Osteoporosis International

, Volume 16, Issue 12, pp 2025–2030

Incidence rates and life-time risk of hip fractures in Mexicans over 50 years of age: a population-based study

Authors

    • Clinical Epidemiology UnitCMN Siglo XXI IMSS-Faculty of Medicine
    • University of California at San Diego, School of MedicineDepartment of Family and Preventive Medicine
  • Pilar Lavielle
    • Clinical Epidemiology UnitCMN Siglo XXI IMSS-Faculty of Medicine
  • Francisco Franco-Marina
    • Faculty of MedicineUNAM
  • Esperanza Ramírez
    • Centro Nacional de Rehabilitación SS
  • Jorge Salmerón
    • Epidemiology and Health Services Research Unit
  • John A. Kanis
    • WHO Collaborating Center for Metabolic Bone DiseasesUniversity of Sheffield Medical School
  • Steven R. Cummings
    • Coordinating Center UCSF
Original Article

DOI: 10.1007/s00198-005-1991-4

Cite this article as:
Clark, P., Lavielle, P., Franco-Marina, F. et al. Osteoporos Int (2005) 16: 2025. doi:10.1007/s00198-005-1991-4

Abstract

The vast majority of hip fractures in the 21st century will occur in the developing countries. The rates and life-time hip fracture risk are not known for Mexico, and for this reason, we studied the incidence of hip fractures, and the remaining life-time probability of having a hip fracture at the age of 50 years in Mexican men and women. All hip fracture cases registered during the year 2000 were collected at all the main tertiary-care hospitals in the two major health systems in México City, Instituto Mexicano del Seguro Social (IMSS) and Ministry of Health (SS), and the diagnosis was validated by chart review in all cases. The annual rates of hip fracture were 169 in women and 98 in men per 100,000 person-years. The life-time probability of having a hip fracture at 50 years of age was 8.5% in Mexican women and 3.8% in Mexican men. We conclude that hip fractures are an important health problem in Mexico and that Mexican health authorities should consider public health programs to prevent hip fractures.

Keywords

EpidemiologyHip fracturesLife-time riskMéxicoPopulation studiesPrevalence

Introduction

Hip fractures are the most serious consequence of osteoporosis. These types of fracture are associated with very high morbidity and mortality rates. Hip fractures are the international barometer of the impact of osteoporosis because of their severe socioeconomic costs and because they are more easily documented than other fractures and can be used to compare rates in different countries [1]. Worldwide projections for the number of hip fractures indicate an increase from 1.2 million in the 1990s to 2.6 million by the year 2025 and to 4.5 million by the year 2050, assuming no change in the age- and sex-specific incidence. These figures could be higher (range of 7.3–21.3 million) if small secular trends occur. Most of these fractures will occur in the developing world [2]. The two main factors contributing to the burden of hip fractures are the increasing population over 65 years of age, and the increase in life expectancy in most populations.

Information about osteoporosis and fractures in México is scarce; according to the World Health Organization (WHO), more than 1 million Mexican women could have osteoporosis [3], representing around 15% of women aged 50 years or more. This figure is similar from a recent study (LAVOS) in which 16% of a random sample of community-dwelling women over 50 years had osteoporosis based on hip bone mineral density (BMD) (approximately 1.5 million) [4]. Reports from the Mexican Health Ministry Epidemiology Surveillance System (SS) in México show that fractures in individuals over 60 years old represented the fourth-most frequent discharge diagnosis between the years 1993 and 2001 (9.9% among 606,104 hospitalized cases) [5,6]. An incidence rate of 160/100,000 fractures per year for individuals aged 50 years and older has been reported at the tertiary-level hospitals of the Mexican Institute of Social Security (IMSS) in 1988 [7]. Population incidence rates of hip fracture have never been determined for México. The objectives of the present study were to determine rates of hip fracture and to estimate life-time risk probability for hip fractures in Mexicans over 50 years of age.

Methods

Setting

The two largest public health care systems within México are the Mexican Institute of Social Security (IMSS) and the Ministry of Health (SS). The IMSS coverage includes all formally employed workers and their families, and delivers health care to nearly 50 million Mexicans (approximately half of the population); the SS delivers health to uninsured people through health care facilities owned and operated by them directly and delivers care to nearly 48 million Mexicans [8]. For the purpose of this study, we included all hospitals with surgical facilities to perform hip surgery from the Mexican Institute of Social Security and the Ministry of Health located in the México City area. Eleven hospitals fulfilled these criteria and nine were selected for the study. One hospital was closed throughout the year 2000. Another hospital did not participate in the study, but usually referred its cases to the other hospitals. It has been estimated that approximately 2% of the population receive health care through private systems [7].

Procedure

From January 1 to December 31, 2000, we identified all hip fracture cases in individuals 50 years and older registered at the emergency rooms and surgical departments of all nine selected hospitals. Charts were retrieved and age, sex, type of fracture, and place of residence were extracted in all cases. The initial diagnoses were verified by surgical reports and/or X-rays by two of the investigators of the study (P.L. and E.R.). Individuals whose residence was not in the México City area were excluded. The following fractures using the ICD-10 classification were included: S72.0 Femoral neck fracture, S72.1 Peritrochanteric fracture (intertrochanteric fracture and trochanteric fracture), and S72.2 Subtrochanteric fracture.

Analysis

Age- and sex-specific hip fracture incidence rates per 100,000 person-years were calculated and presented in 10-year intervals along with their confidence intervals. The numerator was defined as the number of hip fractures registered and confirmed by medical chart reviewing in the hospitals selected for the study. The denominator was obtained from a 10% random sample of the 2000 Mexican Population Census, considering only the population in México City that reported IMSS and Health Ministry health care facilities as their regular source of health care using the sex and age distribution of individuals 50 years and over. To be able to compare our results with other studies, our overall incidence rates were age-adjusted through the direct method, using the 1990 non-Hispanic white US population as standard [9].

For estimating the life-time risk of having a hip fracture in women and men aged 50 years or older, we calculated age- and sex-specific death rates for the whole country, using official death counts and denominators [10]. Life-time risk probabilities were computed from the hazard functions of hip fracture and death using the method of Kanis and colleagues [11].

Results

A total of 1,137 cases of hip fracture were registered in the year 2000 from the selected hospitals, of which 1,067 cases were reviewed (70 charts [6.1%] were not available) including 748 women (70.1%) and 319 men (29.9%) (F/M Ratio 2.3:1)

The annual incidence rates of hip fracture in México City were 169/100,000 person-years for women and 98/100,000 person-years for men. Age- and sex-specific incidence rates of hip fracture per decade are presented in Table 1.
Table 1

Sex- and age-specific incidence rates of hip fractures in México City, 2000*( CI confidence interval)

Age group (years)

Total number of cases

Population (thousands)

Annual rate (per 100,000 person-years)

F/M Radio

Men

Women

Men

Woman

Men (95% CI)

Woman (95% CI)

50–59

29

38

145.99

192.51

20 (13–30)

20 (14–28)

1.1

60–69

52

105

99.70

138.61

52 (36–73)

76 (58–98)

1.46

70–79

95

236

58.62

79.92

162 (115–227)

295 (228–385)

1.80

80+

143

369

20.77

32.44

688 (429–1134)

1,137 (803–1711)

1.65

Total

319

748

325.08

443.48

98 (83–115)

169 (149–190)

-

Hip fracture rates were similar in men and women 50–59 years of age, but after this age, incidence rates were consistently higher in women. Hip fracture rates in both sexes increased in an exponential pattern from 20 per 100,000 for both sexes from ages 50–59 years, to 688/100,000 and 1,137/100,000 in men and women, respectively, in the 80-year-and-older group. The types of hip fractures found were 63% transtrochanteric, 33% cervical and 4% subtrochanteric.

The life-time risk of sustaining a hip fracture at the age of 50 years in Mexicans was 8.5% in women and 3.8% in men.

Discussion

This is the first population-based study of the hip fracture incidence rates and the life-time risk probability of having a hip fracture after 50 years of age in the Mexican population in both sexes. Our results indicate that, for 50-year-old Mexican women living in Mexico City, about one of every 12 will suffer a hip fracture during their remaining life time. Thus, hip fracture is one of the most common and important threats to the health and independence of urban Mexican women. The risks of hip fracture are about half as great in men, but the 4% life-time risk indicates that hip fracture is also a common and important condition in Mexican men.

Rates of hip fracture in Mexico are lower than those reported from population-based studies in Caucasian populations, but slightly higher than rates reported for Hispanics and slightly lower than those reported for Asians living in the USA (Table 2). Mexican rates are lower than those found in northern European countries (Sweden and Norway) but similar to rates observed in southern European countries such as Spain (in the latter case, this is more apparent in subjects 70 years and older). The rates we observed in Mexico are intermediate between the rates reported from the highly developed and westernized regions of Asia, Hong Kong and Singapore and rates in developing regions of Malaysia and Beijing.
Table 2

Age- and sex-specific incidence rates (per 100,000) of hip fracture

Geographic area, years of survey, (reference number)

Women, age (years)

Men, age (years)

Age-standardized rate (per 100,000)

50–59

60–69

70–79

80+

50–59

60–69

70–79

80+

Women

Men

Whites, Olmsted, Minnesota, USA 1989–91 [22]

74

193

568

2,163

36

135

347

1,049

263

142

Blacks, California, USA, 1983–84 [23]

35

80

270

990

46

84

190

816

219*

88*

Asians, California, USA, 1983–84 [23]

17

90

320

1,930

16

49

155

739

338*

104*

Hispanics, California, USA, 1983–84 [23]

16

60

250

960

15

34

150

600

197*

180*

Central Norway, 1983–84 [24]

213

513

1,611

5,689

67

346

867

3,234

1,293*

551*

Malmo, Sweden, 2002 [25]

58

252

930

2,928

86

136

462

1605

-

-

Former West Germany, 1996 [26]

34

106

472

2,006

38

83

239

987

-

-

Former East Germany, 1996 [26]

26

104

422

2,799

39

53

199

856

-

-

Spain, Medos, 1994 [27]

17

74

370

1,960

13

49

203

880

-

-

Australia, 1994–96[28]

25

100

560

2,823

40

55

300

1,666

-

-

Hong Kong, 1997–1998 [29]

24

119

589

2,285

23

71

296

1,285

459

180

Singapore, 1997–1998 [29]

24

138

408

1,369

28

73

210

611

442

164

Thailand, 1997–1998 [29]

34

118

509

741

29

57

185

574

269

114

Malaysia 1997–1998 [29]

18

73

230

644

17

48

96

320

218

88

Beijing, 1996 [30]

65

187

407

628

69

229

380

798

87

96

Argentina, La Plata, 1989–90 [31]

11

102

621

2,807

28

40

207

796

-

-

México City, 2000

20

76

295

1,137

20

52

162

688

203

108

*Data taken from Maggi and colleagues study on cross-national comparisons of fracture rates with standardized figures with 1990 US non-Hispanic population [9]

The Mexican age-adjusted standardized rates with non-Hispanic US population and comparison with some other countries are presented also in Table 2. The trends follow the same pattern as described above.

We found that overall rates in women were about twice those for men; these differences were more pronounced when looking at the age of 70 years and older. The rates increased in an exponential pattern with age in both sexes (Fig. 1.) Our results are consistent with patterns described for hip fractures in most populations over the world where there is a female predominance, with some exceptions such as Beijing, China, and Turkey [1214], and the rates also increase exponentially with age [2,15].
Fig. 1

Hip fractures in Mexicans

The life-time risk of developing a hip fracture in Mexicans 50 years and older is 8.5% in women and 3.8% in men. These results show that Mexicans have a similar life-time risk of hip fracture to the risk found in Hong Kong in Asia, slightly lower than that observed in Spain, Portugal or France, and higher than the risk reported in countries such as Turkey and mainland China, where similar methods have been used to estimate this risk (Table 3).
Table 3

Life-time risk of hip fracture in Mexican population compared with selected countries

Country

Life-time risk at 50 years %

Women

Men

Sweden*

28.5

13.1

USA*

15.8

6

UK*

14

4.8

France*

12.7

3.6

Spain*

12

4.2

Turkey*

1

1.8

China (Hong Kong)*

8.8

4.1

China*

2.4

1.9

Portugal*

10.1

3.6

México

8.5

3.8

Data taken from the study of Kanis 2002 [11]

Many hypotheses have been proposed to explain the wide variation in hip fracture incidence rates and life-time risk found among different countries. Such heterogeneity can be explained mainly by two factors: First, the variation in the rates might reflect true differences between populations (race and genetics) and in their exposure to different risk factors. Risk factors include geographical latitude, sun exposure, Vitamin D deficiency, bone strength and bone geometry, fragility, risk of falling, prenatal nutrition, all of which could affect the risk of fractures in later life. Life-style factors such as physical activity, diet and smoking might also contribute [1,15]. Ecological studies do not suggest important roles for these risk factors with the possible exception of physical activity on the risk of falling [11]. Second, there are some inherent limitations using discharge data and national databases that might partially explain the variation in rates: the accuracy of health statistics may differ from one country to another, and there are no estimates of the error of using these sources. Not all studies included the ICD code for definition of cases; and no validation of cases was attempted in some of the studies. Miscoding, multiple counts, and selection of study populations could also play a role in these differences [11, 13,16]. Nevertheless, there are greater than tenfold differences in fracture incidence, which are much larger than the errors. Indeed, wide variations of this magnitude are reported from prospective studies in Europe using the same methodology [11].

We speculate that the rates of hip fracture in Mexico are similar to those found in some Asians and Hispanics residing in the USA because they share Western lifestyles, including Western diets and more sedentary occupations. Mexican rates are also similar to Spain, perhaps because of the genetic background that Mexicans share with Spaniards (Mexican mestizo is the mixture of Spaniards and the indigenous population of Mesoamerica). Mexican rates are lower than the rates observed in Northern European countries and white Americans in the USA, as expected, since these countries have always reported the highest rates. Similar trends are observed using age-adjusted hip fracture rates, and, therefore, differences in the age structure of the compared populations do not explain the mentioned trends.

This study has several strengths. We double-checked from emergency room lists at all hospitals selected in order not to lose misclassified cases at the discharge registers, and charts were retrieved to verify the diagnoses using surgery logs and/or X-rays. Also, we used comprehensive census data as our denominator and mortality rates for the same year for estimation of life-time risk probabilities. The estimation of life-time risk probabilities using the same methodology to report international variations as used by Kanis and colleagues [11] allowed us to make reasonable comparisons between our population and those reported by the same author.

This study also has limitations. We studied rates in Mexico City and it is possible that rates will be somewhat different in rural areas of Mexico and may differ from north to south, due to latitude, sun exposure or socioeconomic aspects. Our rates are based on the IMSS and SS systems, which cover over 95% of the Mexican population [8]. Although we attempted to include data from the private systems and gathered hip fracture cases from three large private hospitals in Mexico City in the same fashion as we did for the public systems, the lack of a suitable denominator for individuals who use private health care system in México precluded their inclusion in the study. However, the number of cases collected in private hospitals was less than 1% of the total numbers observed in the public systems and this would, therefore, have no significant impact on our estimates of rates.

Despite the above considerations, we believe that hip fracture rates and life-time probabilities are appropriate measures that help characterize the burden of hip fractures in a country. The estimates in the present study, when compared with similar international studies, place the Mexican population at an intermediate risk for hip fractures and at a medium risk for life-time probabilities for the year 2000. Hip fracture risk can also serve as a parameter of fragility fractures, since there appears to be a consistent relationship between hip fracture risk and the risk of other fragility fractures [17].

Developing countries are expected to have a demographic shift over the next 50 years, due to an increasing elderly population. It is expected that 75% of the burden from hip fractures will come from these countries and the remaining 25% from the developed world, in which a stabilization of hip fractures rates or even a decrease in rates has been documented [1820]. In Mexico, there were an estimated 2.6 million women aged 65 years or more in the year 2000. Whereas, the overall population is set to increase by 29% in the year 2005, the female population aged 65 years or more will increase by 270% to 7.1 million. In the year 2050, this figure will reach 21.1 million [21]. Affordable strategies for detection of populations at risk, as well as the development of strategies for primary and secondary prevention, need to be implemented urgently in order to prevent a future epidemic of fragility fractures in México and many other developing counties.

Increasing development and westernization are likely to increase the rate of hip fractures in Mexico, and the growing numbers of elderly in Mexico means that hip fractures will be a growing problem for the Mexican health care system. These results should be used to alert the general population, health institutions, and health policy decision makers in order to start planning and implementing the strategies needed to detect the population at risk and to implement primary and secondary prevention programs and allocation of resources to overcome osteoporotic fractures in México’s future.

Acknowledgements

We are grateful to Helena Johansson for the invaluable assistance estimating the life-time risk probabilities for the Mexican population and to Mrs. Susan Orlofsky for English editing. This study was supported by the University of California Institute for México and the United States (UC-Mexus) CN01/92

Copyright information

© International Osteoporosis Foundation and National Osteoporosis Foundation 2005