Abstract
Introduction
For older men who undergo bone mineral density (BMD) testing, the optimal osteoporosis screening schedule is unknown. Time-to-disease estimates are necessary to inform screening intervals.
Methods
A prospective cohort study of 5,415 community-dwelling men aged ≥65 years without hip or clinical vertebral fracture or antifracture treatment at baseline was conducted. Participants had concurrent BMD and fracture follow-up between 2000 and 2009, and additional fracture follow-up through 2014. Data were analyzed in 2015. Time to incident osteoporosis (lowest T-score ≤ −2.50) for men without baseline osteoporosis, and time to hip or clinical vertebral fracture or major osteoporotic fracture for men without or with baseline osteoporosis, were estimated.
Results
Nine men (0.2%) with BMD T-scores > −1.50 at baseline developed osteoporosis during follow-up. The adjusted estimated time for 10% to develop osteoporosis was 8.5 (95% CI=6.7, 10.9) years for those with moderate osteopenia (lowest T-score, −1.50 to −1.99) and 2.7 (95% CI=2.1, 3.4) years for those with advanced osteopenia (lowest T-score, −2.00 to −2.49) at baseline. The adjusted times for 3% to develop a first hip or clinical vertebral fracture ranged from 7.1 (95% CI=6.0, 8.3) years in men with baseline T-scores > −1.50 to 1.7 (95% CI=1.0, 3.1) years in men with baseline osteoporosis.
Conclusions
Men aged 65 years and older with femoral neck, total hip, and lumbar spine BMD T-scores > −1.50 on a first BMD test were very unlikely to develop osteoporosis during follow-up. Additional BMD testing may be most informative in older men with T-scores ≤ −1.50.
Introduction
Osteoporosis-related fractures impose a significant burden to older men.1 One in four elderly men will sustain a fragility fracture during their remaining lifetime,2 with an exponential increase in fracture incidence after age 75 years.3 Men also have a higher 1-year mortality rate after hip fracture compared with women.4–6 A 2012 RCT of older men with osteoporosis treated with zoledronic acid demonstrated significant reduction in the primary endpoint of radiographic vertebral fracture.7 Similarly, an earlier meta-analysis suggested that risedronate treatment in men aged 55–76 years with osteoporosis by bone mineral density (BMD) criteria or with secondary causes of osteoporosis, with or without existing vertebral fracture, was associated with radiographic vertebral fracture reduction.8 Owing to the serious consequences of fracture and available antifracture treatment, several organizations9–12 have endorsed identification of osteoporosis in older men through BMD testing, but an interval of measurement has not been recommended.
In 2011, the U.S. Preventive Services Task Force concluded evidence was insufficient to assess the balance of benefits and harms of osteoporosis screening in men (Grade I),13 thus highlighting a need for developing a screen/treat strategy to identify elderly men at increased fracture risk. A 2007 cost-effectiveness analysis suggested that a screen/treat strategy may be cost effective for men aged 65 years or older with a prior clinical fracture and for men aged 80 years or older without a prior fracture, assuming a societal willingness to pay $50,000 per quality-adjusted life year gained.14 It is unknown what proportion of men will develop osteoporosis prior to experiencing a fragility fracture. In addition, the time frame for men at specific baseline BMD levels to develop a disabling fracture event is uncertain.
The objective of this study was to inform decisions about BMD testing intervals by estimating the time to development of osteoporosis by BMD criteria and the time to development of hip or clinical vertebral fracture in a cohort of 5,415 men aged 65 years and older without prior hip or clinical vertebral fracture or osteoporosis treatment. Based on a previous study of older women,15 it was hypothesized that men with baseline T-scores ≤ −1.50 would develop osteoporosis significantly more rapidly than men with higher T-scores. Men with baseline osteoporosis were expected to have a more rapid transition to fracture than men with T-scores > −2.50.
Methods
Setting and Participants
The Osteoporotic Fractures in Men (MrOS) Study cohort comprised 5,994 community-dwelling, ambulatory men aged ≥65 years who met these inclusion criteria at baseline:
ability to walk without the assistance of another;
absence of bilateral hip replacements;
ability to provide self-reported data;
planned residence near a clinical site for the duration of the study;
absence of a medical condition that (in the judgment of the investigator) would result in imminent death; and
ability to understand and sign an informed consent.
Participants had to provide written informed consent, complete a self-administered questionnaire, attend the clinic visit, and complete at least the anthropometric, dual energy x-ray absorptiometry (DXA), and vertebral x-ray procedures. Recruitment relied primarily on community-based mailings, supplemented at some sites with local strategies (advertisements, presentations) to enhance the recruitment of minority groups. The study enrolled participants and completed the baseline examination from March 2000 through April 2002.16,17 The study protocol and consent documents were approved by IRB review committees at six clinical sites in the U.S. (Birmingham, AL; Minneapolis, MN; Palo Alto, CA; Pittsburgh, PA; Portland, OR; San Diego, CA).
The plan for the longitudinal analysis was approved by the IRB for the University of North Carolina. DXA BMD imaging was obtained over 8.7 years of follow-up, including the baseline visit, 2000–2002; Year 2, 2002–2003; Year 3, 2003–2005; Year 5, 2005–2006; and Year 7, 2007–2009. Participants were also followed every 4 months by postcard or telephone to ascertain fractures, with fracture follow-up through November 2014. Data were analyzed in 2015. All fractures were adjudicated by central review of radiology reports.16 Men were eligible for the primary analysis if they were representative of the screened population (i.e., they had no history of osteoporosis diagnosis, hip or clinical vertebral fracture, or antifracture treatment at baseline). Men who had osteoporosis without antifracture treatment or hip or clinical vertebral fracture at baseline were eligible to participate in the osteoporosis-to-fracture analysis.
Measures
The 1994 WHO Technical Report18 defined osteoporosis as BMD T-score ([BMD of participant – mean BMD of young reference population]/SD of BMD of reference population) ≤ −2.5 at the lumbar spine, hip, or distal radius. The International Osteoporosis Foundation19 and the International Society for Clinical Densitometry20 recommend that a uniform Caucasian (non–race adjusted) female reference database should be used to calculate T-scores for men of all ethnic groups. The International Osteoporosis Foundation justifies this convention by noting that for any age and BMD at the femoral neck, the risk of hip fracture or a major osteoporotic fracture is approximately the same in men and women.19 Accordingly, T-scores were calculated at the femoral neck, total hip, and lumbar spine using National Health and Nutrition Examination Study III BMD norms for white women aged 20–29 years21,22 in the main analysis.
Participants were categorized into these baseline T-score groups: higher BMD—lowest T-score at the femoral neck, total hip, or lumbar spine > −1.50; moderate osteopenia—lowest T-score at any site −1.50 to −1.99; advanced osteopenia—lowest T-score at any site −2.00 to −2.49; and osteoporosis—lowest T-score at any site ≤ −2.50.
The primary outcome was the time interval for 10% of participants without osteoporosis at baseline to develop osteoporosis before initiation of a Food and Drug Administration–approved agent for the treatment of osteoporosis (bisphosphonate, calcitonin, or teriparatide) and before development of a first hip or clinical vertebral fracture. Because clinicians may potentially treat a patient before BMD is in the osteoporosis range, “antifracture treatment” is used synonymously with the Food and Drug Administration–approved agents for osteoporosis treatment throughout this paper.
The secondary outcome was the time for 3% of participants to develop a first hip or clinical vertebral fracture before initiation of an antifracture treatment agent and before development of osteoporosis (for those without osteoporosis at baseline). This study included both nontraumatic and traumatic fractures in the fracture endpoints because clinical misclassification of fracture etiology is common, and a falsely low fracture count would lead to inappropriately long estimates of time to fracture.
The tertiary outcome was the time for 3% of participants to develop a first nontraumatic or traumatic major osteoporotic (hip, clinical vertebral, proximal humerus, or wrist) fracture before initiation of antifracture treatment and before development of osteoporosis (for those without osteoporosis at baseline).
Statistical Analysis
The time for 10% of men without osteoporosis at baseline to develop osteoporosis prior to hip or clinical vertebral fracture and prior to initiation of antifracture treatment was estimated from parametric log-logistic regression models of the cumulative incidence quantile as defined in Peng and Fine, 23–27 based on interval-censored data. The analysis was stratified by the baseline categories of higher BMD and moderate and advanced osteopenia. The time origin was the first study examination with a BMD measurement that included femoral neck, total hip, and lumbar spine measurements, with follow-up BMD measurements at one or more sites continuing until the study examination preceding death or dropout. In the time-to-osteoporosis analysis, first reported use of antifracture treatment, hip or clinical vertebral fracture, and death were competing risks.
Using separate models for men without versus with osteoporosis at baseline, competing risk analyses were conducted to estimate the cumulative incidence functions for time to first hip or clinical vertebral fracture, and for time to major osteoporotic fracture, based on known fracture dates prior to osteoporosis and treatment. The corresponding time intervals for 3% of participants to transition to each fracture outcome were determined for the higher BMD and moderate and advanced osteopenia groups. First reported use of antifracture treatment, incident osteoporosis (for men without osteoporosis at baseline) and death were competing risks in the time-to-fracture analyses. Results from all models were adjusted for race, and baseline mean-centered age and BMI.
Sensitivity analyses were conducted to evaluate and determine the most appropriate parametric models for estimates of time to osteoporosis and time to fracture. All analyses were performed using SAS, version 9.4.
Results
Of the 5,994 men in the cohort, one was not eligible for this study because he was aged <65 years at baseline. After other eligibility criteria were applied (Appendix Figure 1), baseline characteristics of the 5,415 participants in the analytic cohort were tabulated (Table 1). The mean age of the cohort was 73.6 years, and the mean BMI was 27.4 kg/m2 (overweight). About three fourths of the men (4,203 of 5,415) had higher BMD at baseline, whereas 180 (3.3%) had baseline osteoporosis. Within each baseline T-score range, the following proportions of men developed osteoporosis before initiation of antifracture treatment and before a first hip or clinical vertebral fracture: higher BMD, 9/4,203 (0.2%); moderate osteopenia, 35/680 (5.1%), and advanced osteopenia, 73/352 (20.7%).
Table 1.
Characteristics of the Men in Osteoporotic Fractures in Men (MrOS) Study cohort
| Baseline T-score >−1.50 | Baseline T-score −1.50 to −2.49 | Baseline T-score ≤−2.50 | All participants | |
|---|---|---|---|---|
|
| ||||
| Characteristicsa | N=4,203 (77.6%) | N=1,032 (19.1%) | N=180 (3.3%) | N=5,415 |
| Age, mean years (SD) | 73.2 (5.7) | 74.8 (6.2) | 75.8 (6.3) | 73.6 (5.9) |
| 65–69 | 1,345 (32.0) | 246 (23.8) | 35 (19.4) | 1,626 (30.0) |
| 70–74 | 1,245 (29.6) | 263 (25.5) | 42 (23.3) | 1,550 (28.6) |
| 75–79 | 967 (23.0) | 280 (27.1) | 48 (26.7) | 1,295 (23.9) |
| ≥80 | 646 (15.4) | 243 (23.5) | 55 (30.6) | 944 (17.4) |
| Weight, mean kg (SD) | 85.1 (13.1) | 77.8 (11.4) | 73.0 (10.7) | 83.3 (13.2) |
| BMI, kg/m2 | 27.9 (3.8) | 26.0 (3.3) | 24.9 (3.4) | 27.4 (3.8) |
| <25 | 965 (23.0) | 410 (39.7) | 100 (55.6) | 1,475 (27.2) |
| ≥25 | 3,238 (77.0) | 622 (60.3) | 80 (44.4) | 3,940 (72.8) |
| Race | ||||
| White (Non-Hispanic) | 3,747 (89.2) | 947 (91.8) | 159 (88.3) | 4,853 (89.6) |
| African American | 195 (4.6) | 16 (1.6) | 5 (2.8) | 216 (4.0) |
| Asian | 124 (3.0) | 42 (4.1) | 8 (4.4) | 174 (3.2) |
| Hispanic | 89 (2.1) | 16 (1.6) | 4 (2.2) | 109 (2.0) |
| Other | 48 (1.1) | 11 (1.1) | 4 (2.2) | 63 (1.2) |
| Years of education | ||||
| Less than high school | 266 (6.3) | 65 (6.30) | 13 (7.2) | 344 (6.35) |
| High school/Some college | 1,681 (40.0) | 409 (39.6) | 72 (40.0) | 2,162 (39.9) |
| College/Some grad school | 1,229 (29.2) | 288 (27.9) | 63 (35.0) | 1,580 (29.2) |
| Grad school | 1,027 (24.4) | 270 (26.2) | 32 (17.8) | 1,329 (24.5) |
| Previous fracture after age 50 | ||||
| Yes | 766 (18.2) | 258 (25.0) | 59 (32.8) | 1,083 (20.0) |
| No | 3,437 (81.8) | 774 (75.0) | 121 (67.2) | 4,332 (80.0) |
| Current smoker | ||||
| Current | 140/4,202 (3.3) | 39 (3.8) | 11 (6.1) | 190/5,414 (3.5) |
| Past | 2,488/4,202 (59.2) | 582 (56.4) | 98 (54.4) | 2,056/5414 (38.0) |
| Never | 1,574/4,202 (37.5) | 411 (39.8) | 71 (39.4) | 3,168/5,414 (58.5) |
| Alcohol consumption | ||||
| <21 drinks per week | 4,023 (95.7) | 1,001 (97.0) | 173 (96.1) | 5,197 (96.0) |
| ≥21 drinks per week | 180 (4.28) | 31 (3.00) | 7 (3.89) | 218 (4.03) |
| History of parental hip fracture | ||||
| Yes | 497/2,968 (16.7) | 164/742 (22.1) | 28/146 (19.2) | 689/3,856 (17.9) |
| No | 1,511/2,968 (50.9) | 327/742 (44.1) | 75/146 (51.4) | 1,913/3,856 (49.6) |
| Don’t know | 960/2,968 (32.3) | 251/742 (33.8) | 43/146 (29.5) | 1,254/3,856 (32.5) |
| Rheumatoid arthritis | ||||
| Yes | 221 (5.3) | 44 (4.3) | 6 (3.3) | 271 (5.0) |
| No | 3,982 (94.7) | 988 (95.7) | 174 (96.7) | 5,144 (95.0) |
| Oral glucocorticoid ever-use | ||||
| Yes | 80/3,323 (2.4) | 20/804 (2.5) | 9/128 (7.0) | 109/4,255 (2.6) |
| No | 3,243/3,323 (97.6) | 784/804 (97.5) | 119/128 (93.0) | 4,146/4,255 (97.4) |
| Bone mineral density T-score, mean (SD)b | ||||
| Femoral neck | −0.23 (0.89) | −1.72 (0.44) | −2.38 (0.63) | −0.59 (1.06) |
| range | −4.25, 6.17 | |||
| Total hip | 0.53 (0.96) | −0.98 (0.60) | −1.81 (0.75) | 0.16 (1.14) |
| range | −4.00, 6.73 | |||
| Lumbar spine | 1.67 (2.15) | −0.35 (1.61) | −1.57 (1.70) | 1.17 (2.25) |
| range | −3.93, 15.20 | |||
| Prevalence of osteoporosisc | ||||
| Definition 1 | 0 | 0 | 180 (100.0) | 180 (3.3) |
| Definition 2 | 0 | 184 (17.8) | 180 (100.0) | 364 (6.7) |
| Definition 3 | 0 | 0 | 114 (63.3) | 114 (2.1) |
| Definition 4 | 0 | 0 | 119 (66.1) | 119 (2.2) |
| Hip or clinical vertebral fracture by study endd | 169 (4.0) | 110 (10.7) | 38 (21.1) | 317 (5.9) |
| Hip fracture | 94 (2.2) | 75 (7.3) | 25 (13.9) | 194 (3.6) |
| Clinical vertebral fracture | 75 (1.8) | 35 (3.4) | 13 (7.2) | 123 (2.3) |
| Antifracture treatment by study end | 143 (3.4) | 131 (12.7) | 62 (34.4) | 336 (6.2) |
| Bisphosphonate | 141 (3.4) | 130 (12.6) | 59 (32.7) | 330 (6.1) |
| Calcitonin | 2 (<0.1) | 1 (<0.1) | 2 (1.1) | 5 (<0.1) |
| Teriparatide | 0 (0.0) | 0 (0.0) | 1 (0.5) | 1 (<0.1) |
| Total fracture follow-up time, years | ||||
| Mean (SD) | 11.1 (3.6) | 10.7 (3.8) | 10.2 (4.1) | 11.0 (3.7) |
| Median | 13.0 | 12.7 | 12.2 | 12.9 |
| Minimum | 0.04 | 0.08 | 0.7 | 0.04 |
| Maximum | 14.8 | 14.8 | 14.6 | 14.8 |
| Concurrent BMD and fracture follow-up time, yearse | ||||
| Mean (SD) | 5.2 (2.3) | 4.9 (2.4) | 4.7 (2.6) | 5.1 (2.3) |
| Median | 6.5 | 5.8 | 5.3 | 6.5 |
| Minimum | 0 | 0 | 0 | 0 |
| Maximum | 8.7 | 8.6 | 8.1 | 8.7 |
BMD, bone mineral density
Values are N (% of T-score group) unless otherwise specified. N/non-missing values are presented if data were incomplete except for rheumatoid arthritis (missing value treated as no disease).
Maximum T-score values confirmed as valid by the MrOS Data Coordinating Center
Tabulations of disease events according to the following four definitions of osteoporosis are presented in Appendix Tables 4–7.
Definition 1. Lowest T-score at the femoral neck or total hip or lumbar spine calculated using BMD norms for young white women
Definition 2. Lowest T-score at the femoral neck or total hip or lumbar spine calculated using BMD norms for young white men
Definition 3. Lowest T-score at the femoral neck calculated using BMD norms for young white women
Definition 4. Lowest T-score at the femoral neck or total hip calculated using BMD norms for young white women
Tabulations of first hip or clinical vertebral fracture before a competing risk occurred
Includes participants who had one baseline DXA test, then developed a competing risk before osteoporosis, treatment or major fracture (0 concurrent BMD and fracture follow-up)
Within the higher BMD category, 107 men (3.4% of 3,177) with baseline T-scores ≥ −1.00 and 62 men (6.0% of 1,026) with baseline T-scores between −1.01 and −1.49 had a hip or clinical vertebral fracture by the end of study follow-up (combined results for these T-score groups are in Table 3). Of the total of 371 men without baseline osteoporosis who sustained a first hip (n=219) or clinical vertebral (n=152) fracture, 69 (18.6%) developed osteoporosis before hip or clinical vertebral fracture, and 24 (6.5%) received antifracture treatment prior to the fracture. Of these 371 men, 317 had a first hip or clinical vertebral fracture before a competing risk occurred (Table 1.)
Table 3.
Time for 3% of Men Without Osteoporosisa or Treatment at Baseline to Sustain a Hip or Clinical Vertebral Fracture, According to Baseline T-score Range
| Baseline T-score range | Fracture events, N (%) | Time interval for 3% of participants to sustain a hip or clinical vertebral fractureb | |
|---|---|---|---|
| Unadjusted years (95% CI) | Adjusted years (95% CI) | ||
| >−1.50 | 169/4203 (4.02) | 9.59 (8.69, 10.59) | 7.09 (6.02, 8.34) |
| −1.50 to −1.99 | 68/680 (10.00) | 4.37 (3.41, 5.61) | 4.83 (3.84, 6.09) |
−2.00 to −2.49
42/352 (11.93)
3.51 (2.58, 4.79)
4.06 (3.05, 5.40)
Lowest T-score ≤−2.5 at femoral neck or total hip or lumbar spine, calculated using bone mineral desnity norms for young white women
Estimated time to hip or clinical vertebral fracture (competing risks: incident osteoporosis, antifracture treatment or death), computed by quantile estimates based on the fitted log logistic models, stratified by baseline lowest femoral neck or total hip or lumbar spine T-score (lowest T-score −1.50 to −1.99 vs. −2.00 vs. −2.49), and adjusted for mean-centered age, mean-centered BMI and race.
Note: Estimates greater than 8.7 years (italicized) have questionable reliability due to excessive extrapolation required for 3% to transition to hip or clinical vertebral fracture.
The estimated time for 10% of men without osteoporosis at baseline to subsequently develop osteoporosis varied according to baseline BMD T-score (Table 2 and Figure 1). During follow-up, nine men (0.2% within T-score stratum) with higher baseline BMD developed osteoporosis. For those with moderate osteopenia and advanced osteopenia at baseline, the unadjusted times for 10% to develop osteoporosis were very similar to the adjusted time estimates of 8.5 (95% CI=6.7, 10.9) years and 2.7 (95% CI=2.1, 3.4) years, respectively.
Table 2.
Time for 10% of Men Without Osteoporosisa or Treatment at Baseline to Develop Osteoporosis, According to Baseline T-score Range
| Baseline T-score range | Osteoporosis events, N (%) | Time interval for 10% of participants to develop osteoporosisb | |
|---|---|---|---|
| Unadjusted years (95% CI) | Adjusted years (95% CI) | ||
| >−1.50 | 9/4203(0.21) | ---- | ---- |
| −1.50 to −1.99 | 35/680 (5.15) | 8.57 (6.67, 10.99) | 8.51 (6.67, 10.86) |
| −2.00 to −2.49 | 73/352 (20.74) | 2.59 (2.03, 3.30) | 2.68 (2.12, 3.40) |
Osteoporosis defined as lowest T-score ≤-2.5 at femoral neck or total hip or lumbar spine, calculated using bone mineral density (BMD) norms for young white women
Estimated time to osteoporosis (competing risks: incident hip or clinical vertebral fracture, antifracture treatment or death), computed by quantile estimates based on the fitted log logistic models, stratified by baseline lowest femoral neck, total hip or lumbar spine T-score (lowest T-score −1.50 to −1.99 vs. −2.00 vs. −2.49), and adjusted for mean-centered age, mean-centered BMI and race.
Note: Time estimates for the T-score>−1.50 group could not be calculated due the very small number of endpoint events, leading to excessive extrapolation beyond the observed BMD follow-up of 8.7 years.
Figure 1.
Unadjusted cumulative incidence of osteoporosis according to baseline T-score range.
Note: The proportion of men who had transitioned to osteoporosis is plotted as a function of time. The cumulative incidence curves are estimated by parametric cumulative incidence models for interval-censored data. The dotted horizontal line shows where 10% of men have transitioned to osteoporosis; where this line intersects with each cumulative incidence curve, a vertical dotted line to the x-axis marks the estimated time interval. The time interval for men with baseline T-scores >−1.50 could not be calculated because of the very small number of fracture events (of the 4203 men with baseline T-scores >−1.50, nine [0.2%] developed osteoporosis during 8.7 years of BMD follow-up).
Unadjusted estimates of the cumulative incidence of hip or clinical vertebral fracture as a function of testing interval length were similar to covariate-adjusted estimates (Table 3, Appendix Figure 2). The unadjusted estimated time (9.6 years, 95% CI=8.7 years, 10.6 years) for 3% of men with higher baseline BMD to sustain a hip or clinical vertebral fracture was longer than the adjusted estimate of 7.1 (95% CI=6.0, 8.3) years. Otherwise, the unadjusted estimates of the time for 3% to develop a hip or clinical vertebral fracture were similar to the adjusted times of 4.8 (95% CI=3.8, 6.1) years for those with moderate osteopenia, and 4.1 (95% CI=3.1, 5.4) years for those with advanced osteopenia.
Unadjusted and covariate-adjusted estimates of the cumulative incidence of major osteoporotic fracture as a function of testing interval length were shorter than the analogous times for hip and clinical vertebral fracture (Appendix Table 1). The adjusted time for 3% to develop a major osteoporotic fracture ranged from 4.7 (95% CI=4.0, 5.7) years for men with higher BMD, to 2.6 (95% CI=1.9, 3.6) years for men with advanced osteopenia at baseline.
For all men aged ≥65 years with osteoporosis at baseline, the adjusted time for 3% to sustain a fracture was 1.7 (95% CI=1.0, 3.1) years for hip or clinical vertebral fracture and 1.0 (95% CI=0.5, 1.9) year for major osteoporotic fracture.
Full results of the sensitivity analyses are presented in the Appendix (Appendix Tables 2 and 3 and associated text, Appendix Figure 3). The log-logistic model was found to most closely follow the non-parametric curve and was used in the main analysis.
Discussion
The time to incidence of osteoporosis was estimated according to baseline BMD T-score in older men without a history of hip or clinical vertebral fracture and without prior antifracture pharmacologic treatment. Less than 1% of men with higher BMD (baseline T-score > −1.50 noted in 78% of the cohort) developed osteoporosis during 8.7 years of concurrent BMD and fracture follow-up. A significantly higher proportion of men with moderate or advanced osteopenia developed osteoporosis by the end of follow-up, but more often, hip or clinical vertebral fracture occurred before osteoporosis developed. Twenty-nine percent of the hip and clinical vertebral fractures occurred in men with baseline normal BMD; at this T-score level, clinical practice guidelines do not recommend treatment in the absence of fragility fracture as there is no evidence that pharmacologic treatment prevents clinical fracture events among patients with normal BMD.
These findings for older men are consistent with findings for women aged 67 years and older with T-scores > −1.50 in an earlier study.15 Also similar to longitudinal studies of postmenopausal women,28–30 this study demonstrates that most older men who sustain a hip or clinical vertebral fracture do not have pre-existing osteoporosis. Fragility fractures in the absence of osteoporosis might be more common in men because men report a higher level of trauma associated with their fragility fractures compared with women.31,32 Men with poor physical condition are especially at risk; a 2008 longitudinal analysis demonstrated that older men with the worst performance on a repeated chair stand test had a substantially higher risk of hip fracture compared with men in the fastest quartile of this test (multivariate hazard ratio=8.15, 95% CI=1.46, 6.73).33 Also, men lose bone strength with age as a result of trabecular thinning and changes in cortical bone density, including both cortical thinning and increased porosity that are not easily identified by DXA BMD measurements.34 Like all prognostic tests, BMD testing cannot identify every individual who will have a future fragility fracture; therefore, other screening approaches (risk factor assessment, other imaging methods) and interventions to reduce risk factors for fracture (e.g., fall reduction) that are not related to BMD should continue to be investigated.
Despite supportive recommendations in some clinical practice guidelines,9–11,35,36 no standard of care exists for osteoporosis screening in older men at average risk of fracture. Medicare reimburses DXA scans in men for a limited number of disease indications, not including screening for primary prevention of fractures. Even after a major osteoporotic fracture, men are three times less likely than women to receive BMD testing and are unlikely to receive treatment.37,38 This study identified fracture rates in older men that were higher than rates for younger postmenopausal women,30 for whom policy statements encourage selective BMD screening.11,13,35 Clarification of the optimal target population of men for BMD screening/treatment could potentially have an important population-wide impact on fracture reduction if future RCTs demonstrate primary fracture prevention, especially if treating men with screen-detected osteoporosis at selected BMD thresholds will reduce their risk of hip and clinical vertebral fractures. As the evidence base for BMD screening and treatment continues to expand, the use of BMD testing among older men should be determined by informed decision making with patients and clinicians considering relevant data on BMD testing, clinical risk factors, and treatment.
Limitations
The study has several limitations and strengths. Results are limited to men who have already had one bone density test and therefore whose BMD is known. Precise time estimates were not possible beyond the actual follow-up time in MrOS. Of the 579 (9.7%) of the original MrOS participants who were excluded in this analysis, one third had a history of hip or clinical vertebral fracture or treatment at baseline. Subsequent DXA tests in these men would be for post-fracture surveillance or treatment monitoring and would not meet the epidemiologic definition of screening. The cohort comprised men aged 65 years and older with a low overall prevalence of osteoporosis. The results are not generalizable to postmenopausal women or younger men, or individuals with secondary causes of osteoporosis. Ninety percent of the sample was white; results might differ for nonwhite individuals. The reported time intervals are not based on the time to incident radiographic vertebral fractures. A 2014 analysis of 4,332 MrOS participants with baseline and repeat radiographs found that 192 (4.4%) had an incident radiographic vertebral fracture.39 Treatment of radiographic vertebral fracture has been shown to reduce incident hip and clinical vertebral fractures in postmenopausal women.40 However, because clinicians are unaware of these subclinical fractures, the time of onset of radiographic vertebral fractures is uncertain and cannot be used to calculate an optimal bone density screening interval. This suggests that the reported bone density testing intervals are conservative estimates for identification of patients who are likely to respond to treatment. Falls were not included in the evaluation of clinical risk factors for fracture, to be consistent with the International Osteoporosis Foundation’s caution in considering fall risk in estimating fracture probability because patients selected primarily on the basis of nonskeletal clinical risk factors (including several measures of fall risk) may respond less to antifracture agents compared to patients selected on the basis of low BMD.41,42 Falls are a critical factor in fracture risk, and interventions to reduce falls must be studied and implemented separately to maximize the effectiveness of fracture prevention programs. Strengths of the analysis include the large size of the cohort and the extended BMD and adjudicated fracture follow-up.
Conclusions
Community-dwelling men aged 65 years and older with baseline T-scores > −1.50 had a very low likelihood of developing osteoporosis over an average follow-up of 8.7 years. Ten percent of men with moderate and advanced osteopenia transitioned to osteoporosis in approximately 8 and 2 years, respectively. Most major osteoporotic fractures were not preceded by osteoporosis, suggesting that adjuncts to BMD testing should continue to be investigated to improve fracture prediction in older men.
Supplementary Material
Acknowledgments
We thank John Preisser, PhD, of the University of North Carolina at Chapel Hill Department of Biostatistics for providing statistical code that was adapted for use in the current project; Dr. Preisser received funding from grant number K23RR024685 for this work.
The project described was funded by grant number R01 AG046294 (Gourlay) and grant number K23AG040168 (Dam) from the National Institute on Aging, grant number UL1TR001111 from the National Center for Advancing Translational Sciences, and grant number K24 AR048841 (Lane) from the National Institute of Arthritis and Musculoskeletal and Skin Diseases. Dr. Wilt’s effort was supported in part by the Department of Veterans Affairs (VA) Under Secretary’s Award for Health Services Research and the Minneapolis VA High Value Care Initiative. The work of Drs. Taylor, Fink and Ensrud was supported in part with resources and use of facilities at the Minneapolis VA Medical Center. The Osteoporotic Fractures in Men Study is supported by NIH funding. The following institutes provide support: the National Institute on Aging, the National Institute of Arthritis and Musculoskeletal and Skin Diseases, the National Center for Advancing Translational Sciences, and NIH Roadmap for Medical Research under the following grant numbers: U01 AG027810, U01 AG042124, U01 AG042139, U01 AG042140, U01 AG042143, U01 AG042145, U01 AG042168, U01 AR066160, and UL1 TR000128. The content is solely the responsibility of the authors and does not necessarily reflect the official views of the funding agencies. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or in the decision to submit the manuscript for publication.
Dr. Cawthon reports research grants from GSK, IMS Health, and Merck and consulting for Eli Lilly and KineMed. Dr. Orwoll has received research and consulting support from Merck, Eli Lilly, and Amgen. Dr. Cauley has received consultant support from Merck. Dr. Ensrud serves as a consultant on a Data Monitoring Committee for Merck Sharpe & Dohme. Dr. Kado has consultant support from Kalytera Therapeutics, Takeda Pharmaceuticals, and Amgen. No other financial disclosures were reported by the authors of this paper.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Manthripragada AD, O’Malley CD, Gruntmanis U, Hall JW, Wagman RB, Miller PD. Fracture incidence in a large cohort of men age 30 years and older with osteoporosis. Osteoporos Int. 2015;26(5):1619–1627. doi: 10.1007/s00198-015-3035-z. http://dx.doi.org/10.1007/s00198-015-3035-z. [DOI] [PubMed] [Google Scholar]
- 2.Nguyen ND, Ahlborg HG, Center JR, Eisman JA, Nguyen TV. Residual lifetime risk of fractures in women and men. J Bone Miner Res. 2007;22(6):781–788. doi: 10.1359/jbmr.070315. http://dx.doi.org/10.1359/jbmr.070315. [DOI] [PubMed] [Google Scholar]
- 3.Khosla S, Amin S, Orwoll E. Osteoporosis in men. Endocr Rev. 2008;29(4):441–464. doi: 10.1210/er.2008-0002. http://dx.doi.org/10.1210/er.2008-0002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Haentjens P, Magaziner J, Colon-Emeric CS, et al. Meta-analysis: excess mortality after hip fracture among older women and men. Ann Intern Med. 2010;152(6):380–390. doi: 10.1059/0003-4819-152-6-201003160-00008. http://dx.doi.org/10.7326/0003-4819-152-6-201003160-00008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Morin S, Lix LM, Azimaee M, Metge C, Caetano P, Leslie WD. Mortality rates after incident non-traumatic fractures in older men and women. Osteoporos Int. 2011;22(9):2439–2448. doi: 10.1007/s00198-010-1480-2. http://dx.doi.org/10.1007/s00198-010-1480-2. [DOI] [PubMed] [Google Scholar]
- 6.Bliuc D, Nguyen ND, Milch VE, Nguyen TV, Eisman JA, Center JR. Mortality risk associated with low-trauma osteoporotic fracture and subsequent fracture in men and women. JAMA. 2009;301(5):513–521. doi: 10.1001/jama.2009.50. http://dx.doi.org/10.1001/jama.2009.50. [DOI] [PubMed] [Google Scholar]
- 7.Boonen S, Reginster JY, Kaufman JM, et al. Fracture risk and zoledronic acid therapy in men with osteoporosis. N Engl J Med. 2012;367(18):1714–1723. doi: 10.1056/NEJMoa1204061. http://dx.doi.org/10.1056/NEJMoa1204061. [DOI] [PubMed] [Google Scholar]
- 8.Zhong ZM, Chen JT. Anti-fracture efficacy of risedronic acid in men: A meta-analysis of randomized controlled trials. Clin Drug Investig. 2009;29(5):349–357. doi: 10.2165/00044011-200929050-00007. http://dx.doi.org/10.2165/00044011-200929050-00007. [DOI] [PubMed] [Google Scholar]
- 9.Lim LS, Hoeksema LJ, Sherin K ACPM Prevention Practice Committee. Screening for osteoporosis in the adult U.S. population: ACPM position statement on preventive practice. Am J Prev Med. 2009;36(4):366–375. doi: 10.1016/j.amepre.2009.01.013. http://dx.doi.org/10.1016/j.amepre.2009.01.013. [DOI] [PubMed] [Google Scholar]
- 10.Qaseem A, Snow V, Shekelle P, Hopkins R, Jr, Forciea MA, Owens DK. Screening for osteoporosis in men: a clinical practice guideline from the American College of Physicians. Ann Intern Med. 2008;148(9):680–684. doi: 10.7326/0003-4819-148-9-200805060-00008. http://dx.doi.org/10.7326/0003-4819-148-9-200805060-00008. [DOI] [PubMed] [Google Scholar]
- 11.Runhaar J, Schiphof D, van Meer B, Reijman M, Bierma-Zeinstra SM, Oei EH. How to define subregional osteoarthritis progression using semi-quantitative MRI Osteoarthritis Knee Score (MOAKS) Osteoarthritis Cartilage. 2014;22(10):1533–1536. doi: 10.1016/j.joca.2014.06.022. http://dx.doi.org/10.1016/j.joca.2014.06.022. [DOI] [PubMed] [Google Scholar]
- 12.Lewiecki E, Gordon C, Baim S, et al. International Society for Clinical Densitometry 2007 adult and pediatric official positions. Bone. 2008;43(6):1115–1121. doi: 10.1016/j.bone.2008.08.106. http://dx.doi.org/10.1016/j.bone.2008.08.106. [DOI] [PubMed] [Google Scholar]
- 13.Screening for osteoporosis: U.S. preventive services task force recommendation statement. Ann Intern Med. 2011;154(5):356–364. doi: 10.7326/0003-4819-154-5-201103010-00307. http://dx.doi.org/10.7326/0003-4819-154-5-201103010-00307. [DOI] [PubMed] [Google Scholar]
- 14.Schousboe J, Taylor B, Fink H, et al. Cost-effectiveness of bone densitometry followed by treatment of osteoporosis in older men. JAMA. 2007;298(6):629–637. doi: 10.1001/jama.298.6.629. http://dx.doi.org/10.1001/jama.298.6.629. [DOI] [PubMed] [Google Scholar]
- 15.Gourlay ML, Fine JP, Preisser JS, et al. Bone-density testing interval and transition to osteoporosis in older women. N Engl J Med. 2012;366(3):225–233. doi: 10.1056/NEJMoa1107142. http://dx.doi.org/10.1056/NEJMoa1107142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Orwoll E, Blank JB, Barrett-Connor E, et al. Design and baseline characteristics of the osteoporotic fractures in men (MrOS) study--a large observational study of the determinants of fracture in older men. Contemp Clin Trials. 2005;26(5):569–585. doi: 10.1016/j.cct.2005.05.006. http://dx.doi.org/10.1016/j.cct.2005.05.006. [DOI] [PubMed] [Google Scholar]
- 17.Blank JB, Cawthon PM, Carrion-Petersen ML, et al. Overview of recruitment for the osteoporotic fractures in men study (MrOS) Contemp Clin Trials. 2005;26(5):557–568. doi: 10.1016/j.cct.2005.05.005. http://dx.doi.org/10.1016/j.cct.2005.05.005. [DOI] [PubMed] [Google Scholar]
- 18.WHO. WHO Technical report series no. 843. Geneva: WHO; 1994. Assessment of osteoporotic fracture risk and its role in screening for postmenopausal osteoporosis. [PubMed] [Google Scholar]
- 19.Hernlund E, Svedbom A, Ivergard M, et al. Osteoporosis in the European Union: medical management, epidemiology and economic burden. A report prepared in collaboration with the International Osteoporosis Foundation (IOF) and the European Federation of Pharmaceutical Industry Associations (EFPIA) Arch Osteoporos. 2013;8(1–2):136. doi: 10.1007/s11657-013-0136-1. http://dx.doi.org/10.1007/s11657-013-0136-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Watts NB, Leslie WD, Foldes AJ, Miller PD. 2013 International Society for Clinical Densitometry Position Development Conference: Task Force on Normative Databases. J Clin Densitom. 2013;16(4):472–481. doi: 10.1016/j.jocd.2013.08.001. http://dx.doi.org/10.1016/j.jocd.2013.08.001. [DOI] [PubMed] [Google Scholar]
- 21.Looker A, Wahner H, Dunn W, et al. Updated data on proximal femur bone mineral levels of U.S. adults. Osteoporos Int. 1998;8(5):468–489. doi: 10.1007/s001980050093. http://dx.doi.org/10.1007/s001980050093. [DOI] [PubMed] [Google Scholar]
- 22.Looker AC, Melton LJ, 3rd, Borrud LG, Shepherd JA. Lumbar spine bone mineral density in US adults: demographic patterns and relationship with femur neck skeletal status. Osteoporos Int. 2012;23(4):1351–1360. doi: 10.1007/s00198-011-1693-z. http://dx.doi.org/10.1007/s00198-011-1693-z. [DOI] [PubMed] [Google Scholar]
- 23.Fine J, Gray R. A Proportional Hazards Model for the Subdistribution of a Competing Risk. J Am Stat Assoc. 1999;94(446):496–509. http://dx.doi.org/10.1080/01621459.1999.10474144. [Google Scholar]
- 24.Fine JP. Regression modeling of competing crude failure probabilities. Biostatistics. 2001;2(1):85–97. doi: 10.1093/biostatistics/2.1.85. http://dx.doi.org/10.1093/biostatistics/2.1.85. [DOI] [PubMed] [Google Scholar]
- 25.Jeong JH, Fine JP. Parametric regression on cumulative incidence function. Biostatistics. 2007;8(2):184–196. doi: 10.1093/biostatistics/kxj040. http://dx.doi.org/10.1093/biostatistics/kxj040. [DOI] [PubMed] [Google Scholar]
- 26.Hudgens MG, Li C, Fine JP. Parametric likelihood inference for interval censored competing risks data. Biometrics. 2014;70(1):1–9. doi: 10.1111/biom.12109. http://dx.doi.org/10.1111/biom.12109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Peng L, Fine J. Nonparametric Quantile Inference with Competing-Risks Data. Biometrika. 2007;94(3):735–744. http://dx.doi.org/10.1093/biomet/asm059. [Google Scholar]
- 28.Stone K, Seeley D, Lui L, et al. BMD at multiple sites and risk of fracture of multiple types: long-term results from the Study of Osteoporotic Fractures. J Bone Miner Res. 2003;18(11):1947–1954. doi: 10.1359/jbmr.2003.18.11.1947. http://dx.doi.org/10.1359/jbmr.2003.18.11.1947. [DOI] [PubMed] [Google Scholar]
- 29.Siris ES, Chen YT, Abbott TA, et al. Bone mineral density thresholds for pharmacological intervention to prevent fractures. Arch Intern Med. 2004;164(10):1108–1112. doi: 10.1001/archinte.164.10.1108. http://dx.doi.org/10.1001/archinte.164.10.1108. [DOI] [PubMed] [Google Scholar]
- 30.Gourlay ML, Overman RA, Fine JP, et al. Baseline age and time to major fracture in younger postmenopausal women. Menopause. 2015;22(6):589–597. doi: 10.1097/GME.0000000000000356. http://dx.doi.org/10.1097/GME.0000000000000356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Ensrud KE, Blackwell TL, Cawthon PM, et al. Degree of Trauma Differs for Major Osteoporotic Fracture Events in Older Men Versus Older Women. J Bone Miner Res. 2015 doi: 10.1002/jbmr.2589. http://dx.doi.org/10.1002/jbmr.2589. [DOI] [PMC free article] [PubMed]
- 32.Wang H, Xiang L, Liu J, Zhou Y, Ou L. Gender differences in the clinical characteristics of traumatic spinal fractures among the elderly. Arch Gerontol Geriatr. 2014;59(3):657–664. doi: 10.1016/j.archger.2014.05.004. http://dx.doi.org/10.1016/j.archger.2014.05.004. [DOI] [PubMed] [Google Scholar]
- 33.Cawthon PM, Fullman RL, Marshall L, et al. Physical performance and risk of hip fractures in older men. J Bone Miner Res. 2008;23(7):1037–1044. doi: 10.1359/JBMR.080227. http://dx.doi.org/10.1359/jbmr.080227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Seeman E. Age- and menopause-related bone loss compromise cortical and trabecular microstructure. J Gerontol A Biol Sci Med Sci. 2013;68(10):1218–1225. doi: 10.1093/gerona/glt071. http://dx.doi.org/10.1093/gerona/glt071. [DOI] [PubMed] [Google Scholar]
- 35.Schousboe JT, Shepherd JA, Bilezikian JP, Baim S. Executive summary of the 2013 International Society for Clinical Densitometry Position Development Conference on bone densitometry. J Clin Densitom. 2013;16(4):455–466. doi: 10.1016/j.jocd.2013.08.004. http://dx.doi.org/10.1016/j.jocd.2013.08.004. [DOI] [PubMed] [Google Scholar]
- 36.Watts NB, Adler RA, Bilezikian JP, et al. Osteoporosis in men: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2012;97(6):1802–1822. doi: 10.1210/jc.2011-3045. http://dx.doi.org/10.1210/jc.2011-3045. [DOI] [PubMed] [Google Scholar]
- 37.Feldstein AC, Nichols G, Orwoll E, et al. The near absence of osteoporosis treatment in older men with fractures. Osteoporos Int. 2005;16(8):953–962. doi: 10.1007/s00198-005-1950-0. http://dx.doi.org/10.1007/s00198-005-1950-0. [DOI] [PubMed] [Google Scholar]
- 38.Kiebzak GM, Beinart GA, Perser K, Ambrose CG, Siff SJ, Heggeness MH. Undertreatment of osteoporosis in men with hip fracture. Arch Intern Med. 2002;162(19):2217–2222. doi: 10.1001/archinte.162.19.2217. http://dx.doi.org/10.1001/archinte.162.19.2217. [DOI] [PubMed] [Google Scholar]
- 39.Cawthon PM, Blackwell TL, Marshall LM, et al. Physical performance and radiographic and clinical vertebral fractures in older men. J Bone Miner Res. 2014;29(9):2101–2108. doi: 10.1002/jbmr.2239. http://dx.doi.org/10.1002/jbmr.2239. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Wells GA, Cranney A, Peterson J, et al. Alendronate for the primary and secondary prevention of osteoporotic fractures in postmenopausal women. Cochrane Database Syst Rev. 2008;(1):CD001155. doi: 10.1002/14651858.CD001155.pub2. http://dx.doi.org/10.1002/14651858.cd001155.pub2. [DOI] [PubMed]
- 41.Kanis JA, McCloskey EV, Johansson H, Cooper C, Rizzoli R, Reginster JY. European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int. 2013;24(1):23–57. doi: 10.1007/s00198-012-2074-y. http://dx.doi.org/10.1007/s00198-012-2074-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.McClung MR, Geusens P, Miller PD, et al. Effect of risedronate on the risk of hip fracture in elderly women. Hip Intervention Program Study Group. N Engl J Med. 2001;344(5):333–340. doi: 10.1056/NEJM200102013440503. http://dx.doi.org/10.1056/NEJM200102013440503. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.