Introduction

Osteoporosis is an important health problem mainly because of its association with age-related fractures. Osteoporotic hip fractures represent an important cause of morbidity, mortality, and health care costs for the elderly [1, 2]. This has originated the need to establish strategies of prevention an improvement in the management of these patients.

Most studies of the incidence of hip fractures in the world show geographical important variations. The differences are fewer for vertebral fractures. The northern countries (Sweden, Norway) being most affected, in contrast to the lower incidence in southern countries, like Spain [3–5]. This has probably undervalued this problem in some of these southern countries, resulting in very few health programs that prevent hip fractures. In addition, the incidence of hip fracture increases exponentially with age, especially after age 60 year, and it seems to be generally agreed that the incidence of hip fracture will rise in many countries associated with the ageing of populations. Spain is an aging country. Nevertheless the expected increase of incidence of hip fractures in some countries may have stopped [6], which probably would relate to the efforts carried out for the prevention of the hip fractures [7].

Previous epidemiological studies in Spain have indicated that incidence of hip fractures is between 301/100,000 and 897/100,000 in adults, 65 years and older. This incidence is lower than that of other countries and with moderate region-to-region variation [8]. In a previous study Ferrandez et al. [9] indicated that the incidence of hip fractures per 100000 for the age group >50 years was 195 for women and 73 for men for the 1977–1988 period in Salamanca. The incidence in our region was one of lowest in Europe. To establish preventive strategies, it is necessary to know the present and be able to estimate future incidence. A definitive study focused on the evaluation of the estimated incidence in the future has not been conducted in our country. Therefore, the aim of this study is to determine the real incidence of hip fracture in our region in patients aged 65 years or older between 1994 to 2002, and make a prediction for the future, incidence and number of hip fractures up to the year 2015, in order to bring to attention this important health problem.

Material and methods

Spain is an aging country located in southern Europe, with a population greater than 42 million. Salamanca is a province of northwest Spain of 347,120 habitants (2002), with large variations with other regions in Spain.

In Spain, free health care is provided for all people in public hospitals. The University Hospital of Salamanca is the public hospital of Salamanca that attends to 98% of the population.

Data of fractures

Using the hospital discharge register (100% codification) all patients 65 years or older, with the International Classification of Diseases, version 9 (ICD-9), code-class 820.X (hip fracture) from January 1, 1994 through December 31, 2002 have been identified. Data concerning age, sex, type and mechanism of fracture, day of admission, primary and secondaries diagnoses were collected. The cases owing to tumour processes and caused by high energy trauma have been excluded. The calculation of incidence is taken from the base population offered by the National Institute of Statistics (INE). The incidence data are expressed in the number of cases per 100,000 inhabitants.

Statistical analysis

A descriptive study and an estimate of the effect have been carried out, as well as its development in the period. Hip fracture incidences were calculated for both sexes and were expressed as the number of cases per 100,000 people, by gender and age group. We have performed a direct standardization on the basis of the total population of the study period (1994–2002). An estimate of the effect up to 2015 has been done with a model of simple lineal regression. It was thought the time limit in the year 2015 to get a less misleading estimate.

The model of regression used is

$$Y = \beta _{0} + \beta _{1} X$$
Y :

Incidence

X :

Time in years. (X = 0, 1, 2, 3, 4, 5, 6, 7, 8)

The SPSS 11.0 version, STATGRAPHICS 5.0 and SYSTAT 10.0 packages were used for the statistical analysis.

Results

In people 65 years and older attending the University Hospital of Salamanca, the total number of hip fractures was 2,726, between 1994–2002. The female/male ratio has been constantly maintained during these years (4/1), and a relative risk of 3 (incidence in people 65 years and older). The average age has also been maintained within a tight range with an average of 83.2 (SD 7.4) years of age (Table 1).

Table 1 Analysis of variance of age
Full size table

The hip fracture relative increment has been 84%, and the average increase of 9% per year. The increase was observed in both sex (Table 2 and Fig. 1), in women and in men, in the year 2002 almost double the fractures observed in the year 1994. This relative increase was 81% for women and 98% for men (Tables 3 and 8) (Figs. 4 and 5).

Fig. 1
figure 1

Incidence of hip fractures in Salamanca in women (triangles) and in men (squares)

Full size image
Table 2 Distribution of fractures for year and gender
Full size table
Table 3 Incidence in 5-year age bands, for men
Full size table

The effect has been produced by a notable increase in the incidence, surpassing 315/100,000 (95% CI: 261–408), 65 years of age and older in 1994 up to 496/100,000 (95% CI: 388–535) inhabitants 65 years old and up in the year 2002. This increase is not related with the population ageing. During the period 1994–2002, the elderly population 65 years of age and older had a relative increase of 17% while the relative increase of incidence of hip fractures was 57%, (Table 4) (Fig. 2). The incidence found after the standardization shows itself in the Table 5.

Fig. 2
figure 2

Incidence and hip fractures in Salamanca

Full size image
Table 4 Incidence of hip fractures for year
Full size table
Table 5 Standardized incidence rate by year and gender
Full size table

The result of the simple regression function is shown in Table 6 and Fig. 3. This analysis indicated that prediction of the number of fractures for the year 2015 will reach an incidence of 668/100,000 (95% CI: 516-819) (Fig. 3). These increase represent an enhancement of more than 35% respect to the incidence of hip fracture in 2002.

Fig. 3
figure 3

Observed and estimated incidence, period 1994–2015

Full size image
Table 6 Incidence estimation of hip fracture (1994–2015)
Full size table

Discussion

Hip fractures in the elderly population are one of the most important health problems in this age group [1, 2, 10]. There are numerous studies that emphasize the importance of this pathology, and the exponential increase that is produced [11], though some recent studies have also shown a slowlevelling off of the increase within various countries either due to lessening in global incidence or statistical variance for one of the two sexes [6, 7, 12].

Various studies dealing with the incidence of hip fractures in our country are available [13, 14]. Serra et al. [8] have presented a study on the incidence of these types of fractures for the entire Spanish country, from data furnished by the central government (Ministry of Consumer and Health), thus contributing data on the incidence in middle-aged citizens, within each regions that forms Spain. Their data referring to our region differs from that found by our studies on the incidence of fractures. Here they are reported an the adjusted rate of 518/100,000 for the period 1996–1999. In our study, the maximum incidence was reached in the year 2002 with 495/100,000 65 years of age and older. This discrepancy could be explained by the different way data were collected. The data recounted by Serra et al.were obtained from the information sent by the hospitals to the central government. There are differences as for the codification between every hospital. On the other hand, we do not know if fractures due to tumor processes or to mechanisms of high energy were excluded. In the case of the middle-aged group, Serra et al. found evidence for all of Spain at 82 years of age, while in our study it corresponds to 83 years of age. Recently a study on osteoporotic hip fractures in Spain has been available [15]. The hip fracture rate was 694/100,000 in those 60 years and older. The information was obtained at 77 hospitals. The dates were requested by the hospitals’ departments of orthopedics. There is no information about the origin of the information. This work presents the effect only for the year 2002.

Other studies on the incidence of hip fractures in various regions of Spain are available, offering general results, for brief periods of time [13, 14, 16] (Table 7). In agreement with the study by Serra et al. [8], these studies show differences in the incidence, the highest being in northern regions like Barcelona and lower in southern regions like the Canary Islands. This distribution is similar to what is happening in the rest of the world.

Table 7 Studies of epidemiology of hip fractures in Spain
Full size table

Our study clearly shows an important increment in the incidence of hip fractures in the elderly in Salamanca. This increment is more accentuated if we also consider the published data of Fernandez et al. [9] from our region between 1977–1988. These authors found a total of 1,908 cases of fracture in the elderly (50 years and older), but the incidence of hip fracture in subjects aged under 65 were very low. In our study for the 1994–2002 period, we found 2,726 fracture cases, moreover in the elderly of 65 years of age and older. This supposes to provide data since 1977. From 1977 to 1994 the incidence of hip fractures increased 23%. From 1977 to 2002 the increase in the incidence in older people (65 years and older) was 93%. What this implies is that the increase was dramatically high. The increase is minor when we consider the standardized effect.

The increase in hip fracture cases is greater than expected due to an aging population, suggesting the existence of other factors influencing this higher incidence. What is more, we must assume a progressive increase in the following years with an expectation for the year 2015 of 668/100,000 (95% CI 516–819) in the elderly population (Figs. 4 and 5, Table 8).

Fig. 4
figure 4

Incidence of hip fractures in Salamanca in women by age groups

Full size image
Fig. 5
figure 5

Incidence of hip fractures in Salamanca in men by age groups

Full size image
Table 8 Incidence in 5-year age bands, for women
Full size table

With respect to other European countries, the incidence of hip fractures in our own region is situated at lower levels than those found in Scandinavian countries [17]. In Belgium, a country situated at a latitude closer to Spain, they have found a substantial increase in the number of fracture cases, achieving an incidence of 108/100,000 in 1984, and 141/100,000 in 1996 [18]. In the case of Greece [19], a country much more similar to Spain and in the south of Europe, the data of fracture incidence found in 1992 is very similar to that found in Spain. The average increase per year (9%), was closer than increased observed in Finland (10%) [20], and bigger than the increase reported in other countries [19, 21]. The same way, the estimation of incidence in the future (668/100,000) is similar than the estimation observed by Kannus et al. [20] for Finland (523/100,000).

The geographic differences in which the varied incidence of osteoporosis fracture cases are concerned are very well known. In whatever case this research study does not hope to establish the differences concerning the incidence of hip fractures, but more to the point shows these above-mentioned fractures as an important and growing health problem for our country, and that in our region a dramatic increase in hip fracture cases in the past decade has occurred. This new trend will likely continue to increase.

These increases in the incidence of hip fracture in the elderly population should and must motivate health authorities in our country, to promote the establishment of innovative and effective measures for the prevention of these geriatric concerns, with prevention programs and quality treatment for osteoporosis, as well as programs to help prevent falling and its subsequent consequences for the elderly. Together with these preventative healthcare measures, in agreement with Herrera et al., we should try to improve the health-watch system of the elderly and to establish useful clinical practice guidelines in the hospitals and extra hospital care programs that will lead to the obtaining of the best results in social and health terms.

Conclusions

Our study indicates the existence of a very dramatic increase in the incidence of hip fracture in the province of Salamanca, Spain. This increase cannot only be explained by the aging of the 65-year old elderly population. This information should alert the health authorities to initiate plans of prevention and management of these very debilitating injuries.