BODY WEIGHT and fatness increase with age in women,

0021-972X/97/$03.00/0 Vol. 82, No. 10 Journal of Clinical Endocrinology and Metabolism Printed in U.S.A. Copyright 1997 by The Endocrine Society Regular Exercise and the Age-Related Decline in Resting Metabolic Rate in Women* RACHAEL E. VAN PELT, PAMELA P. JONES, KEVIN P. DAVY, CHRISTOPHER A. DESOUZA, HIROFUMI TANAKA, BRENDA M. DAVY, AND DOUGLAS R. SEALS Human Cardiovascular Research Laboratory, Center for Physical Activity, Disease Prevention, and Aging, Department of Kinesiology (R.E.V.P., P.P.J., K.P.D., C.A.D., H.T., D.R.S.), University of Colorado, Boulder, Colorado 80309; and the Department of Medicine, Divisions of Cardiology and Geriatric Medicine (D.R.S.), and the Department of Pediatrics, Center for Human Nutrition (B.M.D.), University of Colorado Health Sciences Center, Denver, Colorado 80262 ABSTRACT A low resting metabolic rate (RMR) is a risk factor for future weight gain. We tested the hypothesis that the age-related decline in RMR in sedentary women is not observed in women who exercise regularly. Sixty-five healthy, weight-stable women, aged 21 35 or 50 72 yr, were studied: 12 premenopausal and 15 postmenopausal sedentary women, 13 premenopausal and 15 postmenopausal distance runners, and 10 endurance-trained postmenopausal swimmers. RMR was measured by indirect calorimetry (ventilated hood system) after an overnight fast, and values were adjusted for fat mass and fat-free mass (RMR adj ). The RMR adj was 10% lower in the postmenopausal vs. premenopausal sedentary women (52 2 vs. 57 2 Cal/h; P 0.002). In contrast, RMR adj was not significantly different in the premenopausal (59 2 Cal/h) and postmenopausal (57 1 Cal/h) Received March 26, 1997. Revision received June 20, 1997. Accepted June 27, 1997. Address all correspondence and requests for reprints to: Douglas R. Seals, Ph.D., Department of Kinesiology, University of Colorado, Campus Box 354, Boulder, Colorado 80309. E-mail: seals@spot.colorado.edu. * This study was supported by NIH Awards RO1-AG-06537, RO1- AG-13038, and RO1-HL-39966 (to D.R.S.); KO1-AG-00687 and F32-HL- 08834 (to K.P.D.), F32-AG-05705 (to P.P.J.); and F32-AG-05717 (to H.T.) and by Research Supplement to Minority Individuals in Postdoctoral Training Awards RO1-HL-39966 and RO1-AG-13038 (to C.A.D.). distance runners. The postmenopausal swimmers had a RMR adj (57 2 Cal/h) identical to that of the postmenopausal runners, suggesting a generalized influence of the endurance exercise-trained state in postmenopausal women. Group differences in RMR adj were not associated with differences in total energy intake or composition or with plasma concentrations of norepinephrine, T 3,orT 4. However, maximal oxygen consumption (aerobic fitness) accounted for 35% of the individual variance in RMR adj in the overall population (r 0.59; P 0.001). Our results are consistent with the concept that the agerelated decline in RMR in sedentary women is not observed in women who regularly perform endurance exercise. The elevated level of RMR observed in middle-aged and older exercising women may play a role in their lower levels of body weight and fatness compared to those in sedentary women. (J Clin Endocrinol Metab 82: 3208 3212, 1997) BODY WEIGHT and fatness increase with age in women, especially during the postmenopausal years, and this is associated with an increased risk of morbidity and premature mortality (1 3). The increases in body mass and fat mass are probably due to age-related reductions in energy expenditure that are disproportionately greater than the reductions in energy intake that occur with age (4, 5). Physical activity is an important component of daily energy expenditure and declines with age in women (6); thus, this factor probably contributes to the observed weight gain. However, resting metabolic rate (RMR), which accounts for 60 75% of daily energy expenditure (7), decreases with age in women (8, 9) and is thought to play an important role in the ageassociated increases in body weight and fatness. The fact that individuals with low RMR are at greater risk for future weight gain than those with high RMR (10) is consistent with this concept. Compared with that observed in sedentary women, the age-related increases in body weight and in total and regional body fatness are smaller or even absent in women who exercise regularly (11 14). This could be due in part to their high physical activity-related energy expenditure. Additionally, the smaller age-associated increases in body mass and fat mass in highly active women could be related to a maintenance of RMR with advancing age. There is some evidence for this in men (15). Accordingly, the aim of the present investigation was to test the hypothesis that RMR declines with age in sedentary women, but not in women who perform regular endurance-type exercise. Subjects Subjects and Methods Sixty-five healthy women, aged 21 35 or 50 72 yr, were studied: 12 premenopausal and 15 postmenopausal sedentary women, 13 premenopausal and 15 postmenopausal distance runners, and 10 endurancetrained postmenopausal swimmers. The swimmers were included to establish that a higher RMR in postmenopausal runners vs. sedentary women, if observed, was related to endurance training in general (i.e. was not specific to runners). The pre- and postmenopausal runners were matched for age-adjusted competitive performance (both groups, 74 2% of age-adjusted world record times) as described previously by our laboratory (16), ran 54 5 and 43 3 km/week, and had been training for 9 1 and 18 2 yr, respectively. The swimmers were matched to 10 postmenopausal runners for years postmenopausal (8 1 yr both groups), hormone replacement use (5/group), hours of exercise per week (3.7 0.5 vs. 4.2 0.3 h), and relative age-adjusted competitive performance (79 5% vs. 77 5%). The sedentary subjects performed no regular physical activity. All subjects had been weight stable ( 2 kg) during the previous 6 months. 3208

AGING, EXERCISE, AND METABOLIC RATE IN WOMEN 3209 All of the postmenopausal women were at least 2 yr postmenopausal (individual subject range, 2 to 23 yr; mean group range, 8 11 yr), and approximately half of each group were taking estrogen-based hormone supplements (eight sedentary, eight runners, and five swimmers). All premenopausal subjects were eumenorrheic, as assessed by self-report of menstrual cycles, and none was taking oral contraceptives. The range of menstrual cycle length was 21 37 days (mean cycle length, 28 days). All subjects were healthy, as assessed by medical history. Postmenopausal subjects were further evaluated for clinical evidence of cardiopulmonary disease with a physical examination and electrocardiograms during rest and maximal exercise. No subjects reported using any other medications. All subjects were nonsmokers. The nature, purpose, and risks of the study were explained to each subject before written informed consent was obtained. The experimental protocol was approved by the human research committee at the University of Colorado-Boulder. Body mass and composition Total body mass was measured to the nearest 0.1 kg on a physician s balance scale (Detecto, Webb City, MO). Body mass index (BMI) was calculated from weight and height (kg/m 2 ). Total body density was determined by hydrodensitometry as outlined by Brozek et al. (17). Residual volume of the lungs was measured using a nitrogen dilution technique as previously outlined by Wilmore (18). Body fat percentage was then calculated using the equation of Brozek et al. (17). Fat mass and fat-free mass were estimated from the percent fat and body mass based on the two-compartmental model. The waist to hip ratio, a measure of abdominal adiposity, was determined from waist circumference, measured at the narrowest part of the torso, and hip circumference, measured at the maximal extension of the buttocks (19). RMR Subjects were studied after a 12-h overnight fast. RMR was measured in the runners and swimmers 24 h after their last exercise session. All premenopausal women were tested during the follicular phase (days 1 10) of their menstrual cycle. Measurements were performed between 0600 0900 h in a dimly lit room at a comfortable temperature ( 23 C). Subjects remained awake in a semirecumbent position. After a 15-min habituation period in the hood, oxygen consumption and carbon dioxide production were measured each minute for 30 min by indirect calorimetry using a ventilated hood system (DeltaTrac Metabolic Monitor, SensorMedics Corp., Yorba Linda, CA). RMR was then calculated from the average of the 30 min using the Weir formula (20). The reliability of this method was established in our laboratory before the study by testing pilot subjects on 2 consecutive days (n 5). The intraclass correlation was 0.99, with a difference of 3.5% between trials. Aerobic fitness Maximal oxygen consumption (VO 2 max) was used as a measure of aerobic fitness and was determined using an on-line computer-assisted open circuit spirometry system as previously described (16). TABLE 1. Subject characteristics for the sedentary and active women Estimated energy intake Energy intake was determined from 4-day food diaries recorded for 4 consecutive days, including 3 weekdays and 1 weekend day. Each subject weighed (Dayton Hudson diet scale, 8-oz capacity) and recorded all food and beverages consumed. Three-day practice records were completed by each subject before the 4-day records, and a registered dietitian instructed each on ways to improve the accuracy of their diaries. A registered dietitian analyzed all diets for energy and macronutrient intake using the Nutritionist IV (version 3.5.2, The Hearst Corp., San Bruno, CA) computer software program. Plasma norepinephrine and thyroid hormone concentrations All subjects were studied at 0730 h after a 12-h fast. Blood samples were taken after subjects had rested in the supine position for 20 min. Plasma norepinephrine (NE) levels were determined by radioenzymatic assay from an antecubital venous blood sample (21). Plasma total T 4 and total T 3 concentrations were measured using competitive binding RIAs (22, 23). Statistics Group differences for most variables were determined by ANOVA. Group differences in RMR were determined by analysis of covariance, with fat mass and fat-free mass as the two covariates; the adjusted means (RMR adj ) and ses are presented. A Newman-Keuls post-hoc test for multiple comparisons was used to analyze differences among the dependent variables. When a significant interaction was not observed, specific mean comparisons were performed on the main effects. Zero order correlations were performed to determine relations between variables in the overall study population. To identify significant independent predictors of RMR adj, stepwise multiple regression analysis also was performed. Adjusted maximal oxygen consumption values (VO 2 max adj ), covaried for fat mass and fat-free mass, were used in correlational analyses. Because RMR did not differ between users and nonusers of hormone replacement within and across groups, subjects were pooled for all analyses. The level of statistical significance was set at P 0.05. Results Table 1 presents selected characteristics of the sedentary women. The postmenopausal women had higher levels of body mass, BMI, total body fatness, and waist to hip ratio and lower maximal oxygen consumption than the premenopausal women (P 0.05). Fat-free mass did not differ with age. RMR adj was 10% lower in the postmenopausal compared with the premenopausal women (52 2 vs. 57 2 Cal/h; P 0.001; Fig. 1A). Based on analysis of diet records, there were no significant differences in absolute carbohy- Age (yr) 29 1 61 1 a 31 1 58 1 a Body mass (kg) 62.2 3.5 70.5 2.7 a 54.3 2.0 58.9 1.2 a Ht (cm) 166 1 162 3 166 1 167 1 Body mass index (kg/m 2 ) 22.6 1.3 27.0 1.0 a 19.6 0.5 21.0 0.3 % Body fat 27.4 2.3 39.8 1.7 a 15.4 1.3 25.7 1.5 a Fat mass (kg) 17.6 2.5 28.0 2.2 a 8.4 0.8 15.2 1.0 a Fat-free mass (kg) 44.6 1.6 41.2 1.0 45.8 1.6 43.7 1.1 Waist-hip ratio 0.75 0.01 0.83 0.02 a 0.74 0.01 0.74 0.01 Specific training (hr/wk) NA NA 5.3 0.5 4.2 0.3 Other exercise (h/wk) 1.8 0.4 1.5 0.5 6.2 0.9 6.7 1.3 VO 2 max (ml/kg min) 34.3 1.6 22.1 1.3 a 53.0 1.6 38.0 1.6 a Data are the means SE. NA, not applicable. a By two-way ANOVA, P 0.05 vs. younger group (same activity).

3210 VAN PELT ET AL. JCE&M 1997 Vol 82 No 10 FIG. 1. Mean ( SE) values for RMR (Cal/h) adjusted for fat mass and fat-free mass. A, Pre- vs. postmenopausal sedentary women; B, pre- vs. postmenopausal distance runners; C, matched groups of postmenopausal distance runners vs. postmenopausal endurance-trained swimmers. drate, fat, protein, or total estimated energy intake in the premenopausal vs. postmenopausal women, although the postmenopausal women had lower carbohydrate and protein intakes when expressed per kg body mass (BM) (Table 2). Plasma concentrations of norepinephrine, T 3, and T 4 also were not different in the two groups (Table 3). Endurance-trained runners Subject characteristics for the runners are presented in Table 1. The postmenopausal runners demonstrated higher levels of body mass and fatness and lower maximal oxygen consumption than the premenopausal controls (P 0.05). Fat-free mass, BMI, and waist to hip ratio were not different between the groups. In contrast to the age-related difference in RMR adj observed in the sedentary women, RMR adj was not different in the pre- and postmenopausal runners (59 2 vs. 57 1 Cal/h; Fig. 1B). Estimated absolute energy intake and composition also did not differ with age, although the postmenopausal women had lower carbohydrate and protein intakes when normalized per kg BM (Table 2). No differences were observed in plasma norepinephrine, T 3, and T 4 levels between the groups (Table 3). Postmenopausal swimmers vs. runners Table 4 presents characteristics for the postmenopausal swimmers and the matched subgroup of postmenopausal TABLE 2. Estimated daily energy intake from 4-day diet records runners. The two groups did not differ in age, hormone replacement use, fat-free mass, hours of exercise per week (either specific training or other exercise), or relative performance. The swimmers demonstrated a higher body mass, BMI, body fatness, and waist to hip ratio and lower maximal oxygen consumption than the runners (P 0.05). RMR adj was identical in the postmenopausal runners and swimmers (57 2 vs. 57 2 Cal/h; Fig. 1C). No group differences were observed in estimated energy intake and composition or in plasma levels of norepinephrine, T 3,orT 4. Correlates of RMR in the overall population Significant univariate correlates of RMR adj in the overall subject population included: VO 2 max adj (r 0.59; P 0.001; Fig. 2), total energy intake (r 0.48; P 0.001), carbohydrate intake (r 0.50; P 0.001), protein intake (r 0.27; P 0.05), and training hours per week (r 0.29; P 0.05). Of these variables, VO 2 max adj entered the stepwise multiple regression analysis first and accounted for 35% of the overall variance (P 0.001); total energy intake was the only other variable to enter and accounted for an additional 6% of the variance (P 0.05). Discussion The primary new finding from the present study is that a significant decrease in RMR adj with age in healthy sedentary Total Cal 2000 156 1732 75 2403 139 1995 138 CHO (Cal) 1025 68 (51) 956 54 (55) 1491 109 (62) 1138 90 (57) CHO (g/kg BM) 4.2 0.3 3.4 0.2 a 7.2 0.6 b 4.9 0.4 a,b Protein (Cal) 273 20 (14) 264 11 (15) 324 13 (13) 310 18 (15) Protein (g/kg BM) 1.1 0.1 1.0 0.1 a 1.6 0.1 b 1.3 0.1 a,b Fat (Cal) 658 78 (33) 527 43 (30) 594 97 (25) 551 40 (28) Fat (g/kg BM) 1.2 0.1 0.8 0.1 1.3 0.2 1.0 0.1 Values are the mean SE. Percentages are given in parentheses. Significance was determined by two-way ANOVA. Current RDA recommendations for protein, carbohydrate, and fat intake are 0.8 g/kg body mass (BM), less than 50%, CHO (or 4 g/kg BM), and less than 30% respectively. a P 0.05 vs. premenopausal women (same activity). b P 0.05 vs. sedentary women (same age group).

AGING, EXERCISE, AND METABOLIC RATE IN WOMEN 3211 TABLE 3. Plasma concentrations of norepinephrine (NE) and thyroid hormones NE (pg/ml) 426 51 428 34 429 40 472 68 T 3 (ng/dl) 114 6 130 5 113 10 129 4 T 4 ( g/dl) 6.5 0.5 7.6 0.4 6.0 0.4 7.1 0.3 Values are mean SE. Normal ranges: T 3, 90 190 ng/dl; T 4, 4.0 12.0 g/dl. Significance was determined by two-way ANOVA. women is not observed in women who perform endurance exercise on a regular basis. This is supported by the facts that both the postmenopausal runners and swimmers demonstrated levels of RMR adj not different from those in young adult endurance athletes, whereas a significantly lower RM- R adj was observed in the postmenopausal compared to the premenopausal sedentary women. The present data in women are consistent with a previous report by Poehlman and colleagues in men (15). The differences in RMR adj observed with age in the sedentary and active women in the present study, when viewed over time, may have considerable physiological significance. Using the mean RMR adj values in calories per h for the preand postmenopausal women, the net age-related decline in energy expenditure would be approximately 600 Cal/week more in the sedentary compared to the exercising women. Thus, this age-related difference in energy expenditure, if uncompensated completely by a decrease in energy intake, would equate to about a 4 kg/yr greater increase in body weight in the sedentary women. Given this, our findings may have important clinical implications. For example, body weight and fatness increase with age in women (24 26), and recent evidence indicates that even modest weight gain in women with advancing age is associated with markedly increased risks of noninsulindependent diabetes mellitus (27), coronary artery disease (28), and overall mortality (29, 30). We (13, 14) and others (11, 12) have shown that the age-related increases in body weight and body fatness are either smaller or absent in endurance exercise-trained compared to sedentary women. Because RMR accounts for 60 75% of daily energy expenditure and is a risk factor for future body weight gain (10), the higher RMR observed in our exercising postmenopausal women TABLE 4. Subject characteristics for the runners and swimmers Postmenopausal runners (n 10) Postmenopausal swimmers (n 10) Age (yr) 58 2 59 2 Body mass (kg) 59.0 1.3 65.0 2.0 a Ht (cm) 168 1 167 2 Body mass index 21.0 0.2 23.3 0.3 a (kg/m 2 ) % Body fat 23.3 1.7 29.4 2.4 a Fat mass (kg) 13.8 1.1 19.5 2.1 a Fat-free mass (kg) 45.2 1.3 45.5 0.8 Waist/hip ratio 0.74 0.01 0.79 0.02 a Specific training 4.2 0.3 3.7 0.5 (h/wk) Other exercise (h/wk) 6.7 1.3 3.9 0.8 VO 2 max (ml/kg/min) 40.0 1.9 30.7 2.1 a Values are the mean SE. a By one-way ANOVA, P 0.05. FIG. 2. The relationship between RMR and maximal oxygen consumption (aerobic fitness), normalized for fat mass and fat-free mass, in the overall population (A) and the individual subgroups (B). could play a role in their ability to maintain a more favorable body weight and composition with age. Moreover, meeting nutritional requirements is a problem for some older women (31). In this context, the higher RMR in physically active postmenopausal women would allow them to maintain a higher total energy intake and, therefore, have a greater likelihood of meeting their dietary needs. Several factors not directly related to exercise have been reported to be associated with RMR in certain populations or conditions, including thyroid hormones (32, 33), total energy and macronutrient intake (34), and sympathetic nervous system activity (35, 36). In the present study, however, there was

3212 VAN PELT ET AL. JCE&M 1997 Vol 82 No 10 no consistent relation between RMR and any of these putative mechanisms. Fat-free mass and fat mass are important determinants of RMR (7, 8, 33). However, we found that RMR declined with age in the sedentary, but not in the physically active, women after adjusting for these factors. Thus, our findings suggest that RMR per unit of metabolically active tissue is higher in exercise-trained vs. sedentary postmenopausal women. In the present study, RMR adj and aerobic fitness (VO 2 max adj ) were positively related in the overall population (r 0.59) as well as in the various subpopulations (r 0.51 0.76). Aerobic fitness also was the strongest independent predictor of RMR adj, explaining 35% of the overall variance. A significant relationship between VO 2 max and RMR across the adult age range has been found previously in active and sedentary men (37, 38), but not in sedentary women (33). Our results indicate that aerobic fitness is significantly related to RMR among healthy females varying in age and exercise status. We should emphasize at least three limitations of the present study. First, using a model that we have employed in the past (13, 16), we attempted to minimize constitutional differences between the pre- and postmenopausal endurance runners by matching them for age-adjusted performance. Despite this, however, because of the cross-sectional nature of our study design we cannot discount the possibility that genetic or constitutional factors influenced our findings. Secondly, it is unknown whether RMR declines significantly in endurance-trained women beyond the age range studied (i.e. beyond 72 yr of age) or at what point RMR begins to decline during the perimenopausal years (i.e. between 35 50 yr of age) in sedentary women. Lastly, we compared groups who were very different in their activity levels (i.e. sedentary and endurance-trained subjects). Thus, our results do not address the question of the minimum level of habitual exercise that is associated with a diminished age-related decline in RMR. In conclusion, the results of the present study provide experimental evidence that is consistent with the concept that the age-related decline in RMR in sedentary women is not observed in women who regularly perform endurance exercise. The absence of a significant decline in RMR in middleaged and older endurance-trained women may play a role in the maintenance of their lower levels of body weight and fatness compared to those in sedentary women. Acknowledgments The authors thank Drs. Christopher L. Melby and James O. Hill for their consultation throughout the study, and Dr. Melby for his preliminary review of this manuscript. References 1. Durnin JVGA, Womersly J. 1974 Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. Br J Nutr. 32:77 97. 2. Najjar M, Rowland M. 1987 Anthropometric reference data and prevalence of overweight. United States, 1976 1980. Vital Health Statistics 238. 3. Despres J-P, Lamarche B. 1994 Low-intensity endurance exercise training, plasma lipoproteins and the risk of coronary heart disease. J Intern Med. 236:7 22. 4. McGandy R, Barrows C, Spanias A, Meredith A, Stone J, Norris A. 1966 Nutrient intakes and energy expenditure in men of different ages. J Gerontol. 21:581 587. 5. Poehlman ET. 1993 Regulation of energy expenditure in aging humans. J Am Geriat Soc. 42:552 559. 6. Crespo C, Keteyian S, Heath G, Sempos C. 1996 Leisure-time physical activity among US adults. Results from the Third National Health and Nutrition Examination Survey. Arch Intern Med. 156:93 98. 7. Tataranni P, Ravussin E. 1995 Variability in metabolic rate: biological sites of regulation. Int J Obesity. 19:S102 S106. 8. Fukagawa NA, Bandini LG, Young JB. 1990 Effect of age on body composition and resting metabolic rate. Am J Physiol. 259:E233 E238. 9. Vaughan L, Zurlo F, Ravussin E. 1991 Aging and energy expenditure. Am J Clin Nutr. 53:821 825. 10. Ravussin E, Lillioja S, Knowler W, et al. 1988 Reduced rate of energy expenditure as a risk factor for body-weight gain. N Engl J Med. 318:467 472. 11. Drinkwater B, Horvath S, Wells C. 1975 Aerobic power of females, ages 10 to 68. J Gerontol. 30:385 394. 12. Kohrt WM, Malley MT, Dalsky GP, Holloszy JO. 1992 Body composition of healthy sedentary and trained, young and older men and women. Med Sci Sport Exer. 24:832 837. 13. Davy KP, Evans SL, Stevenson ET, Seals DR. 1996 Adiposity and regional body fat distribution in physically active young and middle-aged women. Int J Obesity. 20:777 783. 14. Van Pelt R, Davy K, Stevenson E, Jones P, DeSouza C, Seals D. 1996 Physical activity/aerobic fitness and age-related increases in abdominal and total body fat in women. Physiol. 39:A-88. 15. Poehlman ET, Melby CL, Badylak SF. 1991 Relation of age and physical exercise status on metabolic rate in younger and older healthy men. J Gerontol. 46:B54 B58. 16. Evans S, Davy K, Stevenson E, Seals D. 1995 Physiological determinants of 10-km performance in highly trained female runners of different ages. J Appl Physiol. 78:1931 1941. 17. Brozek J, Grande F, Anderson JT, Keys A. 1963 Densitometric analysis of body composition: revision of some quantitative assumptions. Ann NY Acad Sci. 110:113 140. 18. Wilmore JH. 1969 A simplified technique for determination of residual lung volumes. J Appl Physiol. 27:96 100. 19. Lohman TG, Roche AF, Martorell R. 1988 Anthropometric standardization reference manual. Champaign: Human Kinetics. 20. Weir JBDV. 1949 New methods for calculating metabolic rate with special reference to protein metabolism. J Physiol. 109:1 9. 21. Peuler JD, Johnson GA. 1977 Simultaneous single isotope radioenzyme assay of plasma norepinephrine, epinephrine, and dopamine. Life Sci. 21:625 636. 22. Fang V, Fefetoff S. 1974 Radioimmunoassay for serum triiodothyronine: evaluation of simple techniques to control interference from binding proteins. Clin Chem. 20:1150. 23. Bigos S, Ridgway E, Kourides I, Maloof F. 1978 Spectrum of pituitary alterations with mild and severe thyroid impairment. J Clin Endocrinol Metab. 46:317. 24. Wing RR. 1992 Obesity and weight gain during adulthood: a health problem for the United States women. Womens Health Inst. 2:114 122. 25. Kuczmarski RJ, Flegal KM, Campbell SM, Johnson CL. 1994 Increasing prevalence of overweight among US adults: the National Health and Nutrition Examination Surveys, 1960 to 1991. JAMA. 272:205 211. 26. Going SB, Williams DP, Lohman TG, Hewitt MJ. 1994 Aging, body composition, and physical activity: a review. J Aging Phys Act. 2:38 66. 27. Colditz G, Willet W, Rotnitzky A, Manson J. 1995 Weight gain as a risk factor for clinical diabetes mellitus in women. Ann Intern Med. 122:481 486. 28. Willet W, Manson J, Stampfer M, et al. 1995 Weight, weight change and coronary artery disease in women. JAMA. 273:461 465. 29. Despres JP, Moorjani S, Lupien PJ, Tremblay A, Nadeau A, Bouchard C. 1990 Regional distribution of body fat, plasma lipoproteins, and cardiovascular disease. Arterioscler Thromb. 10:497 511. 30. Kissebah A, Krakower R. 1994 Regional adiposity and mortality. Physiol Rev. 74:791 811. 31. Morley JE. 1986 Nutritional status of the elderly. Am J Med. 81:679 695. 32. Danforth E. 1983 The role of thyroid hormones and insulin in the regulation of energy metabolism. Am J Clin Nutr. 38:1006 1017. 33. Poehlman ET, Goran MI, Gardner AW, et al. 1993 Determinants of the decline in resting metabolic rate in aging females. Am J Physiol. 264:E450 E455. 34. Horton T, Drougas H, Brachy A, Reed G, Peters J, Hill J. 1995 Fat and carbohydrate overfeeding in humans: different effects on energy storage. Am J Clin Nutr. 62:19 29. 35. Kurpad A, Khan K, Calder A, Elia M. 1994 Muscle and whole body metabolism after norepinephrine. Am J Physiol. 266:E877 E884. 36. Welle S, Schwartz R, Statt M. 1991 Reduced metabolic rate during -adrenergic blockade in humans. Metabolism. 40:619 622. 37. Poehlman E, McAuliffe T, Van Houten D, Danforth E. 1990 Influence of age and endurance training on metabolic rate and hormones in healthy men. Am J Physiol. 259:E66 E72. 38. Poehlman ET, Berke EM, Joseph JR. 1992 Influence of aerobic capacity, body composition and thyroid hormones on the age-related decline in resting metabolic rate. Metabolism. 41:915 921.