Modeling adherence and outcome gains from consolidating multi-pill regimens into single-tablet fixed-dose combinations
Secondary cardiovascular prevention is a textbook case of polypharmacy: a post-MI or high-risk patient is commonly prescribed a statin, an ACE-inhibitor or ARB, an antiplatelet agent, and often a beta-blocker — four separate tablets, frequently at different times of day, indefinitely.
Two related but distinct burdens degrade adherence in multi-pill regimens:
Pill count (regimen size): • Each additional medication adds an independent chance of a missed dose, a refill-timing mismatch, or a fill-gap on any given day • Patients on ≥5 chronic medications ("polypharmacy") show significantly higher rates of unintentional non-adherence driven purely by tracking burden, independent of belief-based (intentional) non-adherence
Dosing frequency: • Claxton et al.'s landmark 2001 meta-analysis (calculated from electronic pill-bottle monitoring, not self-report) found average adherence of ~79% for once-daily dosing, falling to ~69% for twice-daily, ~65% for three-times-daily, and ~51% for four-times-daily regimens • The Medication Regimen Complexity Index (MRCI) formally scores regimens on dosage forms, frequency, and additional administration instructions; scores above ~16 are associated with markedly worse adherence and higher hospitalization risk in heart-failure and post-MI cohorts
Baseline in this simulation: A representative 4-drug secondary-prevention regimen (statin + ACE-inhibitor + antiplatelet + beta-blocker) with twice-daily total dosing frequency corresponds to an expected PDC-based adherence of roughly 50–60% at one year — consistent with real-world claims studies of post-MI patients, where only about half remain adherent to all indicated secondary-prevention medications 12 months after discharge.
Not every combination of drugs can be pressed into a single tablet. Before any regulatory step, formulation scientists must confirm the components behave predictably when co-formulated — chemically stable together, absorbed on comparable timescales, and combinable at fixed dose ratios that still cover the range patients actually need.
FDC feasibility screening evaluates several independent compatibility axes before any clinical study begins:
• Elimination half-life alignment: components with wildly different half-lives create a mismatch — the short-acting drug would need re-dosing before the long-acting one, defeating the purpose of a once-daily FDC. Practical FDCs generally combine agents whose half-lives fall within roughly a 2-fold range of one another, or use modified-release formulation to align effective dosing intervals • Absorption / food-effect independence: if one component's absorption is highly food-dependent (e.g., requires an empty stomach) while another requires food to reduce GI upset, co-formulation forces a compromise that can reduce bioavailability of one or both — this single issue screens out roughly 15–20% of otherwise attractive candidate pairs • Chemical/physical compatibility: no degradation-accelerating interactions between active pharmaceutical ingredients (APIs) in the solid dosage form (assessed via forced-degradation and accelerated-stability studies, typically 6 months at 40°C/75% RH) • Fixed dose-ratio feasibility: because a single tablet locks the ratio between components, the FDC must be offered in enough dose-strength variants (typically 3–6 SKUs) to approximate the individual titration ranges patients need — this is why hypertension FDCs are usually reserved for patients already stable on the individual components, not for initial titration
Precedents: HIV medicine pioneered modern single-tablet regimens (STRs) — Biktarvy (bictegravir/emtricitabine/tenofovir alafenamide) and Triumeq (dolutegravir/abacavir/lamivudine) each combine three antiretrovirals with compatible half-lives into one once-daily pill, replacing regimens that once required 6+ tablets across multiple daily dosing windows.
Before an FDC can be marketed, regulators require rigorous proof that the combined tablet delivers each active ingredient into the bloodstream just as reliably as the individual pills taken together. This is bioequivalence testing, governed by strict, internationally harmonized statistical criteria.
Bioequivalence (BE) is a formal pharmacokinetic — not clinical — endpoint. It asks: does the test formulation (FDC tablet) deliver the same rate and extent of absorption as the reference (individual pills, co-administered)?
Standard design — 2-way crossover, single dose, healthy volunteers: 1. Volunteers (typically 24–36, sufficiently powered for the acceptance range) are randomized to receive either the FDC or the individual-pill combination first 2. After a washout period of at least 5 elimination half-lives (to ensure the first dose is fully cleared), volunteers cross over to receive the other formulation 3. Serial blood sampling generates a concentration-time curve for each active ingredient in both study periods 4. Two pharmacokinetic parameters are calculated per drug: Cmax (peak plasma concentration) and AUC (area under the concentration-time curve, total systemic exposure)
Acceptance criterion: • The geometric mean ratio (GMR) of test (FDC) to reference (individual pills) is calculated for both Cmax and AUC, for each active component separately • Bioequivalence is declared only if the 90% confidence interval of the GMR falls entirely within 80.00% to 125.00% — this is the FDA's (and, with minor variations, EMA's) standard acceptance range, applied independently to every active ingredient in a multi-component FDC • A multi-component FDC (e.g., 3 or 4 actives) must pass this test separately for every component — a failure on any single component blocks approval of the combination
Why 80–125%, not 100%: The range reflects accepted normal intra-subject pharmacokinetic variability, not a tolerance for meaningfully different drug exposure. In practice, most well-designed FDCs land GMRs very close to 100%, since they are typically formulated from the same or very similar excipient systems as the originator single-agent products.
Because the bioequivalence bar (80–125% CI) is a pharmacokinetic standard, not an efficacy re-proof, FDCs of already-approved individual agents typically do not require new large-scale clinical efficacy trials — the clinical evidence base of each component drug is presumed to transfer, provided BE is demonstrated. This is what allows fixed-dose combinations to reach patients years faster than a novel single-molecule drug.
Once an FDC is approved and marketed, the practical question becomes empirical: does consolidating pills into one actually change behavior? Comparative real-world studies using Proportion of Days Covered (PDC) as the outcome consistently answer yes — and by a clinically meaningful margin.
Bangalore and colleagues' 2007 meta-analysis (American Journal of Medicine), pooling data across therapeutic areas, found that fixed-dose combination therapy was associated with a 21% lower relative risk of non-adherence compared with the identical drugs taken as free (separate) combinations (RR 0.79, 95% CI 0.67–0.94) — one of the most frequently cited quantitative summaries of the FDC adherence effect.
Cardiovascular-specific evidence: • The UMPIRE trial (Lancet, 2013) randomized over 2,000 patients with established CV disease or high risk to a cardiovascular polypill (aspirin + statin + 2 BP-lowering agents) versus usual care; polypill-assigned patients showed roughly a 15-percentage-point higher rate of adherence to all indicated medications at 15 months • PILL trial and Kanyini GAP trial report broadly consistent effects: absolute PDC or self-reported-adherence improvements in the 10–20 percentage-point range when multi-pill regimens are consolidated
HIV single-tablet regimens (STRs): • Comparative cohort studies of HIV antiretroviral therapy consistently find STR (one pill, once daily — e.g., Biktarvy, Triumeq) adherence in the 85–90% range, versus adherence in the 70s (percent) for multi-tablet regimens covering the same drug classes • STR use is also associated with lower rates of treatment discontinuation and, in several cohort analyses, fewer HIV-related hospitalizations, attributed largely to the adherence effect rather than any pharmacologic difference
Mechanism: The adherence gain from consolidation is best explained as removing, rather than compensating for, the therapy-related WHO dimension barrier (regimen complexity) — patients no longer need to track multiple bottles, multiple refill dates, or multiple dosing times; a single fill event and a single daily action replace what previously required several independent successful actions per day.
An adherence improvement is only clinically meaningful if it translates into better disease control and, ultimately, fewer cardiovascular events. Three major randomized polypill trials — PolyIran, TIPS-3, and HOPE-3 — provide direct outcome evidence, not just surrogate adherence data.
PolyIran (Lancet, 2019): • Cluster-randomized trial in a rural Iranian population (>6,800 participants, ages 50–75) comparing a 4-component polypill (statin + 2 antihypertensives + aspirin) against non-pharmacologic care recommendations • Intention-to-treat analysis: 20% relative reduction in major cardiovascular events over ~5 years • Among participants adherent to ≥80% of prescribed doses, the reduction reached 34% — directly demonstrating that the polypill's clinical benefit scales with the adherence it enables
TIPS-3 (NEJM, 2021): • International trial (>5,700 participants at intermediate cardiovascular risk) testing a polypill (statin + 3 low-dose antihypertensives) with and without aspirin, against placebo • Polypill-alone arm: 21% relative reduction in the composite cardiovascular outcome versus placebo over ~5 years • Polypill plus aspirin arm: further incremental benefit, though with a modest increase in minor bleeding
HOPE-3 (NEJM, 2016): • Enrolled >12,000 intermediate-risk participants without a prior cardiovascular event; tested a fixed combination of candesartan/hydrochlorothiazide plus rosuvastatin versus placebo • Overall population: 21% relative reduction in cardiovascular events over ~5.6 years • Pre-specified subgroup with baseline hypertension (top third of systolic BP): 27% relative reduction — outcome benefit concentrated in patients who most needed BP lowering, consistent with the mechanism (BP and LDL control) rather than a nonspecific "polypill effect"
Surrogate endpoints: Across these and smaller comparative FDC-vs-free-combination studies, BP control rate (proportion reaching goal, typically <140/90 mmHg) improves by roughly 10–15 percentage points, and LDL-goal attainment shows comparable relative gains — both consistent with, and largely explained by, the adherence lift measured in Stage 4.
The original 2003 BMJ "polypill" proposal by Wald and Law predicted an ambitious ~80% reduction in cardiovascular events from a comprehensive polypill strategy. A decade and a half of actual randomized trial evidence (PolyIran, TIPS-3, HOPE-3) instead converges on a more modest but still highly clinically significant 20–34% relative reduction — a useful reminder that adherence-mediated interventions produce real, substantial, but not miraculous gains.
Even a clinically effective, adherence-improving FDC must clear a final bar for widespread adoption: does the health gain justify the cost, relative to the alternative of continuing separate generic components? Cost-per-QALY analyses generally answer this favorably, especially in populations with high baseline cardiovascular risk.
Cost-effectiveness analysis (CEA) for fixed-dose combinations combines three inputs derived from the earlier stages of this pipeline:
1. Incremental drug cost: an FDC tablet is sometimes priced at a modest premium over the sum of its (often generic) individual components, though for off-patent generic FDCs the premium can be negligible or even negative (fewer total dispensing/administration fees)
2. Adherence-mediated event reduction: using the Stage 4 adherence lift (10–20pp PDC improvement) propagated through the Stage 5 trial-derived event-reduction relationship (~20–34% relative CV event reduction at full adherence), a health-economic model estimates avoided myocardial infarctions, strokes, and cardiovascular deaths per 1,000 patients treated
3. Downstream cost offsets: each avoided major cardiovascular event avoids substantial downstream cost — US cost estimates for an acute MI hospitalization plus one year of follow-up care commonly fall in the $10,000–$20,000+ range, before accounting for lost productivity and long-term disability costs
Results across published polypill/FDC cost-effectiveness studies: • Modeling built on the PolyIran and TIPS-3 trial data finds the CV polypill strategy to be highly cost-effective, and in several low- and middle-income country (LMIC) analyses, cost per QALY gained falls below $1,000 — far under the WHO-style heuristic threshold of 1–3× per-capita GDP per QALY gained • In higher-income settings, generic-component FDCs are frequently cost-saving or cost-neutral once avoided hospitalizations are included, because the incremental tablet cost is small relative to individual generic pricing and the adherence-driven event reduction is the dominant economic driver • HIV single-tablet regimens show a parallel pattern: despite sometimes-higher list price than generic multi-tablet alternatives, STRs are frequently found cost-effective or cost-saving in payer models once reduced hospitalization, lower rates of resistance-driven regimen failure, and adherence-linked viral suppression benefits are included
The consistent finding across cardiovascular and HIV FDC literature is that the primary economic lever is not the pill itself but the adherence behavior it enables — cost-effectiveness models are, in essence, adherence-lift models wearing a currency symbol.