The prevalence of vertebral fractures in San Juan, Puerto Rico: a population-based study among females aged 50 years and over.
Abstract: Objective: The participation of Puerto Rico in the Latin American Vertebral Osteoporosis Study (LAVOS) has allowed us to study the magnitude of the problem of osteoporosis in the female population residing in San Juan. The objective of LAVOS was to estimate the prevalence of vertebral fractures in Latin American women using a random sample of females selected from cities across five Latin American countries.

Methods: A probability cluster design was employed to select a random sample of households in San Juan, Puerto Rico, in order to interview 400 females aged 50 years and over. A 30-minute face-to-face interview gathered data on demographics, risk factors, and life styles. Data regarding bone mineral densities of the spine and hip by DXA and lateral dorsolumbar X-rays were obtained using international protocols; digital morphometry was used to determine vertebral deformities, with the results being classified according to Eastell criteria. This article summarizes the main findings observed in San Juan, Puerto Rico.

Results: Overall weighted prevalence of vertebral fractures was 11.2% (95% CI: 8.5%, 14.7%). Age-specific prevalence of vertebral fractures was as follows: 5.4% (95% CI: 2.7%, 10.7%) in the 50-59 years age group, 8.3% (95% CI: 4.4%, 15.1%) in the 60-69 years group, 16.2% (95% CI: 10.5%, 25.0%) in the 70-79 years group, and 22.4% (95% CI: 13.3%, 35.1%) in participants [greater than or equal to] 80 years. Factors significantly associated with vertebral fractures were being 70-79 years old (adjusted [POR.sub.70-79 vs. 50-59] = 2.9; 95% CI: 1.1, 7.4), being [greater than or equal to] 80 years old (adjusted [POR.sub.[greater than or equal to] 80 vs. 50 59] = 3 . 3; 95% CI: 1.2, 9.4), and a T-score [less than or equal to]-2.5 in the lumbar spine (adjusted POR = 2.5; 95% CI: 1.5, 5.7) and in the femoral neck (adjusted POR = 3.5; 95% CI: 1.5, 8.0). Personal history of fractures was marginally associated with vertebral fractures after adjusting for the remaining risk factors (adjusted POR = 2.0; 95% CI: 1.0, 4.0). Nearly 94% of the women with vertebral fractures were not aware of their bone status. Using the WHO classification, we determined that 71% had osteoporosis, 21%, osteopenia, and 8%, normal bone mineral densities. Bone mineral densities in the lumbar spine and femoral neck were significantly lower in women with vertebral fractures. Non-vertebral fractures were more common in women with a personal history of prior vertebral fractures. The prevalence of the most common non-vertebral fractures were hip, 1% (95% CI: 0.4%, 2.7%), and wrist, 5.9% (95% CI: 4.0%, 8.7%), respectively.

Conclusion: This is the first population-based study of osteoporotic fractures in Puerto Rico and should serve as a guide to health providers and policy makers in the prevention and treatment of this disease. [P R Health Sci J 2010;4:377-384]

Key words: Vertebral fracture, Non-vertebral fracture, Risk factors, Osteoporosis, Bone mineral density, LAVOS, Puerto Rico

Objetivo: La participation de Puerto Rico en el Estudio Latinoamericano de Osteoporosis Vertebral (LAVOS, por sus siglas en ingles) permitio estimar la magnitud del problema de osteoporosis en mujeres puertorriquenas residentes de San Juan. El objetivo de LAVOS fue estimar la prevalencia de fracturas vertebrales en una muestra de mujeres latinoamericanas de ciudades de cinco paises latinoamericanos. Metodos: Se utilizo un diseno de muestreo probabilistico por conglomerado para obtener una muestra representativa de viviendas en San Juan, Puerto Rico con el objetivo de entrevistar 400 mujeres de 50 anos o mas. Las participantes completaron una entrevista personal, una radiografia lateral, una densitometria osea y pruebas de laboratorio. Los datos sobre las densidades de la espina lumbar y de la cadera se obtuvieron mediante una densitometria osea, y las radiografias dorsolumbares laterales se realizaron utilizando los protocolos internacionales; la morfometria digital se utilizo para determinar las deformidades vertebrales, cuyos resultados se clasificaron utilizando el criterio de Eastell. Resultados: La prevalencia ajustada para fracturas vertebrales fue 11.2% (IC 95%: 8.5%, 14.7%). Resultados: La prevalencia especifica por edad fue: 5.4% (IC 95%: 2.7%, 10.7%) en el grupo de 50-59 anos de edad, 8.3% (IC 95%: 4.4%, 15.1%) en el grupo de 60-69 anos de edad, 16.2% (IC 95%: 10.5%, 25.0%) en el grupo de 70-79 anos de edad, y 22.4% (IC 95%: 13.3%, 35.1%) en participantes de 80 anos o mas. Los factores asociados significativamente con las fracturas vertebrales fueron tener entre 70 y 79 anos de edad ([POR ajustado.sub.70-79 vs 50-59] = 2.9; IC 95%: 1.1, 7.4), tener 80 anos o mas ([POR ajustado.sub.[greater than or equal to]80 vs. 50-59] = 3.3; IC 95%: 1.2, 9.4), y tener una puntuacion T [less than or equal to]-2.5 en la espina lumbar (POR ajustado = 2.5; IC 95%: 1.5, 5.7) y en el cuello femoral (POR ajustado = 3.5; IC 95%: 1.5, 8.0). El historial personal de fracturas se asocio marginalmente con fracturas vertebrales (POR ajustado = 2.0; IC 95%: 1.0, 4.0). Utilizando la clasificacion de la Organization de la Salud, 71% de las participantes tenia osteoporosis, 21% tenia osteopenia y 8% tenia una densidad osea normal. Las densidades de la espina lumbar y del cuello femoral fueron marginalmente mas bajas en mujeres con fracturas vertebrales. Las fracturas no vertebrales fueron mas comunes en mujeres con un historial previo de fracturas vertebrales. Conclusion: Este es el primer estudio poblacional de fracturas osteoporoticas realizado en Puerto Rico y podria utilizarse como una guia para los proveedores de salud y para el gobierno en la prevention y tratamiento de esta enfermedad.
Article Type: Report
Subject: Bones (Density)
Prevalence studies (Epidemiology)
Authors: Haddock, Lillian
Perez, Cynthia M.
Marrero, Edmir
Clark, Patricia
Cons-Molina, Fidencio
Ragi, Sergio
Suarez, Erick
Pub Date: 12/01/2010
Publication: Name: Puerto Rico Health Sciences Journal Publisher: Universidad de Puerto Rico, Recinto de Ciencias Medicas Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2010 Universidad de Puerto Rico, Recinto de Ciencias Medicas ISSN: 0738-0658
Issue: Date: Dec, 2010 Source Volume: 29 Source Issue: 4
Accession Number: 244280050
Full Text: Epidemiologic studies regarding the prevalence of vertebral or other types of fractures among Latin American women are largely lacking. Haddock and colleagues described the reference values for the bone mineral densities of the lumbar spine and the proximal femur in a sample of 131 healthy Puerto Rican females aged 20 to 69 years (1). The prevalences of osteoporosis and osteopenia in a subsample of 59 women aged 50 to 69 years in which the only risk factor was menopause were 12% and 56%, respectively (2).

Vertebral fractures are the most common osteoporotic fractures and are considered to be the hallmark of osteoporosis. Only about one-third come to medical attention, and in the evaluation of osteoporosis, seldom do physicians order radiographic examination of the thoracolumbar spine (3,4). In addition, there is a failure to recognize radiographic vertebral fractures in a clinical setting (5-6). Their presence predicts future vertebral and non-vertebral fractures (3-4, 7-13). Though emphasis has been given in the world literature to hip fractures because of their high morbidity and mortality (14-19), vertebral fractures are associated with back pain, deformities of the spine, emotional and physical disabilities, impaired quality of life, and increased mortality (20-23). So far no population-based studies of vertebral fractures have been available in Latin America. The need to assess the burden of osteoporosis in the Latin American community was broached in Mexico during a meeting of the Latin American Vertebral Osteoporosis Study (LAVOS) in which members from five Latin American cities (Puebla, Mexico, Buenos Aires, Argentina, Vitoria, Brazil, Barranquilla, Colombia, and San Juan, Puerto Rico) participated (24).

The primary objective of LAVOS was to determine the prevalence of morphometric vertebral fractures in a female population aged 50 years and over in a number of urban cities in Latin America and evaluate the strength of its association with conventional risk factors for fractures and osteoporosis. Secondary objectives included ascertaining both the prevalence of non-vertebral fractures in and the average bone mineral density of women 50 years and over in several Latin American cities. We used morphometry to classify the vertebral deformities in order to compare our findings with those of the Study of Osteoporotic Fractures (SOF) and Beijing's Osteoporosis Project (8, 25).

The aim of this publication is to report the prevalence estimate of morphometric vertebral fractures and their statistical association with conventional risk factors among females living in San Juan, Puerto Rico, using the framework of LAVOS. Prevalence of hip and wrist fractures as well as the mean bone mineral densities in the groups with and without vertebral fractures are also summarized.

Material and methods

Target population

The target population was composed of non-institutionalized females aged 50 years and older residing in San Juan, Puerto Rico. In order to estimate the prevalence of vertebral fractures in this population, a complex sampling design for households was defined. This design included eight strata, formed by four age groups (50-59, 60-69, 70-79, [greater than or equal to] 80 years), and two levels of socioeconomic status (based on median household value according to the Census 2000): above the median and at or below the median. The minimum sample size was estimated to be 400 females distributed equally in each stratum. To determine the overall minimum sample size, the following parameters were assumed: an expected prevalence of morphometric vertebral fracture of 15%, a desired precision of 3.5%, and a 95% confidence level. The sampling frame was determined using the maps of Census tracts of the municipality of San Juan provided by the Puerto Rico Planning Board (26). Taking into consideration this information, the sampling procedure was based on a cluster design (27) where the primary units were the block groups, defined by blocks of households in a specific geographical area.

The first step was the systematic selection of block groups by socioeconomic status and median age. Once the block groups were selected, a random selection of blocks was made in each selected group. Using this method, 50 blocks were selected, ensuring a balanced representation of the two levels of socioeconomic status. For each of the selected blocks, a visit was made to verify the actual number of households. Then, in the selected blocks, eight females were selected among the adult residents, ensuring that the four age groups would be equally represented. In order to attain this equal representation, the number of persons to be interviewed in each selected block was based on the goal of acquiring two females per age group as participants. If more than two females in each group were found, a simple random selection was made; if fewer than two females in a particular age group in a particular block were found, the number in the next block was increased to compensate. At the end of the study, we expected to have an approximately equal number of subjects in each age group.

Data collection

All interviewees were invited to come to the Medical Sciences Campus to fill in the study questionnaire and undergo the following procedures: lateral X-rays of the thoracic and lumbar spine (at the Radiology Department facilities of the Intramural Practice), blood sampling (in the Clinical Research Center), and bone densitometry (at the Endocrinology Unit of the Department of Medicine). All interview data were quality checked upon arriving at the research office. Whenever necessary, participants were recontacted to complete any items in the questionnaire that, for whatever reason, remained unfinished. Each site entered its data using a specifically designed Excel program that was provided by the Coordinating Center in Mexico, after which the files were sent to Mexico.


Face-to-face interviews were done using a standardized questionnaire that elicited demographic information, lifestyles, past and present clinical data, personal and family history of fractures, dietary intake of calcium, and quality of life. The same questionnaire was used by all the participating sites of LAVOS and was based on the questionnaires used in the SOF in the United States, the Study of Vertebral Fractures in Beijing, and the European Vertebral Osteoporosis Study (EVOS).


Lateral radiographs of the thoracic and lumbar spine were taken with a 40" tube-to-film distance, using the same protocols employed in the SOF (28) and in accord with guidelines issued by the National Osteoporosis Foundation Group on Vertebral Fractures (29). Each center sent a package with the thoracic and lumbar X-ray films properly identified for morphometry to Mexicali, Mexico.


For all study sites, morphometric measurements were done by the same center in Mexicali, Mexico. Vertebral dimensions were measured according to the procedure used in the SOF, which procedure consists of the placement of six points defining the margins of each vertebral body using a cursor and backlit digitizing board (8, 30). As part of the study, two investigators from the morphometry center in Mexicali were trained at the University of California in San Francisco, in order to be able to perform the same standardized methodology that was used in the above-mentioned studies. A random sample of 40 X-ray sets was analyzed by both centers to assure quality control after the training, and an inter-observer variation was measured. Using the same set of X-rays, a definition of normal values for morphometry of the different populations at the different sites was expected to be generated during the study.

Bone densitometry

In Puerto Rico, left hip and spine bone densitometry was performed at the Endocrinology Unit (University Hospital in San Juan, Puerto Rico) using a Hologic QDR 1000W. Each study center performed its own quality control program, and cross-calibration was done using the same spine phantom. The coefficient of variation for the spine phantom in the San Juan clinic was 0.41%. The Quality Control Center for Bone Densitometry, coordinated by Dr. Sergio Ragi in Vitoria, Brazil, periodically supervised the centers to assure a level of quality control in accordance with the Latin American Society of Densitometry Quality Control Program. The principal investigator and two technicians were certified by the International Society of Clinical Densitometry.

Biochemical measurements

Blood specimens were collected from participants in a sitting position by certified phlebotomists and shipped to a local reference laboratory accredited by the Clinical Laboratory Improvements Amendments Program. Complete blood count, chemistries (using a multichannel analyzer), and third-generation TSH levels were determined. All study procedures were approved by the Institutional Review Board (IRB) of the University of Puerto Rico, Medical Sciences Campus.

Statistical analysis

Overall and age-specific prevalences of vertebral fractures and osteoporosis were estimated using a logistic regression model, as follows: p=1/[1+exp(-([beta]o)]. The parameter estimation of this model (sso) was performed using a generalized estimating equation to control for the intraclass correlation between subjects of the same block (31). The logistic regression model was also used to estimate the crude and adjusted prevalence odds ratios. The crude prevalence odds ratios (PORC) were estimated with 95% confidence and weighted by the inverse of the probability selection of an individual, defined for each block. The adjusted prevalence odds ratios (PORA) were validated by assessing the interaction terms in the model using the likelihood-ratio test (32). Data management and all statistical analyses were performed using the Stata statistical package for Windows, release 9 (33).


A total of 515 women were invited to participate in the study, 400 (78%) of whom agreed to participate and completed all of the study procedures. Due to the difficulty in recruiting women aged 80 years and over, only 67 subjects in this age group were studied, as a result of which, the other age groups were expanded in order to reach the sample size of 400. The final age distribution of the sample was 112 participants (28%) in the 50 to 59 years age group, 109 (27.3%) in the 60 to 69 years age group, 112 (28%) in the 70 to 79 years age group, and 67 (16.7%) in the 80 years and over age group (Table 1).

Characteristics of the study group

Approximately 73% of participants had 12 or fewer years of schooling, 34.3% were married or cohabitating at the time of the interview, 82% reported an annual household income below $20,000, and the majority (95.3%) had medical insurance (Table 1). The diseases/conditions most frequently self-reported were hypertension (35.3%), osteoarthritis (23.8%), type 2 diabetes mellitus (23%), and osteoporosis (20.5%) (data not shown). Two conditions in which low bone mass can be seen, namely malabsorption and hyperthyroidism, were reported by 11 (2.8%) and 3 (0.8%) participants, respectively. When hypertension was defined as having a history of taking antihypertensive medication, the prevalence increased to 45%. The prevalence of obesity, defined as having a body mass index of at least 30.0 kg/m2, was high (40.8%), whereas that of type 2 diabetes, characterized by a fasting blood glucose of at least 126 mg/dl, was 28.8%. Around 10% of the participants self-reported cardiovascular disease, defined as having a history of angina, nitroglycerine use, coronary artery bypass grafting, myocardial infarction, stroke or peripheral vascular disease, or percutaneous transluminal coronary angioplasty.

Risk factors for fractures and osteoporosis

The prevalence of a self-reported personal history of fractures was 23.7%, and a similar prevalence was self-reported for a family history of fractures (24.3%) (Table 2). The distribution of lifestyle factors was the following: 11.8% were smokers, 25.7% had been smokers, 85.3% had never consumed an alcoholic beverage, and 5.6% consumed more than 10 grams of alcohol daily, 2% were underweight (BMI<18.5 kg/m2 ), 65.2% had never taken hormones, 9.1% currently were taking hormones, 37.9% had had an early menopause, 56.3% had a sedentary life, 18.7% had a history of falls, and 1.5% had a lifetime history of steroid use.

Prevalence of vertebral fractures

Overall prevalence of morphometric vertebral fractures was 11.2% (95% CI: 8.4%, 14.7%) when the age distribution of the population (according to the Census 2000) was considered (Table 3). Prevalence of fractures increased with age, from 5.4% (95% CI: 2.7%, 10.7%) in the 50 to 59 years age group, to 22.4% (95% CI: 13.3%, 35.1%) in the 80 years and older age group. Of the 48 subjects with morphometric vertebral fractures, 34 (70.8%) had only one fracture, and 14 (29.2%) had 2 or more fractures. Nineteen women had had an early menopause, with an average age at menopause of 39.0 [+ or -] 4.7 years. Among these, seven had taken hormones in the past, and three were currently taking hormones. Two women were on steroids, one for rheumatoid arthritis and one for systemic lupus erythematosus. Two had malabsorption, and one had cancer of the cervix and the colon without any clinical evidence of bone metastasis.

Prevalence of non-vertebral fractures

Among the study group, 90 (22.5%) reported having had at least one fracture, of which 5 had a hip fracture, 25 had had a wrist fracture, 6 had had a rib fracture, and 62 had had another type of fractures (i.e., foot, arm) (Table 4). Among women who reported having had one or more vertebral fractures, 19 reported had had fractures: 2 had had a hip fracture, 4 had had a wrist fracture, 3 had had a rib fracture, and 11 had had a finger fracture (Table 4). Among women without vertebral fractures, 70 (20%) reported having had fractures: 3 had had a hip fracture, 20 had had a wrist fracture, 3 had had a rib fracture, and 51 had had other types of fractures. The three who had sustained hip fractures were 76, 83, and 89 years at the time of the interview and had had the fractures at ages 66 (2 hip fractures, 9 months apart from each other), 60, and 87 years, respectively. All had sustained their fractures after a fall, but did not report suffering any other fractures related to that fall. The women who had suffered wrist fractures sustained those fractures anywhere from 1 to 20 years prior to the interview.

The overall weighted prevalence of non-vertebral fractures was 21.7% (95% CI: 17.9%, 26.0%) (Table 5). The weighted prevalence estimate of hip fracture and wrist fracture for the overall group was 1.1% (95% CI: 0.4%, 2.7%) and 5.9% (95% CI: 4.0%, 8.7%), respectively. Among the 80 years and older age group, the prevalence of hip fracture was 4.5% (95% CI: 1.3%, 14.2%).

Bone mineral densities

Table 6 compares the combined bone mineral densities in subjects with and without vertebral fractures in the lumbar spine and the femoral neck. Mean bone mineral densities at both sites were significantly (p<0.05) lower in the subjects with vertebral fractures. This difference in bone mineral density was found only in the 70 to 79 years age group and consisted of 0.122 g/[cm.sup.2] for the lumbar spine and 0.072 g/ [cm.sup.2] for the femoral neck (p<0.05). According to the World Health Organization classification system (using the T-score), 71% of the women had osteoporosis, 21% had osteopenia, and 8% were normal.

Risk factors for vertebral fractures

Multiple logistic regression was used to model vertebral fractures as a function of age, family history of fractures, personal history of fractures, previous history of hormonal therapy, and body mass index. A likelihood-ratio test showed no significant interaction terms (p = 0.13). Vertebral fractures were significantly associated with the 70-79 years age group (adjusted POR 70-79 vs. 50-59 = 2.9, 95% CI: 1.1, 7.4) and the [greater than or equal to] 80 years age group (adjusted POR [greater than or equal to] 80 vs. 50-59 = 3.3, 95% CI: 1.2, 9.4) and was marginally associated with personal history of fractures (adjusted POR = 2.0, 95% CI: 1.0, 4.0) (Table 7). Family history of fractures, no history of hormonal therapy, and a BMI<30.0 kg/m2 were not significantly (p>0.05) associated with vertebral fractures. Laboratory parameters such as TSH, elevated fasting glucose, alkaline phosphatase, serum creatinine, and serum calcium were not associated with vertebral fractures (p>0.05). The statistical association between vertebral fractures and the T-score in the lumbar spine and the femoral neck was also assessed (the 15 patients who still had their menses were excluded in this analysis), and a significant association was found in those patients who had a T-score of [less than or equal to]-2.5 in the lumbar spine (adjusted POR = 2.5, 95% CI: 1.5, 5.7) and femoral neck (adjusted POR = 3.5, 95%CI: 1.5, 8.0) (data not shown).


This is the first population-based study in Puerto Rico to estimate the prevalence of morphometric vertebral and non-vertebral fractures in the female population aged 50 and over living in San Juan. The overall prevalence of morphometric vertebral fractures is lower than that found in the other participating sites of LAVOS (24). However, the prevalence is similar to that reported in SOF and in Beijing although lower than that reported by other European countries (34). The genetic composition of the Puerto Rican population is an admixture of Amerindian, European and African ancestors. Among the study population, 53% classified themselves as Whites, 9.3% as Blacks and 37.8% as Hispanics, replicating the American habit of labeling as Hispanic any individual belonging to any Latin American immigrant population regardless of country of origin or ethnic background. In the 2000 Census, 80% of the population in Puerto Rico stated that their race was White. With the admixture of the various races and the progressive lightening of the skin, it is very difficult to distinguish the race to which any one individual belongs, which is why such differentiation among races has not been made in this study.

Puerto Rican women have frames similar to those of Chinese women, and it is interesting that vertebral fracture prevalence is similar in both groups (ages 50 to 79 years). In the 80 years or older age group, the prevalence is lower than that of both the Chinese (27.1%) and the Caucasian American (33.9%) populations but similar to that of the African American (17.3%) population. In Puerto Rico, 28% of the eligible women refused to participate in the study, many of whom corresponded to the elderly group. Even though the same problem is encountered in many epidemiological studies, this may have resulted in an underestimation of the true prevalence of fractures in the very elderly.

Only 2 (4.2%) of the 48 patients reported having had a vertebral fracture, indicating that 95.8% of these fractures were silent, a much higher percentage than that quoted in the literature. Sixteen had an early menopause, and only two of these subjects took hormones. Since the group without vertebral fractures was larger, a higher number of non-vertebral fractures were reported by this group; nevertheless, percentagewise there was a significant increase in non-vertebral fractures in the group with vertebral fractures. The weighted prevalence for hip fractures was 1.1%, and for wrist fractures, it was 5.9%. This is consistent with previous observations that patients with vertebral fractures are more prone to develop other fractures. As observed in other populations (4,20,25,35-39), the odds of osteoporotic fractures increased with age, low bone mineral density, and history of previous fractures (p<0.05). Cooper and colleagues (4) found that women with vertebral fractures had an earlier menopause, fewer births, and a higher prevalence of clinically diagnosed hyperthyroidism. Even though this was not the case in our study, it is clinically significant that 16 women had an early menopause. In order to elucidate the pathogenesis of fractures, a clinical history and blood samples for chemistries, complete blood cell count, and serum level of thyroid stimulating hormone were taken. No association was found with TSH, alkaline phosphatase, serum creatinine, serum calcium, liver enzymes level, or fasting hyperglycemia. Of the 48 fractures, 3 could be classified as having been due to trauma, while the rest were osteoporotic fractures. The combined bone mineral densities in the lumbar spine and the femoral neck were significantly lower in the group with vertebral fractures (p<0.05). When the World Health Organization (WHO) criteria were applied to the bone mineral densities of the group with vertebral fractures, it was determined that 71% had osteoporosis, 21% had osteopenia, and 8% had normal bone mineral densities. Thus, osteoporotic fractures may occur with normal, osteopenic, and osteoporotic bone mineral densities, as has been shown by the National Osteoporosis Risk Assessment study (40) and the SOF study (41). While in our small group of females the majority had osteoporosis, in the latter group of female populations, the majority had osteopenia. The statistical association between fractures and the T-score in the lumbar spine and femoral neck was also assessed with the logistic model. A significant association was found with a T-score of [greater than or equal to]-2.5 in the lumbar spine and femoral neck, again showing that the vast majority of the group was osteoporotic at the time of each one's individual fracture.

This study is the first to show the burden of osteoporosis in an elderly female population in San Juan, Puerto Rico and should serve as a guide to health providers and health policy makers in the prevention and treatment of this disease. In order to complete the epidemiological profile of osteoporosis in the community residing in Puerto Rico, it is recommended that this study be extended to the rest of the island due to the prevailing admixture of the population in Puerto Rico. Additional studies should also be designed to assess vertebral fractures in females over 80 years old. Such a study should have a higher number of subjects than the current study due to the fact that the members of this population are at high risk of osteoporosis and fractures.


This study was supported by Eli Lilly Export S. A. (Puerto Rico branch). The International Osteoporosis Foundation supported the central activities of the Coordinating Center in Mexico, and grants from different sources supported the other countries' surveys.


(1.) Haddock L, Ortiz V, Vazquez M, Aguilo F, Bernard E, Ayala A, Mejias N. The lumbar and femoral bone densities in a normal Puerto Rican population. P R Health Sci J 1996;15(1):5-11.

(2.) Haddock L. Prevalence of osteopenia and osteoporosis in a normal female Puerto Rican population. P R Health Sci J 1997;16(3):241-244.

(3.) Black DM, Cummings SR, Karpf DB, Cauley JA, Thompson DE, Nevitt MC, Bauer DC, Genant HK, Haskell WL, Marcus R, Ott SM, Torner JC, Quandt SA, Reiss TF, Ensrud KE. Randomized trial of effect of alendronate on risk of fracture in women existing vertebral fractures. Lancet 1996;348:1535-1541.

(4.) Cooper C, Atkinson EJ, O'Fallon WM, Melton LJ 3rd. Incidence of clinically diagnosed vertebral fractures: A population-based study in Rochester, Minnesota, 1985-1989. J Bone Miner Res 1992;7:221-227.

(5.) Gehlbach SH, Bigelow C, Hemisdottir M, May S, Walker M, Kirkwood JR. Recognition of vertebral fracture in a clinical setting. Osteoporos Int 2000;11:577-582.

(6.) Delmas PD, van de Langerijt T, Watts NB, Eastell R, Genant H, Grauer A, Cahall DL; IMPACT Study Group. Underdiagnosis of vertebral fractures is a worldwide problem: The Impact Study. J Bone Miner Res 2005;20:557-563.

(7.) Ross PD, Davis JW, Epstein RS, Wasnich RD. Pre-existing fractures and bone mass predict vertebral fracture incidence in women. Ann Intern Med 1991;114:919-923.

(8.) Black DM, Palermo L, Nevitt MC, Genant H, Epstein R, San Valentin R, Cummings SR. Comparison of methods for defining prevalent vertebral deformities: The Study of Osteoporotic Factures. J Bone Miner Res 1995;10:890-902.

(9.) Burger H, van Daele PL, Algra D, Hofman A, Grobee DE, Schutte HE, Birkenhager JC, Pols HA. Vertebral deformities as predictors of non-vertebral fractures. BMJ 1994;309:991-992.

(10.) Melton LJ 3rd, Atkinson EJ, Cooper C, O'Fallon WM, Riggs BL. Vertebral fractures predict subsequent fractures. Osteoporosis Int 1999;10:214-221.

(11.) Black DM, Arden NK, Palermo L, Pearson J, Cummings SR. Prevalent vertebral deformities predict hip fractures and new vertebral deformities but not wrist fractures. Study of Osteoporotic Fractures Research Group. J Bone Miner Res 1999;14(5):821-828.

(12.) Meyer HE, Henriksen C, Falch JA, Pedersen JI, Tverdal L. Risk factors for hip fracture in a high incidence area: A case-control study for Oslo, Norway. Osteoporos Int 1995;5:239-246.

(13.) Klotzbuecher CM, Ross PD, Landsman PB, Abbott TA, Berger M. Patients with prior fractures have an increased risk of future fractures: A summary of the literature and statistical synthesis. J Bone Miner Res 2000;15(5):721-739.

(14.) Cooper C, Campion G, Melton LJ 3rd. Hip fractures in the elderly: A worldwide projection. Osteoporosis Int 1992;2:285-289.

(15.) Lindsay R The burden of osteoporosis: cost. Am J Med 1995;98 (2A): 9S-11S.

(16.) Barrett-Connor E. The economic and human costs of osteoporotic fracture. Am J Med 1995;98(2A):3S-8S.

(17.) Cooper C, Atkinson EJ, Jacobsen SJ, O'Fallon WM, Melton LJ 3rd. Population-based study of survival after osteoporotic fracture. Am J Epiwedemiol 1993;137:1001-1005.

(18.) Poor G, Jacobsen SJ, Melton LJ. Mortality following hip fracture. In: Vellas BJ, Albarede JL, Garry PJ, eds. Facts and Research in Gerontology. Paris, Serdi; 1994:91-169.

(19.) Magaziner J, Simonsick EM, Kashner TM, Hebel JR, Kenzora JE. Predictors of functional recovery one year following hospital discharge for hip fracture: A prospective study. J Gerontol 1990;45(3):101-107.

(20.) Cook DJ, Guyatt GH, Adachi JD, et al. Quality of life issues in women with vertebral fractures due to osteoporosis. Arthritis Rheum 1993;36:50-56.

(21.) Lyles KW, Gold DT, Shipp KM, Pieper CF, Martinez S, Mulhausen PL. Association of osteoporotic vertebral compression fractures with impaired functional status. Am J Med 1993;94:595-601.

(22.) Ettinger B, Black DM, Nevitt MC, Rundle AC, Cauley JA, Cummings SR, Genant HK. Contribution of vertebral deformities to chronic back pain and disability. J Bone Miner Res 1992;7:449-456.

(23.) Nevitt MC, Ettinger B, Black DM, Stone K, Jamal SA, Ensrud K, Segal M , Genant HK, Cummings SR. The association of radiographically detected vertebral fractures with back pain and function: A prospective study. Ann Intern Med 1998;128(10):793-800.

(24.) Clark P, Cons-Molina F, Deleze M, Ragi S, Haddock L, Zanchette JR , Jaller JJ , Palermo L, Talavera JO, Messina DO, Morales-Torres J, Salmeron J, Navarrete A, Suarez E, Perez CM, Cummings SR. The prevalence of radiographic vertebral fractures in Latin American countries: the Latin American Vertebral Osteoporosis Study (LAVOS). Osteoporosis Int 2009;20:275-282.

(25.) Ling X, Cummings SR, Mingwei Q, Xihe Z, Xioashu C, Nevitt M, Stone K. Vertebral fractures in Beijing, China: The Beijing Osteoporosis Project. J Bone Miner Res 2000;15(10):2019-2025.

(26.) Levy PS, Lemeshow S. Sampling of Populations: Methods and Applications. 4th ed. New York, NY: John Wiley & Sons, Inc.; 2009.

(27.) Kish L. Survey Sampling. New York, NY: John Wiley & Sons, Inc.; 1967.

(28.) Black DM, Palermo L, Nevitt MC, Genant HK, Christensen L, Cummings SR. Defining incident vertebral deformity: A prospective comparison of several approaches. J Bone Miner Res 1999;14:90-101.

(29.) Kiel D. Assessing vertebral fractures. National Osteoporosis Foundation working group on vertebral fractures. J Bone Miner Res 1995;10: 518-523.

(30.) Eastell R, Cedel SL, Wahner HW, Riggs BL, Melton LJ. Classification of vertebral fractures. J Bone Miner Res 1991;6:207-215.

(31.) Snijders TAB, Bosker R. Multilevel Analysis: An introduction to basic and advanced multilevel modeling. London, England: SAGE Publications; Reprinted 2003.

(32.) Hosmer D, Lemeshow S. Applied Logistic Regression. 2nd ed. New York, NY: John Wiley and Sons, Inc.; 2000.

(33.) Statistical package Stata for Windows, release 9. StataCorp LP. College Station, TX.

(34.) O'Neill TW, Felsenberg D, Varlow J, Cooper C, Kanis JA, Silman AJ. The prevalence of vertebral deformity in European men and women: The European Vertebral Osteoporosis Study. J Bone Miner Res 1996;11: 1010-1018.

(35.) The European Prospective Osteoporosis Study (EPOS) Group. The relationship between bone density and incident vertebral fracture in men and women. J Bone Miner Res 2002;17:2214-2221.

(36.) Albrand G, Munoz F, Sornay-Rendu E, DuBoeuf F, Delmas PD. Independent predictors of all osteoporosis-related fractures in healthy postmenopausal women: The OFELY study. Bone 2003;32:78-85.

(37.) Black DM, Steinbuch M, Palermo L, Dargent-Molina P, Lindsay R, Hoseyni MS, Johnell O. An assessment tool for predicting fracture risk in postmenopausal women. Osteoporosis Int 2001;12:519-528.

(38.) Miller PD, Siris ES, Barrett-Connor E, Faulkner KG, Wehren LE, Abbott TA, Chen YT, Berger ML, Santora AC, Sherwood LM. Prediction of fracture risk in postmenopausal White women with peripheral bone densitometry: Evidence from the National Osteoporosis Risk Assessment. J Bone Miner Res 2002;17:2222-2230.

(39.) Seeley DG, Browner WS, Nevitt MC, Genant HK, Scott JC, Cummings SR. Which fractures are associated with low appendicular bone mass in elderly women? The Study of Osteoporotic Fractures Research Group. Ann Intern Med 1991;115:837-842.

(40.) Siris ES, Chen YT, Abbott TA, Barrett-Connor E, Miller PD, Wehren LE, Berger ML. Bone mineral density thresholds for pharmacological intervention to prevent fractures. Arch Intern Med 2004;164:1106-1112.

(41.) Wainwright SA, Marshall LM, Ensrud KE, Cauley JA, Black DM, Hillier TA, Hochberg MC, Vogt MT, Orwoll ES, for the Study of Osteoporotic Fractures Research Group. Hip fractures in women without Osteoporosis. J Clin Endocrinol Metab 2005;90:2787-2793.

Lillian Haddock, MD *; Cynthia M. Perez, PhD ([dagger]); Edmir Marrero, MPH ([dagger]); Patricia Clark *; Fidencio Cons-Molina, MD ([section]); Sergio Ragi, MD **; Erick Suarez, PhD ([daggger])

* University of Puerto Rico School of Medicine and ([dagger]) Graduate School of Public Health, University of Puerto Rico School of Medicine, San Juan, Puerto Rico ([double dagger]) Clinical Epidemiology Unit CMN Siglo XXI-IMSS, Faculty of Medicine UNAM, Mexico City, ([section]) Unidad de Diagnostico de Osteoporosis, Mexicali, Mexico, and ** CEDOES Diagnostico e Pesquisa, Vitoria, Espirito Santo, Brazil

Address correspondence to: Lillian Haddock, MD, School of Medicine Department of Endocrinology, University of Puerto Rico Medical Sciences Campus. Email:
Table 1. Sociodemographic characteristics of the study group
(n = 400)

Characteristics                       n     %

Age group (years)
  50-59                               112   28.0
  60-69                               109   27.3
  70-79                               112   28.0
  [greater than or equal to]80        67    16.7

Years of schooling
  0                                   14    3.5
  1-6                                 84    21.0
  7-12                                193   48.3
  >12                                 105   26.2
  Refused to answer                   4     1.0

Medical insurance
  Yes                                 381   95.3
  No                                  19    4.7

Marital status
  Never married                       34    8.5
  Married/Cohabitating                137   34.3
  Separated                           14    3.5
  Divorced                            77    19.2
  Widowed                             138   34.5

Annual household income
  <$10,000                            242   60.5
  $10,000-$19,999                     86    21.5
  [greater than or equal to]$20,000   59    14.7
  Refused to answer                   13    3.3

Table 2. Distribution of established risk factors for osteoporosis
(n = 400)

Risk Factors

Family history of fractures
  Yes                              97    24.3
  No                               292   73.0
  Do not know                      11    2.7

Personal history of fractures
  Yes                              95    23.7
  No                               205   76.3

Previous history of hormonal
therapy *
  Current                          35    9.1
  In the past                      97    25.2
  Never                            251   65.2
  Do not know                      2     0.5

  Current                          47    11.8
  In the past                      103   25.7
  Never                            250   62.5

Alcohol intake
  Never                            341   85.3
  1-10 g/day                       37    9.3
  11-40 g/day                      19    4.8
  >40 g/day                        3     0.8

Body mass index (kg/[m.sup.2])
  <18.5                            8     2.0
  18.5-24.9                        70    17.5
  25.0-29.9                        159   39.7
  [greater than or equal to]30.0   163   40.8

Early menopause *
  [less than or equal to]45 yr     146   37.9
  >45 yr                           239   62.1

  Sedentary                        225   56.3
  Active                           175   43.7

History of falls
  Yes                              75    18.7
  No                               325   81.3

Steroid therapy
  Yes                              6     1.5
  No                               394   98.5

* 15 women did not answer the question (n = 385) because they had
not yet reached menopause.

Table 3. Weighted prevalence estimate of vertebral fractures

Age group (years)        Prevalence (%)   95% CI([dagger])

50-59 (n = 111)          5.4              2.7-10.7
60-69 (n = 109)          8.3              4.4-15.1
70-79 (n = 111)          16.2             10.1-25.0
[greater than or equal   22.4             13.3-35.1
  to]80 (n = 67)
Overall (n = 398)        11.2             8.4-14.7

([dagger])Using logistic regression model weighting for the
population size according to the Census 2000

Table 4. Distribution of self-reported non-vertebral fractures

Study group           Type of non-    Number of       Percent (%)
                      vertebral       subjects (n)

Total group           Overall         90              22.5
(n = 400)             Hip             5               5.7
                      Wrist           25              27.8
                      Rib             6               6.7
                      Other           62              68.9

With vertebral        Overall         19              39.6
fractures (n = 48)    Hip             2               10.5
                      Wrist           4               21.1
                      Rib             3               15.8
                      Other           11              57.9

Without vertebral     Overall         70              20.0
fractures (n = 350)   Hip             3               4.3
                      Wrist           20              28.6
                      Rib             3               4.3
                      Other           51              72.9

Table 5. Weighted prevalence estimate of non-vertebral fractures

Variable                                    Prevalence     95% CI (a)

Overall (n = 400)                           21.7           17.9-26.0
  50-59 (n = 112)                           16.1           10.8-23.2
  60-69 (n = 109)                           19.3           12.9-27.7
  70-79 (n = 112)                           23.2           15.8-32.7
  [greater than or equal to] 80 (n = 67)    37.3           27.6-50.5
Hip (n = 400)                               1.1            0.4-2.7
  50-59 (n = 112)                           --             --
  60-69 (n = 109)                           --             --
  70-79 (n = 112)                           1.8            0.4-7.6
  [greater than or equal to] 80 (n = 67)    4.5            1.3-4.2
Wrist (n = 400)                             5.9            4.0-8.7
  50-59 (n = 112)                           4.5            2.0-9.5
  60-69 (n = 109)                           3.7            1.4-9.4
  70-79 (n = 112)                           7.1            3.4-14.3
  [greater than or equal to] 80 (n = 67)    11.9           5.7-23.3

(a) Using logistic regression model weighting for the population
size according to the Census 2000.

Table 6. Comparison of the combined bone mineral density
measurements in the lumbar spine and femoral neck in the groups
without and with vertebral fractures

Group              n        Mean (g/cm)      SD       95% CI

Lumbar Spine

No fractures       343      0.91             0.17     0.89, 0.92
Fractures          48       0.82             0.19     0.77, 0.88
Difference                  0.09                      0.03, 0.13 *

Femoral Neck

No fractures       341      0.71             0.13     0.89, 0.92
Fractures          46       0.62             0.13     0.59, 0.66
Difference                  0.09                      0.05, 0.13 *

* Two-sample t-test with equal variances (p<0.05)

Table 7. Logistic regression models for vertebral fractures and
selected risk factors (n = 386) *

Risk Factors                         [OR.sub.crude]    95% CI

Age (years)
  50-59                              1.0               --
  60-69                              1.6               0.6-4.3
  70-79                              3.2               1.3-8.2
  [greater than or equal to] 80      4.7               1.8-12.4

Family history
    of fractures
  No                                 1.0               --
  Yes                                0.9               0.4-1.8

Personal history
    of fractures
  No                                 1.0               --
  Yes                                2.4               1.2-4.7

Previous history
    of hormonal therapy
  Yes                                1.0               --
  No                                 2.0               1.0-4.2

Body mass
    index (kg/[m.sup.2])
  [greater than or equal to] 30.0    1.0               --
  <30.0                              1.4               0.7-2.8

Risk Factors                         [OR.sub.          95% CI
                                     adjusted]         ([dagger])

Age (years)
  50-59                              1.0               --
  60-69                              1.5               0.6-4.3
  70-79                              2.9               1.1-7.4
  [greater than or equal to] 80      3.3               1.2-9.4

Family history
    of fractures
  No                                 1.0               --
  Yes                                1.1               0.5-2.4

Personal history
    of fractures
  No                                 1.0               --
  Yes                                2.0               1.0-4.0

Previous history
    of hormonal therapy
  Yes                                1.0               --
  No                                 1.6               0.7-3.6

Body mass
    index (kg/[m.sup.2])
  [greater than or equal to] 30.0    1.0               --
  <30.0                              1.3               0.7-2.7

* Excluding 10 women who answered "do not know" regarding family
history of fractures, 1 woman who answered "do not know" regarding
previous history of hormonal therapy, and 1 woman who answered "do
not know" to both questions.

([dagger]) Using a logistic regression model weighting for the
population size according to the Census 2000.

([double dagger]) Each variable was adjusted for the others in a
logistic regression model.
Gale Copyright: Copyright 2010 Gale, Cengage Learning. All rights reserved.