JOURNAL OF BONE AND MINERAL RESEARCH

Volume 13, Number 12, 1998

Blackwell Science, Inc.
© 1998 American Society for Bone and Mineral Research

Bone Density and Fracture Risk in Men

L. JOSEPH MELTON, III,1 ELIZABETH J. ATKINSON,1 MICHAEL K. O’CONNOR,2

W. MICHAEL O’FALLON,1 and B. LAWRENCE RIGGS3

ABSTRACT

We evaluated different definitions of osteoporosis in a population-based sample of 348 men (age 22–90 years)
compared with 351 women (age 21–93 years). Thirty-six men (10%) and 46 women (13%) had a history of
osteoporotic fracture (hip, spine, or distal forearm due to moderate trauma at > age 35). In logistic regression
analysis, osteoporotic fracture risk was associated with bone mineral density (BMD) at all sites (neck, trochanter,
total hip, lumbar spine, and total wrist) in both genders ( p < 0.001) except spinal BMD in men. After adjusting
for age, total hip BMD was the strongest predictor of fracture risk in women (odds ratio [OR] per 1 SD decline,
2.4; 95% confidence interval [CI], 1.6–3.7), while wrist BMD was best in men (OR, 1.5; 95% CI, 1.1–2.0). Among
men but not women, bone mineral apparent density (BMAD) was a better predictor of fracture than BMD (wrist
BMAD OR, 1.7; 95% CI, 1.3–2.3). Hip BMD/BMAD decreased linearly from age 20 years onward in both genders,
while spinal BMD/BMAD declined after age 40 in women but not in men. In both genders, total wrist BMD/BMAD
decreased after age 50. By World Health Organization criteria, the age-adjusted prevalence of osteoporosis at the
hip, spine, or wrist was 35% among women >50 years of age. A similar approach (BMD > 2.5 SD below the young
male mean) produced an osteoporosis prevalence rate in men >50 years of age of 19%. Thus, bone density predicts
fracture risk in men as it does in women, and the prevalence of osteoporosis in men, using sex-specific normal
values, is substantial. These observations indicate a need for better prevention and treatment strategies for men.
(J Bone Miner Res 1998;13:1915–1923)

INTRODUCTION bilities for defining osteoporosis in men should be explored:

2.5 SD below the young normal mean for men, 3– 4 SD

I N 1994, A STUDY GROUP of the World Health Organization

(WHO) proposed that osteoporosis in white women be
operationally defined as a bone mineral density (BMD)
below the male mean, and 2.5 SD below the young normal
mean for women.(2) It was thought that the first choice, a
gender-specific cut-off level at –2.5 SD might produce a
level 2.5 SD below the mean for normal young women.(1) It number of affected men much larger than the proportion of
was recognized that this cut-off level was an arbitrary choice the male population that ultimately experiences osteopo-
given the continuous relationship between bone density and rotic fractures, estimated at 13% for the lifetime risk of a
fracture risk in women, but the definition produced a num- hip, spine, or forearm fracture.(3) This problem could be
ber of affected women that was consonant with clinical dealt with either by use of a more stringent cut-off level to
perceptions of the size of the problem.(2) It was not possible define osteoporosis in men (e.g., BMD 3– 4 SD below the
at that time to define osteoporosis in men or nonwhite mean for young men) or, alternatively, by use of the same
women due to the dearth of detailed data on the relation- absolute value of BMD for men as previously chosen to
ship between bone density and fracture risk in these other define osteoporosis for women. At present, however, it is
populations. However, it was suggested that several possi- not clear which of these might be the better choice. The

1
Department of Health Sciences Research, Mayo Clinic and Mayo Foundation, Rochester, Minnesota, U.S.A.
2
Department of Diagnostic Radiology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota, U.S.A.
3
Division of Endocrinology, Metabolism, Nutrition and Internal Medicine, Mayo Clinic and Mayo Foundation, Rochester, Minnesota,
U.S.A.

1915
1916 MELTON ET AL.

objective of the present study was, first, to determine the genograms were not available for review. Consequently, the
association between bone density assessed in various ways diagnosis of vertebral fracture was accepted on the basis of
with fracture risk in white men and then to define normal a radiologist’s report of compression, wedging, or collapse
values and patterns of age-related bone loss in order to of one or more thoracic or lumbar vertebrae. The interview
estimate the prevalence of osteoporosis in men using these and record review were completed independently of any
different definitions. knowledge of each subject’s BMD values. There was gen-
erally good agreement between interview and medical
record data,(5) but where disagreements occurred priority
MATERIALS AND METHODS was given to documented medical history; in the absence of
documentation to the contrary, the subject’s account was
Study subjects
accepted. The duration of contemporary documentation in
Following approval by the Mayo Clinic’s Institutional hand averaged 30.8 years (median, 29 years; range, 1– 81
Review Board, subjects were recruited from an age-strati- years), and ascertainment of clinically evident fractures is
fied random sample of Rochester, Minnesota residents that believed to be complete. Osteoporotic fractures were de-
was selected using the medical records linkage system of the fined as clinically recognized fractures of the hip, spine, or
Rochester Epidemiology Project.(4) Over half of the Roch- distal forearm that resulted from minimal or moderate
ester population is attended annually at the Mayo Clinic, trauma (e.g., a fall from standing height or less) among
and the majority are seen in any 3-year period. Thus, the persons 35 years of age or older.
enumerated population (Rochester women seen in 1990 6
1 year and men seen in 1991 6 1 year) approximates the Bone densitometry
underlying population of the community, including both
free-living and institutionalized individuals. Altogether, BMD in grams per square centimeter was determined for
1138 men were approached for this study, but 239 of them the lumbar spine (L2–L4 in anteroposterior projection),
were ineligible (109 were demented; 13 were radiation proximal femur (total, femoral neck, and intertrochanteric
workers; 91 died before they could be contacted; and 25 regions), and wrist (total) using dual-energy X-ray absorp-
others were so debilitated, e.g., terminal cancer, that they tiometry with the QDR2000 instrument (Hologic,
were not approached; the final ineligible man was involved Waltham, MA, U.S.A.) with software version 5.40. The
in a legal action). Of the 899 eligible men, 348 (39%) coefficients of variation for the spine, hip, and forearm
participated and provided full study data. There were ;50 BMD measurements were 0.6, 1.8, and 0.8%, respectively.
men per decade of age from 20 –29 years to age 80 years and We also estimated volumetric bone mineral apparent den-
over (mean age 6 SD, 55.4 6 19.6 years; range 22–90 sity (BMAD, g/cm3) from these data as previously de-
years), but the participation rate varied by decade (27, 36, scribed,(6) using the following formulas:
43, 50, 62, 45, and 29%, respectively, for age groups 20 –29
through 80 years and over). spine BMAD 5 BMC/A3/2
A total of 938 women aged 20 years and over were
approached for this study, but 126 were ineligible (89 were forearm and hip BMAD 5 BMC/A2
demented and could not give informed consent; 11 were where BMC is the bone mineral content and A is the
pregnant; 9 were radiation workers; 8 were participants in projected bone area. Bone density at each site was catego-
an ongoing clinical trial of osteoporosis prophylaxis; and 9 rized into three groups according to WHO criteria as nor-
died before they could be contacted). Of the 812 eligible mal, low, or osteoporotic(2) relative to young normal means
women, 351 (43%) participated and provided full study and SDs. The means (and SDs) for BMD/BMAD of the
data. There were ;50 women per decade of age from 20 –29 spine, hip, and forearm depended on the reference popu-
years to age 80 years and over. The participation rates by lation and four different ones were compared: the 50 Roch-
age group were 50, 48, 56, 65, 57, 39, and 22%, respectively. ester women 20 –29 years old, all 138 premenopausal
The subjects included 138 premenopausal women (mean women combined, the 48 Rochester men 20 –29 years old,
age, 35.0 6 8.6 years; range, 21–54 years) and 213 post- and the 148 men 20 – 49 years old (Table 1).
menopausal women (mean age, 67.8 6 13.2 years; range,
34 –93 years).
Statistical analysis
All subjects were interviewed in accordance with a stan-
dard protocol in order to collect clinical data, including a The relative risk of an osteoporotic fracture was esti-
history of fractures that was verified by review of each mated with odds ratios (OR) obtained from multiple logis-
subject’s complete (inpatient and outpatient) medical tic regression models, where osteoporotic fracture was the
records in the community. Mayo Clinic records, for exam- endpoint, while age, gender, BMD (spine, hip, wrist), and
ple, contain the details of every inpatient hospitalization at BMAD (spine, hip, wrist) were the potential predictors.
its two large affiliated hospitals in Rochester (Saint Marys Variables were selected in a stepwise fashion, entering only
and Rochester Methodist), every outpatient or office visit at variables that were significant at the 0.05 level after adjust-
the Clinic, emergency rooms or nursing homes, as well as all ing for the other variables in the model. Interactions and
radiographic reports and pathology reports, including au- higher ordered terms were investigated. Kendall’s tau-a was
topsies.(4) The records contained the clinical history and the used to compare logistic models when various bone density
radiologist’s report of each fracture, but the original roent- variables were substituted one for another. Kendall’s tau-a
FRACTURE PREDICTION IN MEN 1917

TABLE 1. YOUNG NORMAL MEANS AND SD FOR BONE MINERAL DENSITY (BMD, G/CM2) AND BONE MINERAL APPARENT
DENSITY (BMAD, G/CM3) ASSESSED AT DIFFERENT SKELETAL SITES USING DIFFERENT REFERENCE POPULATIONS OF ROCHESTER,
MINNESOTA, MEN AND WOMEN

Skeletal site

Reference population Total hip Femoral neck Trochanter AP spine Total wrist

20–29 year male criteria

mean (BMD) 1.077 0.986 0.824 1.100 0.649
SD 0.113 0.109 0.101 0.137 0.055
cut-off 0.795 0.715 0.572 0.756 0.511
mean (BMAD) 0.025 0.174 0.063 0.151 0.020
SD 0.004 0.025 0.009 0.019 0.002
cut-off 0.015 0.111 0.040 0.104 0.016
20–49 year male criteria
mean (BMD) 1.061 0.939 0.817 1.118 0.654
SD 0.140 0.133 0.127 0.156 0.050
cut-off 0.711 0.607 0.500 0.728 0.529
mean (BMAD) 0.024 0.164 0.061 0.153 0.020
SD 0.004 0.027 0.011 0.020 0.002
cut-off 0.014 0.096 0.034 0.102 0.015
20–29 year female criteria
mean (BMD) 0.959 0.903 0.735 1.084 0.549
SD 0.125 0.119 0.111 0.135 0.043
cut-off 0.646 0.606 0.458 0.745 0.441
mean (BMAD) 0.029 0.192 0.074 0.159 0.022
SD 0.004 0.029 0.011 0.018 0.002
cut-off 0.018 0.120 0.048 0.115 0.017
Premenopausal female criteria
mean (BMD) 0.941 0.865 0.723 1.097 0.558
SD 0.124 0.122 0.109 0.128 0.042
cut-off 0.632 0.561 0.450 0.777 0.453
mean (BMAD) 0.029 0.182 0.071 0.162 0.023
SD 0.005 0.032 0.012 0.018 0.002
cut-off 0.017 0.101 0.040 0.116 0.017

Cut-off levels for defining osteoporosis (2.5 SD the below young normal mean) are shown for each set of normal values.

is a rank correlation that compares the predicted probabil- trauma. There were three fractures due to a specific local
ities from a model with the observed responses (whether or pathological process (e.g., bone cyst, metastatic malig-
not an osteoporotic fracture was detected). It is a useful nancy) among the women, including one vertebral fracture
statistic for comparing models—the higher the tau-a, the and one hip fracture, and one pathological fracture among
better the model does at predicting the event. the men. Altogether, there were 89 fractures of the hip,
The prevalence of osteoporosis was determined by gen- spine, or distal forearm among the women, 63 of which
der and decade of age using different sets of normal values were due to minimal or moderate trauma, and 113 such
and summarized by direct age adjustment to the structure fractures among the men, 56 of which were due to minimal
of the white population of the United States in 1990. or moderate trauma; 46 women (13%) and 36 men (10%)
had one or more minimal or moderate trauma fractures at
these sites that occurred at age 35 years or after, thus
RESULTS meeting our definition of an “osteoporotic” fracture.
BMD was associated with the risk of an osteoporotic
At the baseline assessment, 242 (70%) of the men and fracture at all of the sites assessed (total hip, femoral neck,
186 (53%) of the women had experienced one or more trochanteric region, spine, and total wrist) among the
fractures as enumerated in Table 2. The fractures in women women (all p , 0.001) and at all sites among the men (all p
were about equally divided between those due to minimal , 0.001) except for the lumbar spine (not significant). After
or moderate trauma (e.g., a fall from standing height or adjustment for age, the ORs varied by gender and, to a
less) and those due to severe trauma (e.g., motor vehicle lesser degree, with the reference group used to establish
and recreational accidents and falls from heights). Among normal means and SDs (i.e., men aged 20 –29 years, men
men, three-fourths of the fractures were due to severe aged 20 – 49 years, women aged 20 –29 years, or all pre-
1918 MELTON ET AL.

TABLE 2. DISTRIBUTION OF FRACTURES AMONG AGE-STRATIFIED SAMPLES OF ROCHESTER, MINNESOTA, MEN AND WOMEN BY
FRACTURE SITE AND CAUSE

Men Women

Moderate Moderate
trauma Severe trauma trauma Severe trauma

Skeletal site n % n % n % n %

Skull/face 4 2.8 40 10.1 2 1.2 12 7.4

Ribs/sternum/clavicle/scapula 45 31.0 66 16.7 17 10.1 23 14.2
Proximal humerus 2 1.4 2 0.5 12 7.1 1 0.6
Shaft/distal humerus 3 2.1 4 1.0 5 3.0 4 2.5
Shaft/proximal forearm 8 5.5 12 3.0 7 4.2 8 4.9
Distal forearm 21 14.5 40 10.1 43 25.6 14 8.6
Hands/fingers 7 4.8 108 27.3 9 5.4 35 21.6
Vertebra 33 22.8 15 3.8 14 8.3 8 4.9
Pelvis (excluding coccyx) 1 0.7 2 0.5 2 1.2 3 1.9
Proximal femur 2 1.4 2 0.5 6 3.6 2 1.2
Shaft/distal femur 2 1.4 2 0.5 0 0 0 0
Patella 2 1.4 2 0.5 7 4.2 2 1.2
Shaft/proximal tibia/fibula 8 5.5 24 6.1 13 7.7 4 2.5
Ankle 3 2.1 22 5.6 14 8.3 5 3.1
Feet/toes 4 2.8 54 13.7 17 10.1 41 25.3
All sites 145 100.0 395 100.0 168 100.0 162 100.0

menopausal women combined) as shown in Table 3. Gen- determined by BMAD were greater than those for BMD in
erally, there was a stronger relationship between fracture every instance among the men, and in some cases substan-
history and BMD among women than men. Within genders, tially so (Table 3). After adjusting for age, wrist BMAD was
the estimated relative risk of fracture per 1 SD decrease in the strongest predictor of osteoporotic fracture in the men
BMD at each skeletal site was similar regardless of the (OR, 1.7; 95% CI, 1.3–2.3), and none of the other assess-
reference group selected. Likewise, within genders, the ments contributed significantly. However, the model incor-
ORs for fracture were of similar magnitude for BMD as porating age and wrist BMAD (model tau-a 5 0.129) was
assessed across the various skeletal sites (Table 3). After not substantially better than the previous one with age and
adjusting for age, however, total hip BMD was the best wrist BMD (model tau-a 5 0.122).
predictor of osteoporotic fractures in women (OR per 1 SD Altogether, 104 women (30%) and 91 men (26%) had
decline, 2.4; 95% confidence interval [CI], 1.6 –3.7). After experienced some other moderate trauma fracture (not hip,
hip BMD entered the model, none of the other BMD spine, or wrist) at age 35 or thereafter. After adjusting for
measurements independently predicted fracture risk in age, femoral neck BMD was the best predictor of these
women. By contrast, the OR per 1 SD decline in total hip other moderate trauma fractures in women (OR, 1.5; 95%
BMD in men was only 1.3 (95% CI, 0.97–1.7). Total wrist CI, 1.1–2.1), while wrist BMAD was best in men (OR, 1.5;
BMD was the stronger predictor of fractures in men (OR, 95% CI, 1.2–1.9). None of the other bone density assess-
1.5; 95% CI, 1.1–2.0). In these analyses, and those that ments contributed significantly. Again, however, a model
follow, ORs for men are based on the 20- to 29-year-old for men that included age and wrist BMD (tau-a 5 0.091)
male reference population, while those for women are was about as good as the one with age and wrist BMAD
based on the 20- to 29-year-old female reference (tau-a 5 0.106). One hundred and ten of the women (31%)
population. and 206 of the men (59%) had one or more fractures due to
Substituting BMAD for BMD among the women did not severe trauma. Neither age nor the bone density assess-
improve fracture prediction. The estimated OR actually ments predicted severe trauma fractures in women, and
declined for 14 of the 20 combinations of skeletal site and wrist BMAD was the only significant predictor in men (OR,
reference population shown in Table 3. Using the young 1.2; 95% CI 1.04 –1.4).
female reference data, however, the ORs determined for Based on the cross-sectional data from this study, there
BMAD in the women were somewhat greater than those for was a steady loss of hip BMD with age in both sexes
BMD when bone density was measured at the total hip and (Fig. 1A). Bone loss in the lumbar spine, however, did not
total wrist. Nonetheless, after adjusting for age (and age2) begin before age 40 years in women and was not seen at all
hip BMD was still the strongest predictor of fracture risk in in the men (Fig. 1B). There was little loss of wrist BMD
women and none of the other assessments entered the before age 50 years in either gender (Fig. 1C). These pat-
model. For men, the findings were very different. The ORs terns were identical when BMAD was assessed except that
FRACTURE PREDICTION IN MEN 1919

TABLE 3. AGE-ADJUSTED ODDS RATIOS (AND 95% CONFIDENCE INTERVALS) FOR OSTEOPOROTIC FRACTURE PER 1 SD DECREASE
IN BONE MINERAL DENSITY (BMD) OR BONE MINERAL APPARENT DENSITY (BMAD), BY SKELETAL SITE, FOR ROCHESTER,
MINNESOTA, MEN AND WOMEN DEPENDING ON THE CRITERION USED FOR ESTABLISHING NORMAL MEANS AND
STANDARD DEVIATIONS

Skeletal site

Study group and criterion Total hip Femoral neck Trochanter AP spine Total wrist

20–29 year male criteria

women (BMD) 2.20 (1.50, 3.22) 2.25 (1.44, 3.51) 2.06 (1.38, 3.08) 1.67 (1.18, 2.36) 1.77 (1.27, 2.47)
women (BMAD) 2.28 (1.49, 3.48) 1.65 (1.13, 2.42) 1.93 (1.32, 2.82) 1.54 (1.13, 2.10) 1.52 (1.18, 1.96)
men (BMD) 1.29 (0.97, 1.72) 1.10 (0.82, 1.48) 1.34 (1.00, 1.78) 1.15 (0.89, 1.48) 1.51 (1.12, 2.03)
men (BMAD) 1.87 (1.23, 2.83) 1.19 (0.83, 1.70) 1.57 (1.11, 2.21) 1.17 (0.88, 1.54) 1.72 (1.27, 2.33)
20–49 year male criteria
women (BMD) 2.66 (1.66, 4.29) 2.70 (1.56, 4.65) 2.48 (1.50, 4.10) 1.79 (1.21, 2.66) 1.67 (1.24, 2.26)
women (BMAD) 2.44 (1.55, 3.86) 1.72 (1.14, 2.58) 2.15 (1.38, 3.34) 1.59 (1.14, 2.22) 1.57 (1.20, 2.06)
men (BMD) 1.38 (0.97, 1.97) 1.12 (0.78, 1.62) 1.44 (1.01, 2.07) 1.17 (0.88, 1.56) 1.45 (1.11, 1.90)
men (BMAD) 1.97 (1.26, 3.09) 1.20 (0.82, 1.77) 1.69 (1.13, 2.52) 1.18 (0.87, 1.59) 1.80 (1.30, 2.49)
20–29 year female criteria
women (BMD) 2.40 (1.57, 3.68) 2.43 (1.49, 3.95) 2.21 (1.42, 3.42) 1.66 (1.18, 2.33) 1.56 (1.20, 2.03)
women (BMAD) 2.59 (1.59, 4.22) 1.77 (1.15, 2.72) 2.12 (1.37, 3.26) 1.50 (1.12, 2.00) 1.65 (1.22, 2.24)
men (BMD) 1.33 (0.97, 1.83) 1.11 (0.80, 1.54) 1.38 (1.01, 1.88) 1.14 (0.89, 1.47) 1.38 (1.09, 1.74)
men (BMAD) 2.06 (1.27, 3.33) 1.21 (0.81, 1.82) 1.67 (1.13, 2.47) 1.15 (0.89, 1.50) 1.92 (1.33, 2.76)
Premenopausal female criteria
women (BMD) 2.37 (1.56, 3.61) 2.48 (1.50, 4.08) 2.19 (1.42, 3.38) 1.61 (1.17, 2.23) 1.54 (1.20, 1.99)
women (BMAD) 2.70 (1.62, 4.49) 1.89 (1.17, 3.04) 2.38 (1.44, 3.93) 1.52 (1.12, 2.05) 1.74 (1.25, 2.43)
men (BMD) 1.33 (0.97, 1.82) 1.11 (0.80, 1.55) 1.37 (1.00, 1.87) 1.14 (0.90, 1.44) 1.37 (1.09, 1.72)
men (BMAD) 2.13 (1.29, 3.51) 1.24 (0.79, 1.95) 1.81 (1.15, 2.86) 1.16 (0.89, 1.52) 2.05 (1.37, 3.07)

bone density levels were no longer higher in men (data not DISCUSSION
shown).
The age-adjusted (to 1990 U.S. whites) prevalence of
osteoporosis at various skeletal sites using the WHO defi- A definition of osteoporosis in men based on BMD is
nition (BMD . 2.5 SD below the young normal mean) but predicated on showing a strong relationship between bone
different reference groups is shown in Table 4. The preva- density and fracture risk in men like that seen for women.(7)
lence of osteoporosis at the hip, spine, or wrist among Our data confirm earlier reports that bone density is asso-
Rochester women was 35% using the mean values from 20- ciated with fracture risk in men. Among 1355 Japanese–
to 29 year-old women; however, use of this reference pop- American men, the incidence of vertebral fractures was
ulation produced a prevalence of only 3% among Rochester increased significantly by 1.8-fold and 1.5-fold for every 1
men. For the hip sites, the SDs among the 20- to 29-year-old SD decrease in BMD at the proximal and distal radius,
men were quite small and produced a larger-than-expected respectively; the age-adjusted OR for other nonviolent frac-
prevalence of osteoporosis at the femoral neck, trochanter, tures was 1.3 at both sites and was not statistically signifi-
and total hip and an overall prevalence among men of 19%. cant.(8) The age-adjusted OR of a low trauma fracture
The prevalence was much lower at the hip sites when the
among 498 Nebraska men was 1.7 and 1.8 per 1 SD de-
20- to 49-year-old male reference range was used, and the
crease in BMD of the distal radius and ulna, respectively,(9)
overall prevalence in men was then only 13%. If a more
and there was a similar relationship between distal radius
rigorous standard were employed (BMD . 3.0 SD below
BMD and fragility fractures in 654 Swedish men.(10) Our
the mean for 20- to 29-year-old men), then the prevalence
results at the total wrist site are entirely consistent: the
of osteoporosis of the hip, spine, or wrist would be only 9%
in men. The age-adjusted prevalence was only 2% when the age-adjusted OR of a moderate trauma fracture of the hip,
criterion was a BMD level . 4.0 SD below the young male spine, or wrist was 1.5 and of any other moderate trauma
mean (data not shown). Substitution of BMAD for BMD fracture it was 1.4. Comparable results have been reported
reduced some of the gender-specific differences (Table 4). for bone density measured at the calcaneus,(8,11) a site that
Thus, use of the 20- to 29-year-old male reference data was not assessed here. As did we, Nguyen and colleagues
produced an overall prevalence rate of 17% for Rochester found no association between lumbar spine BMD and frac-
women and 15% for Rochester men. Conversely, when the ture risk in 709 Australian men.(12) They did find a 2-fold
20- to 29-year-old female reference data were used, the increase in the risk of an atraumatic fracture with each 1 SD
age-adjusted prevalence of osteoporosis at the hip, spine, or decrease in femoral neck BMD in men, a result that was
wrist was 29% in women and 38% in men. considerably stronger than our age-adjusted OR of only 1.1.
1920 MELTON ET AL.

There is some agreement that the proximal femur is an

optimal site to assess osteoporosis in women(15) because hip
BMD predicts hip fractures better than measurements at
other skeletal sites and predicts fractures in general just as
well as the other measurements.(7) By contrast, bone den-
sity assessed at the wrist was the best predictor of osteopo-
rotic fractures in men in this study, although the advantage
over other sites was not great. It is not clear why this should
be, particularly since men do not have a high likelihood of
experiencing a distal forearm fracture. Nonetheless, unlike
women, fracture prediction in men was not improved by
moving to a hip measurement site. Indeed, at every skeletal
site assessed, there was a weaker relationship between
BMD and fracture risk in men compared with women, as
seen also in some other studies.(8,12) In part, this is due to
the fact that men’s bones are bigger and larger bones are
more resistant to fracture at any given level of bone densi-
ty.(16) Areal bone density does not completely correct for
the men’s larger bone size,(17) but a further correction for
bone volume (e.g., BMAD) did improve fracture prediction
in men, bringing the estimated OR per 1 SD change more
in line with those of Rochester women. Fracture risk in 258
elderly Finnish men and women was almost identical at
comparable levels of calcaneal bone density assessed as
gram per cubic centimeter.(18) Although BMAD predicted
fractures in men better than BMD, the improvement was
modest, and BMD was the better predictor of osteoporotic
fractures in women. Others have also found that the theo-
retical advantages of BMAD over BMD for fracture pre-
diction are slight in practice.(19,20)
The prevalence of osteoporosis depends on the reference
population that is used to determine normal values, the
cut-off level selected to define the condition and the pattern
of bone loss in the community. The choice of a reference
population can have a substantial impact on the estimated
prevalence of osteoporosis.(21,22) We used reference data
from the Rochester population and, within genders, it mat-
tered little whether “young normal” was defined as age
20 –29 years or age 20 – 49 years (or all premenopausal
women combined). For women, the overall prevalence of
osteoporosis at the hip, spine or wrist changed only from
35% to 41% using the different gender-specific standards,
while the men changed from 19% to 13%. This modest
difference is due to the fact that bone loss was not seen at
the spine or wrist in these cross-sectional data before about
age 50 years, despite the linear decrease in hip BMD from
FIG. 1. Distribution of BMD by age among age-stratified
samples of Rochester, Minnesota, men and women for (A) age 20 years onward. Use of male normal values for women,
total hip, (B) lumbar spine, and (C) total wrist. however, resulted in female prevalence estimates that were
extremely high (66 –72%), while female normal values pro-
duced prevalence estimates in men that were very low
(3– 4%). All of these figures are based on use of the WHO
The comparability of these two figures is uncertain since the cut-off level to define osteoporosis at 2.5 SD below the
Australian fracture model also included quadriceps young normal mean.(1) Originally there was some concern
strength and body sway as independent predictors of frac- that using a comparable definition of osteoporosis for men
ture risk. A subsequent univariate analysis of their data (2.5 SD below the young male mean) as for women (2.5 SD
showed an OR of about 1.4,(13) which was similar to our below the young female mean) would produce similar prev-
unadjusted estimate of 1.5. The age-related decline of fem- alence figures in the two genders,(23) whereas the lifetime
oral neck BMD also appeared to account for a doubling of risk of a hip, spine, or forearm fracture was 40% in white
hip fracture risk among 2446 men in Rotterdam, women and only 13% in white men.(3) However, BMD
Netherlands.(14) values declined less rapidly over life in men than women in
FRACTURE PREDICTION IN MEN 1921

TABLE 4. AGE-ADJUSTED* PREVALENCE (%) OF OSTEOPOROSIS (2.5 SD BELOW YOUNG NORMAL MEAN) AMONG ROCHESTER,
MINNESOTA, MEN AND WOMEN $50 YEARS OF AGE DEPENDING ON THE CRITERION USED FOR ESTABLISHING MEANS AND
STANDARD DEVIATIONS

Skeletal site

Any total
Study group and criterion Total hip Femoral neck Trochanter AP spine Total wrist hip/spine/wrist

20–29 year male criteria

women (BMD) 47.7 58.5 43.7 8.0 62.5 65.8
women (BMAD) 3.2 16.5 7.8 2.9 13.4 17.2
men (BMD) 15.8 22.3 11.0 1.4 8.8 19.4
men (BMAD) 7.2 13.0 13.6 1.4 10.8 14.8
20–49 year male criteria
women (BMD) 26.7 28.0 17.8 7.3 72.4 72.4
women (BMAD) 0.5 5.2 2.5 2.4 10.7 13.2
men (BMD) 3.2 6.5 1.4 1.4 12.4 13.3
men (BMAD) 3.2 4.0 4.1 1.4 7.2 9.2
20–29 year female criteria
women (BMD) 13.6 28.0 8.9 7.7 32.9 34.7
women (BMAD) 17.8 24.1 29.3 8.4 22.5 29.0
men (BMD) 1.4 5.8 1.1 1.4 2.0 3.4
men (BMAD) 25.7 27.9 28.3 2.3 25.7 37.9
Premenopausal female criteria
women (BMD) 10.7 15.2 8.9 14.1 38.3 41.3
women (BMAD) 10.0 8.9 8.1 8.4 21.4 25.9
men (BMD) 1.4 2.2 0.7 1.4 2.5 3.8
men (BMAD) 18.3 6.2 13.6 2.9 20.0 29.1

* Both men and women directly age-adjusted to total population distribution of United States whites aged 50 years and over in 1990.

Rochester as seen also by others(24) so that the overall higher (1.08 vs. 1.04 g/cm2), and the SD at this site was
age-adjusted prevalence (35% in Rochester women and much smaller (0.113 vs. 0.144 g/cm2) compared with
19% in Rochester men), appears to be of reasonable mag- NHANES data. The reference data for Rochester men
nitude relative to the lifetime fracture risks previously re- 20 – 49 years of age at the total hip site (mean, 1.06 g/cm2;
ported for each gender. Prevalence estimates were similar SD, 0.140 g/cm2) corresponded more closely to the
when BMAD was used with gender-specific reference val- NHANES III figures and produced a comparable preva-
ues (29% in women and 15% in men). Use of more strin- lence of osteoporosis at the total hip in men of 3%.
gent criteria for men (e.g., BMD 3.0 or 4.0 SD below the It is apparent that reference data from NHANES III, a
young male mean) produced overall prevalence estimates in probability sample of the entire United States population,
men that seem too low (9% and 2%, respectively). It must provide superior normative values for hip BMD than does
be re-emphasized, however, that this empiric approach is a small sample of men from Rochester. However, there are
entirely arbitrary and that better ways might be developed no comparable national data for BMD of the spine or wrist
to select an appropriate cut-off level to define the since those sites were not assessed in NHANES III. This is
condition. important because the prevalence of osteoporosis is ex-
Using the WHO definition, the age-adjusted prevalence pected to be higher if a number of skeletal sites are assessed
of osteoporosis at the total hip alone, based on our gender- simultaneously as demonstrated in this study. None of the
specific reference data for 20- to 29-year-old women, was 48 men aged 20 –29 years and only one of the 50 women
14% for Rochester women compared with a similar 15% aged 20 –29 years had osteoporosis of the hip, spine, or
figure for white women from the Third National Health and wrist. Among women age 50 years and older, however, the
Nutrition Examination Survey (NHANES III), using means age-adjusted prevalence of osteoporosis at the hip, spine, or
and SDs from 409 white women aged 20 –29 years.(25) How- distal forearm was 35%. When the age-specific prevalence
ever, the prevalence was estimated at only 4% for white among Rochester women is extrapolated to the population
men in the NHANES III data, using means and SDs from of the United States in 1990, it suggests that 11.5 million
382 white men 20 –29 years old to establish the osteoporosis white women might be affected. A similar analysis based on
cut-off level. This contrasts sharply with our estimate of an earlier age-stratified sample of Rochester women as-
16% using reference data from 20- to 29-year-old Roches- sessed with dual-photon absorptiometry produced an esti-
ter men. The discrepancy is explained by the fact that the mated 1990 prevalence of 9.4 million white women in the
mean total hip BMD level in the young Rochester men was United States with osteoporosis at the hip, spine, or wrist.(3)
1922 MELTON ET AL.

The estimated number of white men in the United States ACKNOWLEDGMENTS

with osteoporosis in 1990, based on the age-specific preva-
lence of osteoporosis at the hip, spine, or wrist in Rochester We would like to thank Mrs. Veronica L. Gathje and
men, is 4.5 million. Mrs. Margaret F. Holets for help with data collection, Mrs.
The main limitations of this analysis are the relatively Cindy Crowson for assistance with data analysis, and Mrs.
small number of subjects, the marginal response rates, and Mary Roberts for help in preparing the manuscript. This
the cross-sectional nature of the analysis. Low participation work was supported by research grant AR27065 from the
rates in the oldest age groups, for example, raise the pos- National Institute of Arthritis, Musculoskeletal and Skin
sibility that frail individuals with low bone density and high Diseases, United States Public Health Service.
fracture prevalence were disproportionately excluded, lead-
ing to a conservative bias in the fracture ORs. However,
analysis of an earlier sample of women from the community REFERENCES
demonstrated that the prevalence of osteoporotic fractures
was similar whether assessed in study participants alone or 1. World Health Organization 1994 Assessment of fracture risk
and application to screening for postmenopausal osteoporosis.
in participants and nonparticipants combined.(26) In any WHO Technical Report Series 843, Geneva, Switzerland.
event, our cross-sectional estimates of the odds of fracture 2. Kanis JA, Melton LJ III, Christiansen C, Johnston CC, Khal-
per 1 SD change in BMD are entirely consistent with pro- taev N 1994 Perspective: The diagnosis of osteoporosis. J Bone
spective data both in white women and white men. Because Miner Res 9:1137–1141.
3. Melton LJ III, Chrischilles EA, Cooper C, Lane AW, Riggs BL
of the racial composition of the community (96% white in 1992 Perspective: How many women have osteoporosis?
1990), we are unable to address the relationship between J Bone Miner Res 7:1005–1010.
bone density and fracture risk in other ethnic groups. It is 4. Melton LJ III 1996 History of the Rochester Epidemiology
Project. Mayo Clin Proc 71:266 –274.
important that this issue be resolved since expenditures
5. Beard CM, Melton LJ III, Cedel SL, Richelson LS, Riggs BL
for treating osteoporotic fractures in nonwhite women 1990 Ascertainment of risk factors for osteoporosis: Compar-
and men in the United States approached $1 billion in ison of interview data with medical record review. J Bone
1995.(27) Miner Res 5:691– 699.
6. Katzman DK, Bachrach LK, Carter DR, Marcus R 1991 Clin-
We believe that fracture ascertainment in this study is ical and anthropometric correlates of bone mineral acquisition
complete insofar as the fractures were recognized clinically in healthy adolescent girls. J Clin Endocrinol Metab 73:
because we reviewed inpatient and outpatient medical 1332–1339.
records for each subject that spanned 30 years on average. 7. Marshall D, Johnell O, Wedel H 1996 Meta-analysis of how
well measures of bone mineral density predict occurrence of
Undiagnosed vertebral fractures are most likely to have osteoporotic fractures. BMJ 312:1254 –1259.
been missed since only severe deformities come to clinical 8. Ross PD, Yhee Y-K, Davis JW, Wasnich RD 1993 Bone
attention.(28) Of those with 4 SD deformities by vertebral density predicts fracture incidence among elderly men. In:
morphometry in this study (data not shown), 31% of the Christiansen C, Riis B (eds.) Proceedings of the Fourth Inter-
national Symposium on Osteoporosis and Consensus Develop-
men and 45% of the women also had a clinical diagnosis of ment Conference, Hong Kong, 27 March–2 April 1993, Han-
vertebral fracture, while 38% and 59%, respectively, had at delstrykkeriet Aalborg ApS, Aalborg, Denmark.
least one fracture of the spine, hip, or distal forearm. The 9. Stegman MR, Heaney RP, Recker RR 1995 Comparison of
significance of any discrepancies is uncertain because it is speed of sound ultrasound with single photon absorptiometry
for determining fracture odds ratios. J Bone Miner Res
not clear that all of these deformities represent osteopo- 10:346 –352.
rotic vertebral fractures. Consequently, we did not consider 10. Gärdsell P, Johnell O, Nilsson BE 1990 The predictive value of
fracture misclassification to be an important problem here. forearm bone mineral content measurements in men. Bone
11:229 –232.
Likewise, we measured bone density using accurate meth-
11. Cheng S, Suominen H, Era P, Heikkinen E 1994 Bone density
ods, although we recognize that osteoarthritic changes may of the calcaneus and fractures in 75- and 80-year-old men and
cause overestimation of BMD in the lumbar spine and mask women. Osteoporos Int 4:48 –54.
any age-related bone loss.(29 –33) Consequently, the latter 12. Nguyen T, Sambrook P, Kelly P, Jones G, Lord S, Freund J,
Eisman J 1993 Prediction of osteoporotic fractures by postural
approach may be of limited value in assessing osteoporosis instability and bone density. BMJ 307:1111–1115.
in men and older women. In addition, it remains to be 13. Nguyen TV, Eisman JA, Kelly PJ, Sambrook PN 1996 Risk
determined whether BMAD would be more useful than factors for osteoporotic fractures in elderly men. Am J Epide-
BMD in the global assessment of osteoporosis. While miol 144:255–263.
14. De Laet CEDH, van Hout BA, Burger H, Hofman A, Pols
BMAD did not appear to be superior to BMD for fracture HAP 1997 Bone density and risk of hip fracture in men and
prediction in this population, gender-specific differences women: Cross sectional analysis. BMJ 315:221–225.
were minimized by substituting volumetric for areal bone 15. Anonymous 1997 Who are candidates for prevention and treat-
density, as reported by others.(34) Use of BMAD instead ment for osteoporosis? Osteoporos Int 7:1– 6.
16. Melton LJ III, Chao EYS, Lane J 1988 Biomechanical aspects
of BMD has also been shown to reduce the discrepancies of fractures. In: Riggs BL, Melton LJ III (eds.) Osteoporosis:
in bone density values among men and women of other Etiology, Diagnosis, and Management. Raven Press, New
races.(35–37) When these remaining questions are an- York, pp. 111–131.
17. Carter DR, Bouxsein ML, Marcus R 1992 New approaches for
swered, it should be possible to propose an operational
interpreting projected bone densitometry data. J Bone Miner
definition for osteoporosis that addresses the entire Res 7:137–145.
population. 18. Cheng S, Suominen H, Sakari-Rantala R, Laukkanen P, Avi-
FRACTURE PREDICTION IN MEN 1923

kainen V, Heikkinen E 1997 Calcaneal bone mineral density in postmenopausal women. J Clin Endocrinol Metab 72:
predicts fracture occurrence: A five-year follow-up study in 1372–1374.
elderly people. J Bone Miner Res 12:1075–1082. 31. Masud T, Langley S, Wiltshire P, Doyle DV, Spector TD 1993
19. Mazess RB, Barden H, Mautalen C, Vega E 1994 Normaliza- Effect of spinal osteophytosis on bone mineral density mea-
tion of spine densitometry. J Bone Miner Res 9:541–548. surements in vertebral osteoporosis. BMJ 307:172–173.
20. Cummings SR, Marcus R, Palermo L, Ensrud KE, Genant HK, 32. Jones G, Nguyen T, Sambrook PN, Kelly PJ, Eisman JA 1995
the Study of Osteoporotic Fractures Research Group 1994 A longitudinal study of the effect of spinal degenerative disease
Does estimating volumetric bone density of the femoral neck on bone density in the elderly. J Rheumatol 22:932–936.
improve the prediction of hip fracture? A prospective study. 33. von der Recke P, Hansen MA, Overgaard K, Christiansen C
J Bone Miner Res 9:1429 –1432. 1996 The impact of degenerative conditions in the spine on
21. Faulkner KG, Roberts LA, McClung MR 1996 Discrepancies bone mineral density and fracture risk prediction. Osteoporos
in normative data between Lunar and Hologic DXA systems. Int 6:43– 49.
Osteoporos Int 6:432– 436. 34. Faulkner RA, McCulloch RG, Fyke SL, De Coteau WE,
22. Ahmed AIH, Blake GM, Rymer JM, Fogelman I 1997 Screen- McKay HA, Bailey DA, Houston CS, Wilkinson AA 1995
ing for osteopenia and osteoporosis: Do the accepted normal Comparison of areal and estimated volumetric bone mineral
ranges lead to overdiagnosis? Osteoporos Int 7:432– 438. density values between older men and women. Osteoporos Int
23. Melton LJ III 1997 The prevalence of osteoporosis. J Bone 5:271–275.
Miner Res 12:1769 –1771. 35. Marcus R, Greendale G, Blunt BA, Bush TL, Sherman S,
24. Looker AC, Wahner HW, Dunn WL, Calvo MS, Harris TB, Sherwin R, Wahner H, Wells B 1994 Correlates of bone min-
Heyse SP, Johnston CC Jr, Lindsay RL 1995 Proximal femur eral density in the Postmenopausal Estrogen/Progestin Inter-
bone mineral levels of US adults. Osteoporos Int 5:389 – 409. ventions Trial. J Bone Miner Res 9:1467–1476.
25. Looker AC, Orwoll ES, Johnston CC Jr, Lindsay RL, Wahner 36. Cundy T, Cornish J, Evans MC, Gamble G, Stapleton J, Reid
HW, Dunn WL, Calvo MS, Harris TB, Heyse SP 1997 Preva- IR 1995 Sources of interracial variation in bone mineral den-
lence of low femoral bone density in older U.S. adults from sity. J Bone Miner Res 10:368 –373.
NHANES III. J Bone Miner Res 12:1761–1768. 37. Bhudhikanok GS, Wang M-C, Eckert K, Matkin C, Marcus R,
26. Beard CM, Lane AW, O’Fallon WM, Riggs BL, Melton LJ III Bachrach LK 1996 Differences in bone mineral in young Asian
1994 Comparison of respondents and nonrespondents in an and Caucasian Americans may reflect differences in bone size.
osteoporosis study. Ann Epidemiol 4:398 – 403. J Bone Miner Res 11:1545–1556.
27. Ray NF, Chan JK, Thamer M, Melton LJ III 1997 Medical
expenditures for the treatment of osteoporotic fractures in the Address reprint requests to:
United States in 1995: Report from the National Osteoporosis Dr. L. Joseph Melton, III
Foundation. J Bone Miner Res 12:24 –35. Section of Clinical Epidemiology
28. Cooper C, Atkinson EJ, O’Fallon WM, Melton LJ III 1992
Incidence of clinically diagnosed vertebral fractures: A popu- Department of Health Sciences Research
lation-based study in Rochester, Minnesota, 1985– 89. J Bone Mayo Clinic
Miner Res 7:221–227. 200 First Street Southwest
29. Orwoll ES, Oviatt SK, Mann T 1990 The impact of osteophytic Rochester, MN 55905 U.S.A.
and vascular calcifications on vertebral mineral density mea-
surements in men. J Clin Endocrinol Metab 70:1202–1207.
30. Reid IR, Evans MC, Ames R, Wattie DJ 1991 The influence of Received in original form December 18, 1997; in revised form July
osteophytes and aortic calcification on spinal mineral density 8, 1998; accepted August 5, 1998.