Factors Associated with Appendicular Bone Mass in Older Women

Methods

Patients

From September 1986 to October 1988, women who were at least 65 years old were recruited in four areas of the United States: Portland, Oregon; Minneapolis, Minnesota; Baltimore, Maryland; and the Monongahela Valley near Pittsburgh, Pennsylvania. Age-eligible women were identified from membership lists from several sources, as previously reported [1]. We excluded black women because of the reduced incidence of hip fractures in this group [9], and we excluded women who were unable to walk without the assistance of another person or who had bilateral hip replacements.

Measurement of Bone Mass

Bone mass (in g/cm²) was measured using OsteoAnalyzers (Siemens-Osteon, Wahiawa, Hawaii). We scanned three sites: the distal radius, the midradius, and the calcaneus. The protocol for the bone mass measurements has been described elsewhere [1]. To describe the reproducibility of these measurements, a participant in the study was measured five times on 5 consecutive days at each of the four clinical centers. The average coefficients of variation in these older women (standard deviation/mean for each subject) were 1.5% for the distal radius, 2.0% for the midradius, and 1.3% for the calcaneus. To assess variations between scanners at the four centers, two investigators were measured by all four machines; the mean of their coefficients of variation in these younger subjects was 0.4% for the distal radius, 0.5% for the midradius, and 1.2% for the calcaneus [10]. The correlation coefficients between the measurement sites were distal and midradius, 0.75; calcaneus and distal radius, 0.66; and calcaneus and midradius, 0.63.

Predictor Variables

Participants completed a self-administered questionnaire and were interviewed and examined at the clinical center. A selected medical history was obtained, including a history of a physician diagnosis of osteoporosis, spine fracture, arthritis, gastric surgery, hyperthyroidism, and stroke. Reproductive history, including age of last menstrual period, genitourinary surgery (hysterectomy, oophorectomy), number of pregnancies, and breast-feeding, was recorded. Previous fractures in subjects and their parents or sisters were noted.

We collected detailed records about specific health habits including lifetime smoking history, alcohol use, and caffeine intake. Women who had smoked less than 100 cigarettes in their lifetime were considered nonsmokers. We assumed that a cup of coffee contained 95 mg of caffeine and that tea and cola drinks contained 55 mg and 45 mg, respectively.

Participants were asked to bring all prescription and nonprescription medications with them to the interview for verification. The dose and duration of use of sex hormones, diuretics, corticosteroids, thyroid supplements, aluminum-containing antacids, vitamin D, sedative hypnotics, and antiepileptic medications were obtained. The frequency and duration of use of calcium supplements were recorded; use of TUMS (which can be taken as an antacid or calcium supplement) was recorded separately. Responses to the questionnaire were checked by a trained interviewer and were verified against the labels of medications that the subject brought to the interview.

Recent dietary history (past 12 months), particularly calcium, phosphorus, and protein intake, was assessed by a checklist-interview method developed from the HANES-II survey [11]. Food models were used to estimate portion sizes, and foods that account for 80% of calcium intake in most adults were included. This instrument has a correlation of 0.76 with calcium intake assessed by a 7-day diet diary, but it tends to underestimate calcium intake by approximately 150 mg/day [12]. The frequency of milk consumption as a child and young adult was also assessed by questionnaire.

Physical activity was examined with a modified Paffenbarger survey that has been validated in postmenopausal women [13-16]. The type and duration of weight-bearing recreational activities from the previous 12 months were recorded, and these were converted into weekly caloric expenditure. Other current activities, such as stair climbing, walking, and heavy household chores, were included. Intensity-weighted measures were calculated by designating activities as low (for example, walking or gardening), medium (dancing or tennis), or high intensity (jogging or skiing) and multiplying the reported frequency of the activity by 2.5, 5, and 7.5, respectively.

Weight (in light indoor clothes with shoes removed) was recorded with a balance beam scale, and height was measured using a standard held-expiration technique with a wall-mounted Harpenden stadiometer [17]. Maximal right knee extension, triceps (arm extension), and hip abduction torque strength were measured with a hand-held isometric dynamometer (Sparks Instruments and Academics, Coralville, Iowa). Grip strength of the right and left hand was assessed as the average of two attempts with the dynamometer. Waist, hip, and abdominal circumferences were measured using standard methods [18].

Statistical Methods

The cohort was randomly divided into equal-sized (n = 4852) training and validation groups. In the training group, we analyzed potential associations with bone mass in univariate analyses, adjusted for age. We then adjusted associations (P < 0.05) for other plausible confounding effects. For example, associations between the use of thiazide diuretics and bone mass were also adjusted for body weight, because those who were taking diuretics tended to be heavier. Potentially nonlinear associations between bone mass and continuous variables, such as calcium intake, were examined by plotting bone mass against the median of each decile of the predictor variable. Associations were tested for statistical significance with simple linear regression, Student t-test, or analysis of variance.

Variables that were associated with bone mass in univariate analyses (P < 0.05) were examined by multivariate analyses using PROC GLM (SAS Institute, Inc., Cary, North Carolina). Some categories of predictor variables, such as calcium intake and muscle strength, contained several variables associated with bone mass in univariate analyses. After examining these variables for multicollinearity, separate models for strength, family history of fractures, dietary calcium, and physical activity were analyzed to determine the individual variables that explained most of the variance of bone mass within each category. For example, grip strength explained most of the variance for bone mass related to strength; thus, we selected this variable as the measure of strength for subsequent multivariate analyses. Both current dietary calcium and calcium from milk (ingested between ages 18 to 50 years) contributed to bone mass variance; therefore, both were included in the final multivariate models. A history of any maternal fracture after age 50 accounted for most of the bone mass variance from family history, and intensity-weighted lifetime physical activity accounted for most of the variance from physical activity. Selected variables (see Table 2) were then entered into a single multivariate model.

We found that results from multivariate models were very similar in both training and validation groups; thus results are reported for the entire (combined) cohort. Results were generally similar for the three bone mass sites, and for those analyses that were concordant at all three sites, only the results for the distal radius are presented. We presented results at the other two sites only when they differed from the distal radius.

Results

The average age at enrollment was 71.1 years, and the age distributions were similar in the training and validation groups (Table 1). Thirteen percent of patients reported a previous wrist or hip fracture.

Demographic and Anthropometric Data

Bone mass was strongly and inversely associated with age, decreasing by approximately 5% with every additional 5 years of age after 65 years (Table 2). Northern European ancestry, hair color, and educational level were not associated with bone mass (Table 3).

Several anthropometric measurements were strongly associated with bone mass (see Table 2). Both weight [in kilograms] and obesity (as Quetelet index, in kilograms per square meter) were associated with increased bone mass: after adjusting for age, bone mass increased 5.0% for every 10-kg increase in weight (Figure 1). Although weight and Quetelet were highly correlated (r = 0.91), weight was more strongly associated with bone mass than Quetelet index (R² = 0.11 and 0.09, for weight and Quetelet, respectively). After adjustment for age and weight at age 50 years, weight loss after age 50 was associated with lower bone mass; each 10 kg of weight loss was associated with 3.9% lower bone mass.

Grip strength, triceps torque, hip abduction torque, and knee-extension torque were all positively associated with bone mass, even after adjustment for weight. A 5-kg increase in grip strength was associated with a 4.9% increase in distal radius bone mass (see Table 2). Taller women also had higher bone mass; each 10-cm increase in height was associated with a 5.7% increase in bone mass.

Reproductive History and Family History of Fractures

Even after adjusting for age, later menopause was strongly associated with higher bone mass (see Table 2), both in women who took exogenous estrogens and in those who never used estrogen: each 5-year delay in menopause was associated with 1.3% higher bone mass. Surgical menopause, defined as a bilateral oophorectomy with or without hysterectomy before natural menopause, was associated with higher bone mass in unadjusted analyses. However, among women who had never used oral estrogens, no difference existed in the bone mass of women with a surgical or a natural menopause. Although a history of at least one pregnancy was positively associated with bone mass, after adjustment for weight this relation was not significant at the mid- and distal radius. The number of live births was not associated with bone mass. Neither a history of breast-feeding (see Table 3) nor the number of children breast-fed was associated with bone mass.

Age-adjusted bone mass was less in those women who reported a fracture in their mother after age 50 years (see Table 2). Maternal fractures of the spine [as indicated by the presence of a dowager's hump] or wrist were both associated with lower bone mass. Neither a maternal or a sororal history of hip fracture, nor a paternal history of any fracture, was related to bone mass (see Table 3).

Calcium Intake

We found no association between total daily calcium intake (dietary calcium plus supplements) and bone mass (see Table 3). Current calcium intake from food was weakly associated with bone mass in women who did not take calcium supplements; each 400-mg/d increase in dietary calcium was associated with a 1.1% increase in distal radius bone mass. No evidence existed of a threshold effect for the relation between dietary calcium intake and bone mass (Figure 2). Calcium intake from milk as a teenager, between ages 18 and 50, and after age 50 years, adjusted for current calcium intake, was also associated with increased bone mass. For example, women who drank milk at every meal, between ages 18 and 50, had 3.1% higher bone mass compared with those who rarely or never drank milk.

We found no associations between the use of supplemental calcium and bone mass. Because calcium is often recommended as treatment for osteoporosis, we looked separately at women who did not report a history of osteoporosis or fractures after age 50. As TUMS are known to dissolve well in the stomach (Heaney RP. Personal communication), we analyzed their effects separately. However, current or past use of TUMS was not associated with bone mass. Neither current nor past use of vitamin D supplements was associated with bone mass.

Physical Activity, Smoking, Caffeine, and Alcohol

Weight-bearing physical activity was associated with increased bone mass after adjustment for age and weight. A 2000-kcal/wk increase in vigorous activity (approximately 20 minutes of jogging per day) was associated with a 2% increase in distal radius bone mass and a 3.9% increase in calcaneal bone mass.

Smokers had decreased bone mass compared with nonsmokers. After adjusting for age and weight, current smokers had a 4.3% decrease in distal radius bone mass, and past smokers had a 1.7% decrease in bone mass compared with never smokers. However, among current and former smokers, we found no association between amount of tobacco used over a lifetime (in pack-years) and current bone mass, except at the calcaneus where a weak inverse association occurred ( 0.6% per 10 pack-years).

Lifetime caffeine consumption was inversely associated, albeit weakly, with bone mass (see Table 2): 10 000 g of caffeine intake, the equivalent of 10 cups of coffee per day for 30 years, was associated with a 1.1% decrease in distal radius bone mass. Alcohol use, whether measured as lifetime ounces of alcohol consumed or average weekly consumption, was not associated with bone mass in this cohort (see Table 3). The sole exception was that lifetime intake of alcohol was weakly associated with lower bone mass at the calcaneus ( 0.03% per 1000 drinks).

Medical Conditions

Women with a history of gastric surgery had a 9.4% decrease in distal radius bone mass compared with those without such surgery. Because smokers may be more likely to undergo gastrectomy, we also adjusted for smoking history, but this had little effect. Women who reported a history of osteoporosis, or fractures of the spine, hip, or wrist after age 50 had 6.7% lower bone mass than those without a history of osteoporosis. Diabetes was positively associated with bone mass. When adjusted for weight or body mass index, and stratified by insulin use, bone mass in non–insulin-dependent diabetics was 5.3% greater than in nondiabetics. No association existed between insulin-dependent diabetes and bone mass.

A history of osteoarthritis was associated with higher bone mass in unadjusted models, but no relation existed between osteoarthritis and bone mass after adjustment for weight. Women who reported a history of rheumatoid arthritis had significantly less bone mass at the midradius (2.6%) and calcaneus (3.3%), but no effect occurred at the distal radius (see Table 2). After adjustment for weight, a history of hyperthyroidism was not associated with distal radius bone mass but was associated with a 1.9% and a 2.0% decrease in midradius and calcaneus bone mass, respectively. Overall health was not associated with bone mass (see Table 3).

Medications

Because estrogens are often recommended as treatment for osteoporosis, we examined their effects separately in women who did not report a history of osteoporosis or fractures after age 50. Current or past oral estrogen use, and duration of use, were strongly associated with higher bone mass. Among current estrogen users, every 5 years of estrogen use was associated with a 2.4% increase in distal radius bone mass.

After adjusting for age and years of estrogen use, the distal radius and calcaneal bone density of women using estrogen alone was greater than those reporting no current estrogen use or simultaneous use of estrogen and progestin (Figure 3). No overall association existed between progestin use and bone mass. Furthermore, no association existed between bone mass and the use or duration of use of oral contraceptives, except at the calcaneus where ever users had 4.0% higher bone mass compared with never users.

A history of thiazide diuretic use and total number of years of use were positively associated with bone mass. Among current users, each 5-year period of thiazide use was associated with a 1.2% increase in distal radius bone mass. Women currently using thyroid hormone had 3.6% lower bone mass at the midradius after adjusting for weight, but this effect was not present at the distal radius or calcaneus. No association existed between duration of thyroid hormone use and bone mass at any site. Bone mass was decreased in those women who reported a history of current steroid use, but the use of seizure medications or regular use of antacids was not associated with bone mass.

Multivariate Summary

Multivariate analyses, summarized in Tables 2 and 4, showed that increased weight, grip strength, height, age at menopause, non–insulin-dependent diabetes, surgical menopause, as well as use of estrogens and thiazides were associated with increased bone mass at all three sites. Age, caffeine intake, and a history of gastric surgery or maternal fracture were independently associated with decreased bone mass at all three sites. The number of women in the final multivariate model (n = 5430) was less than the total cohort because of the cumulative effect of missing values. Women had difficulty remembering whether their mothers had had fractures, and 24% of the cohort could not provide this information. The total variance (R²) of bone mass explained by the final multivariate models was 20% for the distal radius, 28% for the midradius, and 34% for the calcaneus. At each site, weight and age accounted for the greatest proportions of bone mass variance (distal radius, 8.9% and 3.4%; midradius, 13.2% and 6.4%; and calcaneus, 24% and 3.7%, respectively).

Discussion

This is the largest and most comprehensive cross-sectional study of determinants of bone mass in older women. Our results confirm those of smaller studies and indicate that age, body weight, muscle strength, maternal history of osteoporotic fractures, and use of estrogens are correlated with bone mass. We also encountered a number of new or surprising findings. In particular, we observed that weight was a stronger determinant of appendicular bone mass than obesity and that weight loss after age 50 has a substantial and adverse effect on bone mass. We found that duration of time after menopause has an effect on bone mass that is independent of total age and that pregnancy results in increased bone mass mainly by causing weight gain. Our analyses indicate that the beneficial effect of estrogen replacement is proportional to the duration of treatment, and no additional skeletal benefit occurs from combined estrogen-progestin therapy. We found no evidence that alcohol use in older women was associated with decreased bone mass. Contrary to commonly held beliefs, we found that non–insulin-dependent diabetics had substantially higher bone mass (even after adjusting for weight), that lactation or regular antacid use do not decrease bone mass, and that fair-haired women or those of Northern European ancestry do not have lower bone mass than darker-haired women with other national backgrounds.

Anthropometric Variables

Age and weight were among the strongest predictors of appendicular bone mass. The association between weight and bone mass was stronger than that between bone mass and body mass index, suggesting that mechanical factors may be more important than adiposity. Increased weight or frame size, independent of adiposity, has also been found to correlate with appendicular [19] and axial [20, 21] bone mass in previous studies. Other investigators [19] have found adiposity to be associated with appendicular bone mass. We found height to be independently associated with bone mass. However, taller women have larger bones, and our method of measuring bone density adjusts for area rather than volume of bone scanned. Carter and colleagues [22] have recently shown that certain algorithms, which estimate true density from approximations of the volume of vertebrae, abolish apparent associations between height and spinal bone mass. Greater strength, as measured by both upper and lower body torque, was associated with increased bone mass. A number of previous studies [23-25] have found similar associations between bone mass and strength in postmenopausal women, suggesting that mechanical loading of bone preserves bone mass.

Reproductive and Family History

Other than age at menopause, reproductive history was not related to bone mass. Surgical menopause has been associated with rapid loss of bone mass [26], but we found that surgical menopause had no effect on bone mass after estrogen use was accounted for. Some investigators, but not all, have described an association between pregnancy and increased bone mass [27, 28]. In our study, this association was attributable to increased weight among women who had been pregnant. The effect of breast-feeding on bone mass is also controversial [8, 27, 29-32], and we found no evidence to support an association.

Twin studies have shown a strong familial similarity for bone mass, particularly among monozygotic twins [33]. A strong association also exists between family history of osteoporosis and bone mass [34]. Our findings suggest that maternal history of fractures may be more useful than the fracture history of other first-degree relatives.

Calcium Intake

We did not find a clinically significant association between dietary or supplemental calcium intake and bone mass. The relation between calcium intake and osteoporosis remains controversial [35]. Several investigators have suggested that calcium intake less than 400 mg/d may result in osteopenia [36]. We looked for such a threshold effect and did not find one. We also could not confirm the suggested association between calcium intake as a teenager or young adult and postmenopausal bone mass [37]. Calcium supplements, whether taken as TUMS or other preparations, were not associated with bone mass. The benefits of vitamin D supplements also remain controversial [32, 38], and we found no association between bone mass and vitamin D use or duration of use.

Habits

Although widely believed to prevent osteoporosis, the beneficial effects of weight-bearing exercise on bone in postmenopausal women remain unproven [39, 40]. Several cross-sectional [41, 42] and longitudinal studies [42-44] of exercise have suggested a modest beneficial effect of physical activity, but others have not [45]. We found no beneficial effects of current or previous weight-bearing exercise on appendicular bone mass after controlling for other confounders. However, our results may be limited by poor recall of exercise habits.

Although we assessed alcohol use in several ways, we found no association between bone mass and current or previous use, intensity of use, or total lifetime alcohol intake. Chronic alcohol abuse has been associated with reduced bone mass [46], but the evidence about moderate use is conflicting [8, 32, 47, 48]. Our data suggest that modest alcohol use is not a risk factor for decreased bone mass in our cohort.

Our findings suggest smoking is deleterious to bone mass. After adjusting for several potential confounders, current smokers had less bone mass than nonsmokers. However, no evidence existed that heavy smokers had lower bone mass than moderate smokers. Although some studies have not shown an effect of smoking on bone mass [47], our results are consistent with a number of other studies that have found smoking to be associated with a decreased bone mass [31, 48, 49] or more rapid bone loss [50].

We found a statistically significant inverse association between caffeine consumption and bone mass. Caffeine may increase calcium excretion by the kidney [51, 52], and this may account for the slight decrease in bone mass we found in those women with a greater lifetime intake of caffeine. The clinical significance of this association is unclear because most women do not consume such large quantities of caffeine. Several previous studies [53] have found lower bone mass in postmenopausal women with greater caffeine intake, but others have not [32, 54, 55].

Medical Conditions

We found that gastrectomy was associated with substantially lower bone mass, even after adjusting for smoking history; others have found similar deleterious effects on bone [56, 57]. The association between diabetes and bone mass is controversial [58-60], but our findings suggest that, even after adjusting for weight, bone mass is increased in non–insulin-dependent diabetics. Others have found a relation between osteoarthritis and increased bone mass [61], but we found no association after adjusting for weight. Decreased bone mass has been reported in rheumatoid arthritis [62, 63]. We found a similar relation between self-reported rheumatoid arthritis and bone mass of the midradius and calcaneus but not the distal radius. Although hyperthyroidism was associated with decreased bone mass at the midradius and calcaneus, the bone mass of women who took thyroid hormone did not differ when compared with women who did not take thyroid hormone.

Medications

Similar to a number of other studies [47, 53], estrogen replacement therapy and duration of use were strongly associated with increased bone mass. Smaller studies [64-67] of the impact of progestins on bone have shown either no or slightly beneficial effects on bone mass. Our study suggests that progestins do not add to the protective effects of estrogen on bone.

Our results confirm a number of previous studies [47, 68, 69] showing an association between thiazide diuretic use and increased bone mass. Presumably, this protective effect reflects the blunting of calcium excretion by these agents. We found the beneficial effects of thiazide use to be approximately one half as great as those for estrogen.

The large sample size in our study enabled us to detect even weak associations. For example, we found no association between bone mass and total calcium intake, and the 95% confidence interval (0.2% to 0.5%) for the change in bone mass per 400 mg/d of calcium intake was very narrow. However, some of our statistically significant associations may not result in clinically important effects on fracture risk for an individual woman. Based on our previous estimates of the relation between bone mass and overall risk for fracture [5], we estimate that a 2.1% decrease in distal radius bone mass (which was found in current smokers) would increase the overall risk for fracture by about 5% (assuming that the increased risk for fracture in smokers is mediated entirely through changes in bone mass). A clinically meaningful decrease in overall fracture risk, a 10% to 20% reduction for instance, occurs when distal radius bone mass increases by 5% to 9%.

The variance of bone mass explained by our models was modest but similar to smaller studies of the correlates of bone mass [8]. Although measurement error may account for some of this unexplained variance, it is likely that unmeasured genetic factors account for a substantial proportion [33, 34]. Weight accounted for the greatest proportion of variance at each of the three sites, particularly at the calcaneus, a direct weight-bearing site.

This study has several limitations. Our cohort consisted of generally healthy, independently living volunteers older than 65 years, and these results may not be generalizable to other groups of patients. We assessed a number of health habits by questionnaire, and the results may have been influenced by inaccurate recall [70]. In addition, we analyzed appendicular bone mass, and bone mass of the hip and spine may better predict fractures at those sites. Studies of the determinants of decreased bone mass at these sites are in progress. Finally, exploratory analyses tend to overestimate the strength of association between risk factors and conditions [71]. We tested for this effect by deriving our results in one half of the cohort and testing them in a second half of the cohort and found only a few trivial differences between the results of these two analyses.

Conclusions

This analysis of a large cohort of older women confirmed a number of previously suggested risk factors for decreased bone mass. Increased weight and greater muscle strength were among the strongest predictors of increased bone mass whereas age was one of the strongest predictors of reduced bone mass. Estrogen use, thiazide use, and non–insulin-dependent diabetes were associated with increased bone mass, whereas smoking, maternal fracture, and gastric surgery were associated with decreased bone mass. Caffeine intake was also associated with decreased bone mass, but the clinical significance of this finding is unclear. Unlike a number of smaller studies, we found no association between bone mass and alcohol use, physical activity, or parity. Our findings indicate that, with the exception of a few factors strongly associated with bone mass, most of the individual risk factors we examined do not significantly alter the overall risk for fracture. Finally, the aggregate effect of all these factors was modest, and much of the variance in bone mass among elderly women is not explained by the risk factors we measured.

Appendix. Investigators in the Study of Osteoporotic Fractures Research Group

University of California, San Francisco (Coordinating Center): S.R. Cummings, M.C. Nevitt, D. Black, H.K. Genant, C. Arnaud, W. Browner, K. Faulkner, C. Fox, C. Gluer, S. Harvey, S.B. Hulley, L. Palermo, D. Seeley, and P. Steiger.

University of Maryland: R. Sherwin, J. Scott, K. Fox, J. Lewis, M. Bahr, S. Trusty, B. Hohman, L. Emerson, D. Rebar, and E. Oliner.

University of Minnesota: K. Ensrud, R. Grimm, Jr., C. Bell, D. Thomas, K. Jacobson, S. Jackson, E. Mitson, L. Stocke, and P. Frank.

University of Pittsburgh: J.A. Cauley, L.H. Kuller, L. Harper, M. Nasim, C. Bashada, L. Buck, A. Githens, D. Medve, and S. Rudovsky.

The Kaiser Permanente Center for Health Research, Portland, Oregon: T.M. Vogt, W.M. Vollmer, H. Glauber, E. Orwoll, J. Blank, B. Mastel-Smith, R. Bright, and J. Downing.