Single-Pill Versus Loose-Dose Combination Triple Therapy for Hypertension: Formulary Impact
Published Online: June 20, 2014
Kristine Ogden, BS; Shuchita Kaila, PhD; and Nancy Neil, PhD
Chronic hypertension is a major risk factor for cardiovascular disease (CVD). Systemic hypertension is a pervasive public health concern in the United States, affecting more than 78 million adults.1 It is also the fifth-most costly medical condition in the United States2; direct and indirect medical care costs attributable to hypertension were recently estimated to be $51 billion annually (2009 dollars).1 A strong causal relationship between untreated hypertension and cardiovascular (CV) events has been established.1 Uncontrolled hypertension can lead to cerebrovascular disease, ischemic heart disease, and death.3 By the same token, blood pressure (BP) control is strongly associated with reduced CVD risk and, in turn, lower medical care costs.4 Indeed, the primary goal of treating hypertension is the reduction of CVD morbidity and mortality.5 Sustained antihypertensive therapy is necessary to achieve this goal. However, despite the availability of numerous, highly effective antihypertensive agents, BP remains inadequately controlled in more than half (53.5%) of the individuals with hypertension in the United States.6
Current evidence suggests that the majority of patients treated for hypertension—roughly two-thirds of the diagnosed population—will require combination therapy to achieve targeted BP goals.7-9 Of these, many will require a triple-combination regimen.10 A major obstacle to BP goal attainment is poor regimen adherence and persistence, both of which are exacerbated by regimen complexity.11,12
Ample evidence suggests that simplifying a therapeutic regimen with the use of fixed-dose, single-pill combination (SPC) antihypertensive therapy in place of loose-dose combination (LDC) alternatives can improve regimen adherence and lead to fewer cardiovascular events and lower all-cause medical care costs. For instance, increasing adherence to a oncedaily antihypertensive regimen by as little as 1 pill per week has been shown to reduce multivariate-adjusted mortality risk by 7% (HR 0.93; 95% CI, 0.90-0.096).13 A recent retrospective analysis by Brixner and colleagues used a nationally representative claims database to show that patients receiving SPC regimens had 6.6% greater adherence (P <.001) compared with patients given LDC regimens.14 Similarly, a study by Zeng and colleagues found that patients on SPC regimens were significantly more likely to be adherent (OR 2.915; P <.001) and less likely to discontinue therapy (HR 0.537; P <.001) compared with those on LDC regimens.15 Two recent meta-analyses have reported that, in patients with chronic conditions such as hypertension, SPC regimens significantly improved therapeutic adherence compared with separate LDC regimens of the same drug components. 16,17 A third meta-analysis found that hypertensive patients prescribed SPC therapies had higher treatment adherence and lower all-cause healthcare costs compared with patients given comparable LDC regimens.18 In addition, both greater adherence and greater persistence with hypertensive therapy have been associated with significantly lower risk of CV events, healthcare resource use, and hospitalization compared with nonadherent and/or nonpersistent patients.13,15,19-23
The objective of this analysis is to estimate the budget impact, from the perspective of a commercial health plan, of regimen simplification via greater use of SPC regimens (valsartan/amlodipine/hydrochlorothiazide [HCTZ] or olmesartan/amlodipine/HCTZ) within a formulary of comparable 2- and 3-pill alternative LDC regimens (angiotensin II receptor blockers [ARB] + amlodipine + HCTZ) prescribed for hypertensive patients not controlled on dual therapy. Specifically, our goal is to evaluate the potential of lower medical care costs due to greater adherence (and, thus, reduced risk of CV events) to offset the incrementally higher acquisition costs of SPC therapies versus comparable LDC alternatives.
We developed a Microsoft Excel–based, 1-year budget impact model to consider the effect to a health plan of increasing the use of triple-therapy SPC regimens for hypertensive patients not controlled on dual therapy. In the model, therapeutic adherence and persistence vary by regimen type (daily pill burden) which, in turn, can influence hypertension drug costs, cardiovascular outcomes, and related medical care costs (Figure 1).
Based on the analysis, we estimated that 11,308 patients in a hypothetical plan size of 5 million would be eligible for triple antihypertensive therapy as a 1-, 2- or 3-pill daily regimen of ARB + amlodipine + HCTZ (Table 11,24).
The model considered adherence in terms of patients’ medication possession ratio (MPR), defined as the percentage of time a patient has access to all components of the triple regimen (according to prescription refill patterns). Following Pittman and colleagues,4 patients with low, moderate, and high adherence were assumed to have MPR of 0% to 59%; 60% to 79%; and 80% to 100%, respectively, in a 360-day year. Estimates of high adherence by pill burden were obtained directly from Ferrario and colleagues25 (1 pill) and Panjabi and colleagues26 (2 and 3 pills); sub-analyses of the same data informed assumptions about the distribution of low and moderate adherence for patients on 2- and 3-pill daily regimens27 (Table 2).
Following Yang and colleagues,28 our model assumed that a lower daily pill burden is associated with higher patient persistence with therapy (60% persistence with SPC regimens and 40% persistence with 2-pill daily LDC regimens). In addition, our model assumed that patient persistence with 3-pill daily LDC regimens was incrementally lower (30%) than that reported by Yang and colleagues for 2-pill daily LDC therapies.28
Formulary options for antihypertensive triple therapy in the model included 2- and 3-pill LDC regimens containing ARB + amlodipine + HCTZ as well as SPC regimens with valsartan/amlodipine/HCTZ or olmesartan/amlodipine/ HCTZ. Baseline average wholesale acquisition costs (WACs) were obtained from Medi-Span Price Rx.29 Formulary tier assignments/co-pay for each of these agents were consistent with published sources,30,31 as were percentages of patients with 30- versus 90-day refill schedules32 (Table 3). In the case of valsartan, with planned generic entry following the expiration of its patent, the branded version was assumed to be placed on tier 3 while the generic version was placed on tier 1. Baseline market share for each drug combination in the ARB + amlodipine + HCTZ triple-therapy market was obtained from an analysis of IMS Health Plan Claims Database.24 Market shares unavailable from this source were calculated using actual market volume from IMS Health’s National Prescription Audit (NPA) database.33 Market share for molecules with recent or planned generic entry in 2013 (eg, candesartan, eprosartan, irbesartan, and valsartan) were adjusted so that the total market share assigned to a molecule is split between the generic entry (95%) and its branded counterpart (5%). In addition, the prices of molecules with recent generic entry were calculated to be 10% of the price of the corresponding branded ARB product. These assumptions were intended to reflect the market share and pricing characteristics that generics achieve once they reach a steady state in the marketplace.
A manufacturer rebate was applied to all ARB products assigned tier 2 status (20% for branded olmesartan; 30% for branded telmisartan); no rebate was applied to the generic (tier 1) or tier 3 agents. Our base case scenario assumed a higher rebate for branded telmisartan due to its higher acquisition cost relative to other branded tier 2 alternatives. These rebate levels were assumptions and not reflective of the actual rebate levels that olmesartan and telmisartan manufacturers offer.
One-year risk-adjusted utilization (CV-related hospitalizations and emergency department [ED] visits) and medical care costs based on therapeutic adherence levels were drawn from published estimates (Table 4)4; costs were inflated to 2012 dollars using the medical care component of the Consumer Price Index.34
Two model scenarios were evaluated. In the first scenario (baseline), patients were allocated to a mix of branded and generic LDC (84%) and SPC (16%) triple-therapy daily regimens. The second scenario (increased SPC use) evaluated the budget impact of a 15% shift in the mix of daily regimens, to LDC (69%) and SPC (31%) therapies, with the increased SPC share assigned to olmesartan/ amlodipine/HCTZ (Table 3). Model results compared the estimated number of CV-related events, all-cause healthcare costs, hypertension drug costs, total healthcare costs, health plan impact (net of co-pays, rebates), and the net per member per month (PMPM) impact associated with each modeled scenario.
PDF is available on the last page.