Cost-effectiveness strategies to treat osteoporosis in elderly women.

Background: Comparing the cost-effectiveness of various antiosteoporotic drugs has not been defined.

Methods: We determined the cost-effectiveness of calcitonin, raloxifene, bisphosphates and PTH in a base-case cohort of women aged 65 or older with osteoporosis.

After bone densitometry, women were stratified into groups of treatment or no treatment. Our outcome goal was a value of $100,000 or less per quality-adjusted life years (QALY). A sensitivity analysis varied nonvertebral fracture reduction and compliance between the two most effective strategies to test various cost per QALY thresholds.

Results: Bisphosphonates displayed the most favorable incremental cost saving and prevented more fractures in our base-case analysis. In a sensitivity analysis, virtually all values of bisphosphonates were under $100,000 per QALY and parathyroid hormone (PTH) was between $100,000 and $200,000 per QALY.

Conclusions: Only bisphosphonates are cost-effective for fracture prevention in osteoporotic women aged 65 or older and this economic advantage is also maintained in subsets who have a lower relative risk of future fracture.

Key Words: osteoporosis, cost-effectiveness, sensitivity analysis, fracture reduction efficacy, compliance

**********

Osteoporotic fractures impose a substantial burden on patients and the health care system. The aging of the United States and world population will undoubtedly increase the numbers of these maladies. (1) Moreover, hip fractures are expected to double or triple over the next 50 years, and the incidence of vertebral and wrist fractures will likely follow in a similar manner. (1,2,3) In addition to mortality and impairment in quality of life imposed by these various fractures, (4-12) a considerable financial burden expressed as health costs will be experienced in the 21st century. The estimated annual expenditure for direct medical care of osteoporotic fractures was $17.5 billion in 2002. (13)

Hip fracture is the most expensive of all osteoporotic sequelae and incurs over 60% of direct fracture costs. (14) Fracture prevention at this site has centered around the use of trochanteric padding for impact absorption during a fall, various modalities of fall prevention, and the use of pharmaceutical agents. Although fall prevention programs remain unproven, (15) protective hip pads in community-based patients offer proven protection. (16) The available pharmaceutical agents used to combat osteoporosis display varying degrees of efficacy in fracture reduction at different skeletal sites. This antifracture effect, however, has primarily focused on postmenopausal women with a bone mineral density in the range of osteoporosis. (17-22)

The National Osteoporosis Foundation as well as the U. S. Preventive Services Task Force have recommended a bone mineral density (BMD) determination on all Caucasian women starting at the age of 65 years. (23,24) The high hip fracture rates in our region of Appalachia prompted us to test a model of preventive intervention by screening all women aged 65 or older in our county with a central dual x-ray densitometer (DXA). (25) We then initiated bone remedial therapy in those with hip osteoporosis (bone mineral density T-scores of -2.5 or below) to assess the cost-effectiveness.

Methods

Patient Population

Our base-case cohort consisted of 5,142 resident women aged 65 through 100 in Kanawha County, West Virginia.

Analytical Model

We divided these individuals according to the United States census listing into the following age intervals: 65 to 69, 70 to 74, 75 to 79, 80 to 84, and 85 to 100. (26) Our event pathway in Figure 1 moved women from a healthy state into a state of higher potential fracture in those with hip osteoporosis in 1-year cycles over a 5-year period (2000-2005) while adjusting for expected as well as excess mortality incurred by certain fractures. Our intention was to determine the most cost-effective strategy of each cohort by evaluating the following pathways (1): no treatment; (2) daily oral raloxifene at 60 mg; (3) daily nasal calcitonin at 200 U; (4) daily oral bisphosphonates (5 mg of risedronate or 10 mg of alendronate; (5) or daily injectable recombinant parathyroid hormone (1-34) (PTH) 20 [micro]g for 18 months. The most plausible compliance rates and efficacy of fracture reduction were applied to each strategy.

[FIGURE 1 OMITTED]

Outcomes

Since our analysis was formulated to compare different drug strategies, we used the incremental cost per quality-adjusted life-year (QALY) gained as the primary outcome. A value up to $100,000 per QALY gained was considered an acceptable cost-effective outcome. Of additional importance, the number of fractures prevented was viewed as a secondary outcome.

Sensitivity Analysis

We performed a two-way sensitivity analysis only on those drugs which have a proven fracture reduction at both vertebral and nonvertebral areas. In this instance, the compliance and efficacy of fracture reduction in the nonvertebral sites were varied using the least and most conservative values. Compared with a strategy of no intervention, we varied the efficacy of fracture reduction from 20 to 50% using 5% increments and applied this to compliance rates at 35%, 49% and 65%. These results were used to ascertain their effect on various thresholds of cost-effectiveness.

Costs

The estimated dollar expenses were represented as direct costs for materials and services incurred through the health care system rather than retail charges (personal financial responsibility). All outcome costs were expressed as dollars per capita over a 5-year period. Prevention costs consisted of DXA scans, annual physician visits, drugs and possible complications occurring during treatment. We calculated the intervention costs for each strategy by adding prevention costs to the cost of fractures minus the dollars saved by fracture prevention. A 3.5% annual discount rate was used; this value reflected the average medical inflation index for the previous 7 years through 2000. (27)

Data and Assumptions

The variables employed for fracture prevention, rates and costs are shown in Table 1.

Preventive Intervention

Our model included an initial DXA to screen all women in our county population aged 65 and older. We employed treatment in those with hip osteoporosis and performed a follow-up DXA scan during the third year on those still compliant with therapy. We estimated the age-group prevalence of osteoporosis in women aged 65 and older who had not received previous bone remedial therapy by reviewing 204 consecutive DXA examinations (Hologic 4,500 QDL) of the total hip over a 9-month period. This allowed the assumption in our model that approximately one-fourth of the women would receive benefit from treatment after the age of 65.

Costs of Items and Services

The costs of items for performing our event pathway were based on the Medicare reimbursement rates in 2000. We included baseline comprehensive metabolic panels only when initiating treatment in the bisphosphonate and PTH strategies. On the other hand, prothrombin times and venous Doppler studies were employed in 1.0% of the patients who might experience a venous thrombosis while receiving raloxifene. Furthermore, we added the cost of an annual visit to the physician for each strategy after adjusting for compliance.

Pharmaceutical Costs

The average annual costs to local pharmacies were used to calculate drug expenses. Since PTH did not become available until 2002, we estimated its cost back to 2000. Calcium supplements with vitamin D (total 1,500 mg/daily and 600 IU vitamin D) tablets were administered to each treatment group. With the exception of an 18 month administration of PTH, duration of treatment for each agent was 5 years. When other drugs were needed to remedy a complication incurred by treatment, the adverse event required a significant occurrence in a reported clinical trial over the control group.

Fracture Rates

All fracture incidence rates were projected from the year 2000 through 2005. Since hip, vertebral and certain nonhip, nonvertebral (NH, NV) fractures (pelvis, distal femur, proximal humerus, and tibia) are associated with an excess mortality, we made these annual adjustments in our projections. (5,12) Although we used our county hip fracture incidence rates from the previous 4 years, we estimated fracture rates for vertebral and nonhip nonvertebral sites from published data. Since two-thirds of vertebral fractures are asymptomatic, (33) we based our incidence rates for costs on the remaining one-third. We proportioned NH, NV fractures (ankle, non-ankle tibial-fibular, patella, distal femur, pelvic, wrist, non-wrist radial-ulnar, distal humeral shaft, and proximal humeral) throughout the various age groups.

Fracture Reduction by Treatment

We estimated the proportion of fractures at different sites expected with our prevalence of hip osteoporosis and entered these as potentially preventable for each treatment strategy. (42) Randomized, controlled trials (RCT) were used to ascertain the fracture reduction efficacy of each agent at various sites employing the dose approved by the Food and Drug Administration (FDA) for osteoporosis treatment.

Compliance Rates

We made annual adjustments for the decline in compliance for each treatment strategy. Since published compliance rates were not available for the 5-year span in the raloxifene and bisphosphonate strategies, we provided assumptions for the last 2 years based on the previous 3 years. Although PTH displayed a compliance rate of 79 to 83% in an RCT, (21) we reduced this to 50% based on patients who would be medically ineligible for treatment, eg, chronic renal failure, history of bone irradiation, metastatic cancer to bone, etc.

Fracture Costs

Since most hip fracture mortality occurs primarily during the first post fracture year, we made cost adjustments for reported mortality rates in various age groups using the following monthly intervals: the first 3 months; months 4 through 6; and then months 7 through 12. Although we adjusted 1993 costs for medical inflation to 2000, local hospital Medicare reimbursement ($10,439) and physician fees ($1,704) were used for the year 2000 index hospitalization period. We calculated symptomatic vertebral fracture costs entirely on regional data for the following services and items: the initial and follow-up physician reimbursements; x-rays; medical items (abdominal binders, walkers, etc.); analgesic agents; physical therapy; the proportion of patients requiring hospitalization (28%); and the average number of days in the hospital (6.0). For calculating the costs of various NH. NV fractures, we adjusted reimbursement for medical inflation to the year 2000.

Finally, an adjustment for costs of excess mortality was calculated on a quarterly basis during the first postfracture year for occurrences in the vertebral spine, distal femur proximal humerus, pelvis, and tibia.

Utility

We expressed the QALY gained using the EuroQol system of weighting. These were designated as QALY gained rather than QALY saved by fracture prevention since most patients who die with certain fractures usually do so from associated comorbidities within a short time after the event. (13) It is unlikely, therefore, that fracture prevention will alter mortality. The weight for wrist fractures was used to score other NH, NV fractures; however, we recognize that this can result in a disproportionate weight for other fractures in this category.

Results

Fracture Reduction and Cost-saving

Table 2 lists the fractures prevented with their associated savings for each strategy. In spite of the 25% higher vertebral fracture reduction noted with PTH, a greater fracture prevention occurred with bisphosphonates primarily due to their higher reduction in NH, NV fractures. This finding allowed for more QALY gained in Table 3 and, along with lower drug costs, a better incremental cost per QALY with bisphosphonates. The inability of raloxifene and calcitonin to reduce fractures at the hip and NH, NV areas limited their cost-effectiveness.

Sensitivity Analysis

The compliance rates, drug costs as well as the efficacy of hip and NH, NV fracture reduction were major determinants which played a role in outcomes. In addition, mass screening by DXA consumed between 10 to 21% of our prevention costs. Drug costs and DXA scans are more likely fixed cost determinants, whereas compliance and efficacy of fracture reduction are more susceptible to variance depending on age and fracture incidence in a given population. (43) A multitude of risk factors, some independent of a bone density, exist which are associated with future fracture risk. (29,44,45) Lower fracture rates have been observed among different racial groups at each level of BMD when compared with Caucasian women using the same normative data. (46,47) Therefore, in our sensitivity analysis, we varied compliance and efficacy of nonvertebral fracture reduction to test for various cost-effectiveness thresholds.

Changes in cost per QALY in Figure 2 were more sensitive to the efficacy of nonvertebral fracture reduction than compliance. At comparable levels of compliance and efficacy of nonvertebral fracture reduction, PTH prevented more fractures. This advantage, however, was offset by a higher cost per QALY with PTH. When these variables were applied to the bisphosphonates, 35% of the cost-effectiveness values were under the $50,000 per QALY threshold. A $50,000 per QALY threshold was only exceeded when the efficacies of nonvertebral fracture reduction feel below 42% at the lowest compliance and 35% at the highest compliance. Furthermore, even with the lowest compliance and efficacy of fracture reduction with the bisphosphonates, almost all values fell below the $100,000 per QALY threshold. In contrast, less than 2.0% of the cost per QALY values with PTH were under the $100,000 per QALY threshold, but approximately 55% were below the $150,000 per QALY threshold.

[FIGURE 2 OMITTED]

Although both increases in the efficacy of NH, NV fracture reduction and compliance resulted in lower costs per QALY, increases in efficacy of fracture reduction yielded comparatively more savings (Figs. 3 and 4). From a health payer perspective, osteoporotic populations with higher fracture rates and multiple risk factors would benefit most. On the other hand, focusing on compliance remains within the domain of the health care professional and enhanced public education. The 3.5% discount rate favored the cost per QALY of the bisphosphonates; this was a result of daily treatment with bisphosphonates over 5 years compared with only 18 months of the more expensive PTH.

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

Discussion

These results of cost-effectiveness support the recommendations from the United States Services Task Force and the National Osteoporosis Foundation for the screening of osteoporosis in all women aged 65 and older. (23,24) The bisphosphonate strategy was the only cost-effective strategy among the four we studied to prevent osteoporotic fractures. Pharmaceutical costs, compliance and savings resulting from efficacy of fracture reduction of the individual drugs were the major variables affecting outcomes. Moreover, those agents which also displayed considerable fracture prevention at nonvertebral sites created a profound effect on cost per QALY. The lower drug costs with the bisphosphonate strategy allowed for better cost-effectiveness. Even when we applied the lowest efficacy of nonvertebral fracture reduction through all reasonable ranges of compliance with this strategy, virtually all values were under our threshold goal of $100,000 per QALY. On the other hand, in the PTH strategy, over one-half of the values fell between the $100,000 and $150,000 per QALY thresholds and an additional 40% were between the $150,000 and $200,000 per QALY thresholds. Although the thresholds of $50,000 to $100,000 per QALY are generally considered acceptable, this remains controversial. Some have argued that this figure is currently too low and should be closer to $200,000 per QALY. (48)

No studies of cost-analysis have compared four different drugs, but others have indicated that fracture reduction with bisphosphonate therapy was cost-effective. (41,49-52) These reports, however, focused on fracture reduction in sites limited to the hip, spine plus one to two NH, NV areas or the hip alone. Although these evaluations did not include compliance rates, their cost of drugs. DXA scans, physician visits and effectiveness of fracture reduction with treatment approximated our estimates. One recent report indicated that PTH was cost-effective in the range of $25,000 to $80,000 per QALY gained. (53) As opposed to our model of preventive intervention, this study used a base case cohort of 69-year-old women with a previous vertebral fracture and a femoral neck T-score of -3.0 or less. Enhanced cost-effectiveness was noted especially if PTH treatment was started immediately after fracture occurrence.

This analysis has implications in the management of elderly women who have a potential for osteoporosis and future fracture. First, nonvertebral fractures not only display the highest loss of QALY related to osteoporosis, but also consume the majority of costs. (54) Health care dollars for fracture prevention should be directed toward prescribing drugs with proven substantial fracture reduction not only at the vertebral spine, but also at the hip and NH, NV sites. Second, since increases in compliance with daily oral bisphosphonates further improved cost-effectiveness, we speculate that the once weekly formulations of this drug currently available or future IV preparations given less frequently may provide a significantly higher compliance. (55) No long-term data, however, currently support this. Finally, cost-analysis studies of osteoporosis should be structured on fracture reduction by FDA-approved drug doses and reasonable compliance rates.

Several limitations exist which may affect our outcomes. We did not include the potential savings associated with raloxifene in breast cancer reduction, but no FDA recommendations have been forwarded for this indication. (56) Furthermore, since raloxifene and the bisphosphonates lack 5-year data for compliance, we needed assumptions to complete these strategies. (34-38) Moreover, our assumption of a 35% reduction of hip fracture with PTH injections in our analysis relies on small numbers from a RCT. (21) This efficacy is likely to be higher since PTH has an effect in restoring both trabecular and cortical bone (57); however, stronger confirmation of this from clinical trials is not available. Also, our assumption of continued efficacy throughout a 5-year period with each agent is based on clinical trials of different durations. The nasal calcitonin trial was conducted for 5 years, (20) whereas the raloxifene trial ended at 3 years. (22) We also assumed fracture reduction by PTH administration over an 18-month period would continue without further treatment for the following 42 months. Furthermore, we included only two bisphosphonate compounds (alendronate and risedronate) in our analysis; both have shown a continued or improved fracture reduction at vertebral sites after 5 years or more of continuous treatment. (58,59) Ibandronate, a recently FDA-approved bisphosphonate for osteoporosis treatment, as well as other bisphosphonates, not approved by the FDA for osteoporosis treatment, were not evaluated. Finally, our health state utility scores in our model were based on the first postfracture year and did not include the improved values which occur in subsequent years. (60) This would have yielded a slightly higher cost per QALY.

Several strengths in our model should be noted. In addition to the hip and spine, fracture costs were calculated on several NH, NV areas related to osteoporosis. (42) We applied annual compliance rates, adjusted for expected and excess mortality with certain fractures for each strategy. Only doses approved by the FDA were employed for expected fracture reduction. Finally, the expected osteoporotic fracture rates incorporated into our strategies were based on our county prevalence for osteoporosis. It could be argued that selection bias influenced this prevalence since ambulatory patients were either self referred or sent by a physician suspecting osteoporosis. On the other hand, women likely to have osteoporosis may have been unable to participate in DXA scanning because of frailty or other chronic illnesses. Our prevalence of osteoporosis determined by a DXA of the total hip in 268 women aged 50 and older,* however, was similar to another report in our geographic region which scanned the femoral neck (19.4% versus 20.6%). (23)

As the availability of DXA screening and awareness of therapies for osteoporosis increase, decision making for the allocation of resources in the health care system must focus not only on evidence-based medicine, but also on cost-effectiveness of the treatment. We can only structure effectiveness and compliance based on published clinical trials. Although trial participants may not necessarily represent the reality experienced by the health care practitioner, economic decisions need to be made from these data. Screening and treatment for osteoporosis confers investment benefits for direct health care costs. When future bone remedial agents are evaluated for marketing, a consideration of cost saving related to the efficacy of fracture reduction of various sites should be taken into account.

Conclusion

Using a threshold for cost-effectiveness of $100,000 per QALY gained or less, preventive intervention in women aged 65 or older by universal screening to identify osteoporosis and treatment with bisphosphonates is cost-effective. Calcitonin and raloxifene do not display cost-effectiveness, primarily due to their failure to reduce fractures at the nonvertebral sites. Although PTH does reduce fractures in nonvertebral areas, high drug costs limit its ability to be cost-effective; this agent may demonstrate improved cost-effectiveness if fracture reduction persists with long-term follow-up. Although we modeled cost-effective ratios over a 5-year period, longer estimates (10 or more years) need to be considered particularly in younger age groups who display lower fracture rates. In addition, the use of prescreening evaluations in women to reduce the number of DXA examinations will lower prevention costs, but this strategy results theoretically in less fracture prevention and QALY gained. (61) Instead of relying on the bone mineral density as a surrogate outcome, clinical trials with bone remedial agents are needed to verify whether a prescreening prevention strategy results in significant fracture reduction and satisfactory cost-effective outcomes. Also, since fractures in areas other than the spine contribute appreciably to morbidity cost, future studies on cost-analysis should also include nonvertebral sites. Finally, more data are needed with PTH treatment on fracture reduction in the hip and NH, NV areas.

Acknowledgments

The authors wish to thank Dan Lucas PharmD, Greg Clarke MD, and Patricia Dawson MSLS for their suggestions in preparation and review of this manuscript.

References

1. Melton LJIII. Epidemiology of fractures. In: Riggs BL, Melton LJ eds. Osteoporosis: Etiology, Diagnosis and Management New York, Philadelphia, Lipincott-Raven Publishers, 2nd ed, 1995, pp 225-247.

2. Zuckerman JD. Hip Fracture. N Eng J Med 1995;25:1519-1525.

3. Cooper C, Campion G, Melton LJ III. Hip Fractures in the Elderly: A Worldwide projection. Osteoporos Int 1992;2:285-289.

4. Wolinsky FD, Fitzgerald JF, Stump TE. The effect of hip fracture on mortality, hospitalization, and function status: A prospective study. Am J Public Health 1997;87:398-403.

5. Center JR, Nguyen TV, Schneider D, et al. Mortality after all major types of osteoporotic fracture in men and women: An observational study. Lancet 1999;353:878-882.

6. Crischilles EA, Butler CD, Davis CS, et al. A model of lifetime osteoporosis impact. Arch Int Med 1991;151:2026-2032.

7. Chrischilles E, Shireman T, Wallace R. Costs and health effects of osteoporotic fractures. Bone 1994;14:377-386.

8. Cauley JA, Thompson DE, Ensrud KC, et al. Risk of mortality following clinical fractures. Osteoporos Int 2000;11:556-561.

9. Zimmerman SI, Fox SM, Magazine J. Psychosocial aspects of osteoporosis. Phys Med Rehab Clin N Am 1995;6:441-453.

10. Ross PD. Clinical consequences of vertebral fractures. Amer J Med 1993;30S-43S.

11. Nevitt MC, Ettinger B, Black DM, et al. The association of radiographically detected vertebral fractures with back pain and function: A prospective study. Ann Int Med 1998;128:793-800.

12. Iacovino J. Mortality outcomes after osteoporotic fractures in men and women. Am Acad Insur Med 2001;33:316-320.

13. Melton LJ III. Adverse outcomes of osteoporotic fractures in the general population. J Bone Miner Res 2003;18:1139-1141.

14. Ray NF, Chan JK, Thamer M, et al. Medical expenditures for the treatment of osteoporotic fractures in the United States in 1995: Report from the National Osteoporosis Foundation. J Bone Miner Res 1995;12:24-35.

15. Tinetti ME. Preventing falls in elderly persons. N Eng J Med 2003;348:42-49.

16. Kannus P, Parkkari J, Niemi S, et al. Prevention of hip fracture in elderly people with use of a hip protector. N Eng J Med 2000;343:1506-1513.

17. McClung MR, Geusens P, Miller PD, et al. Effect of risedronate on the risk of hip fracture in elderly women. N Eng J Med 2001;334:333-340.

18. Black DM, Cummings SR, Karpf DB, et al. Randomized trial of effect of alendronate on risk of fracture in women with existing vertebral fractures. Lancet 1996;348:535-541.

19. Guyatt GH, Cranney A, Griffith L, et al. Summary of meta-analysis of therapies for postmenopausal osteoporosis and the relationship of bone mineral density and fractures. Endocrinol Metab Clin North Am 2002;31:659-679.

20. Chesnut CH, Silverman S, Andriano K, et al. A randomized trial of nasal spray salmon calcitonin in postmenopausal women with established osteoporosis: the prevent recurrence of osteoporotic fractures study. Amer J Med 2000;109:267-276.

21. Neer RM, Arnaud CD, Zanchetta JR, et al. Effect of parathyroid hormone (1-34) on fractures and bone mineral density in postmenopausal women with osteoporosis. N Eng J Med 2001;344:1434-1441.

22. Ettinger B, Black DM, Mitlack BH, et al. Reduction of vertebral fracture risk in postmenopausal women treated with raloxifene. JAMA 1999;282:637-645.

23. America's Bone Health: The State of Osteoporosis and Low Bone Mass in our Nation. National Osteoporosis Foundation, Washington, DC; 2002.

24. Nelson HD, Helfand M, Woolf SH, et al. Screening for postmenopausal osteoporosis: A review of the evidence for the U.S. Preventive Services Task Force. Ann Int Med 2002;137:529-541.

25. Pfister AK, McJunkin J, Santrock DA. Hip fracture outcomes and their prevention in Kanawha County, West Virginia. WV Med J 1999;95:170-174.

26. US Centennial Life Tables, 1989-1991. West Virginia Department of Health and Human Resources, Bureau for Public Health, Office of Epidemiology and Health Promotion.

27. Bureau of Economic Analysis. Available at: http://www.newyorkfed.org/research/directors_charts/i-reg_40.pdf. Accessed September 29, 2004.

28. Incidence and Costs to Medicare of Fractures Among Medicare Beneficiaries Aged 65 [greater than or equal to] Years--United States, July 1991-June 1992. MMWR Morb Mort Wkly Rep 1996;45:878-883.

29. Melton LJ, Crowson CS, O'Fallon WM. Fracture incidence in Olmstead County, Minnesota: comparison of urban with rural rates and changes in urban rates over time. Osteoporos Int 1999;9:29-37.

30. Van Der Klift M, De Laet CEDH, McCloskey EV, et al. The incidence of vertebral fractures in men and women. The Rotterdam Study. J Bone Miner Res 2002;17:1051-1056.

31. Black DM, Arden NK, Palermo L, et al. Prevalent vertebral deformities predict hip fractures and new vertebral deformities but not wrist fractures. Study of Osteoporotic Fractures Research Group. J Bone Miner Res 1999;14:821-828.

32. Ross PD, Davis JW, Epstein RS, et al. Pre-existing fractures and bone mass predict vertebral fracture incidence in women. Ann Int Med 1991;114:919-923.

33. Cooper C, Atkinson EJ, O'Fallon WM, et al. Incidence of clinically diagnosed vertebral fractures: A population-based study in Rochester, Minnesota. 1985-1989. J Bone Miner Res 1992;7:221-227.

34. Tosteson ANA, Grove MR, Hammond CS, et al. Early discontinuation of treatment for osteoporosis. Amer J Med 2003;115:209-216.

35. Kayser J, Ettinger B, Pressman A. Postmenopausal hormone support:discontinuation of raloxifene versus estrogen. Menopause 2001;8:328-332.

36. Cole RP, Palushock S, Haboubi A. Osteoporosis management: Physicians recommendations and women's compliance following osteoporosis testing. Womens Health 1999;29:101-115.

37. Cairo JJ, Ishak KJ, Huybrechts KF, et al. Compliance patterns with osteoporosis medications. Osteoporos Int 2003;13 (Suppl 3):S16-S16.

38. Papaionnou A, Ioannidis G, Adachi JD, et al. Adherence to bisphosphonate and hormone replacement therapy in a tertiary care setting of patients in the CANDOO database. Osteoporos Int 2003;14:808-813.

39. Brainsky D, Glick H, Lydick E, et al. The economic cost of hip fractures in community-dwelling residents: A prospective study. J Am Geriatr Soc 1997;45:281-287.

40. Hip fractures in people age 50 and over. Available at: http://www.wws.princeton.edu/cgt-bin/byteserv.prl/ota/ns20/alpha_f.html. Accessed September 27, 2004.

41. Johnell O, Jonsson B, Jonsson L, et al. Cost effectiveness of alendronate (Fosamax) for the treatment of osteoporosis and prevention of fractures. Pharmacoeconomics 2003;21:305-314.

42. Stone KL, Seeley DG, Lui L, et al. BMD at multiple sites and risk of fracture of multiple types: Long-term results from the Study of Osteoporotic Fractures. J Bone Miner Res 2003;18:1947-1954.

43. Kanis JA, Johnell O, De Laet C, et al. International variations in hip fracture probabilities: Implications for risk assessment. J Bone Miner Res 2002;17:1237-1244.

44. Black DM, Steinbuch M, Palermo L, et al. An assessment tool for predicting fracture risk in postmenopausal women. Osteoporos Int 2001;12:519-528.

45. Cummings SR, Nevitt MV, Browner WS, et al. Risk factors for hip fracture in white women. N Eng J Med 1995;332:767-773.

46. Barrett-Conner E, Siris ES, Wehren LE, et al. Osteoporosis and fracture risk in women of different ethnic groups. J Bone Miner Res 2005;20:185-194.

47. Cauley JA, Lui L-Y, Ensrud KE, et al. Bone mineral density and the risk of incident nonspinal fractures in black and white women. JAMA 2005;293:2102-2108.

48. Ubel PA, Hirth RA, Chernew ME, et al. What is the price of life and why doesn't it increase with inflation? Arch Int Med 2003;163:1637-1641.

49. Jonsson B, Christiansen C, Johnell O, et al. Cost effectiveness of fracture prevention in established osteoporosis. Scand J Rheumatol. 1996;103 (Suppl 103):30-38.

50. Kanis JA, Dawson A, Oden A, et al. Cost effectiveness of preventing hip fracture in the general female population. Osteoporos Int 2001;12:356-361.

51. Kanis JA, Borgstrom F, Johnell O, et al. Cost-effectiveness of risedronate for the treatment of osteoporosis and prevention of fractures in postmenopausal women. Osteoporos Int 2004;15:862-871.

52. Brecht JG, Kruse HP, Felsenberg D, et al. Pharmacoeconomic analysis of osteoporosis treatment with risedronate. Int J Clin Pharmacol Res 2003;23:93-105.

53. Lundkvist J, Johnell O, Cooper C, et al. Economic evaluation of parathyroid hormone (PTH) in the treatment of osteoporosis in postmenopausal women. Osteoporos Int. 2005; July 19 [Epub ahead of print].

54. Melton LJ, Gabriel SE, Crowson CS, et al. Cost-equivalence of different osteoporotic fractures. Osteoporos Int 2003;14:383-388.

55. Reid IR, Brown JP, Burckhardt P, et al. Intravenous zoledronic acid in postmenopausal women with low bone mineral density. N Eng J Med 2002;346:653-661.

56. Cummings SR, Duong T, Kenyon E, et al. Multiple outcomes of raloxifene evaluation (MORE) trial. Serum estradiol and the risk of breast cancer during treatment with raloxifene. JAMA 2002;287:216-22.

57. Jiang Y, Zhao JJ, Mitlak BH, et al. Recombinant human parathyroid hormone (1-34) [teripeptide] improves both cortical and cancellous bone structure. J Bone Miner Res 2003;18:1932-1941.

58. Sorenson OH, Crawford GM, Mudler H, et al. Long-term efficacy of risedronate: a 5-year placebo controlled clinical experience. Bone 2003;32:120-126.

59. Bone HG, Hosking D, Devogelaer J, et al. Ten years' experience with alendronate for osteoporosis in postmenopausal women. N Eng J Med 2004;350:1189-1199.

60. Kanis JA, Johnell O, Oden A, et al. The risk and burden of vertebral fractures in Sweden. Osteoporos Int 2003;15:20-26.

61. Richy F, Ethgen O, Bruyere O, et al. Primary prevention of osteoporosis: mass screening scenario or prescreening with questionnaire? An economic perspective. J Bone Miner Res 2004;19:1255-1260.</p> <pre> Giving back involves a certain amount of giving up. --Colin Powell </pre> <p>Alfred K. Pfister, MD, Christine A. Welch, MS, Melissa D. Lester, DO, Mary K. Emmett, PHD, Paul D. Saville, MD, and Shea A. Duerring

From the Department of Medicine, West Virginia University School of Medicine, Charleston, WV; CAMC Health Education and Research Institute; the Center for Health Services and Outcomes Research; and Hampden-Sydney College 

Reprint requests to Alfred K. Pfister, MD, West Virginia University School of Medicine, 4522 MacCorkle Avenue SE, Charleston, WV, 25304. Email: [e-mail]

No financial support was received.

Dr. Pfister has received no propriety support; he has performed as a speaker for Eli Lilly, Inc., Merck & Co., and Aventis. Ms. Chris Welch, Melissa Lester DO, Mary Emmett PhD, Paul Saville MD, and Shay Duerring have no conflicts of interest nor have they received any propriety support. All of the analyses were performed independently by investigators of the CAMC Health Education and Research Institute and West Virginia University. Furthermore, none of the authors received input, comment, or review from any pharmaceutical company or other agency during preparation of this manuscript.

Data collection for this study was approved by the West Virginia University-Charleston Area Medical Center Institutional Review Board.

Accepted October 14, 2005.

RELATED ARTICLE: Key Points

** The best incremental cost per QALY and fracture reduction occurred with the bisphosphonate strategy in our base-case analysis.

** At comparable levels of compliance and efficacy of nonvertebral fracture reduction, PTH prevented more fractures but was not cost-effective.

** In a sensitivity analysis, more cost saving occurred by improved effectiveness of nonvertebral fracture prevention than improved compliance.

** Cost analysis studies of osteoporosis should be structured on fracture reduction not only at the vertebral spine, but also at nonvertebral sites. Also, FDA-approved drug doses and reasonable compliance rates should be factored into these evaluations.

*Unpublished data.

Table 1. Baseline values in decision model

Age groups             65 to 69    70 to 74    75 to 79

Numbers                   4976        5057        4359
Number of DXA scans*        60          35          54
Percent with                15          20          20.4
  osteoporosis
Number with                746       1,011         888
  osteoporosis

Various items and
  services
Central DXA
Physician visit
Comprehensive
  metabolic profile
Venous Doppler
Prothrombin time

Pharmaceutical agents
Calcitonin
Raloxifene
Bisphosphonates
Recombinant
  parathyroid hormone
Warfarin
Enoxaparin
500 mg calcium-200 IU
  vitamin D tablets

                       Cumulative 5-year fracture rates
Age groups             65 to 69    70 to 74    75 to 79

Hip                         59.9       119.7       215
NH, NV                     447         549.1       528.5
Vertebral                   53.8        71.5       114.8

Base-case fracture                             Vertebral
  reduction efficacy
Calcitonin                                           0.34
Raloxifene                                           0.36
Bisphosphonates                                      0.50
Parathyroid hormone                                  0.65

Base-case annual
  compliance over
  5-year
  period [parallel]
Calcitonin                                           0.676 (0.80, 0.73,
                                                     0.68, 0.61, 0.56)
Raloxifene                                           0.62 (0.80, 0.71,
                                                     0.62, 0.53, 0.44)
Bisphosphonates                                      0.49 (0.80, 0.60,
                                                     0.40, 0.35, 0.30)
Parathyroid hormone

                          Mean annual fracture costs
Age groups             65 to 69    70 to 74    75 to 79

Hip                    $25,330     $26,821     $24,299
NH, NV                   $5285       $5410       $5867
Vertebral                $3016       $2952       $2899

Quality of life
  weights
Hip                          0.59        0.59        0.43
N-H NV                       0.75        0.75        0.60
Vertebral                    0.71        0.71        0.57

Age groups             80 to 84                85 & over

Numbers                   2965                    2772
Number of DXA scans*        31                      24
Percent with                25.8                    62.5
  osteoporosis
Number with                765                   1,732
  osteoporosis

Various items and                              Reimbursement
  services
Central DXA                                      $112
Physician visit                                   $32.80
Comprehensive                                     $80.50
  metabolic profile
Venous Doppler                                    $76.96
Prothrombin time                                  $42.00

Pharmaceutical agents                          Costs to pharmacy
Calcitonin                                       $744 per year
Raloxifene                                       $756 per year
Bisphosphonates                                  $588 per year
Recombinant                                    $6,520 per
  parathyroid hormone                            year ([dagger])
Warfarin                                          $63.68 for 100 (5 mg
                                                    tablets)
Enoxaparin                                        $51.33 per 80 mg
                                                    syringe
500 mg calcium-200 IU                             $60.00 per year
  vitamin D tablets

Age groups             80 to 84                85 & over

Hip                        257.2                   345.5
NH, NV                     385.2                   406.4
Vertebral                  130.9                   125.2

Base-case fracture     Hip                     NH NV ([double dagger])
  reduction efficacy
Calcitonin                   0                       0
Raloxifene                   0                       0
Bisphosphonates              0.50                    0.51
Parathyroid hormone          0.35 [section]          0.35

Base-case annual
  compliance over
  5-year
  period ([parallel])
Calcitonin                   0.676 (0.80, 0.73,
                             0.68, 0.61, 0.56)
Raloxifene                   0.62 (0.80, 0.71,
                             0.62, 0.53, 0.44)
Bisphosphonates              0.49 (0.80, 0.60,
                             0.40, 0.35, 0.30)
Parathyroid hormone          0.50

Age groups             80 to 84                85 & over

Hip                    $24,306                 $19,104
NH, NV                   $6043                   $6803
Vertebral                $2759                   $2652

Quality of life
  weights
Hip                          0.43                    0.43
N-H NV                       0.60                    0.60
Vertebral                    0.57                    0.57

Age groups             Reference

Numbers                26
Number of DXA scans*   Unpublished data
Percent with
  osteoporosis
Number with
  osteoporosis

Various items and
  services
Central DXA
Physician visit
Comprehensive
  metabolic profile
Venous Doppler
Prothrombin time

Pharmaceutical agents
Calcitonin
Raloxifene
Bisphosphonates
Recombinant
  parathyroid hormone
Warfarin
Enoxaparin
500 mg calcium-200 IU
  vitamin D tablets

Age groups             Reference

Hip                     5, 12, 25
NH, NV                  5, 12, 28, 29
Vertebral              30-33

Base-case fracture     Reference
  reduction efficacy
Calcitonin             20
Raloxifene             22
Bisphosphonates        17, 18, 19
Parathyroid hormone    21

Base-case annual       Reference
  compliance over
  5-year
  period ([parallel])
Calcitonin             20, 34
Raloxifene             34, 35
Bisphosphonates        34, 36, 37, 38
Parathyroid hormone    Estimated

Age groups             Reference

Hip                    39, 40 and local costs
NH, NV                  7, 28
Vertebral              Local costs

Quality of life        Reference
  weights
Hip                    41
N-H NV                 41 and assumed
Vertebral              41

All reimbursement costs are listed as 2000 undiscounted dollars.
Fracture rates are adjusted for expected as well as excess mortality.
* Bone densitometry performed by central dual x-ray absorptiometry.
([dagger]) Average 2002 undiscounted purchasing price for pharmacies.
([double dagger]) Nonhip nonvertebral fractures.
([section]) Fracture reduction assumed to be that of other peripheral
fractures.
([parallel]) Estimated compliance rates for years 1 through 5 are listed
in parentheses.
NH, NV, nonhip, nonvertebral.

Table 2. Overall fracture reduction and per capita cost-savings by
various therapeutic strategies in women with osteoporosis over a 5-year
period

                                  Hip  NH NV*  Vertebral ([dagger])

Fractures expected                417  622     206
Fractures prevented by treatment
Raloxifene                          0    0      47
Calcitonin                          0    0      55
PTH ([parallel])                   73  105      67
Bisphosphonates                    76  155      50

                                         Savings by fracture
                                  Total  reduction ([double dagger])

Fractures expected                1245   --
Fractures prevented by treatment
Raloxifene                          47   $23
Calcitonin                          55   $27
PTH ([parallel])                   245  $650
Bisphosphonates                    281  $718

* Nonhip nonvertebral.
([dagger]) Symptomatic vertebral fractures.
([double dagger]) Discounted at 3.5%.
([parallel]) Recombinant parathyroid hormone (1-34).

Table 3. Incremental cost-effectiveness ratios using various bone
remedial agents in the prevention of osteoporotic fractures over a
5-year period

                  Costs*
                 ($ per QALY)  QALY ([dagger])  Incremental costs

No intervention   $3041        0.0282            -
                 ($3368)
Raloxifene        $5676        0.0339            $2635
                 ($6255)
Calcitonin        $5761        0.0348              $85
                 ($6356)
PTH               $6833        0.0574            $1072
                 ($7315)
Bisphosphonates   $4200        0.0637           ($2633)
                 ($4623)

                 Incremental QALY  Cost-effectiveness

No intervention  -                 -
Raloxifene       0.0057            Dominated ([double dagger])
Calcitonin       0.0009            Dominated
PTH              0.0226            Dominated
Bisphosphonates  0.0063            Savings

* Cost is listed at a 3.5% discount, whereas numbers in brackets
indicate undiscounted dollars.
([dagger]) Quality-adjusted life-years gained.
([double dagger]) Dominated refers to a less expensive and more
effective strategy available.