SEARCH

SEARCH BY CITATION

Keywords:

  • bone mineral density;
  • diagnosis;
  • female;
  • hip fractures;
  • male;
  • osteoporosis;
  • prospective studies

Abstract

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

In postmenopausal women, the T score for bone mineral density (BMD) is a well-accepted diagnostic criterion for osteoporosis. It is also used to assess fracture risk. However, it is unclear whether in elderly men similar BMD thresholds should be used. Different hypotheses have been proposed for the relation of BMD with hip fracture risk in men. In this study, we tested those hypotheses using a mathematical model and we compared the calculated results with observed prospective data from the Rotterdam study. In the model, we combined the observed femoral neck BMD distribution for men and women with previously derived hip fracture risk functions based on age and BMD. For men, we tested different hypotheses for the relation of BMD with hip fracture risk. The relation of BMD with hip fracture risk is similar in men and women (scenario 1) or the relative risk (RR) per standard deviation (SD) decrease of BMD is either larger or smaller in men than in women (scenario 2a and 2b), or, at a similar absolute fracture risk, men have a higher BMD (scenario 3). In the prospective data, men with a hip fracture had an average BMD that was 0.070 g/cm2 higher than women with a hip fracture. The calculated results from the first scenario were consistent with those data and were also consistent with the observed hip fracture incidence and the observed female-to-male (F/M) risk ratio (1.7). When the RR for each SD decrease of BMD was assumed to be either larger or smaller in men than in women (second scenario), the calculated average BMD difference in men and women became respectively smaller or larger than observed. When men would have a higher fracture risk at similar BMD levels (third scenario), the calculated total number of hip fractures increased and even exceeded that in women, with an F/M risk ratio of 0.94 in our example. In women, a larger proportion of hip fractures occurs at a T score below −2.5 than in men using the same absolute BMD threshold, but using a male-specific T score largely solves this diagnostic problem. Taken together, the average hip fracture risk in men is much lower than in women but appeared to be similar at the same BMD. Therefore, we propose the use of the same absolute BMD thresholds for decisions about interventions.


INTRODUCTION

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

OSTEOPOROTIC FRACTURES and mainly hip fractures cause major morbidity and mortality in the elderly.(1) Consequently, fractures generate substantial costs because of acute hospital treatment and subsequent rehabilitation.(2, 3) Most of those fractures occur in women because they have a higher incidence of fractures at any given age and because of their higher life expectancy.(4) Therefore, most attention in hip fracture prevention was focused on women and only few epidemiological studies have investigated osteoporosis and fractures in men.(5–7) Recently, however, osteoporotic fractures in men have attracted more attention also in intervention trials.(8–10) Although the societal burden is smaller, osteoporotic fractures also have an important impact on men. Therefore, it is necessary to develop diagnostic and intervention standards for men.

Osteoporosis is defined as a condition characterized by low bone mass and microarchitectural deterioration of bone tissue, with a consequent increase in bone fragility and susceptibility to fracture.(11) For diagnostic purposes, osteoporosis in women was defined as a bone mineral density (BMD) of 2.5 SD below the average for young adult women, the so-called T score.(12)

There is uncertainty about how this definition should be used in men. Should the same T score be used in men based on female reference values or should a gender-specific T score be calculated based on the BMD in young adult men and if so what are the appropriate thresholds? Recently, this debate was summarized in a review article published in JBMR.(13) Some studies have suggested that men fracture at the same absolute BMD level as women do,(4,5,13,14) and, therefore, that the same absolute BMD threshold should be used. Other studies suggest a relation between BMD and hip fracture risk that is different in men and women.(13,15,16) Orwoll pointed out that in most studies on fracture patients, men have on average a higher BMD than women.(13) At first sight this appears to be at odds with a relation between BMD and fracture risk that is similar in men and women.

We studied whether the relation of BMD with hip fracture risk is similar in men and women, whether the gradient of risk per SD lower BMD is either higher or lower, or whether men have a similar hip fracture risk at a higher BMD than women. We have approached this question by mathematical modeling, combining the different hip fracture risk assumptions with the known BMD distribution in men and women to calculate the hip fracture distribution by BMD. Subsequently, we compared the calculated results with observed prospective data from the Rotterdam study. We also used the results to discuss diagnostic and intervention BMD thresholds in men.

MATERIALS AND METHODS

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

Input data

In a mathematical model we combined the relation of BMD and hip fracture risk with the BMD distribution to calculate the hip fracture distribution by BMD in the population. Previously, we estimated the hip fracture risk by gender, age, and BMD based on Dutch hip fracture registration data and the distribution of BMD.(4) The resulting risk functions (1-year cumulative hip fracture incidence) were validated over almost 4 years of follow-up in the prospective part of the Rotterdam study.(5) This is a population-based prospective cohort study of the occurrence and determinants of disease and disability in 7983 elderly men and women in a suburb of Rotterdam, The Netherlands. Aims and design of this study were described previously.(17)

We also used the BMD distributions for men and women observed cross-sectionally in the Rotterdam study in a sample of 5814 independently living men and women aged ≥55 years.(4) BMD was measured at the femoral neck using a Lunar DPX-L densitometer (Lunar Corp., Madison, WI, USA).

The model was developed using an Excel spreadsheet (Microsoft Corp., Redmond, WA, USA). Calculations were made for several ages between 65 and 80 years and results are described for age 70 years. We have chosen this age because at 70 years the hip fracture risk begins to increase rapidly.

Modeling assumptions

The baseline hip fracture risk functions by age, gender, and BMD were described previously and are reproduced in the Appendix.(4, 5) These risk functions show an exponential increase in hip fracture risk with lowering BMD and additionally an increase in risk with aging. The curves for men and women (at the same age) almost overlap at the ages of 60–80 years.(4) For women, we used the baseline risk function. For men, we used different scenarios reflecting the different hypotheses about the relation of BMD with hip fracture risk in men.

Scenario 1

In scenario 1, we used the baseline risk function for men. In this scenario the relative risk (RR) for hip fracture per SD decrease in BMD is 2.6 for both men and women and the absolute risk level by BMD is very similar.(4)

Scenario 2

In scenario 2, the risk function was modified to reflect either a higher or lower RR per SD decrease in BMD in men. Arbitrarily, we choose an RR of either 3.6 or 1.6 for men in this second scenario and they will be referred to as scenario 2a and 2b, respectively.

Scenario 3

In scenario 3, the RR was assumed to be the same in men as in women but now the relation of BMD with fracture risk was shifted so that men had a similar fracture risk at a BMD level 0.070 g/cm2 higher than in women. This shift corresponds to the average BMD difference between men and women at the same age.(4)

The relation of femoral neck BMD with the 1-year hip fracture risk at the age of 70 years for women and for the different scenarios in men is shown in Fig. 1.

thumbnail image

Figure FIG. 1.. One-year hip fracture risk by femoral neck BMD at age 70 in women (female symbol), and for the three different scenarios in men (similar risk male symbol 1, higher or lower RR per SD male symbol 2a and male symbol 2b, or a higher risk at the same BMD male symbol 3).

Download figure to PowerPoint

Number of hip fractures

From the BMD distribution at a given age, we derived the proportion of the population at a specific BMD level. Next, we calculated the hip fracture risk that corresponds to this BMD level, as shown in Fig. 1. Finally, we multiplied the proportion of the population at a specific BMD level with the corresponding hip fracture risk to obtain the number of hip fractures at that specific BMD level. When this was done across the whole range of BMD values, we obtained the distribution of hip fractures. The sum of all these values corresponds to the 1-year cumulative incidence at that age.

In women, at the age of 70 years, for example, ∼3% have a BMD of exactly 0.80 g/cm2. This BMD corresponds to a 1-year hip fracture risk of 0.2%. The calculated number of fractures at the age of 70 years contributed by women with a BMD of 0.80 g/cm2, therefore, was 0.006% (3% ∗ 0.2%). When the same was done for all BMD values, the total 1-year incidence for women at the age of 70 years became 0.31% or 3.1/1000.

Comparison with observed data

We compared these distributions to the observed prospective data to determine which of the different scenarios are possible. Femoral neck BMD was measured at baseline between 1990 and 1993 and hip fracture follow-up was obtained as described previously.(5, 18) For this analysis, follow-up started at the time of BMD measurement and ended either at the time of hip fracture, death, or December 31, 1999, whichever occurred first.

We also used the hip fracture distributions to examine diagnostic and intervention BMD thresholds in men. To do this, we calculated at which threshold of BMD a similar proportion of hip fractures would be captured in men as would be in women at a T score of −2.5.

RESULTS

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

Observed BMD in hip fracture cases and controls in the Rotterdam study

Valid hip fracture follow-up in individuals with femoral neck BMD measured at baseline was available for 5794 participants, and 156 hip fractures occurred during an average follow-up of 7 years (range, 0.01–9.4 years). The average baseline BMD in men and women with and without hip fractures during follow-up is given in Table 1. Average BMD was 0.070 g/cm2 (95% CI, 0.025–0.115) higher in male fracture cases than in female fracture cases, and an almost similar difference of 0.065 g/cm2 (95% CI, 0.058–0.072) was observed in controls.

Table Table 1.. Observed Average Baseline Femoral Neck BMD (g/cm2) in Men and Women With and Without Hip Fracture During Follow-up
Thumbnail image of

Observed BMD distributions and hip fracture incidence rates

Figure 2 shows the femoral neck BMD distribution for women and men aged 70 years based on data from the Rotterdam study.(4) The average BMD at the age of 70 years was 0.802 g/cm2 in women and 0.869 g/cm2 in men. In Dutch national registration data the observed 1-year cumulative hip fracture incidence at the age of 70 years was 3.2/1000 in women and 1.9/1000 in men, and the female-to-male (F/M) fracture incidence ratio was 1.7.(4)

thumbnail image

Figure FIG. 2.. The distributions of femoral neck BMD in men (male symbol) and women (female symbol) at age 70.

Download figure to PowerPoint

Calculated hip fracture distributions by BMD

The calculated hip fracture distributions at the age of 70 years for women and for the different scenarios in men are shown in Fig. 3 and detailed results for all ages are shown in Table 2. For women, the calculated average femoral neck BMD in female fracture cases at the age of 70 years was 0.679 g/cm2 and the calculated 1-year hip fracture incidence was 3.1/1000.

Table Table 2.. Women Compared With Men at Different Ages
Thumbnail image of
thumbnail image

Figure FIG. 3.. Hip fracture distribution by femoral neck BMD at age 70 in women (female symbol), and for the three different scenarios in men (similar risk male symbol 1, higher or lower RR per SD male symbol 2a and male symbol 2b, and a higher risk at the same BMD male symbol 3).

Download figure to PowerPoint

Scenario 1

Using the baseline scenario in men (similar fracture risk in men as in women is shown in Fig. 1), the calculated average BMD in 70-year-old male fracture cases was 0.743 g/cm2. This was 0.064 g/cm2 higher than in women. The calculated 1-year incidence in men was 1.9/1000 and the F/M risk ratio was 1.7.

Scenario 2

In scenario 2a, assuming a higher RR in men than in women (RR, 3.6 versus 2.6), the distribution of hip fractures by BMD in men shifts to the left, and the calculated average BMD at the age of 70 years was only 0.021 g/cm2 higher than in women. In scenario 2b, assuming a lower RR in men (RR, 1.6 versus 2.6), the average BMD difference at the age of 70 years was 0.126 g/cm2. With both assumptions, the F/M ratio was 1.7.

Scenario 3

In the third scenario, we assumed that the relation of BMD with fracture risk was shifted to the right in men by 0.070 g/cm2 and that the RR was similar in men and women. Here, the calculated average BMD in fracture cases was higher in men that in women and the same as in scenario 1. But, in scenario 3 the calculated number of hip fractures at the age of 70 years increased to 3.3/1000, exceeding that in women, and the F/M incidence ratio was only 0.94.

Hip fractures and BMD thresholds: Men compared with women

Figure 3 shows that for all scenarios, the proportion of fractures occurring below any specific BMD value was higher in women than in men. In Fig. 4 this is illustrated for a female T score of −2.5, comparing the calculated hip fracture distributions at the age of 70 years for women and for the first scenario in men. The bone densities measured in our female population corresponded to the Lunar U.S. Female Reference Data Base used by the machine,(5) and a female-specific T score of −2.5 corresponded to a BMD of 0.675 g/cm2. In the model, one-half of the hip fractures in women occurred at or below this threshold at the age of 70 years. Using the same absolute BMD value in men, only 32% of the hip fractures were captured. If we wanted a threshold that also captures one-half of the hip fractures in men, the threshold needed to move upward and became 0.740 g/cm2. Based on female reference values, this corresponded to a T score of −2.0, but using male reference data, this corresponded to a T score of −2.7. With a male-specific T score equaling −2.5 (0.769 g/cm2), almost 60% of the hip fractures were captured.

thumbnail image

Figure FIG. 4.. BMD threshold that captures the same proportion of hip fractures in men (male symbol) as a T-score = −2.5 in women (female symbol).

Download figure to PowerPoint

Other ages

The foregoing results were described for men and women aged 70 years. Table 2 also gives these results at other ages. At those other ages, we obtained similar results but at other levels of absolute incidence and average BMD. However, the general conclusions are similar to those at the age of 70 years.

DISCUSSION

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

In participants from the Rotterdam study who suffered a hip fracture during follow-up, the average baseline BMD was 0.070 g/cm2 higher in men than it was in women. Assuming a similar relation between femoral neck BMD and hip fracture risk in men and women, the difference in the calculated average BMD was almost the same in the model. Moreover, the results were also consistent with the observed hip fracture incidence in The Netherlands and with the observed F/M risk ratio of 1.7.(4)

When we assumed that the RR per SD decrease in BMD was either higher or lower in men than in women, the calculated difference in average BMD was respectively smaller or higher than observed. Therefore, we consider this hypothesis less likely and, even if the RR is different in men and women, this difference can only be small.

Assuming that men have a similar hip fracture risk at a higher BMD, the calculated incidence of hip fractures seriously increased and the F/M hip fracture risk ratio even reversed. This does not correspond to observations in The Netherlands and in most other countries,(4,5,19) making this third scenario highly unlikely.

When we repeated the calculations at ages other than 70 years, we confirmed that the scenario in which the relation of BMD with hip fracture risk is very similar in men and women is the most consistent with the prospective observations in our population.

This conclusion is important for the definition of a BMD threshold in men. We illustrated this using a T score of −2.5, the agreed definition of osteoporosis in women.(12) In women aged 70 years, only one-half of the hip fractures occurred below that threshold. This is so because hip fractures also occur in women who do not have osteoporosis. At the age of 70 years, this group of women who do not have osteoporosis is relatively large. With aging, the proportion of women with a T score below −2.5 increases and as a consequence the calculated proportion of hip fractures occurring below that threshold also increased up to 64% at the age of 80 years (Table 2). In men, using the same absolute BMD threshold, the proportion of hip fractures below that female-specific value was much lower.

Capturing a similar proportion of fractures in a population is important if we have the intention to evenly reduce the burden of illness in both men and women. Our analyses show that whatever the proportion of fractures we want to capture, the absolute BMD cut-off value will always be higher in men than in women. The use of a gender-specific T score largely solves this diagnostic problem.(12)

However, when cost-effectiveness of interventions is the goal and assuming equal efficacy in men and women, absolute fracture risks are more important because the numbers needed to treat with an intervention are directly influenced by the fracture incidence in those in whom an intervention is undertaken. This is shown in intervention trials where including populations with a different fracture incidence leads to different numbers needed to treat even with the same intervention.(20, 21) Several recent studies on the cost-effectiveness of interventions have also suggested that the estimation of absolute fracture risk is superior to only measuring the BMD for risk assessment.(22, 23) Because men have, overall, a lower hip fracture incidence than women, fewer men than women reach the required fracture risk threshold to make an intervention cost-effective. Therefore, when cost-effectiveness is the main concern, a similar BMD threshold in men and women should be used to assess the need for intervention and fewer men will be treated than women.

This is a modeling study and we calculated the effects of different assumptions about the relation between femoral neck BMD and hip fracture risk in men and women. Although the model was based on observed data, some caution should be used while interpreting the results. The observed bone density in this study was measured cross-sectionally and, therefore, cohort effects cannot be excluded.

An important reason why men have, on average, a higher areal BMD is because of the fact that men have bigger bones. Because of this larger volume, the areal BMD is artificially larger even with a same volumetric BMD.(24) Evidence has also been provided that the estimated volumetric BMD for vertebral and femoral neck is not essentially different between men and women.(25) However, the aim of this study was not to focus on volumetric BMD and its consequences for bone strength, but to describe a practical model based on the technique most frequently used in clinical practice, the areal BMD measured by DXA. Therefore, the use of areal BMD measurements does not affect the validity of our results in terms of prediction of fractures and/or the definition of thresholds.

Although the Rotterdam study is population-based, a health selection bias is likely because participants had to be able to come to the research center for BMD measurement. Therefore, the average BMD might be overestimated, especially at older ages. However, this would not affect the observed relation between BMD and hip fractures because fractures were observed prospectively.

We conclude that the hip fracture risk in men and women of the same age and at the same absolute BMD is very similar but because of the different BMD distribution, the average BMD in men who fracture their hip is higher than in women. To capture the same proportion of hip fractures in men, the threshold BMD needs to be higher than in women but the use of a gender-specific T score largely solves this for diagnostic purposes. However, for the purpose of interventions, we should use absolute risk rather than RR. The overall hip fracture incidence is lower in men than in women and we propose the use of the same absolute BMD thresholds in men and women for decisions about interventions.

Acknowledgements

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

This work was supported by the Dutch Health Research and Development Council (Zorg Onderzoek Nederland; grant 22000011).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  9. APPENDIX
  • 1
    Melton LJ 1993 Hip fractures: A worldwide problem today and tomorrow. Bone 14: S1S8.
  • 2
    Lindsay R 1995 The burden of osteoporosis: Cost. Am J Med 98: 9S11S.
  • 3
    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.
  • 4
    De Laet CE, van Hout BA, Burger H, Hofman A, Pols HA 1997 Bone density and risk of hip fracture in men and women: Cross sectional analysis. BMJ 315: 221225.
  • 5
    De Laet CEDH, Van Hout BA, Burger H, Hofman A, Weel AEAM, Pols HAP 1998 Hip fracture prediction in elderly men and women: Validation in the Rotterdam study. J Bone Miner Res 13: 15871593.
  • 6
    Nguyen TV, Eisman JA, Kelly PJ, Sambrook PN 1996 Risk factors for osteoporotic fractures in elderly men. Am J Epidemiol 144: 255263.
  • 7
    Mussolino ME, Looker AC, Madans JH, Langlois JA, Orwoll ES 1998 Risk factors for hip fracture in white men: The NHANES I epidemiologic follow-up study. J Bone Miner Res 13: 918924.
  • 8
    Cohen S, Levy RM, Keller M, Boling E, Emkey RD, Greenwald M, Zizic TM, Wallach S, Sewell KL, Lukert BP, Axelrod DW, Chines QQ 1999 Risedronate therapy prevents corticosteroid-induced bone loss. Arthirtis Rheum 42: 23092318.
  • 9
    Saag KG, Emkey R, Schnitzer TJ, Brown JP, Hawkins F, Goemaere S, Thamsborg G, Liberman UA, Delmas PD, Malice M-P, Czachur M, Diafotis AG 1998 Alendronate for the prevention and treatment of glucocorticoid-induced osteoporosis. N Engl J Med 339: 292299.
  • 10
    Adachi JD, Bensen WG, Brown J, Hanley D, Hodsman A, Josse R, Kendler DL, Lentle B, Olszynski W, Ste.-Marie L-G, Tenenhouse A, Chines AA 1997 Intermittent etidronate therapy to prevent corticosteroid-induced osteoporosis. N Engl J Med 337: 382387.
  • 11
    Anonymous 1993 Consensus development conference: Diagnosis, prophylaxis, and treatment of osteoporosis. Am J Med 94: 646650.
  • 12
    Kanis JA 1994 Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: Synopsis of a WHO report. Osteoporos Int 4: 368381.
  • 13
    Orwoll E 2000 Perspective: Assessing bone density in men. J Bone Miner Res 15: 18671870.
  • 14
    Cheng S, Suominen H, Sakari-Rantala R, Laukkanen P, Avikainen V, Heikkinen E 1997 Calcaneal bone mineral density predicts fracture occurrence: A five-year follow-up study in elderly people. J Bone Miner Res 12: 10751082.
  • 15
    Melton LJ, Atkinson EJ, O'Connor MK, O'Fallon WM, Riggs BL 1998 Bone density and fracture risk in men. J Bone Miner Res 13: 19151923.
  • 16
    Melton LJ, Orwoll ES, Wasnich RD 2001 Does bone density predict fractures comparably in men and women? Osteoporos Int 12: 707709.
  • 17
    Hofman A, Grobbee DE, de Jong PT, van den Ouweland FA 1991 Determinants of disease and disability in the elderly: The Rotterdam Elderly Study. Eur J Epidemiol 7: 403422.
  • 18
    Burger H, van Daele PLA, Algra D, van den Ouweland FA, Grobbee DE, Hofman A, van Kuijk C, Schütte HE, Birkenhäger JC, Pols HAP 1994 The association between age and bone mineral density in men and women aged 55 years and over: The Rotterdam Study. Bone Miner 25: 113.
  • 19
    De Laet C, Pols HAP 2000 Fractures in the elderly: Epidemiology and demography. Best Pract Res Clin Endocrinol Metab 14: 171179.
  • 20
    Harris S, Watts N, Genant H, McKeever C, Hangartner T, Keller M, Chesnut C, Brown J, Eriksen E, Hoseyni M, Axelrod D, Miller P 1999 Effects of risedronate treatment on vertebral and nonvertebral fractures in women with postmenopausal osteoporosis: A randomized controlled trial. Vertebral Efficacy With Risedronate Therapy (VERT) Study Group. JAMA 282: 13441352.
  • 21
    Reginster J, Minne H, Sorensen O, Hooper M, Roux C, Brandi M, 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. Vertebral Efficacy with Risedronate Therapy (VERT) Study Group. Osteoporos Int 11: 8391.
  • 22
    Kanis J, Johnell O, Oden A, De Laet C, Jonsson B, Dawson A 2002 Ten-year risk of osteoporotic fracture and the effect of risk factors on screening strategies. Bone 30: 251258.
  • 23
    Kanis J, Oden A, Johnell O, Jonsson B, de Laet C, Dawson A 2001 The burden of osteoporotic fractures: A method for setting intervention thresholds. Osteoporos Int 12: 417427.
  • 24
    Seeman E 1999 The structural basis of bone fragility in men. Bone 25: 143147.
  • 25
    Faulkner RA, McCulloch RG, Fyke SL, De Coteau WE, McKay HA, Bailey DA, Houston CS, Wilkinson AA 1995 Comparison of areal and estimated volumetric bone mineral density values between older men and women. Osteoporos Int 5: 271275.

APPENDIX

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

For the derivation of the bask risk functions, we used the Dutch nationwide hip fracture registration to estimate the hip fracture incidence function by age and gender, and, additionally, we used the assumption of a RR a per lower femoral neck bone density SD (in the baseline scenario, a = 2.6 for both men and women).(4) The formulas were derived for femoral neck BMD measured with a Lunar DPX-L densitometer, although conversion algorithms are available.(5) The 1-year cumulative hip fracture risk is given by

  • equation image

and

  • equation image