- Email: [email protected]
Fractures in the elderly: epidemiology and demography
BaillieÁre's Clinical Endocrinology and Metabolism Vol. 14, No. 2, pp. 171±179, 2000
doi:10.1053/beem.2000.0067, available online at http://www.idealibrary.com on
1 Fractures in the elderly: epidemiology and demography Chris E. D. H. De Laet
MD, PhD
Senior epidemiologist
Huibert A. P. Pols
MD, PhD
Professor of medicine Institute for Medical Technology Assessment, Erasmus University, Rotterdam, The Netherlands Department of Internal Medicine, Erasmus University Medical Center, Rotterdam, The Netherlands
Osteoporosis and osteoporosis-related fractures are a major source of both morbidity and cost in the elderly, the fractures that are most commonly associated with osteoporosis being those of the hip, the distal forearm and the vertebrae, although it is believed that most other fractures occurring in the elderly are also related to osteoporosis. In this review, the incidence of all types of fracture is described based on the available literature, and the foreseeable trends resulting from demographic changes are discussed. Emphasis is given to the epidemiology of hip fracture since this is the most serious consequence of osteoporosis. Hip fractures occur all over the world, most currently occurring in Western countries, mainly Europe and the USA, but it is expected that there will be a large increase in the number of hip fractures in other countries because of demographic changes. The incidence of hip fractures increases exponentially with age, resulting in a 1-year incidence of 1% in women aged 80 in Western countries. Most hip fractures occur in women, but this is again partly due to demography, because of the longer life expectancy of women. Wrist fractures occur more often in women and do not show the same increase with age as hip fractures. The incidence reaches a plateau at age 60±70. Vertebral fractures show a modest increase with age and are again more common in women than men. The incidence of all other fractures increases modestly with age Key words: Colles' fracture; forecasting; fractures; hip fractures; incidence; men; osteoporosis; spinal fractures; women.
Osteoporosis and osteoporosis-related fractures are a major source of both morbidity and cost in the elderly. The most important fractures related to osteoporosis are those of the hip, the distal forearm and the vertebrae, although it is believed that most other fractures occurring in the elderly are also related to osteoporosis.1 Osteoporosis is de®ned as a condition characterized by low bone mass and the micro-architectural deterioration of bone tissue, with a consequent increase in bone fragility and susceptibility to fracture.2 The conventional method of estimating bone mass is by bone densitometry, the World Health Organization de®ning osteoporosis as a bone mineral density (BMD) less than 2.5 standard deviation (SD) below the mean for young adults.3 1521±690X/00/02017109 $35.00/00
c 2000 Harcourt Publishers Ltd. *
172 C. E. D. H. De Laet and H. A. P. Pols
In this review, we will discuss the incidence of fractures and explore the foreseeable trends in the epidemiology of fractures caused by demographic change. Since the most serious consequence of osteoporosis is fracture of the hip, the emphasis will be on this type of fracture. HIP FRACTURES Current incidence Most hip fractures currently occur in Western countries. While the total number of hip fractures worldwide in 1990 was estimated at 1.7 million, 560 000 of these occurred in Europe (including the former Soviet Union) and 360 000 in North America.4 Information about hip fractures is available for most countries since virtually all require hospital admission and treatment, making them relatively easy to detect through hospital discharge registries. In most countries, the incidence of hip fracture shows an exponential increase with age and is in general higher in women than in men after the age of 50. This higher incidence in women is re¯ected in the absolute number of fractures since there are, in most countries, more elderly women than men. Recent ®gures on hip fracture incidence in the USA have been published by Melton et al.5 For Europe, data have been compiled in a 1998 European Commission report on the incidence of hip fractures in the 15 member states of the European Union (EU).6 In this report, data from several studies were assembled, including those from the MEDOS study, focusing on countries around the Mediterranean Sea, and also from several national studies.7±11 The incidence in Europe showed a north±south gradient, the highest incidence being in Scandinavia and a much lower incidence occurring in the countries around the Mediterranean. The incidence in Sweden was highest, the relative incidence compared with that of the USA being 1.3 in women and 1.7 in men (author's own calculations based on the European Commission report).6 In Finland, the rate was comparable to that of the USA for women but slightly higher in men. In the UK, the Netherlands and Germany, the incidence was very similar to that observed in the USA. In Southern European countries, the incidence was much lower, that in France, Greece and Spain being about 70% of that in the USA, while in Italy and Portugal the incidence was as low as 50% of that in the USA. The MEDOS study also contained data on Turkey, a non-EU country partly in Europe but mainly in Asia. These data are quite dierent from those of mainland Europe. Both the absolute and the age-standardized rates were very low, and the increase with age was less pronounced than in European countries.12 The incidence rate was about 10±20% of the equivalent rate in the USA for women, and about 20±30% of that for men (author's own calculations). Less information is available from other continents, but it appears that, in both South America and Asia, the incidence of hip fracture is lower than in North America and Europe. Schwartz et al13 compared data from hospital registers from several continents, attempting to adjust for dierences in coding practice. They concluded that the dierence reported between countries mainly re¯ected a genuine variation in the hip fracture incidence rate. The rate for Beijing was as low as that reported from Turkey in the MEDOS study. The rates for Brazil, Hong Kong and Hungary were somewhat lower than the USA and European averages, while the rate for Iceland equalled the high Scandinavian level. Figures 1 and 2 give a comparison of the one-year cumulative incidence rates per 100 000 for some selected countries based on several studies.5±7,9,13
Fractures in the elderly 173 5000 4500 Sweden Iceland Finland UK Netherlands USA France Italy Portugal Spain Switzerland Hungary Hong Kong Brazil China Turkiye (town)
Incidence/100 000
4000 3500 3000 2500 2000 1500 1000 500 0
55-59
60-64
65-69
70-74
75-79
80-84
85+
Age
Figure 1. One-year cumulative incidence of hip fracture in women per 100 000 in several countries. 5000 4500 Incidence/100 000
4000 3500
Sweden Iceland Finland UK Netherlands USA France Greece Italy Portugal Spain Switzerland Hungary Hong Kong Brazil China Turkiye (town)
3000 2500 2000 1500 1000 500 0
55-59
60-64
65-69
70-74
75-79
80-84
85+
Age
Figure 2. One-year cumulative incidence of hip fracture in men per 100 000 in several countries.
Gender ratio The global female-to-male ratio for the absolute number of hip fractures is around 3.5:1 in most countries, but this high female-to-male ratio is largely due to demography. When the incidence rates are compared, the female-to-male ratio is between 1.6:1 and 2:1 in most Western countries. Only in Sweden and Finland has a lower ratio (1.3:1 and 1.4:1 respectively) been reported. The Turkish data were again strikingly dierent, with a female-to-male ratio of about 1:1 in the towns of Istanbul and Ankara, and a reversal of the ratio (around 0.4) in rural areas, especially in older participants. A reversal of the female-to-male ratio was also observed in China (Beijing).
174 C. E. D. H. De Laet and H. A. P. Pols
The reason for both the huge geographical variation in hip fracture incidence and the reversal of the gender ratio in some countries is largely unknown. It is probably due to a combination of genetic, lifestyle and environmental factors and is clearly a topic for future research. Future trends Future trends are in¯uenced by both demography and changes in age-adjusted incidence. Demography is relatively easy to predict for the near future because all those potentially at risk within the next 50 years have already been born. The only uncertainty with respect to demography is the future mortality rate.14 Trends in the hip fracture incidence rate are not as predictable, especially since there are con¯icting ®ndings from dierent countries. An increase in age-adjusted incidence was described between 1930 and 1980 in several countries. That increase appeared to have levelled o in the 1980s in parts of the USA15,16 as well as in Sweden and the UK.17±19 In other European countries such as the Netherlands, Italy and Finland, the age-adjusted incidence continued to rise during the early 1990s.8,20±22 Because of this uncertainty, most predictions for the future are based on the current incidence rate, taking only demography into account. Taking those assumptions into account, it is believed that the global number of hip fractures worldwide will increase to over 6 million by the year 2050.4 Although the high proportion of hip fractures occurring in Europe will decline over the next few decades because of a demographic evolution in other parts of the world, the absolute number of hip fractures will also continue to rise in Europe to over 1 million in 20504 since, as in other areas of the world, the population in Europe is also ageing, mainly because of the post-Second World War baby boom. Moreover, with the continuing increase in life expectancy in both men and women, there will be a gradual increase in the proportion of very elderly people. Over the next decades, these trends are expected to continue. Estimates of lifetime risk for hip fracture range from 15% to 18% in Northern Europe and the USA, but, taking into account increasing life expectancy, these rates are very likely to be an underestimate.14 The report from the European Commission estimated that the yearly number of hip fractures within the 15 member states would increase to almost 1 million by the year 2050.6 The female-to-male ratio within the EU for the total number of fractures, which was 3.7:1 for the year 2000, was projected to decline to 3.2:1 in the year 2050, again due to demographic changes alone. OTHER FRACTURES The incidence of other fractures is more dicult to determine. Since those fractures are often treated in an outpatient setting, they are poorly registered and cannot easily be traced through hospital discharge registers. Therefore, their incidence can only be assessed in observational studies in which a limited number of care institutions deliver care to a population within a well-de®ned catchment area. For the estimation of the incidence of non-hip fractures, we have used two such studies. In a recently published study, all the fractures that occurred during a 3-year period within one district (Olmsted County) in the USA (3665 fractures) were recorded. 5 We will compare these data with those from a study performed by Singer et al describing over 15 000 fractures in Edinburgh, again within a well-de®ned catchment area.23 These two studies will be referred to as the US and the UK studies.
Fractures in the elderly 175
Fractures of the distal forearm Fractures of the distal forearm are usually caused by a fall on the outstretched arm, and they are also associated with osteoporosis. They typically occur at a younger age than hip fractures, mostly in women, and they do not show the same exponential increase with age as do hip fractures. There is no diagnostic problem, but since they are mostly treated in an outpatient setting, accurate documentation is more dicult. Figure 3 shows the incidence data for wrist fractures from the US and the UK studies. They are remarkably similar, although in the US study the incidence in women rises more rapidly to a plateau, whereas the UK study shows a more gradual increase with age in women. In the age categories over 55, the female-to-male ratio is much higher than it is in hip fractures: at age 55, this ratio is around 6:1, and it increases to 10:1 at ages over 75. Since the increase in incidence with age is much less pronounced than for hip fractures, it is not expected that the incidence of fracture of the distal forearm will change much as the result of demographic changes in the future. 1000 900
Incidence/100 000
800 700 600 USA (women) UK (women) USA (men) UK (men)
500 400 300 200 100 0
55−59
60−64
65−69
70−74
75−79
80−84
>85
Age
Figure 3. One-year cumulative incidence of wrist fracture in men and women per 100 000 person-years in the USA and the UK.
Vertebral fractures Determining the incidence of vertebral fractures is even more complex. Since it is estimated that only one third of radiographically visible deformities readily come to clinical attention, the measured incidence is heavily dependent on its de®nition.24±26 There is currently no universally accepted de®nition of vertebral fractures, although several de®nitions have been proposed.27±31 Moreover, in clinical practice, case ascertainment is very dierent between countries. In the US study, a vertebral fracture was de®ned on the basis of a radiologist's report of compression or collapse of one or more thoracic or lumbar vertebrae during a 3-year period, and it was as such essentially a prevalence study. It is also unclear whether or not these deformities were associated with clinical symptoms. In an earlier study from the same group32, a
176 C. E. D. H. De Laet and H. A. P. Pols
vertebral fracture was de®ned as one which came to clinical attention. In the UK study, the incidence of a vertebral fracture was de®ned as any vertebral fracture reported from a health-care institution, the incidence being extremely low. A comparison was also performed between these data and those from a 6.5-year followup in the Rotterdam study33, in which follow-up spinal radiographs were compared with base-line radiographs using the McCloskey±Kanis algorithm for the morphometric de®nition of vertebral deformities.29 The resulting incidence estimates from these four studies are shown in Figures 4 and 5. Although these ®gures show that, in most studies, the apparent incidence 3500
Incidence/100 000
3000
The Netherlands: morphometric USA radiographic USA clinical UK clinical
2500 2000 1500 1000 500 0 55−59
60−64
65−69
70−74
75−79
80−84
>85
Age
Figure 4. One-year cumulative incidence of vertebral fracture in women per 100 000 in the USA, the UK and the Netherlands. 3500
Incidence/100 000
3000
The Netherlands: morphometric USA radiographic USA clinical UK clinical
2500 2000 1500 1000 500 0 55−59
60−64
65−69
70−74
75−79
80−84
>85
Age
Figure 5. One-year cumulative incidence of vertebral fracture in men per 100 000 in the USA, the UK and the Netherlands.
Fractures in the elderly 177
increases with age, they also show a large dierence in the apparent incidence rate depending on the de®nition used. Therefore, the exact de®nition of a vertebral fracture is extremely important in the interpretation of any reported incidence data. In Europe, a concerted eort is currently ongoing in the form of the European Vertebral Osteoporosis Study (EVOS), in which a very rigorous quality assurance procedure is used to minimize the numbers of false-positive results from radiographic evaluation. Currently, however, only prevalence data are available from this study.34 The female-to-male ratio is around 2:1 at most ages, and it is unlikely that the occurrence of vertebral fractures will alter much in the future because of demographic changes. Fractures other than hip, distal forearm or vertebral Since there is no general consensus on whether other fractures are also related to osteoporosis, we used, as an arbitrary de®nition, all other fractures excluding those of the face, skull and ribs, as well as fractures of ®ngers and toes. Again, data from the USA are compared with those from the UK. Figure 6 shows the estimated incidence of these other fractures. In men, the incidence doubles with age between 55 and 80, the increase being slightly steeper in women. The female-to-male ratio is around 2:1 at 55 and increases slightly with age. 3500
Incidence/100 000
3000
USA (women) UK (women) USA (men) UK (men)
2500 2000 1500 1000 500 0 55−59
60−64
65−69
70−74
75−79
80−84
>85
Age Figure 6. One-year cumulative incidence of other fractures in men and women per 100 000 in the USA and the UK.
SUMMARY AND RESEARCH AGENDA In this chapter, we have reviewed the occurrence of hip, wrist, vertebral and other fractures in the elderly using a worldwide perspective. We have focused this chapter on hip fracture since this is associated with the greatest mortality, important morbidity and cost.
178 C. E. D. H. De Laet and H. A. P. Pols
The incidence of hip fractures is well documented in most countries, increasing exponentially with age up to about 1% per year for women at age 80 in Western countries. There is, however, a large geographical variation in the incidence of hip fracture, a very low incidence rate being reported in China and Turkey. The reason for this variation is largely unknown, and more research is needed in this area. In most areas of the world, hip fractures occur more frequently in women than in men, the typical female-to-male ratio for the occurrence of hip fracture being between 1.6:1 and 2:1. A large increase in the number of hip fractures worldwide is expected as a result of the ageing of the population. A better understanding of the causes of the geographical dierence, as well as of the reversal of the female-to-male ratio in some countries, could help to unravel the aetiological pathways and possibly lead to more eective prevention strategies. Without such intervention, it is expected that over 6 million hip fractures a year will occur by the year 2050, an increase from 1.7 million in 1990. Information on the occurrence of other fractures is more scarce, and more research worldwide is needed. For a description of the occurrence of these fractures, data from two large observational studies from the US and the UK were compared. The incidence of wrist fractures showed only a slight increase with age. They occur much more frequently in women than in men and the female to male ratio rises with age from 6:1 at age 55 to 10:1 at ages over 75. For vertebral fractures the main problem is that there is no universally accepted de®nition. Consequently the reported incidence rates vary greatly as shown by the overview of 4 studies. Agreement on a de®nition of vertebral fracture should be high on the research agenda. For the other fractures, a modest increase with age is observed and this increase is slightly steeper in women. It is unclear whether all those fractures are related to osteoporosis and therefore whether intervention to prevent osteoporosis would reduce these fractures. Here again, more research is needed.
REFERENCES 1. Seeley DG, Browner WS, Nevitt MC et al. Which fractures are associated with low appendicular bone mass in elderly women? The Study of Osteoporotic Fractures Research Group. Annals of Internal Medicine 1991; 115(11): 837±842. * 2. Anonymous. Consensus development conference: diagnosis, prophylaxis, and treatment of osteoporosis. American Journal of Medicine 1993; 94: 646±650. 3. Kanis JA. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report. Osteoporosis International 1994; 4: 368±381. 4. Cooper C, Campion G & Melton LJ III. Hip fractures in the elderly: a world-wide projection. Osteoporosis International 1992; 2(6): 285±289. * 5. Melton LJ III, Crowson CS & O'Fallon WM. Fracture incidence in Olmsted County, Minnesota: comparison of urban with rural rates and changes in urban rates over time. Osteoporosis International 1999; 9(1): 29±37. * 6. European Commission. Report on Osteoporosis in the European Community ± Action for PreventionBrussels: European Commission, 1998. 7. Elors I, Allander E, Kanis JA et al. The variable incidence of hip fracture in southern Europe: the MEDOS Study. Osteoporosis International 1994; 4(5): 253±263. 8. De Laet CEDH, Van Hout BA & Pols HAP. Osteoporosis in the Netherlands: A burden of Illness Study. Rotterdam: Institute for Medical Technology Assessment, 1996. 9. Bacon WE, Maggi S, Looker A et al. International comparison of hip fracture rates in 1988±89. Osteoporosis International 1996; 6(1): 69±75. 10. Hedlund R. Incidence and cause of femur fractures. Stockholm: Karolinska Institute, 1985, PhD thesis. ISBN: 91-628-3328-6.
Fractures in the elderly 179 11. McColl A, Roderick P & Cooper C. Hip fracture incidence and mortality in an English region: a study using routine National Health Service data. Journal of Public Health Medicine 1998; 20(2): 196±205. 12. Ellfors L. Hip fractures: a European perspective. Stockholm, 1999, thesis. *13. Schwartz AV, Kelsey JL, Maggi S et al. International variation in the incidence of hip fractures: crossnational project on osteoporosis for the World Health Organization Program for Research on Aging. Osteoporosis International 1999; 9(3): 242±253. 14. Oden A, Dawson A, Dere W et al. Lifetime risk of hip fractures is underestimated. Osteoporosis International 1998; 8(6): 599±603. 15. Melton LJ III, O'Fallon WM & Riggs BL. Secular trends in the incidence of hip fractures. Calci®ed Tissue International 1987; 41(2): 57±64. 16. Melton LJ III, Therneau TM & Larson DR. Long-term trends in hip fracture prevalence: the in¯uence of hip fracture incidence and survival. Osteoporosis International 1998; 8(1): 68±74. 17. Nungu S, Olerud C & Rehnberg L. The incidence of hip fracture in Uppsala County. Change of time trend in women. Acta Orthopaedica Scandinavica 1993; 64(1): 75±78. 18. Spector TD, Cooper C & Lewis AF. Trends in admissions for hip fracture in England Wales, 1968±85. British Medical Journal 1990; 300: 1173±1174. 19. Melton LJ III. Hip fractures: a worldwide problem today and tomorrow. Bone 1993; 14(1): S1±S8. 20. De Laet CE, van Hout BA, Burger H et al. Bone density and risk of hip fracture in men and women: cross sectional analysis. British Medical Journal 1997; 315(7102): 221±225. 21. Agnusdei D, Camporeale A, Gerardi D et al. Trends in the incidence of hip fracture in Siena, Italy, from 1980 to 1991. Bone 1993; 14(1): S31±S34. 22. Kannus P, Niemi S, Parkkari J et al. Hip fractures in Finland between 1970 and 1997 and predictions for the future. Lancet 1999; 353(9155): 802±805. *23. Singer BR, McLauchlan GJ, Robinson CM & Christie J. Epidemiology of fractures in 15,000 adults: the in¯uence of age and gender. Journal of Bone and Joint Surgery 1998; 808(2): 243±248. 24. Kanis JA & McCloskey EV. Epidemiology of vertebral osteoporosis. Bone 1992; 13: S1±S10. 25. Ross PD. Clinical consequences of vertebral fractures. American Journal of Medicine 1997; 103: 42S±43S. 26. Cooper C, O'Neill T & Silman A. The epidemiology of vertebral fractures. European Vertebral Osteoporosis Study Group. Bone 1993; 14: S89±S97. 27. Black DM, Palermo L, Nevitt MC et al. De®ning incident vertebral deformity: a prospective comparison of several approaches. The Study of Osteoporotic Fractures Research Group. Journal of Bone and Mineral Research 1999; 14(1): 90±101. 28. Genant HK. Assessment of vertebral fractures in osteoporosis research. Journal of Rheumatology 1997; 24(6): 1212±1214. 29. McCloskey EV, Spector TD, Eyres KS et al. The assessment of vertebral deformity: a method for use in population studies and clinical trials. Osteoporosis International 1993; 3(3): 138±147. 30. Genant HK, Jergas M, Palermo L et al. Comparison of semi quantitative visual and quantitative morphometric assessment of prevalent and incident vertebral fractures in osteoporosis. The Study Osteoporotic Fractures Research Group. Journal of Bone and Mineral Research 1996; 11(7): 984±996. 31. Black DM, Palermo L, Nevitt MC et al. Comparison of methods for de®ning prevalent vertebral deformities: the Study of Osteoporotic Fractures. Journal of Bone and Mineral Research 1995; 10(6): 890±902. 32. Melton LJ III, Chrischilles EA, Cooper C et al. Perspective. How many women have osteoporosis? Journal of Bone and Mineral Research 1992; 7: 1005±1010. 33. De Laet CEDH, McCloskey EV, van der Klift M et al. The incidence of vertebral fractures in the Rotterdam Study [abstract]. Calci®ed Tissue International 2000; 66 (supplement 1): S60. 34. O'Neill TW, Felsenberg D, Varlow J et al. The prevalence of vertebral deformity in European men and women: the European Vertebral Osteoporosis Study. Journal of Bone and Mineral Research 1996; 11(7): 1010±1018.
doi:10.1053/beem.2000.0067, available online at http://www.idealibrary.com on
1 Fractures in the elderly: epidemiology and demography Chris E. D. H. De Laet
MD, PhD
Senior epidemiologist
Huibert A. P. Pols
MD, PhD
Professor of medicine Institute for Medical Technology Assessment, Erasmus University, Rotterdam, The Netherlands Department of Internal Medicine, Erasmus University Medical Center, Rotterdam, The Netherlands
Osteoporosis and osteoporosis-related fractures are a major source of both morbidity and cost in the elderly, the fractures that are most commonly associated with osteoporosis being those of the hip, the distal forearm and the vertebrae, although it is believed that most other fractures occurring in the elderly are also related to osteoporosis. In this review, the incidence of all types of fracture is described based on the available literature, and the foreseeable trends resulting from demographic changes are discussed. Emphasis is given to the epidemiology of hip fracture since this is the most serious consequence of osteoporosis. Hip fractures occur all over the world, most currently occurring in Western countries, mainly Europe and the USA, but it is expected that there will be a large increase in the number of hip fractures in other countries because of demographic changes. The incidence of hip fractures increases exponentially with age, resulting in a 1-year incidence of 1% in women aged 80 in Western countries. Most hip fractures occur in women, but this is again partly due to demography, because of the longer life expectancy of women. Wrist fractures occur more often in women and do not show the same increase with age as hip fractures. The incidence reaches a plateau at age 60±70. Vertebral fractures show a modest increase with age and are again more common in women than men. The incidence of all other fractures increases modestly with age Key words: Colles' fracture; forecasting; fractures; hip fractures; incidence; men; osteoporosis; spinal fractures; women.
Osteoporosis and osteoporosis-related fractures are a major source of both morbidity and cost in the elderly. The most important fractures related to osteoporosis are those of the hip, the distal forearm and the vertebrae, although it is believed that most other fractures occurring in the elderly are also related to osteoporosis.1 Osteoporosis is de®ned as a condition characterized by low bone mass and the micro-architectural deterioration of bone tissue, with a consequent increase in bone fragility and susceptibility to fracture.2 The conventional method of estimating bone mass is by bone densitometry, the World Health Organization de®ning osteoporosis as a bone mineral density (BMD) less than 2.5 standard deviation (SD) below the mean for young adults.3 1521±690X/00/02017109 $35.00/00
c 2000 Harcourt Publishers Ltd. *
172 C. E. D. H. De Laet and H. A. P. Pols
In this review, we will discuss the incidence of fractures and explore the foreseeable trends in the epidemiology of fractures caused by demographic change. Since the most serious consequence of osteoporosis is fracture of the hip, the emphasis will be on this type of fracture. HIP FRACTURES Current incidence Most hip fractures currently occur in Western countries. While the total number of hip fractures worldwide in 1990 was estimated at 1.7 million, 560 000 of these occurred in Europe (including the former Soviet Union) and 360 000 in North America.4 Information about hip fractures is available for most countries since virtually all require hospital admission and treatment, making them relatively easy to detect through hospital discharge registries. In most countries, the incidence of hip fracture shows an exponential increase with age and is in general higher in women than in men after the age of 50. This higher incidence in women is re¯ected in the absolute number of fractures since there are, in most countries, more elderly women than men. Recent ®gures on hip fracture incidence in the USA have been published by Melton et al.5 For Europe, data have been compiled in a 1998 European Commission report on the incidence of hip fractures in the 15 member states of the European Union (EU).6 In this report, data from several studies were assembled, including those from the MEDOS study, focusing on countries around the Mediterranean Sea, and also from several national studies.7±11 The incidence in Europe showed a north±south gradient, the highest incidence being in Scandinavia and a much lower incidence occurring in the countries around the Mediterranean. The incidence in Sweden was highest, the relative incidence compared with that of the USA being 1.3 in women and 1.7 in men (author's own calculations based on the European Commission report).6 In Finland, the rate was comparable to that of the USA for women but slightly higher in men. In the UK, the Netherlands and Germany, the incidence was very similar to that observed in the USA. In Southern European countries, the incidence was much lower, that in France, Greece and Spain being about 70% of that in the USA, while in Italy and Portugal the incidence was as low as 50% of that in the USA. The MEDOS study also contained data on Turkey, a non-EU country partly in Europe but mainly in Asia. These data are quite dierent from those of mainland Europe. Both the absolute and the age-standardized rates were very low, and the increase with age was less pronounced than in European countries.12 The incidence rate was about 10±20% of the equivalent rate in the USA for women, and about 20±30% of that for men (author's own calculations). Less information is available from other continents, but it appears that, in both South America and Asia, the incidence of hip fracture is lower than in North America and Europe. Schwartz et al13 compared data from hospital registers from several continents, attempting to adjust for dierences in coding practice. They concluded that the dierence reported between countries mainly re¯ected a genuine variation in the hip fracture incidence rate. The rate for Beijing was as low as that reported from Turkey in the MEDOS study. The rates for Brazil, Hong Kong and Hungary were somewhat lower than the USA and European averages, while the rate for Iceland equalled the high Scandinavian level. Figures 1 and 2 give a comparison of the one-year cumulative incidence rates per 100 000 for some selected countries based on several studies.5±7,9,13
Fractures in the elderly 173 5000 4500 Sweden Iceland Finland UK Netherlands USA France Italy Portugal Spain Switzerland Hungary Hong Kong Brazil China Turkiye (town)
Incidence/100 000
4000 3500 3000 2500 2000 1500 1000 500 0
55-59
60-64
65-69
70-74
75-79
80-84
85+
Age
Figure 1. One-year cumulative incidence of hip fracture in women per 100 000 in several countries. 5000 4500 Incidence/100 000
4000 3500
Sweden Iceland Finland UK Netherlands USA France Greece Italy Portugal Spain Switzerland Hungary Hong Kong Brazil China Turkiye (town)
3000 2500 2000 1500 1000 500 0
55-59
60-64
65-69
70-74
75-79
80-84
85+
Age
Figure 2. One-year cumulative incidence of hip fracture in men per 100 000 in several countries.
Gender ratio The global female-to-male ratio for the absolute number of hip fractures is around 3.5:1 in most countries, but this high female-to-male ratio is largely due to demography. When the incidence rates are compared, the female-to-male ratio is between 1.6:1 and 2:1 in most Western countries. Only in Sweden and Finland has a lower ratio (1.3:1 and 1.4:1 respectively) been reported. The Turkish data were again strikingly dierent, with a female-to-male ratio of about 1:1 in the towns of Istanbul and Ankara, and a reversal of the ratio (around 0.4) in rural areas, especially in older participants. A reversal of the female-to-male ratio was also observed in China (Beijing).
174 C. E. D. H. De Laet and H. A. P. Pols
The reason for both the huge geographical variation in hip fracture incidence and the reversal of the gender ratio in some countries is largely unknown. It is probably due to a combination of genetic, lifestyle and environmental factors and is clearly a topic for future research. Future trends Future trends are in¯uenced by both demography and changes in age-adjusted incidence. Demography is relatively easy to predict for the near future because all those potentially at risk within the next 50 years have already been born. The only uncertainty with respect to demography is the future mortality rate.14 Trends in the hip fracture incidence rate are not as predictable, especially since there are con¯icting ®ndings from dierent countries. An increase in age-adjusted incidence was described between 1930 and 1980 in several countries. That increase appeared to have levelled o in the 1980s in parts of the USA15,16 as well as in Sweden and the UK.17±19 In other European countries such as the Netherlands, Italy and Finland, the age-adjusted incidence continued to rise during the early 1990s.8,20±22 Because of this uncertainty, most predictions for the future are based on the current incidence rate, taking only demography into account. Taking those assumptions into account, it is believed that the global number of hip fractures worldwide will increase to over 6 million by the year 2050.4 Although the high proportion of hip fractures occurring in Europe will decline over the next few decades because of a demographic evolution in other parts of the world, the absolute number of hip fractures will also continue to rise in Europe to over 1 million in 20504 since, as in other areas of the world, the population in Europe is also ageing, mainly because of the post-Second World War baby boom. Moreover, with the continuing increase in life expectancy in both men and women, there will be a gradual increase in the proportion of very elderly people. Over the next decades, these trends are expected to continue. Estimates of lifetime risk for hip fracture range from 15% to 18% in Northern Europe and the USA, but, taking into account increasing life expectancy, these rates are very likely to be an underestimate.14 The report from the European Commission estimated that the yearly number of hip fractures within the 15 member states would increase to almost 1 million by the year 2050.6 The female-to-male ratio within the EU for the total number of fractures, which was 3.7:1 for the year 2000, was projected to decline to 3.2:1 in the year 2050, again due to demographic changes alone. OTHER FRACTURES The incidence of other fractures is more dicult to determine. Since those fractures are often treated in an outpatient setting, they are poorly registered and cannot easily be traced through hospital discharge registers. Therefore, their incidence can only be assessed in observational studies in which a limited number of care institutions deliver care to a population within a well-de®ned catchment area. For the estimation of the incidence of non-hip fractures, we have used two such studies. In a recently published study, all the fractures that occurred during a 3-year period within one district (Olmsted County) in the USA (3665 fractures) were recorded. 5 We will compare these data with those from a study performed by Singer et al describing over 15 000 fractures in Edinburgh, again within a well-de®ned catchment area.23 These two studies will be referred to as the US and the UK studies.
Fractures in the elderly 175
Fractures of the distal forearm Fractures of the distal forearm are usually caused by a fall on the outstretched arm, and they are also associated with osteoporosis. They typically occur at a younger age than hip fractures, mostly in women, and they do not show the same exponential increase with age as do hip fractures. There is no diagnostic problem, but since they are mostly treated in an outpatient setting, accurate documentation is more dicult. Figure 3 shows the incidence data for wrist fractures from the US and the UK studies. They are remarkably similar, although in the US study the incidence in women rises more rapidly to a plateau, whereas the UK study shows a more gradual increase with age in women. In the age categories over 55, the female-to-male ratio is much higher than it is in hip fractures: at age 55, this ratio is around 6:1, and it increases to 10:1 at ages over 75. Since the increase in incidence with age is much less pronounced than for hip fractures, it is not expected that the incidence of fracture of the distal forearm will change much as the result of demographic changes in the future. 1000 900
Incidence/100 000
800 700 600 USA (women) UK (women) USA (men) UK (men)
500 400 300 200 100 0
55−59
60−64
65−69
70−74
75−79
80−84
>85
Age
Figure 3. One-year cumulative incidence of wrist fracture in men and women per 100 000 person-years in the USA and the UK.
Vertebral fractures Determining the incidence of vertebral fractures is even more complex. Since it is estimated that only one third of radiographically visible deformities readily come to clinical attention, the measured incidence is heavily dependent on its de®nition.24±26 There is currently no universally accepted de®nition of vertebral fractures, although several de®nitions have been proposed.27±31 Moreover, in clinical practice, case ascertainment is very dierent between countries. In the US study, a vertebral fracture was de®ned on the basis of a radiologist's report of compression or collapse of one or more thoracic or lumbar vertebrae during a 3-year period, and it was as such essentially a prevalence study. It is also unclear whether or not these deformities were associated with clinical symptoms. In an earlier study from the same group32, a
176 C. E. D. H. De Laet and H. A. P. Pols
vertebral fracture was de®ned as one which came to clinical attention. In the UK study, the incidence of a vertebral fracture was de®ned as any vertebral fracture reported from a health-care institution, the incidence being extremely low. A comparison was also performed between these data and those from a 6.5-year followup in the Rotterdam study33, in which follow-up spinal radiographs were compared with base-line radiographs using the McCloskey±Kanis algorithm for the morphometric de®nition of vertebral deformities.29 The resulting incidence estimates from these four studies are shown in Figures 4 and 5. Although these ®gures show that, in most studies, the apparent incidence 3500
Incidence/100 000
3000
The Netherlands: morphometric USA radiographic USA clinical UK clinical
2500 2000 1500 1000 500 0 55−59
60−64
65−69
70−74
75−79
80−84
>85
Age
Figure 4. One-year cumulative incidence of vertebral fracture in women per 100 000 in the USA, the UK and the Netherlands. 3500
Incidence/100 000
3000
The Netherlands: morphometric USA radiographic USA clinical UK clinical
2500 2000 1500 1000 500 0 55−59
60−64
65−69
70−74
75−79
80−84
>85
Age
Figure 5. One-year cumulative incidence of vertebral fracture in men per 100 000 in the USA, the UK and the Netherlands.
Fractures in the elderly 177
increases with age, they also show a large dierence in the apparent incidence rate depending on the de®nition used. Therefore, the exact de®nition of a vertebral fracture is extremely important in the interpretation of any reported incidence data. In Europe, a concerted eort is currently ongoing in the form of the European Vertebral Osteoporosis Study (EVOS), in which a very rigorous quality assurance procedure is used to minimize the numbers of false-positive results from radiographic evaluation. Currently, however, only prevalence data are available from this study.34 The female-to-male ratio is around 2:1 at most ages, and it is unlikely that the occurrence of vertebral fractures will alter much in the future because of demographic changes. Fractures other than hip, distal forearm or vertebral Since there is no general consensus on whether other fractures are also related to osteoporosis, we used, as an arbitrary de®nition, all other fractures excluding those of the face, skull and ribs, as well as fractures of ®ngers and toes. Again, data from the USA are compared with those from the UK. Figure 6 shows the estimated incidence of these other fractures. In men, the incidence doubles with age between 55 and 80, the increase being slightly steeper in women. The female-to-male ratio is around 2:1 at 55 and increases slightly with age. 3500
Incidence/100 000
3000
USA (women) UK (women) USA (men) UK (men)
2500 2000 1500 1000 500 0 55−59
60−64
65−69
70−74
75−79
80−84
>85
Age Figure 6. One-year cumulative incidence of other fractures in men and women per 100 000 in the USA and the UK.
SUMMARY AND RESEARCH AGENDA In this chapter, we have reviewed the occurrence of hip, wrist, vertebral and other fractures in the elderly using a worldwide perspective. We have focused this chapter on hip fracture since this is associated with the greatest mortality, important morbidity and cost.
178 C. E. D. H. De Laet and H. A. P. Pols
The incidence of hip fractures is well documented in most countries, increasing exponentially with age up to about 1% per year for women at age 80 in Western countries. There is, however, a large geographical variation in the incidence of hip fracture, a very low incidence rate being reported in China and Turkey. The reason for this variation is largely unknown, and more research is needed in this area. In most areas of the world, hip fractures occur more frequently in women than in men, the typical female-to-male ratio for the occurrence of hip fracture being between 1.6:1 and 2:1. A large increase in the number of hip fractures worldwide is expected as a result of the ageing of the population. A better understanding of the causes of the geographical dierence, as well as of the reversal of the female-to-male ratio in some countries, could help to unravel the aetiological pathways and possibly lead to more eective prevention strategies. Without such intervention, it is expected that over 6 million hip fractures a year will occur by the year 2050, an increase from 1.7 million in 1990. Information on the occurrence of other fractures is more scarce, and more research worldwide is needed. For a description of the occurrence of these fractures, data from two large observational studies from the US and the UK were compared. The incidence of wrist fractures showed only a slight increase with age. They occur much more frequently in women than in men and the female to male ratio rises with age from 6:1 at age 55 to 10:1 at ages over 75. For vertebral fractures the main problem is that there is no universally accepted de®nition. Consequently the reported incidence rates vary greatly as shown by the overview of 4 studies. Agreement on a de®nition of vertebral fracture should be high on the research agenda. For the other fractures, a modest increase with age is observed and this increase is slightly steeper in women. It is unclear whether all those fractures are related to osteoporosis and therefore whether intervention to prevent osteoporosis would reduce these fractures. Here again, more research is needed.
REFERENCES 1. Seeley DG, Browner WS, Nevitt MC et al. Which fractures are associated with low appendicular bone mass in elderly women? The Study of Osteoporotic Fractures Research Group. Annals of Internal Medicine 1991; 115(11): 837±842. * 2. Anonymous. Consensus development conference: diagnosis, prophylaxis, and treatment of osteoporosis. American Journal of Medicine 1993; 94: 646±650. 3. Kanis JA. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report. Osteoporosis International 1994; 4: 368±381. 4. Cooper C, Campion G & Melton LJ III. Hip fractures in the elderly: a world-wide projection. Osteoporosis International 1992; 2(6): 285±289. * 5. Melton LJ III, Crowson CS & O'Fallon WM. Fracture incidence in Olmsted County, Minnesota: comparison of urban with rural rates and changes in urban rates over time. Osteoporosis International 1999; 9(1): 29±37. * 6. European Commission. Report on Osteoporosis in the European Community ± Action for PreventionBrussels: European Commission, 1998. 7. Elors I, Allander E, Kanis JA et al. The variable incidence of hip fracture in southern Europe: the MEDOS Study. Osteoporosis International 1994; 4(5): 253±263. 8. De Laet CEDH, Van Hout BA & Pols HAP. Osteoporosis in the Netherlands: A burden of Illness Study. Rotterdam: Institute for Medical Technology Assessment, 1996. 9. Bacon WE, Maggi S, Looker A et al. International comparison of hip fracture rates in 1988±89. Osteoporosis International 1996; 6(1): 69±75. 10. Hedlund R. Incidence and cause of femur fractures. Stockholm: Karolinska Institute, 1985, PhD thesis. ISBN: 91-628-3328-6.
Fractures in the elderly 179 11. McColl A, Roderick P & Cooper C. Hip fracture incidence and mortality in an English region: a study using routine National Health Service data. Journal of Public Health Medicine 1998; 20(2): 196±205. 12. Ellfors L. Hip fractures: a European perspective. Stockholm, 1999, thesis. *13. Schwartz AV, Kelsey JL, Maggi S et al. International variation in the incidence of hip fractures: crossnational project on osteoporosis for the World Health Organization Program for Research on Aging. Osteoporosis International 1999; 9(3): 242±253. 14. Oden A, Dawson A, Dere W et al. Lifetime risk of hip fractures is underestimated. Osteoporosis International 1998; 8(6): 599±603. 15. Melton LJ III, O'Fallon WM & Riggs BL. Secular trends in the incidence of hip fractures. Calci®ed Tissue International 1987; 41(2): 57±64. 16. Melton LJ III, Therneau TM & Larson DR. Long-term trends in hip fracture prevalence: the in¯uence of hip fracture incidence and survival. Osteoporosis International 1998; 8(1): 68±74. 17. Nungu S, Olerud C & Rehnberg L. The incidence of hip fracture in Uppsala County. Change of time trend in women. Acta Orthopaedica Scandinavica 1993; 64(1): 75±78. 18. Spector TD, Cooper C & Lewis AF. Trends in admissions for hip fracture in England Wales, 1968±85. British Medical Journal 1990; 300: 1173±1174. 19. Melton LJ III. Hip fractures: a worldwide problem today and tomorrow. Bone 1993; 14(1): S1±S8. 20. De Laet CE, van Hout BA, Burger H et al. Bone density and risk of hip fracture in men and women: cross sectional analysis. British Medical Journal 1997; 315(7102): 221±225. 21. Agnusdei D, Camporeale A, Gerardi D et al. Trends in the incidence of hip fracture in Siena, Italy, from 1980 to 1991. Bone 1993; 14(1): S31±S34. 22. Kannus P, Niemi S, Parkkari J et al. Hip fractures in Finland between 1970 and 1997 and predictions for the future. Lancet 1999; 353(9155): 802±805. *23. Singer BR, McLauchlan GJ, Robinson CM & Christie J. Epidemiology of fractures in 15,000 adults: the in¯uence of age and gender. Journal of Bone and Joint Surgery 1998; 808(2): 243±248. 24. Kanis JA & McCloskey EV. Epidemiology of vertebral osteoporosis. Bone 1992; 13: S1±S10. 25. Ross PD. Clinical consequences of vertebral fractures. American Journal of Medicine 1997; 103: 42S±43S. 26. Cooper C, O'Neill T & Silman A. The epidemiology of vertebral fractures. European Vertebral Osteoporosis Study Group. Bone 1993; 14: S89±S97. 27. Black DM, Palermo L, Nevitt MC et al. De®ning incident vertebral deformity: a prospective comparison of several approaches. The Study of Osteoporotic Fractures Research Group. Journal of Bone and Mineral Research 1999; 14(1): 90±101. 28. Genant HK. Assessment of vertebral fractures in osteoporosis research. Journal of Rheumatology 1997; 24(6): 1212±1214. 29. McCloskey EV, Spector TD, Eyres KS et al. The assessment of vertebral deformity: a method for use in population studies and clinical trials. Osteoporosis International 1993; 3(3): 138±147. 30. Genant HK, Jergas M, Palermo L et al. Comparison of semi quantitative visual and quantitative morphometric assessment of prevalent and incident vertebral fractures in osteoporosis. The Study Osteoporotic Fractures Research Group. Journal of Bone and Mineral Research 1996; 11(7): 984±996. 31. Black DM, Palermo L, Nevitt MC et al. Comparison of methods for de®ning prevalent vertebral deformities: the Study of Osteoporotic Fractures. Journal of Bone and Mineral Research 1995; 10(6): 890±902. 32. Melton LJ III, Chrischilles EA, Cooper C et al. Perspective. How many women have osteoporosis? Journal of Bone and Mineral Research 1992; 7: 1005±1010. 33. De Laet CEDH, McCloskey EV, van der Klift M et al. The incidence of vertebral fractures in the Rotterdam Study [abstract]. Calci®ed Tissue International 2000; 66 (supplement 1): S60. 34. O'Neill TW, Felsenberg D, Varlow J et al. The prevalence of vertebral deformity in European men and women: the European Vertebral Osteoporosis Study. Journal of Bone and Mineral Research 1996; 11(7): 1010±1018.