Portions of this research were presented as posters: Boonen S, Klemes A, Zhou X, Pack S, Delmas PD. Patients treated with risedronate have a fracture risk similar to patients 20 years younger in age. Presented at the 29th Annual Meeting of the American Society for Bone and Mineral Research; September 16–19, 2007; Honolulu, Hawaii. Lindsay R, Zhou X, James R, Boonen S. Non-vertebral fracture risk of risedronate treated patients is similar to that of untreated patients 10 years younger. IBMS_09. Presented at the 2nd Joint Meeting of the International Bone & Mineral Society and the Australian & New Zealand Bone & Mineral Society, March 21–25, 2009, Syndey, Australia. Lindsay R, Zhou X, James R, Boonen S. Non-vertebral fracture risk of risedronate treated patients is similar to that of untreated patients 10 years younger. Presented at the National Osteoporosis Foundation's 8th International Symposium on Osteoporosis; April 1–5, 2009; Washington, DC.
Assessment of the Relationship Between Age and the Effect of Risedronate Treatment in Women with Postmenopausal Osteoporosis: A Pooled Analysis of Four Studies
Article first published online: 22 MAR 2010
© 2010, Copyright the Authors. Journal compilation © 2010, The American Geriatrics Society
Journal of the American Geriatrics Society
Volume 58, Issue 4, pages 658–663, April 2010
How to Cite
Boonen, S., Klemes, A. B., Zhou, X. and Lindsay, R. (2010), Assessment of the Relationship Between Age and the Effect of Risedronate Treatment in Women with Postmenopausal Osteoporosis: A Pooled Analysis of Four Studies. Journal of the American Geriatrics Society, 58: 658–663. doi: 10.1111/j.1532-5415.2010.02763.x
- Issue published online: 1 APR 2010
- Article first published online: 22 MAR 2010
- fracture risk;
OBJECTIVES: To quantify the effect of age on the incidence of osteoporosis-related fractures and of risedronate treatment on fracture risk in different age groups in women with postmenopausal osteoporosis.
DESIGN: Data from four randomized, double-blind, placebo-controlled, Phase III studies were pooled and analyzed.
PARTICIPANTS: The analysis population (N=3,229) consisted of postmenopausal women with osteoporosis as determined on the basis of prevalent vertebral fractures, low bone mineral density (BMD), or both.
INTERVENTION: Patients had received risedronate 5 mg daily or placebo for 1 to 3 years.
MEASUREMENTS: The endpoints of interest were the incidence of osteoporosis-related fractures, clinical fractures, nonvertebral fractures, and morphometric vertebral fractures. The effect of age on fracture risk and treatment benefit was examined using Cox regression models with age and treatment as explanatory variables. The 3-year fracture risk was estimated for patients in each treatment group at a given age.
RESULTS: Irrespective of treatment, fracture risks were greater in older patients (P<.001). On average, for every 1-year increase in age, a patient's risk for osteoporosis-related fracture increased 3.6% (95% confidence interval=2.3–5.0%). Irrespective of age, risedronate treatment reduced fracture risk 42%. Risedronate-treated patients had fracture risks similar to those of placebo-treated patients 10 to 20 years younger.
CONCLUSION: Patients treated with risedronate have a significantly lower fracture risk, similar to that of untreated patients 10 to 20 years younger.
Worldwide in 2000, there were an estimated 9.0 million osteoporosis-related fractures, including 1.6 million at the hip, 1.7 million at the forearm, and 1.4 million observed clinically in the spine.1 Decreasing fertility and longer life expectancies will result in substantial aging of the population around the globe.2 In the more-developed regions of the world, the percentage of the population aged 65 and older is projected to increase from 14.3% in 2000 to 25.9% in 2050.2 Age is well established as a strong independent predictor of osteoporosis and all types of fracture.3–5 It follows, therefore, that the incidence of hip fracture worldwide is projected to increase from 1.3 to 1.7 million per year in 1990 to between 4.5 million and 6.3 million per year in 2050.6
A recent study of the burden of osteoporosis in the United States indicated that nonvertebral fractures represent 73% of total osteoporosis-related fractures and 94% of their associated costs.7 Hip fracture rates and costs are a valuable measure of the prevalence and financial burden of osteoporosis because they almost always require a hospital stay, and the ensuing records provide ready epidemiological and financial data. Although the incidence of hip fracture appears to be stabilizing in northern Europe8–10 and North America,11 this does not appear to be the case in Asian countries.12–15 For example, the number of hip fractures in Tattori, Japan, were 3.3 times as great in 2004/06 as 1986/88 in women and 2.3 times as great in men.14 Dramatic increases in the number of older adults are expected to occur in Asia in the next half century, and these are expected to contribute to increases in the numbers of hip fracture.6 The annual incidence of hip fracture in Asia is predicted to increase to between 2 million and 3.3 million in 2050, rates more than 5 times as great as those in 1990.5,6
Risedronate is a pyridinyl bisphosphonate that has been shown to be effective in the prevention and treatment of osteoporosis at a daily dose of 5 mg,16–23 a once-a-week dose of 35 mg,24 a dose of 75 mg on 2 consecutive days a month,25 or a once-a-month dose of 150 mg.26 Treatment with risedronate for 36 months reduced the risk of vertebral fractures by 41% and nonvertebral fractures by 39% relative to placebo in women with at least one existing vertebral fracture17 and the risk of hip fractures by 40% relative to placebo in women aged 70 to 79 with osteoporosis.22 Retrospective analyses showed that risedronate significantly reduced the incidence of nonvertebral and clinical vertebral fractures within 6 months.27,28
Given the aging population and the predicted increases in the incidence of osteoporosis and related fractures, information on the relationships between aging and fracture incidence and the effect of treatment on fracture risk as the population ages would be of interest. The large database developed during the risedronate clinical trial program provides the opportunity to explore these relationships further in a large population of postmenopausal women with osteoporosis. The effect of age on fracture risk is frequently examined by holding other risk factors, such as bone mineral density (BMD), constant, although because other risk factors change as patients age, this approach may not be optimal. Using pooled information from four risedronate studies—BMD–North America, BMD–Multinational, VERT–North America, and VERT–Multinational16,17,19,20—analyses were conducted to quantify the effect of age on fracture risk and the benefit of risedronate treatment for different age groups without adjusting for BMD. Prediction of fracture risk in different age groups provides information that clinicians and patients need as they weigh treatment and lifestyle options.
For these analyses, data were combined from four of the six randomized, double-blind, placebo-controlled, parallel-group, phase III studies that constituted the initial risedronate clinical trial program.16,17,19,20,21,23 The four trials used in the analyses were the four used to support regulatory approval of risedronate as a treatment for postmenopausal osteoporosis. These studies were similar with respect to the dosages administered, the assessments performed, and the populations studied. All four studies enrolled postmenopausal women with osteoporosis as determined on the basis of prevalent vertebral fractures, low BMD, or both.16,17,19,20 Data from the other two studies in the risedronate clinical program were not included in the analyses because these studies differed with respect to the age, sex, and disease characteristics of the patients.21,23
The patients received risedronate 5 mg daily (Actonel®, Procter & Gamble Pharmaceuticals, Cincinnati, OH) or placebo for 1 to 3 years, depending on the protocol for each individual study. All patients received calcium supplementation (1,000 mg/d), and patients whose levels of 25-hydroxyvitamin D3 were less than 40 nmol/L at baseline also received up to 500 IU/day of vitamin D. The studies were performed in North America, Europe, and Australia. Patients in the BMD trials were followed for up to 18 months or 2 years, and patients in the VERT trials were followed for 3 years. The local ethics committees or institutional review boards approved the protocols for all four studies. All studies were conducted according to the Declaration of Helsinki, and all patients provided written informed consent. Detailed reports of the efficacy and safety findings from these four pivotal trials have been published previously.16,17,19,20 In all four studies, the overall incidence and types of adverse events noted in patients who received risedronate 5 mg were similar to those observed in patients who received placebo.
The analyses were conducted on an intention-to-treat basis and included data from all patients who had been randomized to treatment and received at least one dose of risedronate 5 mg or placebo. The endpoints of interest were the incidence of osteoporosis-related fractures, clinical fractures, nonvertebral fractures, and morphometric vertebral fractures. Osteoporosis-related fractures were defined as any new morphometric vertebral or radiographically confirmed clinical fracture which include fractures of the hip, pelvis, wrist, humerus, clavicle, or leg, and symptomatic vertebral fractures. Incident new morphometric vertebral fractures were identified on the basis of quantitative morphometry and semiquantitative assessments, as described in the original phase III clinical protocols and reported previously.17
Baseline demographic characteristics were summarized using descriptive statistics. The effect of age on fracture risk and treatment benefit was examined using Cox regression models with age and treatment as explanatory variables. The interaction between treatment and age was tested in the initial model and excluded from the final model because it was not statistically significant.
The age difference between the placebo and risedronate groups for patients with the same fracture risk was estimated on the basis of the final model. The 95% confidence interval (CI) was calculated using Fieller's theorem.29 The 3-year fracture risk was estimated from the final model for patients in each treatment group at a given age.
The adequacy of the Cox regression models was assessed using previously developed graphical and numerical methods.30 All statistical tests were two-sided, with a 5% level of significance.
A total of 3,229 patients were included in this analysis (1,618 placebo, 1,611 risedronate). The treatment groups were similar at baseline with respect to demographic and disease characteristics (Table 1). On average, the patients in each treatment group were 68 years old and had a mean lumbar spine T-score of −2.6. Seventy-two percent of the patients had at least one prevalent vertebral fracture at baseline.
|Characteristic||Placebo (N=1,618)||Risedronate 5 mg (N=1,611)|
|Mean ± SD||67.7 ± 7.6||67.9 ± 7.9|
|Body mass index, kg/m2|
|Mean ± SD||26.0 ± 4.6||25.9 ± 4.6|
|Patients in prevalent vertebral fracture category, n (%)|
|0||458 (28.3)||453 (28.1)|
|≥1||1,157 (71.5)||1,154 (71.6)|
|Femoral neck T-score|
|Mean ± SD||−2.2 ± 0.8||−2.2 ± 0.8|
|Lumbar spine T-score*|
|Mean ± SD||−2.6 ± 1.2||−2.6 ± 1.2|
Irrespective of treatment, fracture risks were significantly greater in older patients for all four fracture types (P<.001; Table 2). On average, for every 1-year increase in age, a patient's risk for any osteoporosis-related fracture increased by 3.6% (95% CI=2.3–5.0%). The findings were similar for clinical fractures, nonvertebral fractures, and morphometric vertebral fractures.
|Type of Osteoporosis- Related Fracture||Relative Risk (95% Confidence Interval) P-Value|
|Any‡||1.04 (1.02–1.05) <.001||0.58 (0.48–0.70) <.001|
|Clinical§||1.04 (1.03–1.06) <.001||0.54 (0.41–0.69) <.001|
|Nonvertebral∥||1.05 (1.03–1.07) <.001||0.59 (0.44–0.79) <.001|
|Morphometric vertebral||1.03 (1.02–1.05) <.001||0.54 (0.43–0.68) <.001|
Irrespective of age, risedronate treatment reduced fracture risk by 41% to 46% across the four fracture types (P<.001; Table 2).
Three-year fracture risks for each fracture type for each age group were markedly greater in the placebo group than in the risedronate group (Table 3, Figure 1A–D). Comparison of the ages of patients in the placebo and risedronate group who were at the same risk of fracture showed that patients in the placebo group were between 10 and 20 years younger than patients in the risedronate group depending on the fracture sites defined (Table 4, Figure 1A–D). The mean difference for any osteoporosis-related fracture was 15.1 years between patients in the control and risedronate groups, 14.4 years for clinical fractures, 10.3 years for nonvertebral fractures, and 19.8 years for morphometric vertebral fractures.
|Type of Osteoporosis-Related Fracture and Age||Fracture Risk, %|
|Morphometric vertebral fracture|
|Type of Osteoporosis- Related Fracture||Age Difference, Years (95% Confidence Interval)|
|Any fracture*||15.1 (9.0–26.0)|
|Clinical fracture†||14.4 (7.7–26.9)|
|Nonvertebral fracture‡||10.3 (4.4–20.2)|
|Morphometric vertebral fracture||19.8 (10.8–42.7)|
The results of these analyses confirm previous findings that fracture risk increases with age.3,4,31 Regression modeling indicated that the risk of fracture increases 4% for every 1-year increase in age in postmenopausal women with osteoporosis. The analyses also showed that patients treated with risedronate had a fracture risk similar to that in placebo-treated patients who were 10 to 20 years younger. The risk of osteoporosis-related fracture in 75-year-old risedronate-treated patients was nearly the same as that in 60-year-old untreated patients.
Turning back the clock on bone fragility is an important step toward preserving independence in older adults. Of noninstitutionalized persons aged 65 and older in the United States, more than 85% live alone or with their spouse, and half of women aged 75 and older live alone.32 Although the population is aging, people are remaining fit and active well into their 70s and even 80s. Advances in healthcare technology, better nutrition, and increases in education levels are postulated as contributors to the progressive declines in the prevalence of chronic disability in recent decades.33 According to a survey of 21,000 people in 21 countries started in 2004/05, a majority of people in their 60s and 70s reported being healthy and in control of their lives, and 86% of those aged 60 to 79 reported being in fair, good, or very good health,34 yet the greatest number of fractures in women worldwide occurs between the ages of 75 and 79,1 when many would be living independently.
Numerous studies have demonstrated substantial reductions in physical function and quality of life after fracture.35–41 Nonvertebral fractures are significant contributors to the morbidity, loss of function, and financial burden associated with fractures. Of the more than 2 million incident fractures estimated to have occurred in 2005 in the U.S. population aged 50 and older, nonvertebral fractures accounted for 73% and represented 94% of the $17 billion cost.7 These fractures have important implications with respect to a patient's quality of life and ability to live independently. In a study of the effect of nonvertebral fractures on health-related quality of life using the mini-Osteoporosis Quality of Life Questionnaire (mini-OQLQ) in 2009 postmenopausal women aged 50 and older seen at a tertiary care facility, multiple regression analyses showed that scores on the total mini-OQLQ and scores for the domains of physical functioning, emotional functioning, and activities of daily living were significantly lower in patients who had experienced nonvertebral fractures than in those who had not.40
Hip fracture is the most severe and economically important consequence of osteoporosis. It is associated with significant excess risk of mortality at least twice that of age-matched population norms, which endures for several years after the index fracture.42,43 Furthermore, hip fracture causes acute pain and loss of function and nearly always requires costly surgery, hospitalization, and lengthy rehabilitation.44,45 Using a Markov simulation based on data from the published literature, one study estimated that, 1 year after fracture, 39% of community dwellers who sustained hip fractures were at home without new services, 26% were at home with new services, 12% were in a nursing facility, and 22% were dead.46 A recent study of 3,533 patients hospitalized for hip fracture in the United Kingdom showed that 43% had to be discharged to a residence providing more care than their usual residence, and 11% died before being discharged.47 Other studies have yielded similar findings. A prospective, controlled study that assessed functional outcome after hip fracture in 159 women found that 30% of women younger than 80 were unable to walk independently 1 year after hip fracture, compared with only 7% of age-matched control women without hip fracture.48 In all patients with hip fracture, the functional decline due to hip fracture over the first 12 months was estimated to be 24%.48 This finding is consistent with that of a prospective study of 275 patients with hip fracture in the United Kingdom,49 which showed that in the 183 (67%) patients surviving 1 year after fracture, mobility and related functions were reduced by 20% to 25%.
Slowing age-related increases in fracture risk would be expected to lead to significant reductions in the economic burden of osteoporosis-related fractures. From 2005 to 2025, the cost of nonvertebral fractures in the United States is predicted to increase from 41% to 60%, depending on fracture site, from $16 billion to $24 billion.7 Growth in the incidence and cost of osteoporosis-related fractures is predicted to be most rapid, more than 87%, in people aged 65 to 74. Given that average life expectancy in the United States is 78 years,50 reducing fracture risk to levels associated with individuals 10 to 20 years younger would result in substantial reductions in healthcare costs, particularly those associated with long-term care.
The results of this study showed that the risk of osteoporosis-related fractures in postmenopausal women with osteoporosis who were treated with risedronate was the same as that in untreated patients 10 to 20 years younger. This slowing of the increase in age-related fracture risk would be expected to reduce the incidence of disabling fractures, preserving self sufficiency and quality of life and significantly reducing healthcare costs.
Conflict of Interest: Steven Boonen and Robert Lindsay are consultants and have received grants and honoraria from Procter and Gamble Pharmaceuticals and sanofi-aventis pharmaceuticals. Andrea Klemes and Xiaojie Zhou are employees of Procter and Gamble. Mary G. Royer is a paid consultant of Procter and Gamble. Dr. Boonen is senior clinical investigator of the Fund for Scientific Research, Flanders, Belgium (F.W.O.-Vlaanderen) and holder of the Leuven University Chair in Metabolic Bone Diseases.
Author Contributions: All authors made substantial contributions to the concept and design of the study, the interpretation of the data, and the drafting of the article, and all gave final approval of the version to be published. Xiaojie Zhou performed the analyses. Mary G. Royer worked from an expanded outline, bibliography, and data tables prepared by the authors to develop the manuscript.
Sponsor's Role: The Alliance for Better Bone Health (Procter and Gamble Pharmaceuticals, Inc., and sanofi-aventis US Inc.) provided funding for this study. The sponsor provided funding for the four pivotal studies used in this analysis and collected and stored the data. The sponsor analyzed the initial data from the individual trials. The sponsor, in conjunction with the investigators, designed the study. The authors had full access to all data from the study and participated in the decision to publish the data.
- 2United Nations. United Nations Deparment of Economic and Social Affairs/Population Division. World Population to 2300. New York: United Nations, 2004.
- 19Risedronate increases bone mineral density at the hip, spine and radius in postmenopausal women with low bone mass (abstract 349). Osteoporosis Int 1998;8 (Suppl 3):111., , et al.
- 29Fieller's theorem. In: Armitage P, Colton T, editors. Encyclopedia of Biostatistics. Chichester: John Wiley, 1998.
- 32Administration on Aging. A Profile of Older Americans: 2007. Washington, DC: U.S. Department of Health and Human Services, 2007.
- 34HSBC Insurance (2007). The future of retirement. The new old age [on-line]. Available at http://www.hsbc.com/1/PA_1_1_S5/content/assets/retirement/2007_for_report.pdf Accessed June 1, 2009.
- 41The impact of incident vertebral and non-vertebral fragility fractures on health-related quality of life in established postmenopausal osteoporosis: Results from the teriparatide randomized, placebo-controlled trial in postmenopausal women. J Rheumatol 2003;30:1579–1583., , et al.
- 50U.S. National Center for Health Statistics. U.S. National Center for Health Statistics, National Vital Statistics Reports (NVSR), Deaths: Final Data for 2005, Vol. 56, No. 10, April 24, 2008 [on-line]. Available at http://www.census.gov/compendia/statab/tables/09s0100.pdf Accessed January 9, 2009.