Learn more: PMC Disclaimer | PMC Copyright Notice
Diabetes and Femoral Neck Strength: Findings from The Hip Strength Across the Menopausal Transition Study
Abstract
Context:
Diabetes mellitus is associated with increased hip fracture risk, despite being associated with higher bone mineral density in the femoral neck.
Objective:
The objective of the study was to test the hypothesis that composite indices of femoral neck strength, which integrate dual-energy x-ray absorptiometry derived femoral neck size, femoral neck areal bone mineral density, and body size and are inversely associated with hip fracture risk, would be lower in diabetics than in nondiabetics and be inversely related to insulin resistance, the primary pathology in type 2 diabetes.
Design:
This was a cross-sectional analysis.
Setting and Participants:
The study consisted of a multisite, multiethnic, community-dwelling sample of 1887 women in pre- or early perimenopause.
Outcome Measurements:
Composite indices for femoral neck strength in different failure modes (axial compression, bending, and impact) were measured.
Results:
Adjusted for age, race/ethnicity, menopausal stage, body mass index, smoking, physical activity, calcium and vitamin D supplementation, and study site, diabetic women had higher femoral neck areal bone mineral density [+0.25 sd, 95% confidence interval (CI) (+0.06, +0.44) sd] but lower composite strength indices [−0.20 sd, 95% CI (−0.38, −0.03) sd for compression, −0.19 sd, 95% CI (−0.38, −0.003) sd for bending, −0.19 sd, 95% CI (−0.37, −0.02) sd for impact] than nondiabetic women. There were graded inverse relationships between homeostasis model-assessed insulin resistance and all three strength indices, adjusted for the same covariates.
Conclusions:
Despite having higher bone density, diabetic women have lower indices of femoral neck strength relative to load, consistent with their documented higher fracture risk. Insulin resistance appears to play an important role in bone strength reduction in diabetes.
Diabetes mellitus, which is rapidly increasing in prevalence worldwide, is associated with not only increased cardiovascular morbidity but also increased risk of hip and other osteoporotic fractures (1–3). Although the consequences of hip fracture are grave due to its high incidence, morbidity, cost, and associated increase in mortality (4, 5), the impact of hip fracture is even greater in diabetic patients because of delayed healing and increased risk of complications (6, 7). Hence, it is imperative to correctly identify and target for preventive interventions individuals (and especially those with diabetes) who are at increased risk for hip fractures.
Low areal bone mineral density (BMD) in the femoral neck, based on measurement by dual-energy x-ray absorptiometry (DXA), is widely used as a validated and reliable predictor of hip fracture risk (8): Individuals with 1 sd lower femoral neck BMD generally have a 2- to 3-fold higher risk of fracture (9). However, several recent longitudinal cohort studies (and a metaanalysis) have demonstrated that despite having higher areal BMD than nondiabetic individuals, patients with diabetes are at increased risk for fracture in the hip (2) and elsewhere (10–12). In addition, a recent analysis of three prospective cohorts showed that diabetic patients had higher fracture risk than nondiabetic individuals for a given femoral neck BMD T score or for a given FRAX (Fracture Risk Algorithm) score, which incorporates important risk factors for osteoporotic fracture including femoral neck BMD T score and body mass index (BMI) (13). The dissociation between BMD and fracture risk suggests that knowledge of BMD alone is not sufficient to adequately assess bone strength and predict future fracture risk.
Independent of BMD, body size, and bone size contribute to fracture risk (14–16): bone size contributes to bone structural strength (just as the strength of engineering structures depends on both material density and structure size), and body size determines the forces that bone is exposed to in a fall. Both body size and bone size are affected in individuals with diabetes (17, 18). But whether the greater BMD and changes in bone size are sufficient to compensate for the higher forces due to higher body mass in diabetes is unknown.
Composite indices of femoral neck strength integrate body size, femoral neck size, and femoral neck areal BMD to capture the major structural contributions to strength relative to loads applied to bone during falls (19). The composite strength indices are inversely associated with incident hip fracture risk in community-dwelling older Caucasian women (19) and in young U.S. Caucasian and Chinese adults (20) and have been used as a measure of femoral neck strength (21–24).
In addition, unlike race/ethnicity differences in areal BMD, race/ethnicity differences in the composite strength indices are consistent with documented race/ethnicity differences in fracture risks (25).
We hypothesized that even if femoral neck BMD were higher, femoral neck composite strength indices would be lower in diabetes and prediabetes than in women with neither condition. We also hypothesized that the femoral neck composite strength indices would be inversely related to insulin resistance, the primary pathology in type 2 diabetes.
Materials and Methods
The Study of Women's Health Across the Nation (SWAN) is a multisite, multiethnic longitudinal study designed to characterize the biological and psychosocial changes that occur during the menopausal transition in a community-based cohort of 3302 women. The eligibility criteria have been described in detail elsewhere (26). Briefly, criteria for entry included age 42–53 yr; intact uterus and at least one ovary; not currently using sex steroid hormone preparations (including oral contraceptives); and at least one menstrual period in the 3 months before screening. Participants were enrolled in 1996–1997 at seven clinical sites: Boston, MA; Chicago, IL; Detroit, MI; Pittsburgh, PA; Los Angeles, CA; Newark, NJ; and Oakland, CA. All sites enrolled Caucasians, and each site also enrolled women belonging to one prespecified minority ethnic group: African American in Boston, Chicago, Detroit, and Pittsburgh; Japanese in Los Angeles; Hispanic in Newark; and Chinese in Oakland. All participants at five of the seven SWAN study sites (excluding the Chicago and Newark sites) who weighed no more than 136 kg (the maximum allowed on the DXA machines) were invited to have DXA scans of the hip and lumbar spine and constituted the SWAN bone cohort of 2413 women. Use of osteoporosis agents was one of the prespecified exclusion criteria, but no one in the cohort used osteoporosis agents at baseline.
Femoral neck size measurements on archived hip DXA scans were made, as part of the SWAN Hip Strength Substudy, on 1940 women in the SWAN bone cohort who had a baseline hip DXA scan and at least two follow-up hip DXA scans by SWAN visit 10 (yr 2006–2007). Women who reported oral steroid use (n = 26), anorexia (n = 17), bulimia (n = 6), or a history of insulin use before age 30 yr (n = 4) were excluded from the analysis, leaving an analytic sample of 1887 women (936 Caucasian, 496 African American, 237 Japanese, and 218 Chinese). All measures used in this analysis come from the SWAN baseline visit, when the participants were in premenopause or early perimenopause and menopause-related bone density decline had not commenced (27).
The SWAN parent study and hip strength substudy protocols were approved by the institutional review board at each site, and all participants gave written informed consent.
Measurements
Standardized interviews and self-reported questionnaires were used to obtain the following information: age (years), smoking status (current, ex-smoker, or never smoking), race/ethnicity, menopausal status (premenopause vs. early perimenopause), level of physical activity [assessed with a modified Baecke interview, which provides a summary score of active living, home, and recreational physical activity (28)], use of calcium and vitamin D supplements (none vs. any), and use of prescription medications. Height and weight were measured using a fixed stadiometer and a digital scale with the participants wearing light clothing and no shoes. BMI was calculated as weight (in kilograms) divided by the square of height (in meters).
Insulin resistance and diabetes status
Serum insulin and glucose was measured from blood drawn after an overnight 12-h fast. Serum insulin was measured using a RIA procedure (Coat-a-Count; Diagnostic Products Corp., Los Angeles, CA). The quality control program for serum insulin in SWAN has been previously described (29). Serum glucose was measured using a hexokinase-coupled reaction (Roche Molecular Biochemicals Diagnostics, Indianapolis, IN). Homeostasis model assessment of insulin resistance (HOMA-IR) was calculated as fasting serum glucose (in milligrams per deciliter) times fasting serum insulin (in microunits per milliliter) divided by the constant 405 (30). HOMA-IR was not calculated in 14 women who used exogenous insulin.
Those who reported the current use of diabetes medications or had fasting glucose 126 mg/dl or higher were classified diabetic; those who did not meet diabetes criteria but had fasting glucose higher than or equal to 100 mg/dl were classified as prediabetic; the rest were classified as nondiabetic (31). It is important to note that the nondiabetic group does not include the prediabetic women.
Composite indices of femoral neck strength
DXA scans of the hip were acquired with Hologic QDR 4500 or Hologic QDR 2000 scanners (Hologic Inc., Waltham, MA) and Osteodyne's hip positioner system (Osteodyne, Inc., Research Triangle, Park, NC). The quality control program for DXA in SWAN has been previously described (27). The projected (areal) BMD in the femoral neck was recorded. Femoral neck axis length (FNAL) is the distance along the femoral neck axis from the lateral margin of the base of the greater trochanter to the apex of the femoral head. Femoral neck width (FNW) is the smallest thickness of the femoral neck along any line perpendicular to the femoral neck axis (Fig. 1.) Measurements of FNAL and FNW were made on the DXA scanner monitor using pixel sizes provided by the manufacturer (Hologic).
Composite indices of femoral neck strength are designed to capture the femoral neck's ability to withstand load during a fall and are computed from the femoral neck size (FNAL and FNW), femoral neck areal BMD, and body size using the following formulas (19):
All three indices were recorded in units of grams per kilogram per meter. With BMD measured in grams per square centimeter, FNW and FNAL in centimeters, weight in kilograms, and height measured in meters, CSI and BSI need to be scaled by 100 to get values in units of grams per kilogram per meter. CSI reflects the ability of the femoral neck to withstand an axial compressive load, BSI reflects its ability to withstand bending forces, and impact strength index reflects the ability of the femoral neck to absorb the energy of impact in a fall from standing height. To examine reproducibility, 20 women volunteers were scanned twice on the same day with repositioning, and femoral neck areal BMD, FNW, and FNAL were measured twice. Intraclass correlation coefficients for the three indices were each greater than 0.98.
Statistical analysis
Means of the composite strength indices and areal BMD in diabetic and prediabetic women were compared with the means in nondiabetic women, using unmatched t tests. We used multiple linear regression to adjust for potential confounders. Because our aim was to examine whether differences in the composite strength indices between diabetic and nondiabetic women were consistent with known differences in hip fracture risk, we adjusted for the following factors that could confound the association between bone strength and hip fracture risk: age; menopause status (premenopause vs. early perimenopause); race/ethnicity; BMI (four level categorical, BMI < 22 kg/m2, 22 = < BMI < 25 kg/m2, 25 = < BMI < 30 kg/m2, 30 = < BMI); DXA scanner type (Hologic models 4500 vs. 2000); and study site. In an additional step, we included controls for smoking status (never, past, current smoking), level of physical activity (above median vs. below median), and use of calcium and vitamin D supplements (none vs. any) to explore their contributions to variation in femoral neck strength across the diabetes spectrum.
The nonparametric LOESS method was used to assess the functional forms of the associations of the strength measures (femoral neck BMD and three composite strength indices) with log-transformed HOMA-IR. LOESS plots revealed that all four relationships were piecewise linear with an inflection (change of slope) at the log HOMA-IR value of 1.4 (corresponding to HOMA-IR of 4.1) (Fig. 2). We then fit piece-wise linear spline models to each of the composite indices and to femoral neck BMD as a function of log(HOMA-IR) with a fixed knot at 1.4, and adjusted for age, menopause status, race/ethnicity, BMI categories, smoking status, physical activity level, supplementary vitamin D, supplementary calcium, type of DXA scanner, and study site.
Results
The study sample was similar to the SWAN bone cohort (from which it was drawn) with respect to major characteristics, except that the study sample included slightly fewer diabetic women and had lower body weight and BMI than the remainder of the bone cohort (Table 1). Of the 1887 women in the study sample, 81 women (4.3%) were diabetic and 291 women (15.4%) were prediabetic. The diabetic women had higher values of HOMA-IR (mean 9.20 μU/ml with sd 6.13) than the prediabetic women (mean 3.69, sd 2.24) or the nondiabetic women (mean 1.90, sd 1.13).
Table 1.
Characteristics of study participants compared with the SWAN bone cohort
| Characteristics | Study sample (n = 1887)a | SWAN bone cohort (n = 2413) |
|---|---|---|
| Age (yr) | 45.9 (2.7) | 45.8 (2.7) |
| Race | ||
| Caucasian (%) | 49.6 | 49.6 |
| African American (%) | 26.3 | 28.4 |
| Chinese (%) | 11.6 | 10.4 |
| Japanese (%) | 12.5 | 11.7 |
| Height (cm) | 162.2 (6.5) | 162.5 (6.6) |
| Weight (kg) | 72.6 (19.3) | 74.3 (21.3) |
| BMI (kg/m2) | ||
| BMI < 22 (%) | 22.4 | 21.5 |
| 22 = < BMI < 25 (%) | 24.0 | 22.9 |
| 25 = < BMI < 30 (%) | 24.3 | 23.9 |
| 30 = < BMI (%) | 29.4 | 31.7 |
| Menopause status | ||
| Premenopausal (%) | 56.2 | 54.1 |
| Early perimenopausal (%) | 43.8 | 45.9 |
| Smoking status | ||
| Never smoked (%) | 59.4 | 57.8 |
| Ex-smoker (%) | 25.6 | 25.9 |
| Current smoker (%) | 15.0 | 16.4 |
| Physical activity | ||
| Above median level (%) | 49.8 | 50.4 |
| Below median level (%) | 50.2 | 49.6 |
| Medications | ||
| Supplementary vitamin D (%) | 38.8 | 38.2 |
| Supplementary calcium (%) | 45.3 | 44.4 |
| Diabetic status | ||
| Diabetes (%) | 4.3 | 5.5 |
| Prediabetes (%) | 15.4 | 15.8 |
| Nondiabetes (%) | 80.3 | 78.6 |
| Femoral neck bone mineral density (g/cm2) | 0.84 (0.13) | 0.84 (0.13) |
| Femoral neck axis length (cm) | 8.96 (0.51) | |
| Femoral neck width (cm) | 2.75 (0.20) | |
| Compression strength index (g/kg · m) | 3.3 (0.64) | |
| Bending strength index (g/kg · m) | 1.02 (0.22) | |
| Impact strength index (g/kg · m) | 0.18 (0.04) | |
For continuous variables, mean and sd are shown. Percentages may not add up to 100 because of rounding errors.
Crude mean femoral neck areal BMD was significantly higher in both prediabetic and diabetic women than in nondiabetic women and also in diabetic women compared with prediabetic women (Table 2, model 1). In sharp contrast, the unadjusted means of all three composite strength indices of the femoral neck were significantly lower in diabetic and prediabetic women than in nondiabetic women and in diabetic than in prediabetic women. Differences in strength index unadjusted means were more than 7% or 0.3 sd between prediabetic and nondiabetic women and more than 16% (0.8 sd) between diabetic and nondiabetic women (Table 2, model 1).
Table 2.
Differences (with 95% confidence intervals) in unadjusted and adjusted means of femoral neck strength measures by diabetes status
| Femoral neck strength measures | Nondiabetica | Prediabetic | Diabetic |
|---|---|---|---|
| Bone mineral density (sd 0.13) | |||
| Model 1 | 0.83 g/cm2 | +0.05 (0.03, 0.06)b | +0.13 (0.11, 0.16)b |
| Model 2 | 0.83 g/cm2 | −0.0001 (−0.01, 0.01) | +0.04 (0.01, 0.06)c |
| Model 3 | 0.83 g/cm2 | 0.002 (−0.01, 0.02) | +0.03 (0.01, 0.06)c |
| Compression strength index (sd 0.64) | |||
| Model 1 | 3.4 g/kg · m | −0.27 (−0.34, −0.19)b | −0.57 (−0.71, −0.43)b |
| Model 2 | 3.4 g/kg · m | −0.05 (−0.11, 0.008) | −0.12 (−0.23, −0.02)d |
| Model 3 | 3.4 g/kg · m | −0.05 (−0.11, 0.02) | −0.13 (−0.24, −0.02)d |
| Bending strength index (sd 0.22) | |||
| Model 1 | 1.04 g/kg · m | −0.069 (−0.096, −0.042)b | −0.171 (−0.220, −0.126)b |
| Model 2 | 1.04 g/kg · m | −0.016 (−0.038, 0.007) | −0.043 (−0.083, −0.003)d |
| Model 3 | 1.04 g/kg · m | −0.016 (−0.039, 0.007) | −0.042 (−0.084, −0.0006)d |
| Impact strength index (sd 0.04) | |||
| Model 1 | 0.19 g/kg · m | −0.015 (−0.020, −0.011)b | −0.033 (−0.041, −0.024)b |
| Model 2 | 0.19 g/kg · m | −0.003 (−0.007, 0.0004) | −0.007 (−0.013, −0.001)d |
| Model 3 | 0.19 g/kg · m | −0.003 (−0.006, 0.001) | −0.007 (−0.014, −0.001)d |
Model 1: unadjusted. Model 2: adjusted for race/ethnicity, menopause status, BMI categories, age, DXA scanner type, and study site. Model 3: adjusted for race/ethnicity, menopause status, BMI categories, age, smoking status, physical activity, supplementary vitamin D use, calcium use, DXA scanner type, and study site. Nonoverlapping confidence intervals in the last two columns are a sufficient (but not necessary) condition for differences between prediabetic and diabetic women in model 1 at the P = 0.05 level significance.
After adjusting for covariates, the magnitude of the differences between prediabetic and nondiabetic women (in both BMD and composite strength indices) diminished and became statistically nonsignificant. However, differences between diabetic and nondiabetic women in femoral neck areal BMD and composite strength indices persisted after adjusting for race/ethnicity, age, menopause status, BMI categories, DXA scanner type, and study site (Table 2, model 2). Differences in adjusted means of the composite strength indices were approximately 0.2 sd in favor of nondiabetic women compared with diabetic women. Further adjusting for factors that could potentially explain observed femoral neck strength differences by diabetes status did not diminish the magnitude of the differences in femoral neck composite strength indices and areal BMD between diabetic and nondiabetic women (Table 2, model 3).
In piece-wise linear regression of each of the femoral neck composite strength indices and BMD as a function of log(HOMA-IR) with a fixed knot at log(HOMA-IR) of 1.4, after adjusting for covariates (age, race/ethnicity, menopause status, BMI categories, smoking status, physical activity, supplementary vitamin D and calcium, DXA scanner type, and study site), the change in slope at the knot became statistically nonsignificant in all four models (P value range 0.10–0.92), and the knot was subsequently dropped. Resulting multiple linear regression models fit simple linear relationships between log HOMA-IR and the four outcomes, adjusted for the same covariates. Strong inverse associations were seen between log(HOMA-IR) and each of the three composite strength indices; every doubling of HOMA-IR was associated with more than 2.4% decline in composite strength indices (Table 3).
Table 3.
Adjusted associations of insulin resistance with femoral neck strengtha
| Femoral neck strength measure | Effect size per doubling of HOMA-IR | 95% Confidence interval | P value |
|---|---|---|---|
| Bone mineral density | 0.005 g/cm2 | (−0.002, 0.012) | 0.14 |
| Compression strength index | −0.086 g/kg · m | (−0.12, −0.06) | <0.001 |
| Bending strength index | −0.030 g/kg · m | (−0.042, −0.019) | <0.001 |
| Impact strength index | −0.005 g/kg · m | (−0.007, −0.003) | <0.001 |
Sensitivity analysis
Because diabetic women had very high values of HOMA-IR, which could influence the association between log(HOMA-IR) and each of the strength measures, we repeated the HOMA-IR analysis as a sensitivity analysis after excluding diabetic women (Table 4). This sensitivity analysis yielded almost identical results, supporting our finding in the main analysis.
Table 4.
Adjusted associations of insulin resistance with femoral neck strength in prediabetic and nondiabetic womena
| Femoral neck strength measure | Effect size per doubling of HOMA-IR | 95% confidence interval | P value |
|---|---|---|---|
| Bone mineral density | 0.004 g/cm2 | (−0.004, 0.011) | 0.37 |
| Compression strength index | −0.091 g/kg · m | (−0.13, −0.06) | <0.001 |
| Bending strength index | −0.033 g/kg · m | (−0.046, −0.020) | <0.001 |
| Impact strength index | −0.006 g/kg · m | (−0.008, −0.004) | <0.001 |
Discussion
In this cross-sectional study of 1887 premenopausal/early perimenopausal women from diverse ethnic groups, we examined differences in composite indices of femoral neck strength by diabetes status. These indices combine femoral neck areal BMD, body height and weight, and femoral neck size to quantify strength (resistance to fracture forces) relative to load (forces placed on the hip during a fall). Consistent with our hypothesis, we found that diabetic women had lower values of composite strength indices than nondiabetic women, adjusted for age, menopause status, BMI, race/ethnicity, smoking, physical activity, use of calcium and vitamin D supplements, type of DXA scanner, and study site.
Our finding of lower femoral neck strength in diabetic women is consistent with a previous study that showed that in SWAN, women with diabetes had a 2-fold increased risk of fracture over the menopausal transition compared with nondiabetic women (10) and other studies that documented the increased rates of fracture in older diabetic individuals (11, 12, 32). Our results suggest that although women with diabetes have higher areal BMD, their femoral neck strength is low relative to the load they bear. We did not, however, find a difference in femoral neck composite strength indices between nondiabetic and prediabetic women, which is consistent with previous studies that have not found differences in fracture risk between prediabetic and nondiabetic individuals (11, 33). Our finding was incongruent with a previous small, computed tomography based study that failed to find difference in bone strength relative to loads between diabetic and nondiabetic controls (34). However, this study had limited power to detect such a difference because of its small sample size.
The current study also demonstrated a strong inverse association between femoral neck composite strength indices and insulin resistance, suggesting that insulin resistance may be a cardinal pathway to lower bone strength and increased fracture risk in type 2 diabetes. Although longitudinal studies have found no increase in fracture risk in participants with impaired glucose tolerance (12, 33, 35), it has been noted that the serum level of osteocalcin (an osteoblast derived bone formation marker) is associated with improved glucose tolerance in healthy men (36). Indeed, efficient insulin signaling in osteoblasts is thought to promote osteoblast function via osteocalcin decarboxylation. Thus, declines in insulin sensitivity may be closely linked to declines in bone strength.
Our study has some limitations that need to be acknowledged. The composite indices of hip strength used in this study are structural measures based on macroscopic measurements from DXA scans and ignore microscopic features such as differences in the quality of cancellous mineralization and microarchitecture, both of which are important determinants of bone strength and are adversely affected by diabetes (37, 38). We also made the simplifying assumption that mineral mass in the femoral neck is largely in the cortex since details of bone composition are not readily available from DXA scans, but diabetes may have differential effects on cortical and trabecular bone (17). Although the strength indices have not been validated against laboratory measures of mechanical strength, they have been shown to be inversely associated with fracture risk in at least two studies. Although we adjusted for major factors known to influence falls and fracture risk, there are several factors we could not control for because the data are not available in SWAN: women with diabetes are at increased risk for fall because of peripheral neuropathy affecting balance, knee osteoarthritis related to excess weight, and retinopathy affecting vision (39, 40). All of these factors would only increase fracture risk in diabetes beyond the increase predicted by their lower values of femoral neck composite strength indices. Lastly, this cohort is limited to women, and studies have shown that there are differences in bone strength between male diabetics and nondiabetics as well (13).
Despite these limitations, this study is, to our knowledge, the first to demonstrate that consistent with their documented higher rates of osteoporotic fracture, diabetic women indeed have lower femoral neck strength relative to the loads they bear. This suggests that unlike areal bone density, the composite indices of femoral neck strength may be able to more correctly stratify individuals by their risk for hip fracture. Further research is needed to examine this contention and demonstrate the fracture prediction ability of the femoral neck composite strength indices in longitudinal cohorts that include both diabetic and nondiabetic men and women.
Acknowledgments
SWAN was supported by the National Institutes of Health (NIH), Department of Health and Human Services, through the National Institute on Aging (NIA), the National Institute of Nursing Research, and the NIH Office of Research on Women's Health (Grants NR004061;, AG012505;, AG012535;, AG012531;, AG012539;, AG012546;, AG012553;, AG012554;, and AG012495). The Hip Strength Through the Menopausal Transition was supported by NIA Grant AG026463. The SWAN Hip Strength Substudy was supported by NIA Grant AG028748. The current study was also supported by NIA Grant AG033067. S.I. was supported by the Veterans Affairs Greater Los Angeles Healthcare System Geriatric Research, Education, and Clinical Center; and the Veterans Affairs Advanced Geriatrics Fellowship.
The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the NIA, National Institute of Nursing Research, Office of Research on Women's Health, Veterans Affairs, or the NIH. Clinical centers included the following: University of Michigan (Ann Arbor, MI), Siobán Harlow, principal investigator, 2011, MaryFran Sowers, principal investigator, 1994–2011; Massachusetts General Hospital (Boston, MA), Joel Finkelstein, principal investigator, 1999 to the present; Robert Neer, principal investigator, 1994–1999; Rush University (Rush University Medical Center, Chicago, IL), Howard Kravitz, principal investigator, 2009 to the present; Lynda Powell, principal investigator, 1994–2009; University of California, Davis/Kaiser, Ellen Gold, principal investigator; University of California, Los Angeles, Gail Greendale, principal investigator; Albert Einstein College of Medicine (Bronx, NY), Carol Derby, principal investigator, 2011, Rachel Wildman, principal investigator, 2010–2011, Nanette Santoro, principal investigator, 2004–2010; University of Medicine and Dentistry, New Jersey Medical School (Newark, NJ), Gerson Weiss, principal investigator, 1994–2004; and the University of Pittsburgh (Pittsburgh, PA), Karen Matthews, principal investigator. The NIH Program Office included the following: NIA (Bethesda, Maryland), Sherry Sherman, 1994 to the present; Marcia Ory, 1994–2001; and National Institute of Nursing Research (Bethesda, Maryland) Program Officers. The Central Laboratory included the following: University of Michigan, Ann Arbor, Daniel McConnell (Central Ligand Assay Satellite Services). The Coordinating Center included the following: University of Pittsburgh (Pittsburgh, PA), Kim Sutton-Tyrrell, principal investigator, 2001 to the present; New England Research Institutes (Watertown, MA), Sonja McKinlay, principal investigator, 1995–2001. The Steering Committee included the following: Susan Johnson, current chair; and Chris Gallagher, former chair. We thank the study staff at each site and all the women who participated in SWAN.
Disclosure Summary: There is no conflict of interest to be disclosed for any of the authors regarding this manuscript.
Footnotes
Abbreviations:
- BMD
- Bone mineral density
- BMI
- body mass index
- BSI
- bending strength index
- CSI
- compression strength index
- DXA
- dual-energy x-ray absorptiometry
- FNAL
- femoral neck axis length
- FNW
- femoral neck width
- HOMA-IR
- homeostasis model assessment of insulin resistance
- SWAN
- Study of Women's Health Across the Nation.