may seek care in an intermittent manner, rendering the traditional cascade model too simplistic as an individual level monitoring tool [3, 9, 10].

According to an expert panel of the International Association of Physicians.

in AIDS Care, physicians should prescribe ART to PLWH immediately after their HIV diagnosis, regardless of CD4 count [38]. This is because lower CD4 counts weaken the immune system and increase the risk of opportunistic infections, AIDS-related diseases, and even non-AIDS-related diseases caused by chronic inflammation [15, 38]. Unfortunately, some countries continue to set national threshold guidelines for physicians. In South Africa and in many Latin American countries, for example, ART prescription protocol dictates that CD4 counts must first reach a suboptimal level (e.g., less than 500 cells per milliliter) [28, 42]. While practices like these are often a function of limited resources and are intended to prioritize care to PLWH most in need, a recent cost-benefits analysis indicates that early ART prescription is actually more cost-effective, even in low-income countries where medical resources are limited [43].

The final step in the continuum is also the primary goal of HIV treatment and public health interventions—viral suppression through ART. The CDC considers PLWH virally suppressed if their most recent viral load, measured within the past year, is less than 200 copies per milliliter (c/mL) [1]. However, there are an estimated 70% of PLWH in the US [1] who are not virally suppressed, including both PLWH who are not aware of their infection status and PLWH who have been diagnosed but are not in care. Internationally, not all countries and provinces use the same cutoff for viral suppression, which makes comparison more difficult. However, when using a more restrictive 50 c/mL as the cutoff, viral suppression estimates range from 25% in the US to 62% in Australia [19]. Using this comparative method, the US once again ranks lowest with respect to HIV outcomes.

Yet, for PLWH who are diagnosed, retained in care, and adhering to ART—in other words, PLWH who have completed each step of the HIV care continuum—90% have achieved viral suppression in the US [35]. Recent studies have also demonstrated that simple, cross-sectional measures of viral suppression are prone to misclassification [44, 45]. As for RiC, viral suppression is not constant once achieved, and patients often transition between suppressed and non-suppressed states, even over periods as short as 1 year.

The updated NHAS has set a goal for increasing the proportion of viral suppression among diagnosed PLWH to 80% by 2020 [6]. In order to achieve this target, HIV providers must work to link and retain PLWH in care. Beyond individual health benefits, success in this step is vital to success of a test-and-treat strategy, or “treatment as prevention” method, and is vital in preventing additional infections [2, 8, 45]. This method has proved effective even in rural African countries, where one community-based HIV testing and counseling intervention resulted in a 77% viral suppression rate for participants who were retained for a measured one-year period [46].

While many PLWH in the US have become eligible for health insurance as a result of the ACA, some HIV providers have expressed concern that patients will fall out of care, especially those who have previously received specialized services for PLWH through the Health Resources and Services Administration (HRSA)’s Ryan White HIV/AIDS Program (RWHAP) [47, 48]. First created in 1990 to address what was then termed the “AIDS crisis,” today the RWHAP serves over half a million PLWH who are either uninsured or underinsured, providing both supportive services (e.g., case management) and direct medical services (e.g., funding to medical clinics), including access to ART via the AIDS Drug Assistance Program [48]. Thus, while the ACA provides a tremendous source of new and additional coverage, RWHAP will remain a valuable “payer of last resort” for low income PLWH [47, 48], especially in states that have not expanded Medicaid to all childless, non-disabled adults under age 65 with incomes at or below 138% of the Federal Poverty Level [49]. In fact, data from the most recent RWHAP Annual Client-Level Data Report [47] indicates that a quarter of PLWH in the US still lack health insurance, highlighting the need for continued funding of the program. In light of the RWHAP’s significant role in providing HIV care to PLWH in the US, current literature urges HIV providers to advocate for continued funding of RWHAP [50‐53], which has not been reauthorized since 2013 and relies on yearly appropriations from Congress [54].

At the diagnosis stage of the continuum, the US is on par with other high-income countries and provinces. However, the number of PLWH who are lost to care at the LTC stage subsequently positions the US behind other countries like Australia and Denmark, and the gap widens at each step. Ultimately, this should be a call-to-action for the US, which excels in HIV testing and diagnosis but falls behind other countries with regards to care. Until the US addresses all steps in the continuum of care, NHAS and UNAIDS goals for 90% viral suppression will be unattainable.