This an Epidemiology in Public Health Assignment due today!

A B S T R A C T

July 2001, Vol. 91, No. 71056 American Journal of Public Health

Jogging and Bone Mineral Density in Men: Results From NHANES III

Michael E. Mussolino, MA, Anne C. Looker, PhD, and Eric S. Orwoll, MD

Michael E. Mussolino and Anne C. Looker are with the National Center for Health Statistics, Centers for Disease Control and Prevention, Hyattsville, Md. Eric S. Orwoll is with Oregon Health Sciences Uni- versity, Portland.

Requests for reprints should be sent to Michael E. Mussolino, MA, Division of Epidemiology, Na- tional Center for Health Statistics, 6525 Belcrest Rd, Suite 730, Hyattsville, MD 20782 (e-mail: [email protected]).

This article was accepted January 12, 2001.

Objectives. This cross-sectional population-based study assessed the as- sociation of jogging with femoral bone mineral density (BMD) in men.

Methods. Data are from a nationally representative sample of 4254 men aged 20 to 59 years from the Third National Health and Nutrition Examination Sur- vey (NHANES III). Total femoral BMD was measured by dual energy x-ray ab- sorptiometry. Jogging was self-reported.

Results. Jogging (any vs none) was strongly associated with higher BMD in multivariate models (P < .01) for both young and middle-aged men. Men who jogged 9 or more times per month had higher BMD levels than those who jogged only 1 to 8 times per month (P = .01).

Conclusions. Jogging is associated with higher femoral neck BMD in men. Additional large-scale studies that mea- sure all aspects of jogging are war- ranted. (Am J Public Health. 2001;91: 1056–1059)

Jogging is enjoyed by millions of peo- ple in the United States. Individuals who un- dertake high-impact activities like jogging on a regular basis may have higher bone mineral density (BMD), particularly at the femoral neck.1 Higher levels of weight-bearing or vig- orous physical activity have been shown to be associated with a lower incidence of hip fractures.2,3 A number of studies have inves- tigated the effect of jogging on BMD, with many focusing on women.4–9 Among men, the results for jogging and BMD have been inconsistent. Some studies have found higher BMD among joggers in the femoral neck, lumbar spine, or total body,1,10–12 while others have found an association only at lower limb sites.13,14 Other studies have found that BMD among runners is the same as or lower than in nonrunners.15,16

A problem of the studies of both men and women has been the use of self-selected ath- letes, which, coupled with small sample sizes, likely limits generalizability. Other analyses of BMD and physical activity have combined jog- ging with other strenuous activities.17 The Third National Health and Nutrition Examination Survey (NHANES III), a large-scale national study, provides an opportunity to investigate the relation between jogging and BMD in a representative sample of young and middle- aged men with a wide variation in self-reported jogging frequency that likely encompasses the casual jogger as well as the more serious run- ner. Thus, our results may be more broadly gen- eralizable than those of previous studies.

Methods

Data for these analyses were taken from NHANES III, which collected data from a na- tional probability sample of the civilian non- institutionalized US population from 1988 to 1994.18 Analyses were limited to 4603 non- Hispanic White, non-Hispanic Black, and Mex- ican American men whose BMD levels were

measured and who were aged 20 to 59 years at the time of the NHANES III examination, be- cause relatively few study respondents 60 years or older reported jogging. Men of other race/ ethnicity groups were omitted because of the small number of observations. Also excluded were 3 men with missing values on the jog- ging variables and 346 men with unknown val- ues on any of the other variables assessed in the study. A total of 4254 men were included in the analysis (954 joggers and 3300 nonjog- gers). Because nonjoggers in the main analy- sis may have participated in other leisure ac- tivities, we also did a subanalysis in which nonjoggers were restricted to those who re- ported no leisure activities (n = 577).

Measurement of BMD

BMD was measured by trained examin- ers in mobile examination centers. Total femoral bone density was measured by dual energy x-ray absorptiometry (Hologic QDR- 1000; Hologic, Inc, Waltham, Mass). Scans were reviewed by consultants at the Mayo Clinic, Rochester, Minn, for quality control.19

Jogging Assessment

For data on self-reported jogging, the fol- lowing questions were asked: “In the past month, did you jog or run?” and “In the past month, how often did you jog or run?” Jog- ging was categorized as 9 or more times per month, 1 to 8 times per month, and none.

July 2001, Vol. 91, No. 7 American Journal of Public Health 1057

TABLE 1—Characteristics of 4254 Male Joggers and Nonjoggers Aged 20 to 59 Yearsa: Third National Health and Nutrition Examination Survey

Jogging Status Any Noneb

N 954 3300 Age at interview, y 33.4* 38.2 Weight, kg 81.4* 83.2 Height, cm 176.9 176.8 Body mass index, kg/m2 26.0* 26.6 Total femoral bone mineral density, g/cm2 1.069* 1.018 Smoking status, %

Current 23.2* 40.0 Former 28.1 25.4 Never 48.6* 34.6

Alcohol, drinks/mo 11.6 13.4 Any chronic conditions, % 4.2* 8.7 Weight loss from maximum ≥ 10%, % 26.1 24.9 Dietary variables

Food energy, kcal/day 2840.7 2828.5 Calcium, mg/day 1016.5 1036.4 Protein, g/day 105.0 104.7

Self-assessed health status, % Excellent 34.2* 20.1 Very good 39.3 34.6 Good 21.7* 34.4 Fair 4.4* 9.5 Poor 0.4* 1.4

aControlling for age at interview. bOf the 3300 nonjoggers, 2723 were “active nonjoggers” (i.e., reported other physical

activity) and 577 were “sedentary nonjoggers” (i.e., did not report any physical activity). *P < .05 (t test).

Other Baseline Variables

Information on smoking status (current, former, or never), alcohol consumption (drinks per month), self-assessed health status, food energy, calcium consumption, protein intake, weight history (including maximum weight), and chronic conditions related to secondary osteoporosis were obtained by interview. We calculated body mass index (BMI) from mea- surements of the subjects’ height and weight. The nutritional variables were determined on the basis of what the subject recalled eating in the past 24 hours. The chronic-conditions vari- able was based on self-reported doctor’s diag- noses of congestive heart failure, stroke, chronic bronchitis, goiter, other thyroid dis- ease, and diabetes.

Statistical Analysis

Statistical analyses were performed with linear regression procedures in SAS20 and SU- DAAN.21 We used sample weights when cal- culating point estimates, so estimates are rep- resentative of the civilian noninstitutionalized US population at the time of NHANES III. Weighted multivariate regression analyses were performed for the total sample and stratified by age (20–39 and 40–59 years). All models were adjusted for age at interview, race/eth-

nicity, BMI, food energy, calcium consump- tion, protein intake, smoking status, alcohol consumption, chronic conditions, and weight change, unless otherwise indicated.

Results

Baseline characteristics of men by jog- ging status (any vs none) are shown in Table 1. Joggers tended to be younger, so all subsequent comparisons were made after age was con- trolled for. Joggers weighed less than nonjog- gers and were more likely to have never smoked, have no chronic conditions, and be in excellent health. Overall, 22.3% of men re- ported jogging in the past month. Mean femoral BMD was 5.0% higher among joggers than among nonjoggers; when joggers were com- pared with sedentary nonjoggers, the figure rose to 7.7% (1.069 g/cm2 vs 0.993 g/cm2).

In the multivariate models, men who re- ported any jogging had a significantly higher BMD than those who did not jog (P < .001). Results of categorizing jogging status by fre- quency are shown in Table 2. Men who re- ported jogging 9 or more times per month or 1 to 8 times per month had higher BMD lev- els than men who did not jog (P < .01). Fur- thermore, men who jogged 9 or more times per month had significantly higher BMD lev-

els than those who jogged 1 to 8 times per month (P=.01) (data not shown). Age-specific analyses revealed similar results for ages 20 to 39, while only the category of 9 or more times per month was significant for ages 40 to 59. The relatively small number of joggers in this age group may have led to the reduced P val- ues in the category of 1 to 8 times per month.

The interaction of race/ethnicity and jog- ging group was not significant (P = .56), sug- gesting that the effect of jogging frequency is the same in all 3 race/ethnicity groups.

In a subanalysis to assess the effect of jog- ging frequency on BMD, we found that the BMD levels of those who jogged frequently (i.e., >20 times/month) were similar to the lev- els of those who jogged less frequently, sug- gesting that there may be a ceiling for the graded effect beyond which no further benefit occurs.

When treated as a continuous variable in the multivariate analyses, jogging remained a strong predictor of higher BMD (P<.01) (data not shown). When the self-assessed health sta- tus variable was added to the regression mod- els, jogging remained a statistically significant predictor of BMD (P < .01).

We also investigated the impact of total number of leisure activities on the relationship between jogging and BMD. Leisure activities varied, but walking and gardening/yard work were the most reported activities among both joggers (68.0% and 61.4%, respectively) and physically active nonjoggers (56.8% and 66.1%, respectively). Joggers reported a larger number of leisure activities than nonjoggers (mean = 4.8 vs 2.6). Joggers also reported a greater number of weight-bearing activities (mean = 2.9 vs 1.1). In regression models ad- justed for age and total number of leisure ac- tivities, BMD remained significantly higher for joggers (P < .01). Results were similar after age and number of weight-bearing activities were adjusted for (P < .01).

Discussion

Our results from this large nationally rep- resentative sample are consistent with previ- ous research suggesting that jogging is asso- ciated with higher femoral BMD in men. Adjustment for established BMD risk factors did not appreciably alter the conclusions. Fur- thermore, among the 954 joggers in the study, 47.3% reported running 8 or fewer times per month, suggesting that even infrequent jog- ging may be beneficial to BMD.

In our study, we were able to examine one aspect of jogging on BMD: frequency. Other aspects of jogging, such as distance or pace, could not be assessed because NHANES III did not include questions on these items. To

July 2001, Vol. 91, No. 71058 American Journal of Public Health

TABLE 2—Frequency of Jogging and Bone Mineral Density (BMD) for Men Aged 20 to 59 Yearsa: Third National Health and Nutrition Examination Survey

Jogging Frequency n BMD (g/cm2) β P

Ages 20–39 ≥9 times/mo 350 1.113 .0705 .0000 1–8 times/mo 405 1.083 .0401 .0015 None 1796 1.043 (Reference) (Reference)

Ages 40–59 ≥9 times/mo 98 1.076 .0651 .0021 1–8 times/mo 101 1.034 .0236 .1487 None 1504 1.011 (Reference) (Reference)

Ages 20–59 ≥9 times/mo 448 1.104 .0685 .0000 1–8 times/mo 506 1.071 .0354 .0010 None 3300 1.036 (Reference) (Reference)

aModels were adjusted for age at interview, race/ethnicity, body mass index, food energy, calcium consumption, protein intake, smoking status, alcohol consumption, chronic conditions, and weight change.

our knowledge, jogging pace has not been ex- amined in previous studies. Several studies have examined the effect of jogging distance on BMD in men, and all but one report22 sug- gested that there may be a ceiling beyond which additional distance does not improve BMD. For example, one study reported that men who ran more than 64 km per week had signifi- cantly lower vertebral BMD than nonrunners,23

and another concluded that male long-distance runners had reduced BMD and increased bone turnover compared with controls.16 Mac- Dougall et al.14 found no further increase in lower-leg BMD in male runners who ran more than 20 miles per week compared with those who ran less; in fact, increased distance was potentially detrimental (i.e., mean BMD in those who ran 60–75 miles per week was sim- ilar to that in controls). Our results showing similar femoral BMD in those who jogged fre- quently (i.e., >20 times/month) and those who jogged less frequently, while exploratory in na- ture, were consistent with these studies. The mechanism underlying this ceiling effect is probably multifactorial and may include factors such as body weight14 or sex hormone status.24

Our study focused on the proximal femur, which is a skeletal site that receives increased loading from jogging. The effect of jogging on other skeletal sites, which may not be as di- rectly loaded, has varied. One longitudinal study found significantly higher lumbar spine BMD levels in older male runners,11 but other studies have found similar lumbar spine BMD levels in runners and controls.1,15 Running also does not appear to affect BMD in the forearm.3

Our study has the advantage of including joggers with widely diverse jogging frequency. Thus, our results may be more applicable to the general population than those of previous studies, which tended to focus on more dedi-

cated or elite athletes. However, our study has several limitations. It is cross-sectional in na- ture and therefore cannot provide definitive ev- idence that jogging caused the higher BMD levels observed in male joggers. The dose– response seen for jogging frequency is sug- gestive in this regard, but it is also possible that self-selection played a role (i.e., those who chose to jog had higher BMD before com- mencing to jog). Results did not change when we included in the analysis self-reported health status or the presence of chronic conditions, which suggests that the relationship between jogging and BMD is not due to differences in health status between joggers and nonjoggers. However, self-selection could also occur be- cause those who choose to jog are better adapted physically to perform this activity (i.e., their musculoskeletal system, including BMD, is better suited to jogging). This would not be addressed by the overall health status or chronic condition variables.

We also used a self-reported assessment of jogging in our study rather than a direct mea- surement. However, other data suggest that self-reported physical activity is reasonably valid. For example, in the predictive models of cardiorespiratory fitness that also included weight, resting heart rate, and current smok- ing, self-reported physical activity was found to be the principal contributor.25 The associa- tions consistently found between self-reported physical activity and lower risk of several dis- eases also support the validity of the measure.25

Our study looked at leisure-time physical ac- tivity only, so it could underestimate total phys- ical activity. However, most people do not have jobs requiring regular physical exertion.25

We conclude that jogging is associated with significantly higher femoral BMD in men. This effect was graded in nature and was ob-

served in both younger and middle-aged men. This finding may have public health signifi- cance, since femoral BMD is a strong predic- tor of hip fracture, which is the most devastat- ing consequence of osteoporosis from a public health standpoint. Most important, our data suggest that even a modest frequency of jog- ging is associated with higher femoral BMD and thus may provide some protection against osteoporosis and fracture. Additional large- scale studies that measure all aspects of jog- ging are warranted.

Contributors M. E. Mussolino was primarily responsible for the study design, writing, and data analyses. A. C. Looker provided input into the initial study design, wrote parts of the “Discussion” section, and revised other parts of the manuscript. E. S. Orwoll provided guidance on interpreting data and contributed to the analysis of the impact of other leisure activities on the relation- ship between jogging and BMD. All authors partici- pated in preparing the manuscript.

References 1. Brahm H, Strom H, Piehl-Aulin K, Mallmin H,

Ljunghall S. Bone metabolism in endurance trained athletes: a comparison to population- based controls based on DXA, SXA, quantitative ultrasound, and biochemical markers. Calcif Tis- sue Int. 1997;61:448–454.

2. Kujala UM, Kaprio J, Kannus P, Sarna S, Koskenvuo M. Physical activity and osteoporotic hip fracture risk in men. Arch Intern Med. 2000; 160:705–708.

3. Joakimsen RM, Magnus JH, Fonnebo V. Phys- ical activity and predisposition for hip fractures: a review. Osteoporos Int. 1997;7:503–513.

4. Petit MA, Prior JC, Barr SI. Running and ovu- lation positively change cancellous bone in pre- menopausal women. Med Sci Sports Exerc. 1999;31:780–787.

5. Emslander HC, Sinaki M, Muhs JM, et al. Bone mass and muscle strength in female college ath- letes (runners and swimmers). Mayo Clin Proc. 1998;73:1151–1160.

6. Taaffe DR, Robinson TL, Snow CM, Marcus R. High-impact exercise promotes bone gain in well-trained female athletes. J Bone Miner Res. 1997;12:255–260.

7. Etherington J, Harris PA, Nandra D, et al. The effect of weight-bearing exercise on bone min- eral density: a study of female ex-elite athletes and the general population. J Bone Miner Res. 1996;11:1333–1338.

8. Snow-Harter C, Bouxsein ML, Lewis BT, Carter DR, Marcus R. Effects of resistance and en- durance exercise on bone mineral status of young women: a randomized exercise interven- tion trial. J Bone Miner Res. 1992;7:761–769.

9. Jonsson B, Ringsberg K, Josefsson PO, Johnell O, Birch-Jensen M. Effects of physical activity on bone mineral content and muscle strength in women: a cross-sectional study. Bone. 1992;13: 191–195.

10. Need AG, Wishart JM, Scopacasa F, Horowitz M, Morris HA, Nordin BEC. Effect of physical

July 2001, Vol. 91, No. 7 American Journal of Public Health 1059

activity on femoral bone density in men. BMJ. 1995;310:1501–1502.

11. Lane NE, Oehlert JW, Bloch DA, Fries JF. The relationship of running to osteoarthritis of the knee and hip and bone mineral density of the lumbar spine: a 9 year longitudinal study. J Rheumatol. 1998;25:334–341.

12. Stewart AD, Hannan J. Total and regional bone density in male runners, cyclists, and controls. Med Sci Sports Exerc. 2000;32:1373–1377.

13. Bennell KL, Malcolm SA, Khan KM, et al. Bone mass and bone turnover in power athletes, endurance athletes and controls: a 12-month lon- gitudinal study. Bone. 1997;20:477–484.

14. MacDougall JD, Webber CE, Martin J, et al. Re- lationship among running mileage, bone den- sity, and serum testosterone in male runners. J Appl Physiol. 1992;73:1165–1170.

15. Goodpaster BH, Costill DL, Trappe SW, Hughes GM. The relationship of sustained exercise train-

ing and bone mineral density in aging male run- ners. Scand J Med Sci Sports. 1996;6:216–221.

16. Hetland ML, Haarbo J, Christiansen C. Low bone mass and high bone turnover in male long distance runners. J Clin Endocrinol Metab. 1993;77:770–775.

17. Greendale GA, Barrett-Connor E, Edelstein S, Ingles S, Haile R. Lifetime leisure exercise and osteoporosis. The Rancho Bernardo Study. Am J Epidemiol. 1995;141:951–959.

18. Plan and operation of the Third National Health and Nutrition Examination Survey, 1988–94. National Center for Health Statistics. Vital Health Stat 1. 1994;No. 32.

19. Wahner HW, Looker A, Dunn WL, Hauser MF, Walters LC, Novak C. Quality control of bone densitometry in a national health survey (NHANES III) using three mobile examination centers. J Bone Miner Res. 1994;9:951–960.

20. SAS/STAT User’s Guide, Version 6. 4th ed. Cary, NC: SAS Institute Inc; 1990.

21. Shah BV, Barnwell BG, Bieler GS. SUDAAN User’s Manual: Software for Analysis of Corre- lated Data, Release 6.40. Research Triangle Park, NC: Research Triangle Institute; 1995.

22. Lane NE, Bloch DA, Jones HH, Marshall WH Jr, Wood PD, Fries JF. Long-distance running, bone density, and osteoarthritis. JAMA. 1986; 255:1147–1151.

23. Bilanin JE, Blanchard MS, Russek-Cohen E. Lower vertebral bone density in male long dis- tance runners. Med Sci Sports Exerc. 1989;21: 66–70.

24. Cooper CS, Taaffe DR, Guido D, Packer E, Hol- loway L, Marcus R. Relationship of chronic en- durance exercise to the somatotropic and sex hormone status of older men. Eur J Endocrinol. 1998;138:517–523.

25. Physical Activity and Health: A Report of the Surgeon General. Atlanta, Ga: National Center for Chronic Disease Prevention and Health Pro- motion; 1996.