SEARCH

SEARCH BY CITATION

Keywords:

  • osteoporosis;
  • epidemiology;
  • fractures;
  • health services and economics;
  • race/ethnicity

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  9. APPENDIX 1.
  10. APPENDIX 2.
  11. APPENDIX 3.
  12. APPENDIX 4.

This study predicts the burden of incident osteoporosis-related fractures and costs in the United States, by sex, age group, race/ethnicity, and fracture type, from 2005 to 2025. Total fractures were >2 million, costing nearly $17 billion in 2005. Men account for >25% of the burden. Rapid growth in the disease burden is projected among nonwhite populations.

Introduction: The aging of the U.S. population will likely lead to greater prevalence of osteoporosis. Policy makers require precise projections of the disease burden by demographic subgroups and skeletal sites to effectively target osteoporosis intervention and treatment programs.

Materials and Methods: A state transition Markov decision model was used to estimate total incident fractures and costs by age, sex, race/ethnicity, and skeletal site for the U.S. population ≥50 years of age for 2005–2025.

Results: More than 2 million incident fractures at a cost of $17 billion are predicted for 2005. Total costs including prevalent fractures are more than $19 billion. Men account for 29% of fractures and 25% of costs. Total incident fractures by skeletal site were vertebral (27%), wrist (19%), hip (14%), pelvic (7%), and other (33%). Total costs by fracture type were vertebral (6%), hip (72%), wrist (3%), pelvic (5%), and other (14%). By 2025, annual fractures and costs are projected to rise by almost 50%. The most rapid growth is estimated for people 65–74 years of age, with an increase >87%. An increase of nearly 175% is projected for Hispanic and other subpopulations.

Conclusions: Osteoporosis prevention, treatment, and education efforts should address all skeletal sites, not just hip and vertebral, and appropriate attention is warranted for men and diverse race/ethnicity subgroups.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  9. APPENDIX 1.
  10. APPENDIX 2.
  11. APPENDIX 3.
  12. APPENDIX 4.

Osteoporosis is a common disease that manifests itself as fractures occurring at multiple skeletal sites, most often at the spine, hip, or wrist, and causes significant morbidity and mortality.(1) Prevalence of osteoporosis in the United States is estimated to increase from ∼10 million to >14 million people in 2020 (based on 2000 census data).(2) Although osteoporosis is less prevalent in men than women, it is estimated that 30% of all hip fractures occur in men.(3) Moreover, studies have shown that the fracture-related mortality rate is higher in men than in women.(4) The U.S. population ≥50 years of age is predicted to increase by 60% between 2000 and 2025, eventually reaching 121.3 million people,(5) with the largest increases expected to occur in nonwhite populations.(5) The projected increase in the elderly U.S. population will likely lead to an increasingly frail population at greater risk of falls and fracture. However, there are no firm estimates of the future fracture burden and its related costs by sex and race/ethnicity.

Current direct medical costs of osteoporosis in the United States have been estimated at 13.7–20.3 billion dollars (adjusted to 2005 dollars).(6–10) Previous studies on the cost burden of osteoporosis used data from the 1980s and early 1990s,(6–10) with none providing potential cost implications of expected demographic changes over the next two decades. Information on the burden of osteoporosis across race/ethnicity and across age groups, sex, and fracture sites is also lacking. In addition, incidence and cost of fractures at sites other than spine, hip, or wrist have not been well documented.

New information on the current and future costs of osteoporosis in the United States will aid clinicians, policy makers, and health care organizations to assess the importance of interventions to reduce osteoporosis-related fractures and associated costs and to identify resource needs, such as rehabilitation programs and long-term care. This study evaluates osteoporosis-related fractures and associated costs across race/ethnicity, age groups, sex, and fracture types in the United States during 2005 and projects costs and fracture incidence to the year 2025.

MATERIAL AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  9. APPENDIX 1.
  10. APPENDIX 2.
  11. APPENDIX 3.
  12. APPENDIX 4.

Model

A Markov state-transition model of osteoporosis was used to estimate the incidence of fractures and related costs in the U.S. population 50–99 years of age.(11) The model has been validated using age-specific hip, spine, and wrist fracture data and age-specific mortality data for women in the United States, Canada, and Sweden(11) and used in published economic analyses.(12–14) The model simulates the natural history of disease by tracking fracture events and costs for cohorts of patients as they transition across health outcome states, using 1-year cycles. These health states include nonfracture or “healthy”; fracture (hip, second hip, vertebral, wrist, pelvic, other); postfracture (for hip, second hip, vertebral, and pelvic fracture); and death (caused by natural causes or after a hip or pelvic fracture). The transition probabilities are based on fracture incidence and mortality rates. In the primary analysis, patient cohorts transition from an initial healthy state to one of the fracture states, the death state, or they can remain in the healthy state in consecutive cycles. Long-term consequences after a hip, pelvic, or vertebral fracture are captured using postfracture health states. The model limits patients to one fracture per cycle and to two hip fractures per lifetime.

Data sources

As detailed below, key data sources for the model included fracture incidence rates, mortality rates, U.S. population estimates, and estimated unit costs for each fracture type.

Epidemiology

The epidemiologic data sources by age, sex, and race/ethnicity included fracture incidence rates for each fracture site,(15,16) mortality rates,(17,18) and population estimates.(19) The race/ethnicity groups were those defined by the U.S. Census Bureau—white non-Hispanic, black, Hispanic (any race), and other (Asian/Pacific Islander, American Indian, other). Population mortality rates and total population estimates for 2005–2025 were specific to the four race/ethnicity groups by age and sex.(17–19) To estimate the full burden of the disease, fracture sites included those most commonly associated with osteoporosis (i.e., hip, vertebral, wrist) and other “nontraditional” skeletal sites (i.e., pelvic, other [humerus, clavicle, hands/fingers, patella, tibia, fibula]) associated with fragility (low- or nontraumatic) fractures, particularly among the elderly.(20,21) Hip fracture incidence rates (where hip fractures were defined using primary ICD-9 codes 820.0x, 820.2x, 820.8x, closed only, and excluding trauma-related cases according to E-codes) were estimated by age, sex, and race/ethnicity from the 2001 Nationwide Inpatient Sample (NIS) database, a large U.S. hospital discharge database,(16) whereas non-hip fracture incidence rates were obtained from a published epidemiology study.(15) Non-hip fracture incidence rates for nonwhite populations were calculated by applying adjustment factors to the fracture incidence rates for whites. These adjustment factors were derived by taking the ratio of hip fracture incidence rates for each nonwhite race/ethnicity group to the white hip fracture incidence rates. (See appendix for fracture incidence rates.)

Unit costs

Unit costs for each fracture site were estimated by age, sex and race/ethnicity, based on inpatient, outpatient, and long-term care cost components. Race-specific cost data were available only for inpatient facility costs, which were estimated from the 2001 NIS database.(16) Hip fracture data used for the inpatient hip cost estimation were consistent with the data used to estimate hip fracture incidence rates. For non-hip fractures, the data used for inpatient cost estimates included hospital discharges with primary diagnoses of 805.0x, 805.2x, 805.4x, 805.8x (closed only) for vertebral fractures; 813.2x, 813.4x, 813.8x (closed only) for wrist/forearm fracture; 808.0x, 808.2x, 808.4x, 808.8x (closed only) for pelvic fracture; and 810.0x, 811.0x, 812.0x, 812.2x, 812.4x, 814.0x, 815.0x, 821.0x, 822.0x, 823.0x, 823.2x, 823.4x, 823.8x (closed only) for “other” fractures. After excluding trauma-related cases, pathologic fractures, cases with secondary diagnosis of fracture, and high-cost outliers, mean inpatient facility charges were converted to costs using a national cost-to-charge ratio (CCR) of 0.61.(7) Outpatient care costs were obtained from published studies.(6,9,22,23) Long-term care costs for hip and pelvic fractures were estimated using published long-term care treatment pathways(24) and nursing home per diem costs.(25) For all other fracture types, long-term care costs were estimated by taking a national proportion of long-term care to inpatient care costs for each fracture type(6) and applying the ratio to the respective inpatient cost estimate. (See appendix for unit cost example.)

Calculation of base-year 2005 fracture incidence and costs

The model was run for 400 unique patient populations (50 age groups × 2 sexes × 4 race/ethnicity groups) for one cycle to estimate base year 2005 incident fractures and costs for each patient population. The total numbers of incident fractures by age/sex/race were estimated by multiplying fracture incidence rates by corresponding population totals. Total costs of incident fractures were calculated by multiplying the number of incident fractures by their corresponding unit cost, or average cost, per fracture type. Results were aggregated across four age groups (50–64, 65–74, 75–84, and ≥85 years) and reported by race and sex.

Cost and fracture projections through 2025

Projections for 2006–2025 were performed by multiplying each base-year fracture and cost distribution by its corresponding annual population distribution.(19) The population projections for age, sex, and race/ethnicity served as future year weights to the base-year distribution for the total United States. Changes in total fractures and costs over time were determined by population growth and demographic changes for age, sex, and race/ethnicity distribution. All projections assumed that medical practice patterns and medical input prices in 2005 remained constant over time.

Sensitivity analysis

The robustness of the base-case analysis was evaluated using one-way sensitivity analyses. Although unit costs of fractures are important model parameters, the linear structure of the model would yield proportionate changes in total costs for a given change in unit costs, if imposed across all fracture types and populations. Similarly, proportionate changes in total incident fractures and total costs would result from a given change in all fracture incidence rates. Therefore, we focused the sensitivity analyses on important subcomponents of unit costs and on changes in incidence rates by fracture type and for subpopulations.

To evaluate the impact of changes in unit costs of fracture, a low-cost scenario was created by reducing the CCR value from 0.61 to 0.50 and applying it to inpatient facility charges. This reduction lowered the estimate for inpatient costs below the lower limit of the 95% CI for hospital charges. For all other cost categories, the costs were reduced by 25%, which is commonly done when measures of statistical variation are unavailable. A high-cost scenario was created by increasing all noninpatient facility costs by 25%, whereas the base-case inpatient facility costs were used to be conservative for total upper-bound unit costs.

The effects on the base case analysis from changes in fracture incidence rates were assessed by imposing changes on incidence rates by fracture type and for subpopulations and by applying a declining fracture trend across all fracture incidence rates. First, we changed the base case hip fracture incidence rates by ±25% for all populations and reran the model to estimate low and high hip fracture incidence rate scenarios. Second, to assess effects from using non-hip fracture rates from a single geographic area for the total United States, changes of ±25% were imposed. Third, low and high fracture incidence rate scenarios for nonwhite race/ethnicity groups were created by changing fracture incidence rates by ±25% for all fracture types. A fourth analysis applied a 4-year linear decline in all incidence rates through 2008, and then leveling off, to capture observed trends in hip fracture incidence rates from an analysis of NIS 1998–2001 hospital discharge data.

The final sensitivity analysis involved the inclusion of costs from prevalent fractures (i.e., among those individuals in 2005 having survived fractures in previous years), which were not included in the base model. A 5-year natural history period was used to construct current costs of previous hip, spine, and pelvic fractures. The costs in the base year caused by previous fractures for each individual age cohort were estimated from the simulation of the natural history of disease in previous or younger cohorts. The detailed methodology is described elsewhere.(26)

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  9. APPENDIX 1.
  10. APPENDIX 2.
  11. APPENDIX 3.
  12. APPENDIX 4.

Base year 2005

In 2005, the model predicts total fractures to exceed 2.0 million (Fig. 1A) and costs to be more than $16.9 billion (Fig. 1B). The overall distribution of incident fractures shows that “other” fractures comprise the largest share (33%), followed by vertebral fractures (27%), wrist fractures (19%), hip fractures (14%), and pelvic fractures (7%). Nonvertebral fractures, therefore, represent 73% of the total, whereas the share of nontraditional osteoporosis-related fractures (i.e., “other” and pelvic) is 40%. The total cost distribution by fracture type is skewed toward hip fractures, which account for 72% of total costs (but only 14% of fractures), followed by “other” fractures (14%), vertebral fractures (6%), pelvic fractures (5%), and wrist fractures (3%). Thus, nonvertebral fractures account for 94% of total costs, with the nontraditional osteoporosis-related fractures making up 19% of the total. The overall distribution of fracture costs by site of care is 57% for inpatient care, 13% for outpatient care, and 30% for long-term care (LTC; data not shown).

thumbnail image

Figure Figure 1. (A) Fracture distribution by type, 2005. (B) Cost distribution by type, 2005.

Download figure to PowerPoint

The majority of fractures (71%) occur among women (Table 1). The distribution of fractures in women differs slightly from the overall combined male and female distribution, where women have larger shares of hip (15%) and wrist fractures (22%), but proportionately fewer “other” fractures (29%). By age group, the number of fractures is highest among 50–64 (26%) and 75–84 year olds (31%). As a group, women 65+ years of age accounted for 74% of all fractures. The distribution of fractures by race/ethnicity among women reveals that whites have 89% of the total, followed by black (4%), Hispanic (4%), and other women (3%).

Table Table 1.. Incident Fractures by Fracture Type, Sex, Age, and Race for 2005
Thumbnail image of

Men are predicted to incur almost 595,000 fractures (29%) in 2005 (Table 1). The fracture distribution among men is weighted more toward “other” fractures (44%), but less toward hip (12%), wrist (12%), and pelvic fractures (5%) compared with women. Vertebral fractures make up the same proportion of fractures in men and in women (27%). Unlike women, total fractures decline within older age groups in men. About 39% of fractures were incurred by men 50–64 of age, whereas only 17% were in men ≥85 years of age. Fractures in men 65+ years comprise 61% of the fracture total. The distribution of incident fractures across race/ethnicity groups differs in men versus women, as proportionately more fractures occur among nonwhites (20% in men versus only 11% in women).

Over three quarters of the total costs of incident fractures were among women ($12.8 billion) in 2005 (Table 2). The costs are distributed across fracture types similarly to the combined sex distribution. By age group, total costs increase substantially with higher age. Among women, only 11% and 13% of fracture costs occur in the age groups 50–64 and 65–74 years, respectively; whereas 36% and 40% of costs occur in age groups 75–84 and ≥85 years of age, respectively. The senior female population 65+ years of age bears the overwhelming share of total costs (89%). The disproportionately large share of costs in women ≥75 years of age reflects this group's higher mix of more expensive hip and pelvic fractures (i.e., 34% of total fractures in this age group) compared with women <65 years of age (where only 9% of fractures are hip or pelvic). Total costs by race/ethnicity follow the same distribution as incident fractures, where ∼90% of the total accrues to whites.

Men account for 24% ($4.1 billion) of total costs (Table 2). The distribution of costs by fracture differs for men compared with women, because a lower proportion results from hip fractures (67%), whereas a higher share is found for vertebral (10%) and “other” fractures (15%). Costs generally increase with age, although costs within the male 50- to 64-year-old group are higher than those for male 65–74 year olds ($790 million [19%] versus $667 million [16%]; Table 2). Men 65+ years of age account for 81% of men's total costs. By race/ethnicity, ∼84% of men's costs are caused by fractures in white men, who comprise 80% of men's fractures.

Table Table 2.. Fracture Costs by Fracture Type, Sex, Age, and Race for 2005
Thumbnail image of

Projections 2006–2025

By 2025, fractures and costs are projected to grow by >48% to >3 million fractures, incurring $25.3 billion in costs (Figs. 2A–2C). The cumulative cost of incident fractures is predicted to rise from $209 billion during 2006–2015 to $228 billion for 2016–2025 (annual data not shown). During 2006–2015, annual fracture incidence will rise by 22%, whereas annual costs will increase >20%. Across fracture types, the largest changes are predicted for pelvic fractures, where incidence increases by 56% and costs rise by 60% between 2005 and 2025 (Fig. 2A). By race/ethnicity, the proportion of fractures and costs among the nonwhite population will increase from 14% and 12% in 2005, respectively, to 21% and 19% in 2025 (Figs. 2B and 2C).

thumbnail image

Figure Figure 2. (A) Growth in fractures and costs from 2005 to 2025, by fracture site. (B) Fracture and cost shares by race: 2005, 2025. (C) Fracture and cost shares by sex: 2005, 2025.

Download figure to PowerPoint

The projected effects of changes in population demographics on fracture costs by race/ethnicity, and particularly for nonwhite populations, are reflected in Table 3. The most rapid increase is projected to occur in the Hispanic and other subpopulations. The annual costs for Hispanics grew from $754 million in 2005 to over $2 billion per year by 2025 for an increase of 175%. Similarly, the other population shows cost increases of 175%, starting from a smaller 2005 total of $502 million and rising to more than $1.38 billion per year in 2025 (Table 3). Costs within the other population are projected to surpass those in the black population by 2021.

Table Table 3.. Costs by Race/Ethnicity: 2005, 2010, 2015, 2020, and 2025*
Thumbnail image of

Sensitivity analyses

The low-cost scenario reduces total 2005 and 2025 fracture costs by 22%, and the high-cost scenario increases costs by 12% (Table 4). In the base case, adjustment of fracture incidence rates for each fracture type by ±25% produces the largest impact on the number of incident fractures for the “other” (±8%) and vertebral (±7%) fracture sites, and it produces the largest impact on costs for hip fractures (±18%), followed by “other” fractures (±3%). Second, changes in all non-hip fracture incidence rates of ±25% result in changes of ±21% in fractures, but only ±7% in costs. Third, among the nonwhite population, the change in all fracture incidence rates of ±25% reveals a 3% impact on total fractures and costs in 2005 and a 5% impact by 2025. Fourth, when a declining trend was imposed on fracture incidence rates through 2008, the initial effect on base-year 2005 fractures, and costs was a reduction of 4%, with an impact on projected fractures and costs of 15% in 2025. The scenario including current-year costs caused by prevalent fractures (hip, vertebral, pelvic) during the previous 5 years increased the 2005 cost estimate by 14% to almost $19.3 billion. The combined cumulative cost of both incident and prevalent fractures was predicted to rise from $215 billion during 2006–2015 to $259 billion during 2016–2025 (annual data not shown).

Table Table 4.. Sensitivity Analyses for Base Year 2005 and 2025
Thumbnail image of

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  9. APPENDIX 1.
  10. APPENDIX 2.
  11. APPENDIX 3.
  12. APPENDIX 4.

This study used a validated model of osteoporosis, populated with recent epidemiologic and cost data, to estimate the total number of incident fractures and associated costs for the U.S. population ≥50 years of age from 2005 through 2025. This study adds to previous research on the burden of osteoporosis by providing current and projected annual estimates of incident fractures and costs by race/ethnicity, age, sex, and by fracture type. Several earlier studies used an assortment of databases from the 1980s through mid-1990s to estimate the cost burden of disease in the United States.(6,7,9,10) However, none has provided the overall fracture incidence by sex, race/ethnicity, age, and fracture type alongside the cost estimates. Ray et al.(6) estimated the 1995 costs of prevalent and incident fractures by sex, race/ethnicity, age, and fracture type. However, estimates on fracture incidence across all skeletal sites by sex, race/ethnicity, and age were not provided; nor did that study project estimates of the disease burden for the near future. Using a Markov model based on fracture-related functional impairment, Chrischilles et al.(8) produced 10-year projections of incident fractures and costs, although the analysis was limited to three fracture sites (hip, spine, distal forearm) and to white women.

For 2005, the total burden of osteoporosis is estimated to be >2 million incident fractures at a cost of almost $17 billion in the United States (more than $19 billion if costs of prevalent fractures are included), which is consistent with earlier estimates.(6–10) Nonvertebral fractures represent 73% of total incident fractures and 94% of total costs, with nontraditional osteoporosis-related sites (pelvic and “other”) accounting for 40% of fractures and almost one fifth of the total costs. These results illustrate the importance of osteoporosis treatments with shown efficacy beyond the spine and across multiple skeletal sites such as bisphosphonates,(27–29) PTH,(30) and strontium ranelate.(31) Prevention, treatment, and education efforts should be concentrated on all skeletal sites, rather than just hip and vertebral sites. Men account for 29% of fractures and costs, showing that this disease is not restricted to women.

An assessment of costs by site of care reveals that the majority of costs are incurred by inpatient care (57%) and LTC (30%) versus 13% by outpatient care. These estimates by site of care fall within those found in earlier studies.(6,7) The disproportionate contribution to costs of different age groups in this study shows that most of the burden of osteoporosis (70% and 87% of fractures and costs, respectively) is incurred by those ≥65 years of age (the Medicare population), and that most costs are related to inpatient care and LTC. Earlier research suggests that osteoporosis diagnosis and treatment may be budget neutral or produce savings for the Medicare program.(14) However, LTC costs will pose a significant burden on state Medicaid programs. Patients 50–64 years of age are predicted to account for 30% of fractures but only 13% ($2.2 billion) of total costs in 2005.

By 2025, the burden in the United States is projected to grow by almost 50% to >3 million fractures and $25.3 billion each year. By race/ethnicity group, our model predicts a 2.7-fold increase in fracture incidence and costs for Hispanic and other subpopulations. Over the next 20 years, the nonwhite population will comprise a growing proportion of total osteoporosis fractures and related costs. The prevalence and impact of chronic diseases other than osteoporosis, such as diabetes, hypertension, cardiovascular disease, and asthma, have also been shown to vary across racial and ethnic subgroups.(32–35) These variations translate to clinical differences in treatment strategies and risk factors for hard-to-reach and at-risk populations.

Important limitations of the study include the data sources used in the model regarding fracture incidence rates and unit costs of fractures. The projection results also were dependent on U.S. Census Bureau population projections. Over- or underestimation of these inputs would directly influence the base-case estimates. The key assumptions for fracture incidence rates were the use of non-hip fracture incidence rates from a single county (Olmsted County, MN, USA) to represent the entire United States and the application of adjustment factors to these rates to derive non-hip fractures incidence rates for nonwhite populations. Our use of race-specific hip fracture incidence rates (calculated from national hospital discharge data) as adjustment factors to derive nonwhite, non-hip fracture incidence rates was based on the assumption that the ratio of non-hip to hip fractures was constant across race/ethnicity. No other U.S. fracture incidence data for all non-hip fracture sites by sex were available, underscoring the need for additional research in this area, although our model can easily accommodate newer data from future studies. Because the model is linear, reductions across all fractures types would result in proportional changes in incidence and costs; therefore, our sensitivity analyses were conducted on subgroups and cost components. For non-hip fractures, the impacts on total incident fractures were less than proportionate, but the effects on costs were tempered by the 28% share of total costs for non-hip fractures, resulting in a modest change in base-case costs. The use of alternative adjustment factors to derive race-specific non-hip fracture incidence rates for nonwhite populations led to relatively small changes to the base case. We also evaluated the impact of using static fracture incidence rates for the base-year and for future projections by imposing a declining fracture trend across all fracture sites for a 4-year period. The results from this scenario indicate that the base-year estimates are reasonably stable, but the projections are sensitive to secular trends in fracture incidence rates.

The derivation of unit costs involved combining published sources for outpatient care and inpatient physician care with calculated estimates for long-term and inpatient facility care. It is noteworthy that our unit cost estimates for hip fracture—the most important cost driver—in almost every subgroup were considerably below the estimates used in previous studies (when updated to 2005 dollars).(6,10,36) Our inpatient unit costs excluded trauma-related cases, pathological fractures, and cases with fractures as a secondary diagnosis. Also, we did not capture indirect costs such as lost productivity, unpaid caregiver time, transportation costs, and other social services. To address the uncertainty in unit costs, we applied a lower cost-to-charge ratio of 0.5 to convert inpatient charges to costs, and we reduced all other cost components by 25% for a low-cost scenario. In a high-cost scenario, inpatient costs were held at the base level, whereas all other costs were increased by 25%. The effects on the base results were less than proportionate but still were somewhat sensitive to these assumptions.

Another limitation concerns the projection method used in the study. The projection results from 2006 through 2025 were based on population estimates by age, sex, and race/ethnicity. Therefore, the accuracy of our projections is strongly dependent on the accuracy of the population projections by the U.S. Census Bureau. However, we did not explore alternative population projection series in our study.

In conclusion, the burden of osteoporosis for 2005 is estimated to be >2 million incident fractures, creating direct medical costs of $17 billion for the United States. Nonvertebral fractures account for 94% of costs and 73% of fractures, with nontraditional osteoporosis-related skeletal sites comprising 40% of fractures and almost 20% of costs. Almost 30% of fractures and a quarter of the total cost burden are borne by men. By 2025, annual fractures and costs are projected to grow by 50% and will surpass 3 million and $25 billion, respectively. Over time, the nonwhite population is predicted to suffer a rapidly increasing share of the disease burden. The broad and asymmetric impact of disease across age, sex, race/ethnicity, and skeletal sites found in this study underscores the need for targeted disease intervention and treatment programs to populations with growing and unrecognized needs. Our study results may assist health care policy-makers and providers in allocating scarce resources for this disease in more cost-effective ways. By doing so, the cost and fracture trajectories predicted for the U.S. population may be favorably altered.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  9. APPENDIX 1.
  10. APPENDIX 2.
  11. APPENDIX 3.
  12. APPENDIX 4.

We are thankful to Eric Balda, Procter & Gamble, and Douglas Smith, Computer Consultant and Brainbench Excel MVP, Abundant Solutions, for programming support. The Alliance for Better Bone Health (Procter & Gamble Pharmaceuticals, Cincinnati, OH, USA, and sanofi-aventis, Bridgewater, NJ, USA) sponsored all work by Dr Burge, Dr King, and data programmers, but did not provide any funding for this study to Drs Dawson-Hughes, Solomon, Tosteson, or Wong. Dr Tosteson was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (AR048094) and the National Institute on Aging (AG12262). Work by Dr Solomon was supported by the National Institutes of Health (NIH AR-48616). The sponsor's role included review of the manuscript. All team members had full access to all data and analytic models.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  9. APPENDIX 1.
  10. APPENDIX 2.
  11. APPENDIX 3.
  12. APPENDIX 4.
  • 1
    Barrett-Connor E 1995 The economic and human costs of osteoporotic fracture. Am J Med 98: 3S8S.
  • 2
    National Osteoporosis Foundation 2002 America's Bone Health: The State of Osteoporosis and Low Bone Mass in Our Nation. Washington, DC: National Osteoporosis Foundation.
  • 3
    Campion JM, Maricic MJ 2003 Osteoporosis in men. Am Fam Physician 67: 15211526.
  • 4
    Olszynski WP, Shawn Davison K, Adachi JD, Brown JP, Cummings SR, Hanley DA, Harris ST, Hodsman AB, Kendler D, McClung M, Miller PD, Yuen CK 2004 Osteoporosis in men: Epidemiology, diagnosis, prevention, and treatment. Clin Ther 26: 1528.
  • 5
    Day JC 1996 Population Projections of the United States by Age, Sex, Race, and Hispanic Origin: 1995 to 2050. Washington, DC: U.S. Government Printing Office.
  • 6
    Ray NF, Chan JK, Thamer M, Melton LJ III 1997 Medical expenditures for the treatment of osteoporotic fractures in the United States in 1995: Report from the National Osteoporosis Foundation. J Bone Miner Res 12: 2435.
  • 7
    Hoerger TJ, Downs KE, Lakshmanan MC, Lindrooth RC, Plouffe L Jr, Wendling B, West SL, Ohsfeldt RL 1999 Healthcare use among U.S. women aged 45 and older: Total costs and costs for selected postmenopausal health risks. J Womens Health Gend Based Med 8: 10771089.
  • 8
    Chrischilles E, Shireman T, Wallace R 1994 Costs and health effects of osteoporotic fractures. Bone 15: 377386.
  • 9
    Phillips S, Fox N, Jacobs J, Wright WE 1988 The direct medical costs of osteoporosis for American women aged 45 and older, 1986. Bone 9: 271279.
  • 10
    Holbrook TL, Grazier K, Kelsey JL, Stauffer RN 1984. Frequency of Occurence, Impact and Cost of Selected Musculoskeletal Conditions in the United States. Chicago, IL: American Academy of Orthopedic Surgeons.
  • 11
    Tosteson AN, Jonsson B, Grima DT, O'Brien BJ, Black DM, Adachi JD 2001 Challenges for model-based economic evaluations of postmenopausal osteoporosis interventions. Osteoporos Int 12: 849857.
  • 12
    Grima DT, Burge RT, Becker D, Tosteson AN 2002 Short-term cost-effectiveness of bisphosphonate therapies for post-menopausal osteoporotic women at high risk of fracture. Pharma Therap 27: 448455.
  • 13
    Brecht JG, Kruse HP, Felsenberg D, Mohrke W, Oestreich A, Huppertz E 2003 Pharmacoeconomic analysis of osteoporosis treatment with risedronate. Int J Clin Pharmacol Res 23: 93105.
  • 14
    King AB, Saag KG, Burge RT, Pisu M, Goel N 2005 Fracture reduction affects Medicare economics (FRAME): Impact of increased osteoporosis diagnosis and treatment. Osteoporos Int 16: 15451557.
  • 15
    Melton LJ III, Crowson CS, O'Fallon WM 1999 Fracture incidence in Olmsted County, Minnesota: Comparison of urban with rural rates and changes in urban rates over time. Osteoporos Int 9: 2937.
  • 16
    Nationwide Inpatient Sample 2001 Release 10, Healthcare Cost and Utilization Project. Rockville, MD: Agency for Healthcare Research and Quality.
  • 17
    Arias E 2004 United States life tables, 2001. Natl Vital Stat Rep 52: 138.
  • 18
    National Center for Health Statistics 1998 U.S. Decennial Life Tables for 1989-91, vol II. Hyattsville, MD: National Center for Health Statistics.
  • 19
    U.S. Census Bureau 2004 U.S. interim projections by age, sex, race, and Hispanic origin 2000-2050. Available at http://www.census.gov/ipc/www/usinterimproj/. Accessed November 28, 2005.
  • 20
    Klotzbuecher CM, Ross PD, Landsman PB, Abbott TA, Berger M 2000 Patients with prior fractures have an increased risk of future fractures: A summary of the literature and statistical synthesis. J Bone Miner Res 15: 721739.
  • 21
    U.S. Department of Health and Human Services 2004 Bone Health and Osteoporosis: A Report of the Surgeon General. Rockville, MD: U.S. Department of Health and Human Services, Office of the Surgeon General.
  • 22
    Osteoporosis 1998 Review of the evidence for prevention, diagnosis and treatment and cost-effectiveness analysis. Osteoporos Int 8: S7S80.
  • 23
    U.S. Congress Office of Technical Assistance 1994 Hip fracture outcomes in people age fifty and over. Washington, DC: U.S. Congress Office of Technical Assistance.
  • 24
    Braithwaite RS, Col NF, Wong JB 2003 Estimating hip fracture morbidity, mortality and costs. J Am Geriatr Soc 51: 364370.
  • 25
    Grabowski DC, Feng Z, Intrator O, Mor V 2004 Recent trends in state nursing home payment policies. Health Aff (Millwood) W4: 363373.
  • 26
    Burge RT, King AB, Balda E, Worley D 2003 Methodology for estimating current and future burden of osteoporosis in state populations: Application to Florida in 2000 through 2025. Value Health 6: 574583.
  • 27
    Harris ST, Watts NB, Genant HK, McKeever CD, Hangartner T, Keller M, Chesnut CH III, Brown J, Eriksen EF, Hoseyni MS, Axelrod DW, Miller PD 1999 Effects of risedronate treatment on vertebral and nonvertebral fractures in women with postmenopausal osteoporosis: A randomized controlled trial. JAMA 282: 13441352.
  • 28
    Reginster J, Minne HW, Sorensen OH, Hooper M, Roux C, Brandi ML, Lund B, Ethgen D, Pack S, Roumagnac I, Eastell R 2000 Randomized trial of the effects of risedronate on vertebral fractures in women with established postmenopausal osteoporosis. Osteoporos Int 11: 8391.
  • 29
    Black DM, Thompson DE, Bauer DC, Ensrud K, Musliner T, Hochberg MC, Nevitt MC, Suryawanshi S, Cummings SR 2000 Fracture risk reduction with alendronate in women with osteoporosis: The Fracture Intervention Trial. FIT Research Group. J Clin Endocrinol Metab 85: 41184124.
  • 30
    Neer RM, Arnaud CD, Zanchetta JR, Prince R, Gaich GA, Reginster JY, Hodsman AB, Eriksen EF, Ish-Shalom S, Genant HK, Wang O, Mitlak BH 2001 Effect of parathyroid hormone (1-34) on fractures and bone mineral density in postmenopausal women with osteoporosis. N Engl J Med 344: 14341441.
  • 31
    Reginster JY, Seeman E, De Vernejoul MC, Adami S, Compston J, Phenekos C, Devogelaer JP, Curiel MD, Sawicki A, Goemaere S, Sorensen OH, Felsenberg D, Meunier PJ 2005 Strontium ranelate reduces the risk of nonvertebral fractures in post-menopausal women with osteoporosis: Tropos Study. J Clin Endocrinol Metab 90: 28162822.
  • 32
    Jiang HJ, Andrews R, Stryer D, Friedman B 2005 Racial/ethnic disparities in potentially preventable readmissions: The case of diabetes. Am J Public Health 95: 15611567.
  • 33
    Oparil S, Wright JT Jr 2005 Ethnicity and blood pressure. J Clin Hypertens (Greenwich) 7: 357364.
  • 34
    Saha C, Riner ME, Liu G 2005 Individual and neighborhood-level factors in predicting asthma. Arch Pediatr Adolesc Med 159: 759763.
  • 35
    Hicks LS, Shaykevich S, Bates DW, Ayanian JZ 2005 Determinants of racial/ethnic differences in blood pressure management among hypertensive patients. BMC Cardiovasc Disord 5: 16.
  • 36
    U.S. Congress Office of Technology Assessment. 1994 Hip Fracture Outcomes in People Age 50 and Over—Background Paper. Washington, DC: U.S. Government Printing Office.

APPENDIX 1.

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIAL AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  9. APPENDIX 1.
  10. APPENDIX 2.
  11. APPENDIX 3.
  12. APPENDIX 4.
Table  . Unit Costs of Fracture for White Race, by Sex and Age Group, Total United States, 2005
thumbnail image