Health Economics Metrics

Economic reasoning foundations

• Opportunity cost — the value of the best alternative forgone; why fixed budgets make every choice a displacement
• Discounting and time preference — present values, the 3.5% Green Book/NICE rate
• Analysis perspective — payer vs provider vs societal: whose costs count
• Time horizon — how long to count costs and effects, and horizon gaming
• Marginal vs average cost — why freeing a bed doesn't save its average cost
• Cash-releasing vs non-cash-releasing savings — the honesty test for every "time saved" claim
• Sensitivity analysis — tornado diagrams; which assumption carries your case
• Probabilistic sensitivity analysis — Monte Carlo, CEACs, probability of being right
• Expected value of perfect information — pricing the pilot before you run it
• Dominance and the efficiency frontier — eliminating options nobody should pick

Outcome measures

• Quality-adjusted life year (QALY) — the common currency of health value
• Disability-adjusted life year (DALY) — the burden-side mirror; global health's metric
• EQ-5D — the instrument behind most QALY utility weights
• Incremental cost-effectiveness ratio (ICER) — extra cost per extra unit of health
• Willingness-to-pay thresholds — NICE £20–30k/QALY and the world's other lines
• Net monetary benefit (NMB) — value minus cost, done properly
• Life-years gained — survival math, and the evLYG equity variant
• Health-adjusted life expectancy (HALE) — population-level healthy-years accounting
• QALY shortfall and severity modifiers — why sicker populations' QALYs count more

Economic analysis types

• Cost-effectiveness analysis (CEA) — cost per natural unit of outcome
• Cost-utility analysis (CUA) — cost per QALY; comparing unlike interventions
• Cost-benefit analysis (CBA) — everything in money; Green Book NPV
• Cost-minimization analysis (CMA) — cheapest option, after proving equivalence
• Cost-consequence analysis (CCA) — the disaggregated table; NICE's preference for digital health
• Budget impact analysis (BIA) — affordability, as distinct from value
• Return on investment (ROI) — the shared metric, with declared parameters
• Social return on investment (SROI) — monetizing what markets don't price

Health system operational economics

• Bed days saved — the workhorse benefit, and its valuation traps
• Length of stay — the hospital's cycle time
• Readmission rate — the health system's change failure rate
• Did-not-attend (DNA) rate — missed appointments; the purest waste metric
• Emergency attendance avoidance — upstream intervention economics
• National tariff and unit costs — the NHS price book and costing infrastructure
• Referral to treatment (RTT) — the 18-week standard as a lead-time metric
• Waiting list impact — converting saved hours into patients seen
• Practitioner time — valuing bottleneck capacity, not wages
• Workforce retention — turnover costs and burnout economics
• Avoidable outsourcing costs — repatriating premium-rate work
• Downstream resource optimization — unblocking the role everyone waits on
• Earlier intervention — the economics of treating before progression
• Value-generating capacity (operational turnaround) — minting capacity without hiring
• Hard cash-releasing savings (deficit defense) — deleting budget lines; the CFO's metric

HTA frameworks and prevention economics

• Health technology assessment (HTA) — NICE, ICER (US), CADTH: who decides what's worth buying
• NICE Evidence Standards Framework — risk-tiered evidence requirements for digital health
• Germany's DiGA fast-track — apps on prescription; provisional listing with an evidence deadline
• Number needed to treat (NNT) — effort-per-benefit units that keep claims honest
• Prevention economics — why prevention is cost-effective but rarely cost-saving
• Screening economics — Wilson–Jungner, PPV collapse at low prevalence, alert fatigue
• Avoided downstream costs — cost offsets and the rules that make them credible

Software engineering and digital delivery

• Cost of delay — £/week or QALYs/week of non-delivery; the master bridge metric
• DORA metrics — delivery performance, translated into health-economics terms
• Flow metrics — Little's Law, WIP, flow efficiency; the shared queueing math of hospitals and pipelines
• WSJF and CD3 — value-density prioritization; the backlog as a QALY league table
• SPACE and DevEx — multi-dimensional productivity; the EQ-5D lesson for engineering metrics
• Technical debt — principal, interest, and chronic-disease economics for codebases
• Total cost of ownership (TCO) — maintenance is 50–80%; the naive-drug-price error in software
• Cloud unit economics (FinOps) — cost per unit of output; the digital service's reference cost
• Build vs buy — risk-adjusted comparison with the delay term priced
• Benefits realization — auditing that forecast benefits actually happened
• GDS service metrics — cost per transaction, satisfaction, completion, take-up

AI acceleration

• AI developer productivity — Copilot RCT vs METR RCT; efficacy vs effectiveness
• AI return on investment — the 95%-no-return finding and what the 5% did differently
• Inference unit economics — cost per token, and modeling relentless price decline
• AI quality metrics — hallucination rates as harm rates with a price
• Clinical AI evaluation — sensitivity, specificity, AUROC, and why prevalence rules the economics
• AI regulatory evaluation — FDA SaMD, PCCPs, and the economics of model updates

Consumer health apps and devices

• Engagement metrics — engagement as clinical dose
• Retention and churn — the law of attrition; retention curves as treatment windows
• Activation and uptake — the front gates of the value funnel
• Adherence and persistence — MPR, PDC, effective engagement, minimum effective dose
• Patient-reported outcomes — PROMs, PREMs, and the MCID honesty bar
• Digital endpoints and biomarkers — from sensor telemetry to regulator-grade evidence
• Wearable validation — MAPE, agreement statistics, wear-time, completeness
• Remote patient monitoring economics — CPT code stacks and hospital-at-home substitution
• Health app unit economics — CAC, LTV, PMPM, and ROI vs VOI
• Reach and equity — RE-AIM; population impact = reach × effectiveness

Benchmark freshness

Many quoted figures refresh annually (NHS unit costs, payment scheme prices, DORA clusters, DiGA counts, LLM prices). Each doc dates its benchmarks in-line; re-verify before using in a live business case.