2.4 Economic Burden of Fractures
Key Takeaways
Cost estimates depend on the perspective, population, time horizon and included services.
A fracture site can account for a larger share of costs than events. Compare the shares using their stated denominators.
A model projection is conditional on the stated assumptions. Different cost scopes cannot be treated as observed trends.
Clinical outcomes after a fracture liaison intervention and its local financial effects require different evidence.
Identify whose costs are counted
Fractures create costs for health systems, patients and caregivers. An estimate depends on the perspective, population, fracture sites, period and included services. The cost of hospital treatment in the index year answers a different question from the cost of rehabilitation, long-term care and caregiver time after discharge. Separate resources consumed from changes in health or independence. Also record the price year when comparing monetary estimates. They are related measures of burden with different units.
| Type of estimate | What to check |
|---|---|
| Direct medical costs | Check whether inpatient care, outpatient visits, surgery, rehabilitation and long-term medical or institutional care are counted and for how long. |
| Societal cost | Check which direct costs are included and whether the analysis values time away from work and unpaid caregiving. |
| Costs attributed to incident fractures | Check how new events were defined, how repeat visits were assigned, and whether follow-up costs of earlier fractures were counted separately. |
| Projection | Check the model's assumptions for population, prices, fracture rates, services and treatment uptake. A future value is conditional on those assumptions. |
Use a study-specific comparison
A model by Burge and colleagues estimated more than two million osteoporosis-related fractures and almost 17 billion U.S. dollars in related costs in the United States in 2005. Hip fractures were about 14% of modeled events and 72% of costs. The shares use two different denominators. Dividing 72% by 14% gives about 5.1. That compares the model's average cost per hip fracture with its average across all included fractures. The ratio does not mean that 72% of patients fractured a hip or that every hip fracture cost the same amount. Hip fractures can use substantial acute and post-acute resources, and the mix varies with clinical and social needs.
In the same paper, changing the analytic scenario to include current-year costs from certain fractures occurring during the prior five years increased the estimate for 2005 to almost 19.3 billion U.S. dollars. The estimate changed in part because the cost boundary changed. It would be misleading to label the difference as an increase in fracture incidence between two years. When reading a headline, ask which cases and costs the investigators counted.
Distinguish a projection from surveillance
A separate analysis by Lewiecki and colleagues modeled the burden among Medicare-eligible U.S. women. In its status quo scenario, estimated direct medical costs were 48.8 billion U.S. dollars for 2018 and 81.5 billion for 2040. Including relevant indirect costs gave 57.0 billion and 95.2 billion, respectively. These are values within that model. Its scope, assumptions and cost categories differ from those of the 2005 model. Subtracting the estimates across the papers would not give an observed change in spending over time.
For the 2018 estimate within the Lewiecki model, the difference between the broader and direct totals is 8.2 billion U.S. dollars. That difference illustrates the contribution of the modeled indirect categories. Before using a projection for planning, review uncertainty and check whether local age distributions, service use and prices resemble the assumptions. Forecast costs and actual payments can differ.
Connect burden to a prevention decision
After a low-trauma fracture, a fracture liaison service can identify patients, coordinate assessment and help arrange prevention and follow-up. The Norwegian NoFRACT evaluation found lower subsequent fragility fracture risk after implementing a standardized service. That outcome was measured in its care setting and does not establish the savings for a different payer. An economic assessment would also need service implementation costs, local fracture and treatment costs, follow-up time and an appropriate comparator.
For example, a hospital considering coordinated follow-up can track how many eligible patients are identified and evaluated, and how many experience further fractures. It can also measure care costs over a stated period. An increase in completed evaluations measures implementation. A change in fractures or resource use measures a different outcome and requires a design that considers changes in who was treated, time trends and follow-up. Distinguish a program's effect on care from its financial effect when reporting results.
Burge U.S. fracture cost model · Lewiecki U.S. scenario model · NoFRACT evaluation · CDC hip fracture utilization analysis
In a U.S. model, hip fractures accounted for 14% of events and 72% of costs. What does the ratio 72 ÷ 14 describe?
The percentage of hip fractures that required surgery.
About 5.1 times the overall modeled average cost per fracture for the average hip event.
The risk of hip fracture following a previous fracture.
About 5.1 times the cost of every individual nonhip fracture.
In one 2018 scenario, direct costs were estimated at 48.8 billion dollars and the total including indirect costs at 57.0 billion. What does the difference represent?
Observed growth in annual medical payments over one year.
The observed cost of fracture liaison services in the United States.
8.2 billion dollars of included indirect costs within that model.
8.2 million incident hip fractures.
A study in another health system found fewer fractures after introducing a fracture liaison service. What would a local payer need to estimate its own savings?
Use the published fracture result alone to claim the same savings.
Count how many DXA scans occurred and treat that as net savings.
Apply the modeled 2005 hip fracture cost share directly to the local program.
Estimate local program and care costs and compare clinical outcomes and resource use over a defined period.
Sections you finish are checked off in the contents.