The objective of the current study was to investigate the association between age and weight-related quality of life in a broad range of overweight/obese individuals. Participants included 9,991 overweight and obese adults from a cross sectional database (mean age = 44.9, mean BMI = 38.3, 75.3% women, 73% white). Participants completed the Impact of Weight on Quality of Life-Lite (IWQOL-Lite), a measure of weight-related quality of life. For the total sample, weight-related quality of life was more impaired with increasing age for physical function, sexual life, and work. However, increasing age was associated with less impairment for self-esteem and public distress. On the sexual life domain there was an interaction between age and gender. Men showed a steady decline in sexual life with increasing age, whereas women showed reduced scores on sexual life in all age groups beyond age 18–24.9. Of note, women's scores on all IWQOL-Lite domains were significantly lower (more impaired) than men's. Thus, there are both positive as well as negative consequences of increasing age with respect to the impact of weight on quality of life in overweight and obese persons.
Health Related Quality of Life (HRQOL) has received much attention with respect to obesity (1,2,3). Although degree of obesity has been associated with HRQOL, other factors also seem to influence this relationship, such as gender, race, comorbid conditions, and treatment-seeking status (4). However, the effects of age on HRQOL in overweight and obese persons have been relatively unexplored.
Using a nationally representative sample in Taiwan, Huang and colleagues examined the relationship between HRQOL and BMI by age, gender, and status of chronic condition. They found that as age increased, excess weight was associated with decreased physical HRQOL but not mental HRQOL (5). Larsson and colleagues (6) in a Swedish study found that among men and women aged 16–34, obesity had a negative effect on the physical aspects of HRQOL. However, among 35–64 year olds, HRQOL varied by gender. In this age group, obese women reported decreased HRQOL on all scales compared to healthy weight women, while obese men reported only decreased physical functioning and general health perception.
HRQOL can be assessed with generic measures that assess the broad aspects of HRQOL and are applicable to any population. Disease-specific measures are designed to assess the complaints and characteristics associated with a specific disease, such as obesity, and are usually more sensitive than generic measures (7). In the studies cited above, generic measures were used to examine the role of age in obesity and HRQOL. The purpose of the present study was to explore the relationship of age to HRQOL in a broad range of overweight/obese individuals using a weight-related measure of HRQOL. Based upon the studies cited above and our clinical experiences with obese patients, we hypothesized that domains related to the physical aspects of HRQOL (i.e., physical function and work) would be adversely affected by age, whereas the psychosocial domains (i.e., self-esteem) would be positively affected by age. We had no specific hypotheses concerning other domains. Based on the Larsson study above as well as other studies (4,8), we also anticipated that men and women would exhibit differences in HRQOL and that these differences might vary by age.
Methods and Procedures
Data for this study were obtained from 9,991 overweight (BMI of 25–29.9 kg/m2) and obese (BMI ≥30 kg/m2) adults (i.e., ≥18 years old) who completed a weight-related measure of HRQOL (Impact of Weight on Quality of Life-Lite (IWQOL-Lite (9))) in a variety of settings. These data are part of the current IWQOL-Lite normative database (10) and include 742 (7.4%) community volunteers, 4,170 (41.7%) participants in double-blind weight loss medication trials, 2,708 (27.1%) participants in outpatient weight reduction studies, 736 (7.4%) participants in an intensive day treatment program for weight loss, and 1,635 (16.4%) gastric bypass candidates. All participants enrolled in weight loss programs/studies completed the IWQOL-Lite prior to initiation of treatment.
The IWQOL-Lite is a 31-item self-report measure of weight-related quality of life measure (9). Its items begin with the phrase “Because of my weight.” The IWQOL-Lite includes five subscales (physical function, self-esteem, sexual life, public distress, and work) as well as a total score. Physical function assesses mobility and physical comfort, self-esteem assesses self-consciousness and devaluation of self, sexual life assesses avoidance of and difficulty with sexual relations as well as lack of sexual enjoyment, public distress assesses ridicule/discrimination as well as physically fitting into one's environment, and work assesses energy level, productivity, and perceived opportunities at work.
Alpha coefficients for the IWQOL-Lite range from 0.90–0.96, and one-week test–retest coefficients from 0.81–0.94 (11). The scale structure of the IWQOL-Lite has been verified with confirmatory factor analysis (9). IWQOL-Lite scores have been shown to be sensitive to changes in weight. Scores on the IWQOL-Lite range from 0–100, with 100 representing the best and 0 the worst quality of life.
IWQOL-Lite scores have been shown previously to vary by gender (4) and BMI (9), with women and individuals with higher BMI's having poorer quality of life. Therefore, a two-way analysis of covariance was performed comparing each of the five IWQOL-Lite scales on gender and age group (18–24.9, 25–34.9, 35–44.9, 45–54.9, 55–64.9, 65+) covarying for BMI. Partial eta-squared coefficients were calculated for main effects (i.e., gender and age group) and interactions (i.e., gender-by-age group), representing the portion of the variance accounted for by age and gender. Cohen (12) describes an eta-squared of 0.0099 as a “small effect”, 0.0588 as a “medium effect,” and 0.1379 as a “large effect.” Tukey's honestly significant differences post hoc tests were performed to compare age groups.
The sample was composed of 9,991 individuals of whom 7,524 (75.3%) were women. Age ranged from 18–90 years, with a mean age of 44.9 (s.d. = 11.7) years. The mean BMI was 38.3 kg/m2 (s.d. = 8.3).
Data on the racial background were available for 92.4% of the men and 87.5% of the women. Of men reporting race, 9.6% were African American, 2.9% were Hispanic, 1.0% were Asian, 83.9% were white, and 2.6% were other races. Of women reporting race, 11.6% were African American, 2.8% were Hispanic, 0.5% were Asian, 81.7% were white, and 3.3% were other races.
Consistent with previous research (4), women reported significantly lower scores (i.e., greater impairments) than men on all IWQOL-Lite scales: physical function: F(1, 9961) = 100.9, P < 0.001, η2 = .01; self-esteem: F(1, 9961) = 438.72, P < 0.001, η2 = .04; sexual life: F(1, 9596) = 169.87, P < 0.001, η2 = .02; public distress: F(1, 9960) = 45.32, P < 0.001, η2 = .01; work: F(1, 9818) = 42.56, P < 0.001, η2 = .004. Figure 1 presents BMI-adjusted means and standard errors by gender for the IWQOL-Lite scales. The greatest difference between men and women occurred on the self-esteem scale (4% of the variance in self-esteem was accounted for by gender).
Figure 2 presents results for the effects of age on HRQOL. Age had a significant main effect for all five IWQOL-Lite scales: physical function: F(5, 9961) = 59.12, P < 0.001, η2 = .03; self-esteem: F(5, 9961) = 32.4, P < 0.001, η2 = .02; sexual life: F(5, 9596) = 12.18, P < 0.001, η2 = .006; public distress: F(5, 9960) = 7.99, P < 0.001, η2 = .004; and work: F(5, 9818) = 42.563, P < 0.001, η2 = .004. For some domains, increasing age was associated with better weight-related quality of life, while on other domains increasing age was associated with reduced quality of life. As hypothesized, weight-related quality of life was more impaired (i.e., scores decreased) with increasing age for physical function and work, while it was less impaired (i.e., scores increased) with increasing age on the self-esteem scale. Increasing age was also associated with reduced sexual life scores as well as increased public distress scores (i.e., less distress out in public). The strongest relationships between age and weight-related quality of life were for physical function and self-esteem, accounting for 3 and 2% of the variance, respectively. The relationship between age and weight-related quality of life was markedly lower for the other scales, albeit statistically significant. For comparative purposes, BMI accounted for between 8% (sexual life) and 44% (public distress) of the variance in IWQOL-Lite scores.
The only significant gender-by-age group interaction was for the sexual life scale, F(5, 9596) = 4.61, P < 0.001, η2 = .002. Although women's scores on the sexual life scale were lower (indicating greater impairment) than men's in all age groups, a different pattern of results was observed for men and women. For women, the best sexual quality of life was obtained by those in the 18–24.9 age group, with scores dropping off substantially and similarly for all other age groups. For men, the best sexual quality of life was obtained by those in the 18–24.9 age group, with a gradual and steady decline in increasing age groups Figure 3.
To our knowledge, this is the first paper to explore the effects of age on weight-related quality of life in overweight and obese persons. One of the most interesting findings in this study is that age affects the various domains of quality of life differently. Although some domains, particularly those involving physical function, become more impaired with increasing age, others actually improve (self-esteem and public distress). Whereas self-esteem involves one's evaluation of oneself, public distress relates more to how others react to the obese person (i.e., ridicule, teasing, and discrimination). One can readily see why these two elements may covary because as self-esteem increases, the negative impact of unfavorable public feedback would reasonably diminish. The question remains though why weight-related self-esteem would be higher with increasing age in this obese population. Perhaps, this finding can be explained by developmental differences across the lifespan with respect to self-esteem and by extension public distress. Studies of the normative trajectory of self-esteem across the lifespan report that self-esteem gradually increases throughout adulthood and begins to drop off at around age 70 (13,14). As global self-esteem tends to increase with increasing age, it would follow that weight-related self-esteem and feelings about being out in public might also improve. However, these findings should be interpreted with caution as the effects of specific factors known to influence psychosocial well-being such as age of onset of obesity and weight cycling were not assessed in the present study. For example, it is unknown whether someone who was obese in childhood may have a different trajectory of weight-related self-esteem and public distress than someone who became obese in adulthood. Further research is required to more fully understand these relationships.
Interesting differences between men and women were noted in the sexual life domain. Women scored lower overall than did men, suggesting that for women being overweight was more likely to result in diminished sexual quality of life. These findings are consistent with previous reports of poorer body image (15), higher prevalence of sexual difficulties (16), and decreased sexual desire and interest (17) in women. However, a recent study of sexual behavior and BMI in a nationally representative database of women aged 15–44 found that overweight and obese women were more likely to report ever having had male sexual intercourse than women with lower BMI's (18). Of particular note in the present study is the different trajectory for sexual life scores in men and women. For women, we noted higher levels of quality of life in the 18–24.9 age group followed by a rapid decline, whereas for men the decline with age appeared more gradual. One might speculate that these differences may be mediated by factors other than weight. For example, the 18–24.9 age group might typically be considered a prechild-bearing period for women, coinciding with a time when sexual activity is high and socially acceptable and women have fewer family demands. The present data do not allow us to control for childbearing as a possible mediator of this effect. For men, the gradual decline over time may represent a more typically male pattern that may be mediated more physically. Typically, male sexual performance (and related factors) decrease gradually beginning around the time of sexual maturity and continues throughout the lifespan.
The current findings have implications for weight loss treatment and public health awareness. Health care providers and weight loss programs could emphasize the importance of remaining physically active and agile so that physical functioning, mobility, and work would become less impaired with increasing age. A recent study examining the relationship between physical activity and weight-related quality of life in obese persons underscores the value of physical activity for these individuals (19). In addition, the positive benefits of aging with respect to self-esteem and public distress could be emphasized as something to look forward to. Furthermore, younger individuals could be taught specific strategies to improve their ability to cope with the psychological and social challenges associated with obesity, rather than wait for age to exert its positive effects. For example, problem-focused coping strategies, such as social trust (the person turns to others for support) and fighting spirit (problems are regarded as something to be resolved), could potentially allow obese individuals to better cope with the psychological and social challenges of obesity. Ryden and colleagues (20) reported that the use of these particular strategies was inversely related to obesity distress.
Strengths of this research include the large sample size and the fairly diverse sample of overweight and obese individuals recruited from a wide variety of settings. Nonetheless, about three-quarters of the sample were composed of white women, which may limit the generalizability of the current findings. Another limiting factor is the use of cross-sectional data only. As such, the differences in HRQOL observed across age groups in the current report may not accurately reflect actual longitudinal changes in HRQOL associated with aging and may represent more of a cohort effect (21). In addition, as was mentioned above, no data on age of obesity onset were available.
In conclusion, increasing age seems to be associated with both positive as well as negative consequences with respect to weight-related quality of life in overweight and obese persons. While some decline over the lifespan may be inevitable, especially with respect to physical function, increasing self-esteem and comfort in public may perhaps provide a buffer against this physical decline.
We thank Dr Martin Binks and Dr Truls Østbye, colleagues from Duke University Medical Center, for their helpful comments on revisions to this manuscript.