Increasing BMI is associated with a progressive reduction in physical quality of life among overweight middle-aged men

We assessed whether increasing body mass index (BMI) affects health-related quality of life in a group of 38 overweight (BMI 25–30 kg/m2) middle-aged (45.9 ± 5.4 years) men, recruited in Auckland (New Zealand). Health-related quality of life was assessed with SF-36v2 at 0, 12, and 30 weeks. Increasing BMI was associated with a progressive reduction in physical component summary score (p = 0.008), as well as lower general health (p = 0.036), physical functioning (p = 0.024), and bodily pain (p = 0.030) scores. Stratified analyses confirmed these findings: participants who were more overweight (n = 19; BMI 27.5–30 kg/m2) had poorer physical component summary (p = 0.005), physical functioning (p = 0.040), bodily pain (p = 0.044), and general health (p = 0.073) scores than the less overweight (n = 19; BMI 25–27.5 kg/m2). Increasing BMI is associated with a progressive reduction in physical quality of life, even within a relatively narrow BMI range encompassing only overweight middle-aged men.

BMI vs HRQL. Univariate analyses showed that BMI was negatively correlated with physical component summary (r 5 20.43; p 5 0.007), physical functioning (r 5 20.34; p 5 0.035), and bodily pain (r 5 20.34; p 5 0.037) scores, also tending towards a similar association with general health (r 5 20.30; p 5 0.065) and vitality (r 5 20.29; p 5 0.081) scores. These results were corroborated by multivariate models, which showed that greater BMI was associated with poorer physical quality of life. Thus, increasing BMI was associated with a progressive reduction in physical component summary score (p 5 0.008), as well as lower general health (p 5 0.036), physical functioning (p 5 0.024), and bodily pain (p 5 0.030) scores.
The stratified analyses confirmed the findings that increasing BMI among overweight men was associated with a progressive reduction in physical quality of life. Participants who were more overweight (n 5 19; BMI 27.5-30 kg/m 2 ) had poorer physical quality of life than those were less overweight (n 5 19; BMI 25-27.5 kg/m 2 ) ( Table 1). This was illustrated by differences in the physical component summary score (p 5 0.005), as well as physical functioning (p 5 0.040), bodily pain (p 5 0.044), and general health (p 5 0.073) scores (Table 1). There were no differences between groups in mental health domains.

Discussion
This study shows that increasing BMI is associated with a progressive reduction in physical quality of life among middle-aged men. Importantly, this was observed to occur within a relatively narrow BMI range, encompassing only overweight men. Further, our findings were observed in an otherwise healthy group of individuals, with the possible confounding effects of physical activity levels and significant comorbidities accounted for.
The increased morbidity and mortality associated with obesity is widely recognized. However, recent studies have suggested that this is not the case for those who are overweight when compared to normal-weight individuals [13][14][15][16] . Similarly, the HRQL literature on the effects of overweight is conflicting, and some studies found no effects of overweight on HRQL in comparison to those of normal weight [6][7][8] . However, irrespective of whether HRQL in a group of overweight males actually differs to that of normal-weight controls, our data do show that increasing BMI is associated with poorer outcomes in physical domains, even within a group consisting solely of overweight individuals.
We did not observe any effects of BMI on mental health domains assessed. These findings are consistent with a number of previous studies showing that physical quality of life is far more affected by increasing BMI than mental health domains 6,17,18 . However, the observed reduction in physical functioning in our overweight cohort may lead to decreased energy expenditure, resulting in increasing mismatch in energy balance and additional weight gain 19 . Thus, the observed impairments in physical health in overweight individuals may eventually lead to the onset of obesity. As a result, while we did not observe any effects of increasing BMI on mental health domains, progression to obesity would likely result in adverse effects on mental health 20 .
The main limitation of our study was our relatively small number of participants (n 5 38). However, our findings are particularly robust as each participant underwent three assessments over a 30week period. Thus, our study design minimized the potential effects of temporal variations in HRQL that might have been experienced by individual participants. Nonetheless, we studied a relatively narrow range of individuals (overweight males living in a large urban centre, mostly of New Zealand European ethnicity), which may limit wider applicability of our findings.
In summary, we found that physical quality of life is affected by increasing overweight levels, in the absence of overt obesity. There is considerable focus on public health interventions targeting those who are obese. Our study shows that overweight non-obese individuals also display impairments in physical quality of life. It is important therefore that such aspects are considered in the health management of overweight individuals, not only to improve quality of life, but also to prevent further reductions in physical activity levels and halt a consequent progression to obesity.

Methods
This study encompassed the post hoc analysis of data from a 30-week randomized, double-blinded, placebo-controlled, crossover trial examining the effects of olive leaf extract on insulin sensitivity 9 . All participants were overweight men (BMI 25-30 kg/ m 2 ) aged 35-55 years, who were recruited in February 2011 via advertisements in local newspapers that circulate freely in the central Auckland metropolitan area (New Zealand). Note that only males were recruited to the clinical trial so that the effects of the menstrual cycle on insulin sensitivity could be avoided. Exclusion criteria were: drug use (including tobacco), diabetes, or being on medications likely to affect insulin sensitivity. Subjects taking antihypertensive or lipid-lowering medications were recruited, but were required to have been on a stable dose for at least 6 months prior to start of the study. These subjects were also encouraged not to change dose throughout the trial, and doses were checked at each assessment. All participants were asked not to make any substantial alterations to their lifestyle for the duration of the trial. Specifically, participants were instructed not to make changes to their diet and physical activity levels.
Participants were assessed at baseline and after 12 and 30 weeks. Clinical assessments were carried out at the Maurice & Agnes Paykel Clinical Research Unit (Liggins Institute, University of Auckland). HRQL was assessed using the SF-36v2 Health Survey (New Zealand/Australia adaptation), based on subjective measures of wellbeing. The SF-36v2 is a validated tool that measures perception of health on eight multi-item dimensions covering functional status, wellbeing, and overall evaluation of health 10 . The SF-36v2 assesses 8 domains, including 4 physical (general health, physical functioning, bodily pain, and role limitations related to physical problems) and 4 mental (mental health, vitality, social functioning, and role limitations related to emotional problems) 10 . Parameters summarizing both physical and mental domains were also obtained. There are a number of items that make up each individual domain. These have been presented and discussed by Ward 10 , and are summarized in Figure 1.
Auxological assessment included height measurement using a Harpenden stadiometer. Weight was measured using whole-body dual-energy X-ray absorptiometry (DXA, Lunar Prodigy 2000, General Electric, Madison, USA). Physical activity levels were assessed using the International Physical Activity Questionnaire (IPAQ) 11 , covering four domains of physical activity: work-related, transportation, housework/ gardening, and leisure time.
Univariate analyses were initially performed using simple correlations or Spearman's rank correlations. SF-36v2 data for the study cohort were compared to normative data for New Zealand men aged 45-64 years 12 using two-sample t-tests. Random effect mixed models with repeated measures were used to evaluate the association of BMI with SF-36v2 outcomes. Randomization sequences, time period, and on-going use of medication (for cholesterol or hypertension) were accounted for, with IPAQ score also included as a co-variate. Stratified analyses were also carried out dividing the study cohort in half, separating participants according to BMI. Univariate analyses were carried out in Minitab v.16 (Pennsylvania State University, State College, PA, USA), while multivariate analyses were performed in SAS v.9.3 (SAS Institute Inc. Cary, NC, USA). Demographic data are presented as means 6 standard deviation, while outcome data are presented as model-adjusted means (estimated marginal means adjusted for the confounding factors in the models), with associated 95% confidence intervals. All statistical tests were two-sided and a 5% significance level maintained throughout the analyses.
Ethics approval for this study was provided by the Northern Y Regional Ethics Committee (New Zealand Ministry of Health), and written informed consent was obtained from all participants. This study was registered with the Australian New Zealand Clinical Trials Registry (#336317).