Epidemiology of hip fractures in Belgrade, Serbia Montenegro, 1990–2000
- First Online:
- Cite this article as:
- Lešić, A., Jarebinski, M., Pekmezović, T. et al. Arch Orthop Trauma Surg (2007) 127: 179. doi:10.1007/s00402-006-0234-2
- 61 Views
This study retrospectively determined the incidence rates of hip fractures in Belgrade, Serbia and Montenegro, during the period 1990–2000.
Materials and methods
All patients with hip fractures treated at all Belgrade hospitals were identified from the Republic of Serbia’s Ministry of Health National Health Care database. Patient demographics, type of hip fracture, and details of the mechanism of injury were collected. The annual incidence rates were calculated with interpolation according to the Belgrade population census of 1991 and 2002.
There were a total of 8,904 hip fractures with a mean annual incidence of 51.7 per 100,000 adults (62.2 females and 35.5 males). Mean age at the time of fracture was 67 years (72.6 for females and 59.3 for males), with 64.7% of all fractures occurring in women. There was a significant increase in hip fracture incidence rates over the observed period in females (P = 0.006), but not in males (P = 0.962). Trochanteric fractures predominated, accounting for 53% compared with cervical fractures. In patients over 50 years of age there was an exponential increase in the incidence of hip fractures in both sexes; though more so in females. 91% of hip fractures occurred in these older patients with incidence rates of 143.6 per 100,000 (185.9 for female and 92.2 for male patients). The most common mechanism of injury in the older group was low-energy trauma (70.3%) resulting from a fall from standing height onto a flat surface (same level). Standardizing incidence rates in the older age group to the US 1985 white population gave values of 228 per 100,000 females and 96 per 100,000 males. These incidence rates are similar to those reported in Italy, France and Great Britain, but lower than those in Scandinavian countries.
In view of growing population numbers and an increase in the proportion of patients aged over 60 years, we can expect an increase in the prevalence of osteoporosis and an increase in the incidence of fragility hip fractures in the future, with resource implications.
KeywordsHip fracturesEpidemiologyBelgradeSerbia Montenegro
We are currently mid-way through the UN and WHO-endorsed Bone and Joint decade 2000–2010 , which was set up to raise awareness of the growing burden of musculoskeletal disorders on society, healthcare systems and the individual, as well as to promote their prevention and treatment. Part of this increased awareness has been the recognition of osteoporosis, and its associated fragility fractures, as a major healthcare problem, possibly second only to cardiovascular disease . It is currently estimated that, worldwide about one in three women and one in eight men over the age of 50 are at risk of having an osteoporotic fracture during their lifetime .
Hip fractures are probably the most serious consequence of osteoporosis, and a leading cause of morbidity and mortality in the older population, with half of previously independent individuals becoming partly or totally dependent, and 5–20% dying within 1 year following injury . The incidence of hip fractures increases exponentially with age over 50 years, and thus as the global population steadily grows older, hip fractures may reach epidemic levels in the future [4–6].
Incidence rates for hip fracture vary around the world, with higher rates being reported in Scandinavian countries, the United States and Western Europe, and lower rates in the Far East and Africa [7–9]. The only previous hip fracture data, from the former Yugoslavia, come from studies by Matkovic et al.  and Johnell et al. . In the first study, two small subpopulation groups in Istra and Podravina were compared for rates of proximal femoral fracture and daily calcium dietary intake. Unfortunately, this provided very limited data, and we were not able to usefully combine it with our own. No comparison was made with the data from the second study, as the source of information is not known.
This paper represents the first comprehensive study of the population in the Belgrade region of Serbia Montenegro, with the purpose of establishing hip fracture incidence rates and trends, and to define the possible at-risk population groups according to demographic characteristics and the mechanisms of injury.
Materials and methods
All patients with hip fractures, treated for the first time at all the Belgrade hospitals between 1990 and 2000, were retrospectively identified using the International Classification of Diseases, Trauma and Cause of Death (codes 820.0 and S72.0 and revised codes 820.1-9 and S72.1-10), from the Republic of Serbia’s Ministry of Health National Health Care database. Patients were treated at, the Institute for Orthopedic Surgery and Trauma at the Clinical Center of Serbia, the Institute for Orthopedic and Surgical Diseases at Banjica, and the trauma wards of the Belgrade Clinical Center.
Patient demographics, the type of fracture (whether cervical or trochanteric), and details of the mechanism of injury were obtained from hospital discharge data, as coded by the responsible orthopaedic teams. The annual incidence rates were calculated with interpolation according to the Belgrade population census of 1991 and 2002, and expressed per 100,000 adults per year. For persons over the age of 50 years, hip fracture rates were standardised according to the US white population of 1985 , so as to allow comparisons to make with other hip fracture incidence rates in the literature . Incidence rates were calculated with confidence intervals (CI) based on a probability of 95%, and any significant change in annual incidence was determined by testing the coefficients for linear regression .
During the period 1990–2000, 8,904 hip fractures were treated by hospitals in Belgrade; 6,001 patients (67.4%) were female and 2,903 (32.6%) were male. The mean age of patients with hip fractures was 67 ± 6.9 years, with female patients significantly older (72.6 ± 9.8 years) than male patients (59.3 ± 11.5 years) (P < 0.005). There was a higher rate of trochanteric fractures compared with cervical fractures (53% compared with 47%, respectively).
Annual frequency and incidence (per 100,000 adult population) of hip fractures in Belgrade from 1990 to 2000 by sex
Total (M + F)
Though there was no overall statistically significant annual increase in fracture rate (P = 0.085), there was a significant annual increase in female hip fractures (y = 53.7 + 2.026x, P = 0.006); (males: y = 34.2-0.020x, P = 0.962).
Mean frequency and incidence (per 100,000 adult population) of hip fractures in Belgrade from 1990 to 2000 by age group
Age groups (years)
Up to 19
Hip fracture incidence rates (per 100,000 adult population) by sex in patients aged 50 years and over, standardized to the 1985 US white population
Oslo, Norway 1978–1979
Malmo, Sweden 1987–1991
Rochester, USA 1965–1974
Australia, NSW 1989–1990
Vaud, Switzerland 1986–1991
Picardy, France 1987
Southampton Great Britain 1986
Belgrade, Serbia Montenegro 1990–2000
Tottori Prefecture, Japan 1986–1987
Analysis of the mechanisms of injury revealed that low-energy trauma resulting from a fall from standing height onto a flat surface (at the same level), and traffic accidents emerged as the two main causes of hip fractures in the Belgrade population. Again, a difference was observed between the age groups; in patients less than 50 years traffic accidents were the most frequent cause of hip fracture (59.7%), while falls onto a flat surface predominated in those aged 50 years and over (70.3%).
Over the 11-year study period, the number of hip fractures increased in both sexes. Though this increase was only statistically significant in females, the total adult annual increase in incidence rates approached significance (P = 0.085). This increasing trend has also been noted by other authors [4, 5], while some notably Scandinavian studies, have shown no such increase [14, 15]. Females represented the majority of hip fracture patients treated in Belgrade hospitals (67.4 vs 32.6% males), with their dominance starting around the age of 60 and becoming more prominent in older age groups. Similar female/male ratios are found in many series with females typically representing 66–78% of hip fracture patients [14–17]. In some Far Eastern and one Southern European study, however, the reverse has been found to the case; with males showing an overall predominance [18–20]. Our population had higher rates of trochanteric hip fractures compared with cervical fractures (53 vs 47%), without any significant differences in this distribution between age and sex; mirroring findings in other studies [14–16].
The mean age in our hip fracture patients (females 72 years, males 59 years) was lower than in some European countries (78–82 years for women and 71–78 for men) [14, 21, 22], but higher than for the Far Eastern women (62–67 years) [9, 18, 23]. This may reflect the differences in life expectancy of people in different regions, although variations in the incidence and mean age of patients, within different areas of the same territorial region such as a single country, point towards other contributing factors such as differences in occupation, environment factors, and lifestyles .
The increasing rate of hip fracture incidence with increasing age is strongly associated with increases in the incidence of osteoporosis. In addition, with increasing age, hip fractures occur more commonly than do other osteoporosis-related fractures, such as the spine, proximal humerus and wrist . They account for more than 30% of all hospital treated fractures and over 50% of all hospital inpatient days spent by fracture patients [24, 25]. Bearing in mind that there has been an increase in the population numbers and proportion of patients aged over 60 years (from 16% in 1991 to 22% in 2002), according to the Belgrade population census figures, as well as a tendency towards further increases in the elderly population, we can expect an increase in the prevalence of osteoporosis and an increase in the number of hip fracture patients in our region in the future. Estimates predict an increased annual global burden, from 1.7 million hip fractures in 1990 to 6.26 million hip fractures by 2050 .
The use of standardised hip fracture incidence rates enable comparisons to be made between the population groups of different countries, regardless of their age ranges. However, it would be better if these standardisations were made to a more general population group, such as a European or World population group, rather than taking a specific population such as the 1985 white US population as a standard.
According to Table 3, the incidence of hip fractures in Scandinavian countries, especially Norway, is among the highest in the world . However, neither Norwegian, Swedish nor Finnish studies show trends towards future increases in incidence rates [14, 15, 27, 28]. Possible explanations for this lie with, the successful prevention and treatment of osteoporosis, an increase in the non-Scandinavian immigrant population group with a possible lower hip fracture risk, a decrease in use of long-acting sedatives leading to a lower risk of falls, and a better general state of health within the older population.
In our study, low-energy trauma such as a fall onto a flat surface in the home, yard or on the street, was the cause of hip fracture in 65.6% of all the hospitalised patients; 16.0% in the under 50s age group and 70.3% in older patients. Traffic accidents were the cause of hip fractures in almost two thirds of all patients under 50 years. This is similar to the findings from all the literature, except that from the Far East, where, bicycle accidents were the most common cause of hip fracture [8, 22, 29, 30].
The common risk factors for hip fractures can be separated into three interacting components: the bone density, trauma risk, and the effectiveness of the neuromuscular response in defending the skeletal system. Cooper et al.  showed that a reduced muscle mass is a significant independent risk factor for hip fractures, and in people over 75 years of age osteoporosis may be less important than an adequate protective neuromuscular response. Cummings et al.  also stressed the importance of other risk factors for hip fracture aside from osteoporosis.
Bearing in mind that this study has shown an increased risk of hip fracture in the elderly and especially the female population, the battle against these risk factors is of the utmost importance. Recommendations for osteoporosis prevention and improvements in bone density, such as the promotion of physical activity , improvement in dietary habits, reduction in the consumption of alcohol and cessation of smoking, in addition to anti-osteoporotic drug therapies, are therefore very much encouraged, for the benefit of the individual as well as for reducing the high health care costs that hip fracture treatment incurs.
We accept that our data set may not have been complete, as it is based on inpatient hospital information, hence, those patients with hip fractures not admitted to hospital and not treated in the orthopaedic units, may have been missed. However, we feel that this represents only a very small number of patients. We are also aware that there are potential errors related to information obtained from hospital discharge databases, such as possible incorrect coding, resulting in under-reporting; or “double-counting” of patients who are re-admitted, resulting in over-reporting. However, we feel that these errors are largely overcome by virtue that our data set includes all the available data in the region, and not from a single institution, and is part of the national policy-forming data collection system in place in Serbia Montenegro.