Methods

Ethics and planning

The study was designed in consultation with multidisciplinary team members, including Indigenous Liaison Officers from the Better Cardiac Care and Aboriginal and Torres Strait Islander Cardiology Outreach services at Cairns Hospital, and was informed by the Australian Institute of Aboriginal and Torres Strait Islander Studies (AIATSIS) Code of Ethics for Aboriginal and Torres Strait Islander Research. An Aboriginal and Torres Strait Islander co-author (Iamalaig and Kaantju woman; National Centre for Aboriginal and Torres Strait Islander Wellbeing Research, Australian National University) contributed to culturally informed study design, interpretation, and framing. Ethics approval was obtained from the London School of Hygiene and Tropical Medicine (approval no. 28464) and the James Cook University Human Research Ethics Committee, including Aboriginal and Torres Strait Islander ethics subcommittee approval (24H-9669).

Setting and design

A cohort state-transition Markov model with annual cycles was constructed using Microsoft Excel (Microsoft Corporation, USA) (Table 2, appendix 1). This is similar in structure to previously published peer-reviewed Australian cardiovascular economic models in Aboriginal and Torres Strait Islander populations [10]. The population, defined in Table 3, were Aboriginal and Torres Strait Islander Australians with CVD and hypercholesterolaemia based on AIHW estimates. The perspective taken was that of the Australian Healthcare system. The intervention was inclisiran with Statin. The comparator was current first-line PBS funded, guideline-directed Australian standard-of-care (ie, statins alone). Ezetimibe's real-world uptake is estimated at <10% in this population and was therefore not explicitly modelled as a separate treatment pathway [6]. As no data exists on the real-world use of inclisiran in Aboriginal and Torres Strait Islander Australians, adherence was assumed at 100% across both arms to isolate pharmacological treatment effect and avoid confounding from differential adherence assumptions. The modelled outcomes were standard no event, fatal CVD event, non-fatal CVD event or fatal-other event. Time was cycled annually to 25 years (the difference between average Aboriginal and Torres Strait Islander Australian life expectancy and first CVD event). The primary outcome was the Incremental cost-effectiveness ratio (ICER) which reflected cost-per-quality-adjusted life year (QALY) gained. QALYs were used to align with conventions in healthcare economic evaluation reporting. Secondary outcomes were CVD events and QALY's gained. Proposed health states and transitions are defined in Fig. 1.

Intervention | Model variables | Health state | Sensitivity Range | Source: | Values: | Distribution:
Early Inclisiran with Standard Therapy | Annual Transition Probabilities | No event | ±5% |  | 61.11% | Beta
Non-fatal CVD event | ±5% | Pooled ORION 9/10/11 trials | 36.02% | Beta
Fatal CVD event | ±5% | Pooled ORION 9/10/11 trials | 2.85% | Beta
Fatal other event | ±5% |  | 1.54% | Beta
Utility weights | No event | ±5% | VALIANT trials | 0.8 | Beta
Non-fatal CVD event | ±5% | VALIANT trials | 0.7 | Beta
Fatal CVD event and other events | ±5% |  | 0 | Beta
Cost Parameters | Statin + Inclisiran | ±5% | PBS | $3922 | Gamma
No event | ±5% | Australian DRG data | $4878 | Gamma
Non-fatal CVD event | ±5% | Australian DRG data | $21,943 | Gamma
 | Fatal CVD event | ±5% | Australian DRG data | $76,927 | Gamma
 | Fatal Other event | ±5% | Australian DRG data | $76,927 | Gamma
Standard Therapy | Annual Transition Probabilities | No event | ±5% |  | 57.33% | Beta
Non-fatal CVD event | ±5% | Australian Indigenous Burden of Disease study | 38.39% | Beta
Fatal CVD event | ±5% | Australian Indigenous Burden of Disease study | 2.74% | Beta
Fatal other event | ±5% | Australian Indigenous Population Statistics | 1.54% | Beta
Utility weights and Cost parameters | Same as Inclisiran arm | 

Cohort | Number at-risk | CVD-related hospitalisations | CVD-related deaths | CVD-related DALY's | Annual CVD Healthcare costs (AUD)
Aboriginal and Torres Strait Islander Australians with CVD | 42,000 | 17,275 | 1150 | 24,612 | $424,800,000
Aboriginal and Torres Strait Islander Australians with CVD attributable to Hypercholesterolaemia (target population) | 12,180 | 5010 | 334 | 7.137 | $123,192,000

Fig. 1: Proposed Markov Model health states and transitions.

Transition probabilities – intervention arm

Model inputs for lipid-lowering efficacy and cardiovascular events were derived from pooled patient-level analyses of the ORION 9/10/11 RCT's [6]. These observed trial data were subsequently extrapolated within a validated Markov framework to estimate long-term cardiovascular outcomes, QALYs, and costs. Long-term projections therefore represent modelled estimates beyond the duration of trial follow-up. Similar methodology has been used in similar peer-reviewed Australian inclisiran cost-effectiveness study for non-Aboriginal and Torres Strait Islander Australians [11]. The inclisiran arm had 17 fatal CVD events in 1833 patients (0.93%) versus placebo arm (15/1827, 0.82%); an absolute risk increase of 0.11%. The inclisiran arm had 114 non-fatal CVD events (6.22%) versus placebo arm (157/1827, 8.59%); an absolute risk reduction (ARR) of 2.37% [6]. These estimates were used to inform transition probabilities. Importantly, these trial-derived estimates were applied as relative risk inputs rather than assumed absolute effects in the base model to preserve consistency with modelling conventions. MACE definitions were aligned with ORION trials to ensure consistency [6].

Transition probabilities – comparator arm

Aboriginal and Torres Strait Islander Australian's with CVD and hypercholesterolaemia have no published transition probability data. Best available national administrative and epidemiological data from Australian Government sources were used as calibrated population-level proxies as presented in Table 3 (data accessed between August-October 2024).

Of the 3.2% (812,728) total Aboriginal and Torres Strait Islander Australians [11], approximately 42,000 have known CVD accounting for 17,275 CVD-related hospitalisations, 1150 CVD-related deaths, 24,612 CVD-related QALY's gained [12], and 3.6% of the total $11.8 billion Australian CVD healthcare expenditure ($424,800,000) annually [1]. The AIHW estimates the population attributable fraction of hypercholesterolaemia to this CVD burden in Aboriginal and Torres Strait Islander Australians is 29% [1]. Thus, the baseline population entering our model were the estimated 12,180 Aboriginal and Torres Strait Islander Australians with CVD and hypercholesterolaemia of whom 334 experience fatal CVD events and 4676 non-fatal CVD events yearly (Table 3). The fatal-other event rate reflected age-adjusted mortality rates and was held constant between arms to avoid competing risk bias and isolate cardiovascular-specific effects. [11].

Costs and discounting

All costs were reported in 2025 Australian dollars (AUD). Annual statin therapy costs were estimated at AUD 223 per person using weighted average generic pricing [13]. Annual inclisiran acquisition cost was AUD 3699 per person (July 2025 pricing [14]. Costs associated with no event, non-fatal CVD events, and fatal CVD events were derived from Australian diagnosis-related group (DRG) funding data and aligned with previous Australian cardiovascular economic evaluations [11,14]. The Pharmaceutical Benefits Advisory Committee (PBAC) has no defined willingness-to-pay threshold (WTPT) to determine whether an intervention is cost-effective. Whilst other factors including clinical need and equity are also considered, generally <$50,000AUD/QALY WTPT is commonly used as a reference benchmark in Australian health economic evaluations [15]. Costs and outcomes were discounted at 5% per annum.

Utilities

No published health-state utility estimates were identified specifically for Aboriginal and Torres Strait Islander Australians with established CVD and hypercholesterolaemia. Utility estimates were therefore informed by the VALIANT cohort and prior Australian cardiovascular models [11,14,16]. Utility values were 0.80 for established CVD without recurrent event and 0.70 following non-fatal recurrent CVD events. Utility values for cardiovascular and non-cardiovascular death were 0.

Sensitivity analysis

Deterministic one-way sensitivity analyses were performed to assess model robustness. Key parameters varied included inclisiran cost, treatment duration, discount rate, utility estimates, comparator event rates, and treatment-effect assumptions. Unless otherwise specified, parameters were varied by ±5%. As robust variance estimates were unavailable for several key model inputs, broader (±20%) scenario analyses were additionally performed for the principal model drivers to further explore uncertainty in the absence of probabilistic sensitivity analysis. Probabilistic sensitivity analysis was not performed because robust variance estimates were unavailable for several key population-level parameters derived from administrative and epidemiological datasets.

Model validation

Face validity was assessed through clinical review by cardiovascular and Indigenous health experts involved in study design with knowledge in healthcare economic analysis. The systematic reporting on validation efforts for our Markov model are presented via the Assessment of the Validation Status of Health-Economic decision models (AdViSHE) tool (appendix 2) [17].