# Health Economics: Cost-Effectiveness and Cost-Benefit Analysis

## Overview

Health economic evaluation provides systematic frameworks for comparing the costs and outcomes of health interventions, enabling rational resource allocation in settings where resources are inherently limited. These methods are essential for decision-making in public health, clinical medicine, and health policy. Four main types of economic evaluation exist: cost-minimization analysis, cost-effectiveness analysis, cost-utility analysis, and cost-benefit analysis. Preventive medicine physicians must be able to interpret and apply economic evidence to guide program and policy decisions.

## Types of Economic Evaluation

### Cost-Minimization Analysis (CMA)

Cost-minimization analysis applies when two interventions produce equivalent outcomes, simplifying the comparison to costs alone. The goal is identifying the least expensive option. This approach is rarely used in practice because truly equivalent outcomes are uncommon. An example would be comparing a generic drug to its brand-name equivalent when bioequivalence has been conclusively established.

### Cost-Effectiveness Analysis (CEA)

Cost-effectiveness analysis compares interventions by calculating the cost per unit of health outcome achieved. Outcomes are measured in natural units — cost per life-year gained, cost per case prevented, or cost per unit of clinical improvement. The central metric is the incremental cost-effectiveness ratio (ICER), calculated as the difference in costs divided by the difference in effects between two interventions. A limitation is that CEA cannot compare across different health conditions when outcomes are measured in different natural units. It remains the most commonly used approach for evaluating clinical preventive services.

### Cost-Utility Analysis (CUA)

Cost-utility analysis is a subtype of CEA that uses quality-adjusted life-years (QALYs) as the universal outcome measure. Because QALYs combine both length and quality of life, this approach allows comparison across different diseases and interventions. The ICER is expressed as cost per QALY gained. Cost-utility analysis is the dominant framework for health technology assessment worldwide.

### Cost-Benefit Analysis (CBA)

Cost-benefit analysis measures both costs and outcomes in monetary terms. Health outcomes must be converted to dollar values using approaches such as willingness-to-pay surveys or the human capital method. The advantage is that CBA allows comparison across sectors entirely — health versus education versus transportation, for example. Net benefit equals the monetary value of benefits minus costs. The approach remains controversial because it requires assigning explicit monetary value to human life and health.

| Analysis Type | Outcome Measure | Result Expressed As | Cross-Condition Comparison | Example |
|---|---|---|---|---|
| Cost-Minimization (CMA) | Assumed equivalent | Cost difference only | No | Generic vs. brand-name drug |
| Cost-Effectiveness (CEA) | Natural units (life-years, cases prevented) | Cost per unit of effect (ICER) | No (different units) | Cost per life-year gained from screening |
| Cost-Utility (CUA) | QALYs | Cost per QALY gained (ICER) | Yes | Comparing cancer screening vs. statin therapy |
| Cost-Benefit (CBA) | Monetary (dollars) | Net benefit (benefits − costs) | Yes (including non-health sectors) | Health program vs. education program |

## Key Concepts

### Quality-Adjusted Life-Years (QALYs)

QALYs combine quantity and quality of life into a single metric. One QALY equals one year of life lived in perfect health. Health utility scores range from 0 (representing death) to 1 (perfect health), with some health states valued as worse than death (negative utility). Methods for measuring utility include the standard gamble (a choice between a certain outcome and a gamble between perfect health and death), time trade-off (willingness to trade life-years for improved health state), the visual analog scale (rating on a 0-100 scale), and multi-attribute instruments such as EQ-5D, SF-6D, and HUI that derive utilities from population survey data.

### Disability-Adjusted Life-Years (DALYs)

DALYs measure disease burden as years of healthy life lost. A DALY consists of years of life lost due to premature death (YLL) plus years lived with disability (YLD). DALYs are used primarily in global health contexts by the WHO and the Global Burden of Disease study. Unlike QALYs, which measure health gained, DALYs measure health lost. Disability weights are assigned through expert panels and population surveys.

### Willingness-to-Pay (WTP) Thresholds

The willingness-to-pay threshold represents the maximum amount a decision-maker will pay for one additional unit of health outcome. In the United States, the commonly cited range is $50,000 to $150,000 per QALY gained. The WHO-CHOICE framework historically used one to three times GDP per capita per DALY averted. These thresholds are not fixed rules but guides for decision-making. Interventions with ICERs below the threshold are generally considered "cost-effective."

### Incremental Cost-Effectiveness Ratio (ICER)

The ICER equals the difference in costs divided by the difference in effectiveness between two alternatives. Interpretation depends on which quadrant of the cost-effectiveness plane the result falls in. A dominant intervention is both more effective and less costly — it should always be adopted. A dominated intervention is less effective and more costly — it should always be rejected. The most common scenario places an intervention in the northeast quadrant (more effective but more costly), where the ICER is compared to the WTP threshold to determine whether the additional benefit justifies the additional cost. The southwest quadrant (less effective and less costly) is rarely of clinical interest.

### Cost-Effectiveness Plane

The cost-effectiveness plane plots incremental cost on the y-axis against incremental effectiveness on the x-axis, creating four quadrants. The northeast quadrant (more effective, more costly) represents the most common real-world scenario. The southeast quadrant (more effective, less costly) represents dominance. The northwest quadrant (less effective, more costly) represents being dominated. The southwest quadrant represents less effective and less costly alternatives.

## Analytical Methods

### Perspective

The perspective of the analysis determines which costs and benefits are counted. The societal perspective is the broadest, including all costs and benefits regardless of who bears them — patient costs, payer costs, employer productivity losses, and more. The healthcare system or payer perspective includes only direct medical costs. The patient perspective focuses on out-of-pocket costs and health outcomes. The Second Panel on Cost-Effectiveness in Health and Medicine recommends conducting analyses from both a healthcare sector and a societal perspective.

### Time Horizon

The time horizon must be long enough to capture all relevant costs and outcomes. Chronic disease prevention programs may require a lifetime horizon. Short time horizons risk missing the long-term benefits that make prevention programs worthwhile.

### Discounting

Future costs and outcomes are valued less than present ones, reflecting time preference. The standard discount rate in the United States is 3% per year, applied to both costs and health outcomes. Discounting significantly penalizes prevention programs whose benefits accrue far in the future. Sensitivity analyses should test alternative discount rates, typically 0% and 5%.

### Sensitivity Analysis

One-way sensitivity analysis varies one parameter at a time to assess its impact on results. Multi-way analysis varies two or more parameters simultaneously. Probabilistic sensitivity analysis (PSA) assigns probability distributions to all uncertain parameters and uses Monte Carlo simulation to characterize overall uncertainty. A tornado diagram displays the relative impact of each parameter on the ICER, ranking them from most to least influential. A cost-effectiveness acceptability curve (CEAC) plots the probability that an intervention is cost-effective across a range of WTP thresholds.

### Decision Models

Decision trees are suitable for short-term, one-time decisions with discrete outcomes. Markov models handle chronic diseases with recurring transitions between health states over time. Microsimulation models simulate individual patients, allowing heterogeneous characteristics and complex disease histories. Dynamic transmission models are required for infectious diseases where an intervention affects disease transmission at the population level.

## Budget Impact Analysis

Budget impact analysis is distinct from cost-effectiveness analysis. Rather than assessing value, it estimates the financial consequences of adopting a new intervention on a specific budget, such as a Medicaid program or health system. It has a short-term focus, typically one to five years, and addresses affordability rather than value. This analysis is critical for real-world implementation decisions because even highly cost-effective interventions may be unaffordable in the short term.

## Common Pitfalls and Limitations

QALYs may systematically disadvantage elderly and disabled populations who have fewer potential QALYs to gain regardless of intervention effectiveness. Industry-funded analyses may incorporate assumptions that favor the sponsor's product. Model structure and parameter choices drive results substantially, making sensitivity analyses essential for credibility. Standard CEA does not address equity — a QALY gained in a disadvantaged population receives the same weight as one gained in an advantaged population. Emerging approaches such as distributional cost-effectiveness analysis (DCEA) and equity-weighted frameworks attempt to incorporate equity considerations.

<image>A cost-effectiveness plane divided into four quadrants with the x-axis showing incremental effectiveness and y-axis showing incremental cost. The northeast quadrant is labeled "more effective, more costly" with a willingness-to-pay threshold line. The southeast quadrant is labeled "dominant (more effective, less costly)." The northwest quadrant is labeled "dominated (less effective, more costly)." The southwest quadrant is labeled "less effective, less costly." Scattered points represent hypothetical interventions. A dashed diagonal line represents the WTP threshold. Medical health economics education illustration.</image>

<image>A tornado diagram showing one-way sensitivity analysis results for a hypothetical cost-effectiveness analysis. Horizontal bars represent different model parameters (discount rate, treatment efficacy, baseline risk, costs, utility weights) arranged from most influential at top to least at bottom. Each bar shows the range of ICER values as the parameter varies across its plausible range. A vertical line indicates the base-case ICER. Health economics teaching illustration with clear labels.</image>

<image>A diagram explaining QALYs visually, showing two scenarios for a patient: (1) natural history without intervention showing declining health utility over time leading to death, and (2) with intervention showing improved health utility and extended survival. The shaded area between the two curves represents QALYs gained. X-axis shows time in years, y-axis shows health utility from 0 to 1. Clear annotations label each component. Medical education illustration for health economics.</image>

## Clinical Pearls

An intervention that is cost-effective is not necessarily affordable — budget impact analysis separately addresses the affordability question. Discounting at 3% significantly reduces the apparent benefit of prevention programs whose payoff lies decades in the future, such as childhood vaccination or cancer prevention. The widely cited $50,000 per QALY threshold originated from the approximate cost of dialysis in the 1970s and was never formally adopted as policy. The prevention paradox applies to economic evaluation: interventions with modest individual benefit but large population impact may prove highly cost-effective. For board preparation, be able to distinguish CEA from CBA from CUA, interpret an ICER, and explain why discounting matters for the evaluation of prevention programs.

## References
- Drummond MF, et al. Methods for the Economic Evaluation of Health Care Programmes. 4th ed. Oxford University Press; 2015.
- Sanders GD, et al. Recommendations for conduct, methodological practices, and reporting of cost-effectiveness analyses: Second Panel on Cost-Effectiveness in Health and Medicine. JAMA. 2016;316(10):1093-1103.
- Neumann PJ, et al. Cost-effectiveness in Health and Medicine. 2nd ed. Oxford University Press; 2016.
- Murray CJL, et al. Measuring the global burden of disease. N Engl J Med. 2013;369(5):448-457.
- Weinstein MC, Stason WB. Foundations of cost-effectiveness analysis for health and medical practices. N Engl J Med. 1977;296(13):716-721.