Concept Architecture
Concept
Theoretically, Predicted Cost is the expected healthcare expenditure for an individual, provider or population estimated from a statistical or actuarial prediction model. It represents the cost anticipated on the basis of observed risk factors rather than actual expenditure and forms a central component of risk adjustment, economic modelling and healthcare financing. Predicted costs enable fair comparisons between populations by accounting for differences in demographic and clinical characteristics.
Mathematically, Predicted Cost is represented as the fitted value generated by a statistical prediction model. In health economics, the estimate is commonly obtained from multiple linear regression, generalised linear models or other predictive modelling techniques in which healthcare expenditure is expressed as a function of demographic, diagnostic and utilisation variables. The fitted value represents the model's estimate of expected expenditure conditional on the observed characteristics.
In practice, predicted costs are calculated by applying estimated model coefficients to patient-level data derived from administrative claims, electronic health records or survey datasets. The predicted values are used in capitation payment systems, risk adjustment, budget forecasting, provider benchmarking and health economic evaluations. Comparison of predicted and observed costs also provides information on model performance and resource utilisation.
Purpose
Used to estimate expected healthcare expenditure, support risk-adjusted payment systems, forecast healthcare budgets, evaluate provider performance, compare healthcare populations and inform resource allocation decisions.
Mathematical Formulae
Primary Formula
?? = ??? + ???X?? + ???X?? + ? + ???X??
where:
- ?? = predicted healthcare cost for individual i
- X??X? = explanatory variables
- ??????? = estimated model coefficients
Supporting Formulae
Residual:
e? = Y? ? ??
Mean Predicted Cost:
?? = (1/n) ? ? ??
Related Mathematical Methods
- Multiple linear regression
- Generalised linear models
- Risk adjustment
- Predictive modelling
- Ordinary least squares estimation
- Maximum likelihood estimation
Example
A healthcare expenditure model is estimated as:
Predicted Cost = 600 + (45 ? Age) + (900 ? Diabetes)
For a 68-year-old patient with diabetes:
Predicted Cost = 600 + (45 ? 68) + (900 ? 1)
= 600 + 3,060 + 900
= �4,560
The model therefore predicts annual healthcare expenditure of �4,560 for this individual.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| SUMPRODUCT | =SUMPRODUCT(B2:E2,$J$2:$J$5)+$J$1 | Calculates predicted healthcare costs using estimated coefficients. |
| LINEST | =LINEST(B2:B1000,C2:F1000,TRUE,TRUE) | Estimates regression coefficients used to predict costs. |
| FORECAST.LINEAR | =FORECAST.LINEAR(G2,B2:B1000,C2:C1000) | Generates predicted values for simplified prediction models. |
| RSQ | =RSQ(B2:B1000,H2:H1000) | Assesses agreement between observed and predicted healthcare costs. |
VBA (Optional)
A VBA routine can automatically calculate predicted healthcare costs for all individuals by applying stored model coefficients to updated patient data.
Sources
- Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation.
- Iezzoni LI. Risk Adjustment for Measuring Health Care Outcomes.
- van de Ven WPMM, Ellis RP. Risk Adjustment in Competitive Health Plan Markets.
- Drummond MF, Sculpher MJ, Claxton K, Stoddart GL, Torrance GW. Methods for the Economic Evaluation of Health Care Programmes.
- ISPOR Good Practice Reports.
Related Concepts (2)
Library
Publications
1
The Economics of Health and Health Care — Folland, Goodman, Stano & Danagoulian, 9th Edition ed., 2024 (Routledge)
The market-leading general health economics textbook, giving comprehensive coverage of health economics through core economic themes and balancing theory, empirical evidence and public policy. The ninth edition adds chapters on health disparities and pandemic economics.
BookView source →
Frequently Asked Questions (6)
What is a predicted cost?
An estimate of expected future healthcare expenditure generated using a risk adjustment model incorporating demographics, diagnoses, and prior utilisation.
Source: Ellis 2008
What future spending does a predicted cost estimate?
A predicted cost is an estimate of a person's expected future healthcare spending, generated by a risk adjustment model. The model draws on inputs such as demographics, recorded diagnoses, and prior use of care to project what that person is likely to cost. It matters because it lets payments to insurers or providers reflect the expected burden of those they cover, compensating fairly for taking on sicker members. It is the output of risk adjustment, turning a person's characteristics into a forecast figure. Forecasting a person's likely future health costs is what it does. Ellis (2008) sets out such models.
Source: Ellis 2008
How is a predicted cost generated?
A predicted cost is generated using a risk adjustment model that incorporates demographics, diagnoses, and prior utilisation, so these inputs are combined to estimate expected future healthcare expenditure. So a predicted cost is generated from a model using these inputs, which is why they matter, since they inform the estimate, and a predicted cost is generated by a risk adjustment model that draws on demographics, diagnoses, and prior utilisation to estimate an individual's expected future healthcare expenditure.
Source: Ellis 2008
What does a predicted cost estimate?
A predicted cost estimates expected future healthcare expenditure, so it gives a forecast of what an individual's care is likely to cost, based on the risk adjustment model's inputs. So a predicted cost estimates future spending, which is why it is used in risk adjustment, since forecasting cost supports adjustment, and a predicted cost estimates an individual's expected future healthcare expenditure, produced from demographics, diagnoses, and prior utilisation within the risk adjustment model.
Source: Ellis 2008
What inputs does a predicted cost use?
A predicted cost uses demographics, diagnoses, and prior utilisation as inputs to the risk adjustment model, so these characteristics together inform the estimate of expected future healthcare expenditure. So a predicted cost uses demographics, diagnoses, and prior use, which is why these are incorporated, since they predict cost, and a predicted cost uses the demographics, diagnoses, and prior utilisation that the risk adjustment model combines to estimate expected future healthcare expenditure.
Source: Ellis 2008
How does a predicted cost relate to risk adjustment?
A predicted cost relates to risk adjustment as its output: risk adjustment estimates expected costs to adjust payments, and the predicted cost is the estimate the model produces. So a predicted cost is what risk adjustment computes, which is why they are connected, since the model yields the predicted cost, and a predicted cost is the estimate of expected future healthcare expenditure that a risk adjustment model produces, used to adjust payments for the expected costliness of a population.
Source: Ellis 2008
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 16 Jan 2026
Content version: 1.0.0
Canonical Identity
- Persistent URI
- https://healtheconomics.wiki/concept/predicted-cost
- Term code
- HS-HP-HI-136
Stable URI · Machine-readable · Resolvable · CC BY 4.0