Physical Frailty and Body Composition in Obese Elderly Men and Women


  • Dennis T. Villareal,

    Corresponding author
    1. Division of Geriatrics and Nutritional Science, Department of Internal Medicine, Washington University School of Medicine, St. Louis, Missouri
    Search for more papers by this author
  • Marian Banks,

    1. Division of Geriatrics and Nutritional Science, Department of Internal Medicine, Washington University School of Medicine, St. Louis, Missouri
    Search for more papers by this author
  • Catherine Siener,

    1. Division of Geriatrics and Nutritional Science, Department of Program in Physical Therapy, Washington University School of Medicine, St. Louis, Missouri
    Search for more papers by this author
  • David R. Sinacore,

    1. Division of Geriatrics and Nutritional Science, Department of Internal Medicine, Washington University School of Medicine, St. Louis, Missouri
    2. Division of Geriatrics and Nutritional Science, Department of Program in Physical Therapy, Washington University School of Medicine, St. Louis, Missouri
    Search for more papers by this author
  • Samuel Klein

    1. Division of Geriatrics and Nutritional Science, Department of Internal Medicine, Washington University School of Medicine, St. Louis, Missouri
    Search for more papers by this author

  • The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Washington University Older Adult Health Center, 4488 Forest Park Boulevard, St. Louis, MO 63108. E-mail:


Objective: To evaluate the prevalence of frailty and interrelationships among body composition, physical function, and quality of life in community-dwelling obese elderly (OE) persons.

Research Methods and Procedures: Fifty-two OE, 52 nonobese frail, and 52 nonobese nonfrail subjects, matched for age and sex, were studied. Subjective and objective measures of functional status were evaluated by using the physical performance test, exercise stress test, lower extremity (LE) strength, gait speed, static and dynamic balance, functional status questionnaires, and health-related quality-of-life questionnaire (Medical Outcomes Short Form). Body composition was evaluated by using DXA, and muscle quality was evaluated by determining the ratio of LE strength to LE lean mass.

Results: Among OE subjects, 96% met our standard criteria for mild to moderate frailty. Compared with the nonobese nonfrail group, the OE and nonobese frail groups had lower and similar scores in physical performance test, peak aerobic power, and functional status questionnaire, and exhibited similar impairments in strength, walking speed, balance, and health-related quality of life. Although absolute fat-free mass (FFM) was greater, the percentage body weight as FFM and muscle quality was lower in the OE group than in the other two groups.

Discussion: Physical frailty, which predisposes to loss of independence, is common in community-living OE men and women. Physical frailty in OE subjects was associated with low percentage FFM, poor muscle quality, and decreased quality of life. These findings suggest that weight loss therapy may be particularly important in OE persons to improve physical function, in addition to improving the medical complications associated with obesity.


The number of obese elderly (OE)1 persons in the United States is rapidly increasing because of an increase in the elderly population and the percentage of elderly persons who are obese (1). Obesity has important functional implications in elderly persons because it worsens the age-related decline in physical function, which can lead to frailty and loss of independence. In fact, obesity may be the single greatest cause of disability in elderly persons (2,3,4,5) and is associated with increased rates of nursing home admissions (6).

Aging is associated with a decrease in skeletal muscle mass and an increase in body fat (7). Disability occurs when the diminished muscle mass causes an inability to perform normal daily activities (8). However, the precise relationship between lean body mass and fat mass that results in disability is not clear because of conflicting data from different studies; some studies suggest that sarcopenia is a major predictor of functional limitations (9), whereas others suggest that fat mass is most important (5,10,11,12). In addition, little is known about body compositional changes that might contribute to disability in OE persons, who may be at particularly high risk for disability because of aging-related decline in muscle mass and muscle quality (13,14) in the face of excessive body fat.

The major aims of the present study were to determine the interrelationship among body composition, physical function, and quality of life in community-dwelling OE persons. We hypothesized that despite higher fat-free mass (FFM) (15), OE subjects are at high risk for frailty and impaired quality of life because of smaller FFM relative to body weight and poor muscle quality.

Research Methods and Procedures


A total of 156 elderly (age > 65 years old) men and women, recruited from the St. Louis metropolitan area by advertisements, mass mailing, and regional presentations, participated in this study. Fifty-two subjects were obese (BMI ≥ 30 kg/m2) and were matched for age and sex with 52 frail and 52 nonfrail nonobese (BMI < 30 kg/m2) subjects. The criteria for frailty are described below. All subjects were sedentary (i.e., did not participate in regular exercise more than twice a week) and had a stable body weight (±2 kg) over the past year. Subjects completed a comprehensive medical evaluation, which included a history and physical examination, an electrocardiogram, standard blood and urine tests, fasting lipid profile, and an oral glucose tolerance test. Subjects with impaired glucose tolerance, diabetes, diagnosis of cancer within the last five years, use of corticosteroids, androgens, estrogens, and bone acting drugs (e.g., bisphosphonates), cigarette use within the previous year, sensory impairment or physical disability that would interfere with physical testing, or significant cognitive dysfunction were excluded. Informed consent was obtained from all subjects before enrollment into the study, which was approved by the Institutional Review Board and the General Clinical Research Center Scientific Advisory Committee of the Washington University School of Medicine.

Study Measurements

Body Composition Analyses

Total body fat mass, FFM, and bone mineral content were measured by using whole-body DXA (QDR-1000/W; Hologic Inc., Waltham, MA). Bone mineral density (BMD) of the total body, lumbar spine (L2 to L4), and total hip were determined. Regional measurement of the lower extremities (LEs) was made by using a line bisecting the femoral neck (16). The bone mineral-free portion of the LE was used as a measure of LE skeletal muscle mass (17,18). Phantom calibration was performed before each scan. The coefficients of variation for measuring fat mass, bone mineral mass, and FFM were 1.6 ± 1.0%, 0.8 ± 03%, and 1.8 ± 0.9%, respectively. The coefficients of variation for BMD were 0.6 ± 0.2% for whole body, 1.6 ± 1.0% for lumbar spine, and 1.2 ± 0.5% for total hip (19).

Blood Pressure

Systolic and diastolic sitting blood pressures were determined as the average of two readings taken 10 minutes apart after the subject sat quietly for 10 minutes. If the first and second measurements differed by >20 mm, a third measurement was done after another 5 minutes of quiet sitting. An appropriate cuff size was selected for each subject based on arm circumference.

Assessment of Physical Frailty

Modified Physical Performance Test (PPT)

An objective evaluation of overall physical function was obtained by using a modification of the PPT. The standard PPT, developed by Reuben et al. (20), correlates with degree of disability, loss of independence, and mortality (20,21). We modified the PPT to provide more focus on gross motor function by substituting a chair-rise task and a balance task for the writing and simulated eating tasks described in the original PPT (22). The modified PPT includes seven standardized tasks that are timed (50-foot walk, putting on and removing a laboratory coat, picking up a penny from the floor, standing up five times from a 16-in. chair, lifting a 7-pound book to a shelf, climbing one flight of stairs, and standing with feet in side-by-side, semitandem, and full-tandem positions with eyes open) and two additional tasks (climbing up and down four flights of stairs and performing a 360° turn). The score for each item ranges between 0 and 4, with 36 representing a perfect total score for the test. We have successfully used this modified PPT in previous studies conducted in frail elderly subjects (23,24,25,26,27). Test-retest reliability for the total modified PPT score in our laboratory for this population is 0.96 (28).

Peak Aerobic Power (Vo2peak)

Vo2peak was assessed during graded treadmill walking. During a 3- to 5-minute warm-up on the treadmill at 0% grade, the speed was adjusted to identify the fastest comfortable walking speed for the individual. Speed was then held constant during the test, and elevation was progressively increased by 2% to 3% every 2 minutes. As a safety measure, participants were allowed to lightly hold on to a handrail to maintain their balance during the test. Cardiorespiratory data were collected at 30-second intervals by using a computerized system (29). The test was terminated when the participant became too fatigued to continue.

Activities of Daily Living

The Older American Resources and Services Instrument was used to collect information about the use of human assistance or assistive technology for basic activities of daily living and instrumental activities of daily living (30). The Physical Function subscale of the Functional Status Questionnaire (FSQ) (31) was used to assess an individual's difficulty with task performance during the previous month. The FSQ subscale has a possible maximum score of 36.

Definition of Frailty

Because there are no established criteria or criterion standard for frailty, we included measures with established predictive validity for disability and mortality in older populations. Accordingly, as in our previous studies (23,24,25,26), individuals were considered to be mild to moderately frail if they met at least two of the following three criteria: modified PPT score between 18 and 32, Vo2peak of 11 to 18 mL/kg per minute, and report of difficulty or need for assistance with two or more instrumental activities of daily living or one basic activity of daily living.

Specific Physical Function Measures

Tests of strength, balance, and gait were used to evaluate specific deficits associated with physical frailty. These tests correlate closely with scores obtained from PPTs (27).


Knee extensor and flexor strength were evaluated by using a Cybex isokinetic dynamometer while subjects were seated with their back supported and hips placed at 120° of flexion, as previously described (32). Tests were performed at an angular velocity of 60°/s.


Static balance was assessed by having subjects stand on one leg unsupported for a maximum of 30 seconds on each side. These measures were taken during quiet standing allowing for only minimal fluctuation of ankle position or obvious toe clawing, without hopping to maintain balance and without undue upper extremity movement. The test was terminated if the other foot touched the floor or contacted the stance leg/foot, if ankle motion was excessive, or if hopping occurred. Dynamic balance was assessed by using an obstacle course (27). The course consisted of rising from a standard 18-in. chair without using arms, walking forward ∼ 6 feet, stepping over a 2- × 2-in. obstacle, walking forward another ∼6 feet, ascending a 6-in.-high curb, turning around, stepping down off the curb, and returning to the chair, as quickly as possible, safely. The obstacle was stepped over on the return trip as well. The time required to complete the course correlates closely with physical frailty (27).

Ambulatory Ability

Walking speed was measured as the time needed to walk 25 feet as rapidly and safely as possible.

Muscle Quality

Muscle quality is defined by the relationship between muscle strength and muscle mass (force per cross-sectional area of muscle). LE muscle quality was assessed by calculating the ratio of isokinetic torque at the knee (in Newton-meter) during knee extension and knee flexion (13,14), determined by Cybex dynamometer, to LE lean mass (in kilograms), determined by DXA.

Health-Related Quality-of-Life Assessment

The Medical Outcomes Short Form (SF-36) Health Survey was used as a measure of disability to evaluate perception of general health status, functional ability, and well-being. This survey has been validated by several studies (33) and consists of eight domains: physical functioning, role limitations due to physical problems, social functioning, bodily pain, general mental health, role limitations due to emotional problems, vitality, and general health perceptions.

Health Status

Because chronic health conditions (e.g., hypertension, arthritis) may themselves contribute to frailty, the number of chronic conditions was evaluated in each subject. The presence of chronic health conditions was determined by medical history, use of medications, and complete physical examination. These chronic health conditions included history of hypertension, coronary artery disease, atrial fibrillation, heart failure, arthritis, Parkinson's disease, chronic obstructive lung disease, and joint replacement (25).

Statistical Analyses

Comparisons of means among the three groups were made by using two-way ANOVA (group × gender) for quantitative variables and Tukey's test for post hoc analyses. For categorical variables, χ2 test was used to compare the groups. Covariance analyses were used to adjust frailty and physical performance measures for chronic health conditions (i.e., chronic health condition was included as a covariate in the analyses). All analyses were done by using the SPSS/PC statistical program (version for Windows; SPSS, Inc., Chicago, IL). Statistical significance was defined as an α level at or below 0.05. All data are presented as means ± SE. Subanalyses by gender did not change the results (i.e., no significant group × gender interaction effects); therefore, data for women and men were combined for each group.


OE, nonobese frail (NO-F), and nonobese nonfrail (NO-NF) groups were carefully matched for age and sex and had normal blood pressure and serum lipid profile (Table 1). Body composition of the study subjects is shown in Table 1. Average BMI in the NO-NF and NO-F groups was similar and much lower than the BMI of the OE group. The OE group had larger amounts of total fat and percentage body weight as fat than the two nonobese groups. Although the OE group had the largest amount of total FFM, FFM expressed relative to body weight (i.e., percent FFM) was lower in the OE group than the other groups. Value for percentage FFM was intermediate for the NO-F group. Total and regional BMD was higher in the OE group than in the other two groups of subjects.

Table 1.  Characteristics and body composition of subjects
  • *

    Value significantly different from NO-F and NO-NF groups, p < 0.05.

  • Value significantly different from OE and NO-NF groups, p < 0.05.

 Age (years)76.0 ± 0.877.3 ± 0.576.5 ± 0.9
 Women/men (n)25/2725/2725/27
 Systolic blood pressure142 ± 3149 ± 3147 ± 3
 Diastolic blood pressure81 ± 280 ± 281 ± 1
 Total cholesterol (mg/dL)192.3 ± 5.7235 ± 15.4218.0 ± 15.3
 HDL cholesterol (mg/dL)52.6 ± 1.751.2 ± 1.647.2 ± 1.5
 LDL cholesterol (mg/dL)110.9 ± 3.6127.4 ± 5.0123.4 ± 4.8
 Triglycerides (mg/dL)124.5 ± 9.1132.7 ± 10.5142.5 ± 9.8
Body composition
 BMI (kg/m2)25.9 ± 0.425.0 ± 0.436.6 ± 0.6*
 Weight (kg)72.7 ± 1.371.1 ± 1.396.5 ± 2.3*
 Total fat (kg)23.0 ± 1.025.2 ± 1.040.4 ± 1.3*
 Total fat (%)31.0 ± 1.136.0 ± 1.242.5 ± 1.2*
 Total FFM (kg)49.0 ± 1.545.5 ± 1.354.0 ± 1.7*
 Total FFM (%)70.0 ± 1.164.1 ± 1.257.1 ± 1.2*
 Total body (g/cm2)1.09 ± 0.161.06 ± 0.141.14 ± 0.16*
 Lumbar spine (g/cm2)1.01 ± 0.031.00 ± 0.041.12 ± 0.03*
 Total hip (g/cm2)0.84 ± 0.030.79 ± 0.020.93 ± 0.02*

Among our 52 OE subjects, 50 (96%) met the standard criteria for mild to moderate frailty. Average scores for our three measures of physical frailty, namely PPT, Vo2peak, and FSQ, were lower in the OE group than in the NO-NF group but were similar to values observed in the NO-F group (Table 2). Specific tests of physical function (i.e., LE strength, walking speed, and static and dynamic balance) were also lower in the OE group than in the NO-NF group but were similar to values observed in the NO-F group (Table 2), which further supports the notion that the OE group was frail. Moreover, muscle strength was lower in the OE group than in the NO-NF group, even though the OE group had a greater amount of FFM

Table 2.  Frailty, strength, gait, balance, and quality of life of subjects
  • *

    Value significantly different from NO-NF group, p < 0.05.

Measures of frailty
 PPT34.4 ± 0.529.3 ± 0.7*27.8 ± 0.8*
 Vo2peak (mL/kg per minute)20.0 ± 0.614.9 ± 0.4*14.6 ± 0.5*
 Functional status questionnaire33.9 ± 0.430.1 ± 0.5*29.2 ± 0.5*
Strength, gait, balance
 Knee extension (Nm)109.5 ± 5.087.1 ± 3.8*93.1 ± 5.5*
 Knee flexion (Nm)75.6 ± 3.359.6 ± 3.0*60.4 ± 3.6*
 Walking speed (m/min)99.2 ± 5.383.1 ± 2.7*74.0 ± 2.9*
 One leg limb stand (seconds)12.3 ± 2.25.8 ± 0.9*4.5 ± 0.9*
 Obstacle course (seconds)9.7 ± 0.413.2 ± 0.6*14.6 ± 0.8*
SF-36 quality of life (%)
 Role limitations—physical93.2 ± 2.977.0 ± 4.8*78.1 ± 5.1*
 Role limitations—emotional95.1 ± 1.786.8 ± 4.788.4 ± 4.1
 Social functioning97.4 ± 1.392.7 ± 2.494.5 ± 1.9
 Bodily pain82.4 ± 2.373.7 ± 3.776.8 ± 3.1
 Mental health86.1 ± 1.481.3 ± 1.783.5 ± 1.9
 Vitality58.2 ± 1.951.0 ± 1.9*52.5 ± 2.6*
 Health perceptions63.6 ± 1.758.4 ± 1.962.0 ± 2.0
 Change in health45.4 ± 1.940.4 ± 2.7*33.7 ± 2.6*
 Physical function85.9 ± 1.971.5 ± 3.9*62.7 ± 3.1*

Average LE lean mass was greater in the OE group (8.5 ± 4.0 kg) than in the NO-F (7.0 ± 2.5 kg) and NO-NF (6.5 ± 2.0 kg) groups (p < 0.01). The OE group had the poorest muscle quality (i.e., lowest LE muscle torque/LE lean mass) among the three groups when computed for either knee flexion or knee extension (Figure 1).

Figure 1.

Muscle quality during knee extension (A) and during knee flexion (B) in NO-NF, NO-F, and OE subjects. (*) Value significantly different from NO-NF and OE groups, p < 0.001. (†) Value significantly different from NO-NF and NO-F groups, p < 0.001.

Physical subscales in the SF-36 (physical function, role limitations-physical, vitality, and change in health) were impaired in the OE group compared with the NO-NF group (Table 2), indicating excess disabilities as reflected in quality-of-life responses. OE values for these parameters were similar to those in the NO-F group.

The number of chronic health conditions was higher in the OE and NO-F groups than in the NO-NF group (1.7 ± 1.2 and 1.9 ± 1.1 vs. 0.9 ± 0.9, respectively; p < 0.01). After adjustment for the number of chronic conditions by covariance analyses (Table 3), the average scores for measures of frailty, strength, walking speed, and balance were still lower in the OE than in the NO-NF group but similar to the values obtained in the NO-F group.

Table 3.  Frailty, strength, gait, and balance of subjects after controlling for chronic health conditions*
  • *

    Covariance analyses were used to control for chronic health conditions.

  • Value significantly different from NO-NF group, p < 0.05.

Measures of frailty
 PPT33.6 ± 1.129.6 ± 0.728.0 ± 0.8
 Vo2peak (mL/kg per minute)19.8 ± 0.915.1 ± 0.414.7 ± 0.4
 FSQ33.6 ± 1.030.2 ± 0.529.3 ± 0.5
Strength, gait, balance
 Knee extension (Nm)109.5 ± 5.087.1 ± 3.893.1 ± 5.5
 Knee flexion (Nm)75.6 ± 3.359.6 ± 3.060.4 ± 3.6
 Walking speed (m/min)99.2 ± 5.383.1 ± 2.774.0 ± 2.9
 One-leg limb stand (seconds)12.4 ± 1.76.0 ± 1.04.7 ± 1.1
 Obstacle course (seconds)10.0 ± 1.113.0 ± 0.614.3 ± 0.7


In this study, we evaluated the relationship between body composition and physical frailty in lean and OE adults and the effect of obesity on physical function and quality of life in elderly men and women. We used validated subjective and objective measures that predict loss of independence to assess physical frailty and DXA to measure fat mass and FFM (20,21,22). Our findings suggest that obesity is an important cause of physical dysfunction in community-dwelling elderly persons; almost all of our obese subjects had both subjective and objective evidence of physical frailty. In addition, physical frailty in OE subjects was associated with a low percentage body weight as FFM and poor muscle quality.

The results of our study suggest that frailty in OE persons is caused by both a smaller relative muscle mass and lower muscle quality. Although the absolute amount of FFM was greater in the OE than the NO-NF and NO-F groups, the proportion of body weight as FFM was lowest in the OE group, suggesting that relative sarcopenia contributes to frailty. Moreover, these data suggest that obesity and excess body weight might help reduce muscle loss normally associated with aging, but this greater muscle mass is not adequate to maintain good physical function. In addition, compared with the NO-NF and NO-F groups, the OE subjects had the poorest muscle quality, assessed as the ratio of LE muscle torque per kilogram of LE lean mass. Although aging itself is associated with a decline in muscle quality (13,34,35), our findings suggest that obesity is also associated with smaller muscle torque in elderly subjects. The mechanism(s) responsible for aging- and obesity- related alterations in muscle quality is not known, but could include a reduction in type II fibers, increased connective tissue content, fatty infiltration, impaired muscle energetics, and altered substrate metabolism (13,36,37,38).

In concordance with the results from previous studies conducted in younger adults (39), our data demonstrated that obesity was associated with significant impairment in health-related quality of life. Specifically, scores in the SF-36 physical function domains (i.e., physical functioning, role limitations due to physical problems, and vitality) were lower in OE than in NO-NF subjects. It is likely that these domains reflect a general adverse effect of physical frailty on healthy-related quality of life, because a similar impairment in function was observed in the NO-F group. Nonetheless, our results suggest that impaired quality of life is common in OE men and women and should be considered an important health outcome if weight loss therapy is implemented.

Bone loss is a prominent feature of aging, and low BMD, which is a major predictor of fractures (40), is common in the elderly population. Therefore, inducing weight loss in OE persons may have adverse medical consequences because weight loss decreases bone mass (41). However, we found that our OE subjects had greater bone mass than both our NO-F and NO-NF groups. It is not known whether this greater amount of bone mass is enough to prevent osteopenia or osteoporosis and minimize fracture risk after diet-induced weight loss.

The results of the present study may not necessarily apply to the general community-dwelling OE population. Our sample of OE subjects represented those who were willing and able to come to our center to participate in a research study. Therefore, it is possible that the extent of physical frailty in a randomly selected sample of community-living OE persons may be even greater than that observed in our study. In addition, the conclusions of our study were limited due to its cross-sectional design, which can identify associations between body composition and physical function but cannot determine cause-and-effect relationships.

In summary, the results of the present study supported the notion that obesity worsens the age-related decline in physical function and that subjective and objective physical frailty in OE persons living in the community is common. Moreover, physical frailty in OE persons was associated with high body fat mass, low FFM relative to body weight, and poor muscle quality. These findings underscore the importance of considering weight loss therapy to improve physical function in OE persons, as well as possibly to prevent or improve the medical complications associated with obesity, which is the major goal of weight loss therapy in obese young adults. However, the effect of weight loss on functional status in OE men and women has not been carefully studied. In addition, therapeutic approaches must consider the potentially adverse effects of weight loss on muscle (42) and bone (43) masses.


This research was supported by the following awards from NIH: DK37948, AG13629, RR16191 (Research Career Development Award), RR00036 (General Clinical Research Center), and DK56341 (Clinical Nutrition Research Unit).


  • 1

    Nonstandard abbreviations: OE, obese elderly; FFM, fat-free mass; BMD, bone mineral density; LE, lower extremity; PPT, physical performance test; Vo2peak, peak aerobic power; FSQ, Functional Status Questionnaire; SF-36, Medical Outcomes Short Form; NO-F, nonobese frail; NO-NF, nonobese nonfrail.