点击显示 收起
1 From the University of Pittsburgh (JSL and ABN); Wake Forest University, Winston-Salem, NC (SBK); the University of Tennessee, Memphis (FAT); the National Institute on Aging, Bethesda, MD (TH); the National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD (JE); the Intramural Research Program, National Institute on Aging, Baltimore (EMS); and the University of California, San Francisco (SMR)
2 Supported by the National Institute on Aging (contracts N01-AG-6-2106, N01-AG-6-2101, and N01-AG-6-2103). 3 Reprints not available. Address correspondence to JS Lee, Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, 130 North Bellefield Avenue, 5th floor, Pittsburgh, PA 15213. E-mail: leej{at}edc.pitt.edu.
ABSTRACT
Background: Many older adults desire to lose weight, yet the proportion with a health-related weight-loss indication, weight-loss strategies, and success is unknown.
Objective: We examined the associations of reported intention to lose weight with health-related indications for weight loss, diet quality, physical activity, and weight-loss success in well-functioning older adults.
Design: This prospective, community-based cohort included 2708 elderly persons aged 70–79 y at baseline. We determined indication for weight loss by using the modified National Institutes of Health guidelines, diet quality by using the Healthy Eating Index, and weight-loss intention and physical activity by using questionnaires. Measured weight change over 1 y was assessed.
Results: Twenty-seven percent of participants reported an intention to lose weight, and 67% of those participants had an indication for weight loss. Participants who reported a weight-loss intention were heavier than those who did not, had more depressive symptoms, and were more likely to be dissatisfied with their weight, regardless of weight-loss indication. Participants with an intention to lose weight reported better eating behaviors and a more active lifestyle than did participants without a weight-loss intention, independent of other health conditions. No significant difference in actual weight loss was found between participants intending and not intending to lose weight, regardless of indication for weight loss.
Conclusions: Despite being associated with healthier behaviors, the intention to lose weight did not predict greater weight loss in this well-functioning elderly cohort. More attention needs to be focused on the necessity and efficacy of specific strategies for weight loss in older adults.
Key Words: Weight-loss intention • weight-loss indication • diet quality • physical activity • Health ABC study
INTRODUCTION
Overweight and obesity are common in older adults and are increasing in prevalence (1–4). In 2000, 40% of adults aged 65 y were overweight [defined as a body mass index (BMI; in kg/m2) between 25 and 29.9] and 18% were obese (defined as a BMI 30), which reflect respective increases of 10% and 50% since 1990 (2–4). A larger percentage (30.6%) of older adults report the desire to lose weight than ever before (4), yet little is known about the attributes of older adults attempting to lose weight (5).
Currently, weight loss is recommended for overweight and obese older adults in the same manner as in younger adults. According to the National Institutes of Health (NIH) Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: the Evidence Report (6), a clinical indication for weight loss for older adults is determined by the degree of overweight or obesity and the presence of cardiovascular disease risk factors or comorbidities along with careful examination of the benefits and risks of weight loss. Weight loss in the elderly results from a wide array of risk factors, which range from physiologic changes with aging to chronic diseases and psychological problems, and can be either intentional or unintentional. Unintentional weight loss is commonly observed in older adults, is often associated with more severe disease or with unrecognized health problems, and predicts higher hospitalization, disability, and mortality rates (7–17). Limited understanding of the differences between intentional and unintentional weight loss in older adults has contributed to uncertainty about weight management guidelines and concern that even intentional weight loss could be detrimental (7, 16, 18). Many older adults try to lose weight because of a complication of obesity. Better understanding of weight-loss intention and related behaviors in relation to health-related weight-loss indication is critical to develop appropriately tailored weight-management guidelines that address unique physiologic and medical conditions of older adults who want or need to lose weight (15).
The present study examined reported intention to lose weight in relation to health-related weight-loss indication, weight-loss strategies, and success in weight loss in older adults with the use of data from the Health, Aging, and Body Composition (Health ABC) study. We were particularly interested in examining the following: 1) whether sociodemographic, health, and weight-related characteristics differ by weight-loss intention in older adults with or without health-related weight-loss indication; 2) whether eating behavior and physical activity differ by weight-loss intention; and 3) the degree of weight loss achieved. We hypothesized that older adults with weight-loss intention would report healthier eating habits and exercise patterns and would be more likely to lose weight than would older adults without a weight-loss intention, regardless of weight-loss indication.
SUBJECTS AND METHODS
Study population
The Health ABC study is a longitudinal investigation of the relation between changes in body composition and functional decline. Study eligibility criteria included age 70–79 y during the recruitment period and having no difficulty with activities of daily living or lower-extremity functions, such as difficulty walking one-quarter mile or climbing 10 steps without resting. Exclusion criteria included recent treatment for cancer, participation in a lifestyle intervention trial, or intention to move out of the study location within 3 y of baseline. Participants were recruited from a random sample of Medicare beneficiaries and supplemented by community-based recruitment of black participants in designated ZIP code areas in and around Pittsburgh and Memphis. The main study cohort consisted of 3075 black (42%) and white men and women (52%); the analytic sample included 2708 participants with complete information on weight-loss intention, indications for weight loss, dietary intake, and physical activity. Excluded participants were more likely to be black, poor, and less educated and to have a higher mean BMI. All measurements used for this study were collected at baseline except for dietary intake, which was collected in the first follow-up examination. All procedures were in accordance with the ethical standards of the Institutional Review Boards of the University of Pittsburgh, the University of Tennessee, and the University of California, San Francisco, and these boards approved the protocol and informed consent forms.
Weight-loss intention and weight-loss indication
Weight-loss intention was considered present when a participant responded yes to the question "Are you trying to lose weight at the present time?" at baseline. A weight-loss indication was determined with the use of the modified NIH clinical guidelines, The Practical Guide: Identification, Evaluation, and Treatment of Overweight and Obesity in Adults (19). The criteria used to determine whether a participant had a weight-loss indication included 1) a BMI 30 and 2) a BMI of 25–29.9 [or waist circumference > 35 inches (88.9 cm) for women and >40 inches (101.6 cm) for men] with 2 of the following selected risk factors: established coronary diseases (myocardial infarction, angina pectoris, coronary artery surgery, or other procedures), peripheral arterial disease, type 2 diabetes, knee osteoarthritis, cigarette smoking, hypertension, LDL-cholesterol concentration 160 mg/dL, HDL-cholesterol concentration 35 mg/dL, and fasting plasma glucose concentration of 110–125 mg/dL. Disease was adjudicated by self-reported physician-diagnosed diseases, medication use, and clinic assessment at baseline.
We had 4 mutually exclusive categories of weight-loss intention in relation to weight-loss indication as determined by NIH clinical guidelines for weight loss: 1) weight loss indicated and intended, 2) weight loss indicated but not intended, 3) weight loss not indicated but intended, and 4) weight loss neither indicated nor intended.
Healthy Eating Index
A modified Block food-frequency questionnaire was administered by a trained dietary interviewer to estimate usual nutrient and food group intake at the first annual follow-up examination. The food-frequency questionnaire, developed and modified for the Health ABC study by Block Dietary Data Systems (Berkeley, CA), derives from the National Health and Nutrition Examination Survey III study with the use of food lists from the National Health and Nutrition Examination Survey III 24-h dietary recall data for adults aged >60 y who were either non-Hispanic white or black and resided in either the Northeast or South. A total of 108 food items were included. Interviews were monitored throughout the study to ensure consistent high-quality data-collection procedures. Wood blocks, real food models, and flash cards were used to help participants estimate portion size. Nutrient and food group intakes, as well as number of servings of the Food Guide Pyramid food groups, were estimated by Block Dietary Data Systems with the use of the Food Guide Pyramid–recommended serving sizes.
A Healthy Eating Index (HEI) was calculated to measure the amount of variety in the diet and compliance with specific dietary guidelines (20). The HEI consisted of 10 components: 5 measured conformity to the sex- and age-specific serving recommendations from the US Department of Agriculture’s Food Guide Pyramid for grains, fruit, vegetables, dairy, and meat, and 5 assessed intakes of total fat, saturated fat, cholesterol, sodium, and dietary variety. Each component was scored from 0 to 10 with higher scores indicating better compliance with recommended intake range or amount. Total HEI score ranged from 0 to 100 and were grouped into 3 categories for analysis: good (>80), needs improvement (51–80), and poor (<51).
Physical activity and exercise
Physical activity and exercise were ascertained by using a standardized, interviewer-administered instrument derived from the leisure-time physical activity questionnaire (21), which was supplemented with questions on lower-intensity activities commonly performed by older adults. For each activity, participants were asked whether they had done the activity 10 times in the past 12 mo. If yes, they were asked whether they had done the activity in the past 7 d, and, if so, how many times they did the activity and the average length of each session. Intensity level was also queried for the moderate- and vigorous-exercise activities and for walking. Approximate metabolic equivalent unit (MET) values were assigned to each of the activity categories to calculate a weekly energy expenditure estimate in kcal · kg–1 · wk–1 (22) for each type of activity. Total physical activity (in kJ/wk) comprised the sum of these estimates over all activities performed multiplied by body weight. Similarly, total exercise and different intensity level of exercise (in kJ/wk) were calculated by using the number of minutes, body weight, and METs. Also, a variable for total time spent per week performing exercise, including brisk walking and aerobics or calisthenics, weight training, and high-intensity exercise (eg, bicycling, swimming, jogging, racquet sports, or using a stair-stepping, rowing, or cross-country ski machine), was created.
Measured weight change
While wearing a standard clinic gown and without shoes, study participants were weighed. Weight was measured in kilograms with the use of a standard balance beam scale and was recorded to the nearest 0.1 kg. Scale calibration was checked monthly against known weight at clinics and was also checked yearly by the local Department of Weights and Measures. Weight change over 1 y was either calculated as a weight change proportional to baseline weight or was defined according to 3 groups: 1) loss (>5%), 2) stable (±5% weight change), and 3) gain (>5%).
Sociodemographic and weight-related characteristics
Participants provided information on sociodemographic variables (age, race, sex, study site, education, and family income for the past year), past medical history (adjudicated by self-reported physician-diagnosed diseases, medication use, and clinic assessment), and various lifestyle factors (smoking, physical activity, and eating behavior). Participants were also asked about their weight perception (underweight, about the right weight, or overweight), weight satisfaction (very satisfied, moderately satisfied, a little satisfied, or not at all satisfied), and weight trajectory during adulthood [stayed about the same (within 10 lb, or 4.5 kg), gradual gain of >10 lb, gradual loss of <10 lb, marked loss and then kept weight off, and repeatedly gone up and down again). Depressive symptoms were assessed by using the Center for Epidemiologic Studies Depression scale, a 20-item instrument designed to measure depressive symptoms experienced during the previous week (23). Participants scoring >15 of a maximum score of 60 were judged to have higher symptomatology. Personal mastery was assessed by using 2 questions derived from work by Pearlin and Schooler (24) on personal control and self-efficacy. Participants who agreed strongly with the statement "I can do anything that I really set my mind to" and disagreed strongly with the statement "I often feel helpless in dealing with the problems of life" were regarded as having high personal mastery.
Statistical analysis
Descriptive statistics of the study population were analyzed by weight-loss intention groups. Differences in means and proportions of baseline characteristics by weight-loss intention and weight-loss indication were tested with the use of analysis of variance and chi-square tests. Analysis of covariance was used to determine the main effects of weight-loss intention and weight-loss indication, as well as potential effects of interactions between them on HEI scores, amount of total physical activity, and proportional change in body weight over 1 y after control for potential confounders, including age, sex, race, study site, and education. Adjusted means across weight-loss intention groups were computed and presented in the text and tables. There were, however, no statistically significant interactions between weight-loss intention and weight-loss indication. Because of highly skewed distributions, variables for the amount and duration of exercise were categorized into 5 groups as follows: for the amount of exercise, 1) 0 kJ/wk, 2) 2092.9 kJ/wk (499.9 kcal/wk), 3) 2093.0–4185.9 kJ/wk (500.0–999.9 kcal/wk), 4) 4186.0–6277.9 kJ/wk (1000.0–1499.9 kcal/wk), and 5) 6278.0 kJ/wk (1500.0 kcal/wk); for the duration of exercise, 1) 0 min/wk, 2) 99.9 min/wk, 3) 100.0–199.9 min/wk, 4) 200.0–299.9 min/wk, and 5) 300.0 min/wk. Ordinal logistic regression procedures were used to examine whether older adults with a weight-loss intention were more likely to report greater amount and longer duration of exercise than were older adults without weight-loss intention after control for potential confounders, including weight-loss indication, age, sex, race, study site, and education. Logistic regression procedures were also used to assess whether older adults with an intention to lose weight were more likely to have 1-y measured weight change, either gain or loss >5%, than were older adults without an intention to lose weight. There were no significant interactions between weight-loss intention and indication except for the amount of moderate-level exercise. All statistical analyses were performed by using SAS 8.02 (25). Statistical significance was set at P < 0.05.
RESULTS
Prevalence of weight-loss intention and indication
Twenty-seven percent of participants reported that they were currently trying to lose weight, and 47% of the sample had an indication for weight loss. Only 38% of participants with an indication for weight loss were actually trying to lose weight. Seventeen percent of participants without an indication for weight loss were also trying to lose weight (Table 1).
View this table:
TABLE 1. Prevalence of self-reported weight-loss intention and of weight-loss indication in 2708 participants in the Health, Aging, and Body Composition study1
Sociodemographic and weight-related characteristics by weight-loss intention type
Participants with an indication for weight loss were significantly more likely to be men, black, poor, consuming special diets, and less educated than were participants without an indication for weight loss (Table 2). Participants trying to lose weight were significantly more likely to be women and to have graduated from high school than were participants without a weight-loss intention. Participants trying to lose weight also had a significantly higher mean BMI, significantly poorer weight satisfaction, and a significantly higher frequency of perceiving themselves as overweight and were significantly more likely to have high depressive symptoms and low personal mastery.
View this table:
TABLE 2. Baseline characteristics of 2708 participants by weight-loss intention type: the Health, Aging, and Body Composition study
Healthy Eating Index by weight-loss intention type
Total and component scores of HEI by weight-loss intention groups after control for age, sex, race, study site, and education are shown in Table 3. Participants with an indication for weight loss did not differ significantly in total HEI score from participants without an indication but had significantly lower component HEI scores for grain and dairy foods. Regardless of indication for weight loss, participants trying to lose weight had significantly higher total and component HEI scores for fruit, total fat, and saturated fat. Also, the proportion of subjects in the good HEI category was significantly higher in the participants with weight-loss intention than in participants without intention (chi-square = 6.25, P = 0.044).
View this table:
TABLE 3. Healthy eating index (HEI) scores in 2708 participants by weight-loss intention type: the Health, Aging, and Body Composition study1
Physical activity and exercise by weight-loss intention type
Participants with an indication for weight loss did not have a significantly different amount of total physical activity than did participants without an indication for weight loss. Regardless of weight-loss indication, participants trying to lose weight did not have a significantly higher total physical activity than did participants without a weight-loss intention [27294.6 ± 956.4 (adjusted
View this table:
TABLE 4. Percentages of various levels and durations of exercise in 2708 participants by weight-loss intention type: the Health, Aging, and Body Composition study1
1-y Weight change by weight-loss intention type
Data on 1-y weight change by weight-loss intention groups are shown in Table 5. Participants with an indication for weight loss had a significantly greater adjusted mean proportional measured weight change (loss) than participants without indication for weight loss. Regardless of indication for weight loss, however, the mean proportional weight changes and proportions of participants who remained weight stable, gained weight, or lost weight over 1 y were remarkably similar between participants with and without an intention to lose weight.
View this table:
TABLE 5. Proportional weight changes over 1 y and percentages of participants who gained weight, maintained a stable weight, or lost weight by weight-loss intention type: the Health, Aging, and Body Composition study1
DISCUSSION
Many studies have shown that weight-loss intention and weight-control behaviors are common among adolescents and young adults, particularly among females (26–32). Our study showed that many older persons in this well-functioning, community-dwelling cohort also desired weight loss and followed overall healthy behaviors, particularly in participants who had weight-related health risks.
About one-half of the cohort had an indication for weight loss, but only one-third of them were actually trying to lose weight. Participants who had an indication for weight loss were more likely to report trying to lose weight than were participants who did not have an indication for weight loss. Approximately 10% were trying to lose weight even though they did not have weight-loss indication. These findings are similar to those of several national surveys showing that not all overweight people are trying to lose weight and that a considerable proportion of persons who are trying to lose weight are not overweight (26, 28, 30). Similar to previous findings across age subgroups (26, 28, 30, 33), the percentages of participants in the present study who were trying to lose weight varied by sociodemographic category, but trying to lose weight was most strongly related with weight status and weight-related characteristics. Participants with a weight-loss intention tended to be heavier than were participants without an intention, had more depressive symptoms, and were more dissatisfied with their current weight, regardless of indication for weight loss. The results imply that the persons trying to lose weight, for the most part, consisted of older adults who had serious concerns to control their apparent or foreseen adverse weight and health conditions, but that a sizable proportion also had continuing concerns about their weight, perhaps because of appearance and a desire to feel young. This understanding should be incorporated into clinical practice. Health care providers should recognize the need to advise older adults about the current weight status and related risks appropriately. Such efforts could help older adults who want or need to lose weight make appropriate decisions regarding controlling their weight to maximize the health effects of attempted weight loss and to reduce detrimental weight-related risks.
Consuming fewer calories and regular physical activity are key factors in successful weight loss (34). Results from previous studies on weight-control behaviors have shown that people trying to lose weight engage in behaviors that create a negative energy balance by eating less or by increasing their physical activity (28–30), but that most people, especially older adults trying to lose weight, do not follow the recommended combination of reducing energy intake and engaging in leisure-time physical activity of 150 min/wk (30). The present study, however, showed that older adults trying to lose weight have overall better eating and physical activity behavior than do older adults without a weight-loss intention, regardless of adverse physical, psychological, and health conditions related to poorer health in later life (35–39).
In particular, increased exercise, the least popular strategy to lose weight in older adults (30), differed dramatically by weight-loss intention type. Older adults with an intention to lose weight were more likely to engage in higher levels of total exercise, in particular higher-intensity exercise, than were older adults without an intention. This pattern does not necessarily run parallel with the current public health guidelines that suggest that amount of activity is more important than the specific manner in which the activity is performed (ie, mode, intensity, or duration of the bouts of activity). Previous studies provided mixed results as to whether the type, duration, and intensity of exercise make differences in the amount of weight loss or change in lean or fat mass (40–44). The limited research done in older adults has found no differential effect of exercise intensity on weight loss and body composition (42). More research is needed to better elucidate the health effects of different exercise patterns practiced by older adults with weight-loss intention.
The quantity and quality of dietary intake assessed by HEI did not dramatically differ by weight-loss intention. However, a slightly greater percentage of participants who intended to lose weight fell in the range of having good eating habits. Conceivably, healthy eating might be a common strategy not only for weight loss but also for health promotion even in older adults (29, 30). This cohort of well-functioning older adults had good dietary behavior comparable with recommended intakes.
Despite healthy behaviors, the intention to lose weight did not predict greater weight loss in this well-functioning cohort of older adults. Our study supports the findings of Gregg et al (5) that indicate weight-loss intention might reflect an overall healthy lifestyle associated with lower mortality, independent of actual achieved weight loss. Whether the reduced mortality or other health benefits of attempted weight loss are not modulated by actual achieved weight loss remains to be determined. Some cautions are warranted in interpreting these results. The achieved measured weight change was calculated as the weight difference between baseline and 1 y later. The 1-y time frame might not adequately capture weight fluctuation throughout the year or the underlying physiologic and biological changes that accompanied aging and other health conditions. However, it is notable that participants trying to lose weight were actually less likely to lose weight, but they were more likely to maintain baseline weight. Healthier diet and exercise pattern in this age range might attenuate the loss of muscle mass with aging without weight loss (45, 46), which suggests that attention to weight can result in a more stable weight. Without more information on weight-loss intention about goals, time period, weight-change patterns (such as weight regain or weight cycling), and the cause of age-related skeletal muscle mass loss, however, we could not determine whether this result reflected failure of intended weight loss, short-lived success with rebound weight gain, or desirable weight change pattern known to be related with lower mortality in older adults (47).
The results of this study should be viewed with its limitations in mind. Generalization of these findings to the older population at large is limited because study participants were well functioning and relatively healthy. The NIH weight-loss guidelines have been criticized as not being directly applicable to the older population. Several criteria of the guidelines do not have established validity for use in older adults (39, 40). For example, waist circumference cutoffs might not be appropriate for older adults because abdominal fat is known to accumulate with increasing age. Most of the data used in this study was cross-sectional, which does not allow a determination of whether any observed relation between different weight-loss intention types and eating and exercise patterns is causal. It is possible that weight-loss intention might simply reflect an overall healthy lifestyle that results from lifelong health behaviors (5). Measures of eating and exercise patterns in this study relied on self-report and, therefore, might be subject to response biases such as social desirability, including underreporting of dietary intake, overreporting of exercise, or both. Because life-course trajectories of health behaviors tend to demonstrate continuity and momentum across situations and through life transitions (48), understanding early life experiences related to weight-loss intention might be important to better understand the relation between weight-loss intention and its effect on lifestyle and achieved weight loss.
Because many older adults desire or need to lose weight, more attention needs to be given to examining the necessity and efficacy of specific recommendations for weight loss in older adults. Observational studies, such as the Health ABC study, can be used to evaluate the long-term health effects of weight loss on weight change, body composition, functional limitation, and longevity, which can enlighten the approach to weight management in older adults (7, 9, 11, 12, 17). However, given the strong association between weight-loss intention and health-related weight-loss indication in older adults, clinical trials could be necessary to evaluate the risks and benefits of intentional weight loss (17, 18, 33). In the meantime, under the guidance of health care professionals, older adults might need to alter their diet and exercise patterns, remain nonobese, or maintain stable weight for continued health, independence, and successful aging (6, 37, 49–54).
ACKNOWLEDGMENTS
JSL and ABN were responsible for the study design. JSL provided data analysis and drafted the article. ABN, SBK, TH, FAT, EMS, JE, and SMR provided critical revision of the article for important intellectual content. SMR provided administrative, technical, or logistic support. None of the authors had any possible conflicts of interest.
REFERENCES