The majority of studies detailing hip fracture rates have been performed in the last five decades. Scandinavia has the highest reported incidence of hip fracture worldwide. There are a large number of studies looking at incidence rates as well as secular trends in this geographically northern region. The incidence rates vary from North to South Europe, with the highest being in Sweden and Norway and the lowest in France and Switzerland. From Norway, the reported age-standardized annual incidence rate of hip fracture is 920/100 000 in women and 399.3/100 000 in men and that in Switzerland is 346/10000 and 137.8/100 000 in women and men, respectively. A study on secular trends from Uppsala, Sweden, (1965–1980) showed an annual increase of 2.2% for age- and sex-adjusted hip fracture rates, which increased from 430/100 000 in 1965 to 650/100 000 in 1980.27
The age-specific incidence increased especially in the group aged ≥85 years, in whom fractures of the femoral neck were three times more common and trochanteric fractures four times more common in 1980 than in 1965. Studies from Malmo, Sweden, showed an exponential increase in hip fracture incidence from 1950 to 1985 in both men and women over age 50, increasing from an annual age-adjusted incidence of 150–390/100 000 in men and 300–830/100 000 in women.28
Increases were seen for both trochanteric and cervical fractures. However, the most recent data from Malmo from 1992–1995 shows that this fracture rate is now steady, in line with many of the Northern American studies.29
If the data is broken down according to fracture subtypes, however, the incidence of cervical fractures had decreased (210–170/100000 in men and 420–410/100000 in women), whereas there was still a small rise in trochanteric fractures (180–190/100000 in men and 410–440/100000 in women).
The incidence of hip fractures in Oslo, Norway, between 1978 and 1997 was assessed using electronic diagnosis registers.30
The age-adjusted fracture rates per 10000 were 118.0 and 44.0 in 1996/97, 124.3 and 44.9 in 1988/89, and 104.5 and 35.8 in 1978/79 for women and men, respectively, indicating that the incidence of hip fractures in Oslo has not changed significantly during the last decade. Denmark used its national patient register to look at hip fracture incidence between 1987 and 1997 in Viborg County.31
The incidence of age-adjusted first hip fractures increased significantly by 18 and 8 per 100 000 per year for women and men, respectively; peri-trochanteric fractures increased by 10 per 100 000 per year There are a number of studies from Finland, all using the Finnish National Hospital discharge register. The first study drew its hip fracture data from the entire >50-year-old population. Between 1970–997, the age-adjusted hip fracture rate increased from 292/100 000 to 467/100 000 in women and from 112/100 000 to 233/100 000 in men.23
In a further follow-up study using the same population from 1997–2004, the age-adjusted incidence fell nationally by 2.4% annually in women and 0.9% in men in 2004.32
A smaller study in central Finland between 1992–2003 showed that the age-adjusted rates increased considerably from 2.0 per 1000 person-years to 3.9 per 1000 person-years in men and from 2.8 per 1000 person-years to 5.6 per 1000 person-years in women.33
There is no obvious explanation why central Finland should have increasing rates compared to the rest of the country. One possibility is that only two time points were examined in this study, whereas most other studies have examined the secular trends; it is possible that a sharp rise in the rates may have occurred during the early part of the study period and then stabilized.
Overall, the epidemiological data from Scandinavia indicates that although early studies (dating to the1950s) described an increase in hip fracture incidence, the rates appear to have fallen in the most recent periods. The decline in fracture rate appears to have occurred several years later in Finland compared to Sweden and Norway. The rise in the incidence of hip fracture in Finland from the early 1970s until the late 1990s has been followed by declining fracture rates. The exact reasons for this are unknown, but a cohort effect toward a healthier aging population and increased average body weight and improved functional ability among elderly Finns cannot be ruled out as possible causes.34
Data from central Europe includes studies from the UK, Netherlands, Germany, Switzerland, Austria, and Hungary. The highest number of studies comes from the UK. The first study on this subject assessed hip fracture data from the hospital inpatient enquiry for England and Wales between 1968–78.35
In this study, the age-specific rates increased steadily in women by 61% and in men by 73% until 1979. No further increases occurred in either sex up to 1985. The Oxford Record Linkage Study which looked at the period 1968–1986 found a similar pattern, although there was a more continuous trend.22
Age, cohort, and period modelling were used in this study to look at the incidence rates. There was a clear cohort effect in both the studies and the difference in incidence rates was apparent from births in 1883 to 1917 in addition to the age. A birth cohort effect was confirmed in subsequent analyses of the Framingham data, showing that in progressive birth cohorts from 1887–1921 there were age-specific increases in fracture incidence rate ratios up from 1.0 to 1.2–1.4 in women and 1.0 to 2.0 in men.21
The most recent study from the UK looked at hospital episode statistics from 1989–1998.36
Age-standardized incidence rates increased by 32% in women and 38% in men up to 1991–92 and thereafter remained stable. In the Netherlands, data from the Dutch Medical Registry shows that between 1972–1987 the age-adjusted incidence of hip fractures rose linearly from 479/100 000 to 669/100 000 per year in women and from 198/100 000 to 308/100 000 per year in men aged ≥65 years.37
In a later study conducted between 1986 and 2002 using the Dutch Medical Registry, the age incidence of hip fracture increased linearly from 1986–1993 in patients over 45 years;38
after this the incidence decreased by 0.5% annually until 2002. In Germany, between 1995 and 2004 (in a study using the national hospital discharge register) the age- and sex-adjusted hip fracture incidence increased by 0.5% per year in women and 0.7% per year in men.39
In women aged ≥40 years there was a tendency for a decrease up to the age of 74 years, but there was a pronounced increase in patients over 75 years. Interestingly, the increase was significantly higher in Eastern Germany compared to Western Germany, particularly in the older age-groups, which suggests that the differences between the East and West decreased over time.
Hip fracture trends in Geneva, Switzerland, between 1991 and 2000 have been studied using computerized medical records from the main hospital. The study found a significant decline of 1.4% annually for age-adjusted rates in women, but the rates remained stable in men.40
In neighbouring Austria, between 1994 and 2006, after adjustment for age and sex, the incidence rates rose from 471/100 000 to 567/100 000 per year in men and from 637/100 000 to 759/100 000 per year in women.41
This rise is predominantly accounted for by the rise seen in patients aged over 80 years. Hip fracture rates in Hungary are available for the period 1999–2003 from the National Health Insurance Fund database. This database covers the whole population of Hungary.42
The age-adjusted incidence of hip fracture was 430/100 000 in women and 223/100 000 in men; these rates have remained stable over this period. A lesser number of studies are available from southern Europe; two of these have been chosen to represent hip fractures from this region. An Italian study looked at the incidence of hip fracture in the county of Sienna from 1980–1991 using records from the orthopedic departments of various hospitals.43
During this 12-year period, the temporal trend rose linearly in men from 57.5/100 000 person-years to 108.9/100 000 person-years: a 7.4% annual increase. In females, no significant trend was observed. The overall incidence rate during this period was 157/100 000, much lower than that in northern or central European countries. Another study from Spain looked at hip fracture trends in northern Spain between 1988 and 2002 using clinical records from all hospitals in the region of Cantabria.44
Whilst the crude hip fracture incidence increased during this period, no significant changes were noticed following adjustment for age. Neither was there a noticeable trend in age-specific incidence rates. The crude rate increased mainly among with a more noticeable rise in cervical fractures as opposed to trochanteric fractures.