Real-world adherence and persistence for newly-prescribed HIV treatment: single versus multiple tablet regimen comparison among US medicaid beneficiaries

At the end of 2015, ~ 1.1 million people aged ≥ 13 years were living with human immunodeficiency virus (HIV) in the United States [1]. While estimated annual HIV infections in the United States have declined by 10% from 2010 to 2014 [1], the condition remains a serious public health concern given the burden it imposes on patients and the health care system.

HIV treatment typically includes a combination of antiretroviral therapy (ART) regimens to prevent HIV disease progression and transmission [2]. ART has brought about a substantial decrease in mortality due to HIV infection, changing it from a rapidly lethal disease into a chronic manageable condition [3]. In fact, the use of ART has significantly shown an improvement in the life-expectancy and quality of life (QoL) among people infected with HIV [4, 5]. Moreover, early initiation of ART has shown improvement in clinical outcomes and reduction in sexual transmission of HIV through viral suppression [5]. The Department of Health and Human Services (DHHS) panel on antiretroviral guidelines for adults and adolescents also recommends immediate initiation of ART for all people living with HIV, regardless of CD4 count, to reduce the morbidity and mortality associated with HIV infection [6]. These guidelines recommend initiating ART in treatment naïve patients with a regimen consisting of two nucleoside reverse transcriptase inhibitors (NRTIs) along with a third drug: preferably an integrase strand transfer inhibitor (INSTI), with non-nucleoside reverse transcriptase inhibitors (NNRTI) or boosted protease inhibitors (PI) as alternatives to INSTIs [5‐7].

Combination ART was initially characterized by high pill burden and multiple daily doses [8‐10]. Over the past two decades, more potent, more convenient, and less-toxic ART regimens have been developed; treatment regimens of up to 20 pills per day have been largely replaced by once-daily, single-tablet regimens (STRs) [10, 11]. As of March 2016, six STRs—elvitegravir (EVG)/cobicistat (COBI)/emtricitabine (FTC)/tenofovir alafenamide fumarate (TAF), EVG/COBI/FTC/tenofovir disoproxil fumarate (TDF), dolutegravir (DTG)/abacavir (ABC)//lamivudine (3TC), FTC/rilpivirine (RPV)/TAF, FTC/RPV/TDF, and efavirenz (EFV)/FTC/TDF—had been approved by the US Food and Drug Administration for HIV treatment [12]. STRs are single-dose units administered once-daily, whereas multi-tablet regimens (MTRs) require multiple dosing units per day [4]. STRs may improve adherence, thus resulting in greater efficacy, reduced risk of virologic failure, and prevention of the emergence of drug resistance [13, 14]. The decision to prescribe STRs vs MTRs is based to a large extent on patient and clinician preference as well as, in some instances, medication cost and positioning of products on payer formularies [15, 16]. Moreover, HIV infection requires a high level of adherence to achieve viral suppression, presenting a challenge for HIV management, particularly among patients who face barriers such as limited provider and pharmacy services, low social support, and substance use [17, 18]. Reducing pill burden and structural barriers may therefore help patients achieve increased adherence and improved health outcomes.

Prior research suggests that adherence and persistence may be higher among patients treated with STRs compared with those treated with MTRs [13, 19]. However, with the advent of new treatment options with greater effectiveness and tolerability, the extent to which adherence and persistence varies between and within STRs and MTRs has not been fully investigated. Amongst third agents, INSTIs are the most preferred as evident in studies that showed it be superior or equivalent to other third agents in safety and efficiency. In addition, it is reported that INSTIs are more tolerable to patients, in turn reducing discontinuation rates [20‐23]. However, for patients that are at higher risk of non-adherence, PIs as the third agent would be most appropriate because of their genetic barrier to resistance [24]. For backbones, tenofovir-containing nucleoside has shown higher safety and tolerability then abacavir-lamivudine [25]. In addition, tenofovir/emtricitabine has been found to have a lower rate of discontinuation than co-formulated abacavir/lamivudine [20]. Hence, this retrospective claims-based study was conducted with an aim to assess real world adherence and persistence for newly prescribed HIV treatment comparing STRs versus MTRs, backbones, and third agents using the Truven Health Medicaid database.

For the adherence study, 1744 patients met inclusion criteria, of whom 1290 (74.0%) initiated STRs and 454 (26%) initiated MTRs. For the persistence study, 2409 patients met inclusion criteria, of whom 1782 (74.0%) initiated STRs and 627 (26.0%) initiated MTRs. Patients on STRs and MTRs were further stratified by specific ARTs to compare persistence within STRs and within MTRs (Fig. 1).

The discontinuation caused by poor tolerability, poor adherence, or complexity carries risks of the development of toxicity and resistance associated with compromised future treatment options. Improved rates of adherence and persistence are critically important for the long-term therapy of HIV patients [32] and improved health outcomes. In this retrospective claims-based study including several thousand Medicaid recipients newly initiating ART, treatment adherence and persistence were examined and compared among patients prescribed STRs and MTRs.

ART adherence has been shown to improve health outcomes and QoL, including HIV transmission, viral load suppression, drug resistance prevention, and survival rates [4]. Most research examining adherence to ART defines adherence based on either medication possession ratio (MPR) or proportion of days covered (PDC) [27, 33‐35]. MPR and PDC are the conventional measure of adherence but are not necessarily the gold standard. The current study’s approach of defining adherence as a < 5 day gap in fills over 30-day intervals during the 6-month follow-up period is more stringent in allowing for examination of the proportion of patients who are adherent and non-adherent at different time points from therapy initiation date. The current study found that during the overall 6-month follow-up period, the proportion of adherent patients was higher among STR initiators (22.7%) compared to MTR initiators (11.7%); this is similar to the finding from Hines et al., where the same measure of adherence was applied and the proportion of adherent patients for STR initiators and MTR initiators were 24.9% and 11.7%, respectively [36]. In support of our study, an observational study of 755 people living with HIV recruited from community services found that patients taking STRs were more likely to adhere to their medication compared to those on MTRs [5]. Lowering pill burden is associated with improved adherence and viral suppression, and consequently QoL [4, 37, 38]. In further support of our study findings Sutton et al. showed that an STR cohort had significantly better adherence when compared to an MTR cohort; additionally, risk of hospitalization was lower in the STR cohort [33]. Although our study did not examine barriers to adherence, a study by Chen et al. reported that MTR patients were more likely to report barriers including scheduling, side-effects, and confusion over their prescription as reasons for missing a dose compared to those prescribed STRs [5]. Comparing the STRs, the adherence was highest among those prescribed EVG/COBI/FTC/TAF in contrast to all other STRs.

Adherence and persistence are also considered cornerstones of effective ART therapy; hence, it is recommended that ART regimens should be efficacious, with no virologic failures, well-tolerated, and without toxicity, to improve adherence, and, in turn, improve long-term persistence [39, 40]. The unadjusted analysis results in the current study show that a lower proportion of STR patients discontinued their first-line therapy (36.3%) compared to MTR patients (48.8%). Likewise, the adjusted analysis revealed that MTR patients had a 1.6 times greater hazard of treatment discontinuation as compared to STR patients. In support of this finding, a recent claims-based study among Medicaid beneficiaries also revealed that persistence is greater with less toxic ART regimens that include fewer pills. The study further reported that there was a 29% reduced hazard of non-persistence with STRs compared with regimens containing > 6 ART pills per day [41]. Another study also reported that STR discontinuation was lower (i.e., few patients discontinued the STR throughout the study), and further found that patients who switched from an MTR to an STR were able to maintain virologic suppression [14]. Although outside the scope of the current analysis, future research should investigate reasons for discontinuation and switching of therapy. Past research has reported that factors including pill burden, poor tolerability, risk of resistance due to complexity, toxicity, regimen performance, drug efficacy, and virologic failure could be possible reasons for higher rates of discontinuation among MTR patients [20, 37, 40]. Furthermore, the current study also showed that among STRs, patients who were treated with EVG/COBI/FTC/TAF had greater persistence compared to those treated with other STRs. Since EVG/COBI/FTC/TAF has been found to have higher efficacy due to fewer drug-related adverse events compared to EVG/COBI/FTC/TDF among treatment naïve patients [42], providers presumably switched regimens based on those data as well as changes to HIV treatment guidelines.

The current study examined regimen components in addition to specific STRs and MTRs. Separate comparisons were performed among patients prescribed third agents and backbones to explicitly explore the benefits derived from each individual component. Our study found that patients who were treated with regimens including an FTC/TAF backbone had higher persistence compared to regimens with ABC/3TC and FTC/TDF, providing one of the first real world studies assessing persistence of the FTC/TAF backbone, which was approved in 2016. While reasons for persistence could not be assessed in this analysis, the higher persistence of FTC/TAF likely relates to clinical trial data showing high efficacy and bone and renal safety advantages compared to prior regimens, and the 2018 IAS-USA guidelines recommended TAF but not TDF as a component of initial suggested regimens [43, 44].

The findings from our study should be viewed in the context of study and claims data limitations. While claims data are extremely valuable for the efficient and effective examination of health care outcomes, they are primarily collected for business purposes rather than research purposes. Therefore, analyses may be subject to inherent limitations of the source administrative claims data, such as coding errors or diagnoses entered for administrative processing rather than clinical completeness. Moreover, the presence of a claim for a filled prescription does not indicate that the medication was consumed or taken as prescribed. Also, medications filled over-the-counter or provided as samples by the physician are not observed in claims data. Certain information that could influence study outcomes is not readily available in claims data, such as clinical and disease-specific parameters including HIV viral load, which could be used as a surrogate for adherence. In addition, although the study adjusted for observed baseline characteristics in the multivariable analyses, some unobserved confounders may remain. Further, given the study timeframe, newly-available HIV regimens were not included. Lastly, this study was limited in that health claims may be subject to error; however, given that ART is only used for HIV and that at least two ART claims were required for inclusion, it is reasonable to conclude that patients identified in the Truven Medicaid population by ART use are HIV patients.

In this analysis of real world claims data associated with US Medicaid recipients newly initiating ART, adherence and persistence was greater among patients initiating STRs than those initiating MTRs. In fact, the proportion of patients who were adherent was almost double for those initiating STRs compared to MTRs. Additionally, median treatment persistence was also higher among STR initiators. Regimens containing EVG/COBI as a third agent and FTC/TAF as a backbone showed higher persistence than other third agents and backbones, respectively. In addition, both adherence and persistence were greater with single-tablet EVG/COBI/FTC/TAF than with other STRs. As new regimens with improved safety and tolerability profiles become available, continued research on persistence and adherence for HIV treatments is warranted to inform better therapeutic management.