Volume 25, Issue 1 , Pages 1-6, January 2011
Accuracy, determinants, and consequences of body weight self-perception in type 2 diabetes: the Fremantle Diabetes Study☆
Article Outline
Abstract
Objective
To assess the accuracy, determinants, and consequences of body weight self-perception in type 2 diabetes.
Methods
We studied 1272 community-based patients and a 518-patient overweight/obese subset who returned for ≥4 annual reviews. Multiple logistic regression was used to identify baseline predictors of correct weight self-perception and to determine whether correct weight self-perception predicted future weight loss. Overweight and obesity were defined as body mass indices of 25.0–29.9 and ≥30.0 kg/m2, respectively.
Results
Of the patients who were overweight (40.0%) or obese (41.8%) at baseline, 52.8% and 83.7%, respectively, correctly self-identified their weight category. Overweight/obese participants who self-identified correctly were more likely to have been informed they were overweight (P<.001), predominantly by their general practitioner (80.1%). Overweight participants had less self-awareness if they were not abdominally obese, did not speak English fluently, were male, or had a low income. Obese participants were more likely to consider themselves overweight if they had better diabetes knowledge and higher educational attainment. Correct weight self-perception did not influence subsequent weight loss.
Conclusions
Health care professionals can facilitate body weight self-awareness in type 2 diabetes. Education programmes should recognise the impact of gender and socio-demographic variables on accurate weight self-perception.
Keywords: Type 2 diabetes, Body weight, Overweight, Obesity, Self-perception
1. Introduction
Weight management is regarded as a key therapeutic strategy in type 2 diabetes (American Diabetes Association, 2006). Reductions in body weight can improve not only glycaemic control in patients who are overweight or obese but also other important co-morbidities such as hypertension and dyslipidaemia (Hensrud, 2001). Self-perception of weight appropriateness is important in the promotion and management of weight loss (Chang & Christakis, 2003). However, the results of studies from the US (Paeratakul, White, Williamson, Ryan, & Bray, 2002), Europe (Blokstra, Burns, & Seidell, 1999), and Australia (Donath, 2000) suggest that there is an inconsistent relationship between self-reported and actual body weight in the general population. There is a risk that normal-weight individuals who consider themselves overweight may diet unnecessarily, while those who are overweight but who judge themselves to be of normal weight may lack the motivation necessary for beneficial weight reduction practices.
There are scant published data relating to weight self-perception in diabetes. One recent study of 573 diabetic patients from a collection of US general medicine practices found that many overweight and obese subjects overestimated the weight that would be healthiest for their height (McTigue et al., 2006), but the response rate was <25%, weight and height were self-reported, and there was a lack of differentiation between type 1 and type 2 diabetes as well as a limited range of potentially predictive variables. Given the importance of self-management in contemporary diabetes care especially in type 2 diabetes (Mensing et al., 2006), there is a need for more data assessing the frequency and determinants of inaccurate weight perception with a view to developing strategies that facilitate appropriate weight loss in type 2 diabetes.
The aims of the present study were therefore to (i) determine whether the perception–reality mismatch found in the general population applies in type 2 diabetic patients from the community; (ii) assess key demographic, socio-economic, and health-related variables associated with weight self-perception; and (iii) assess whether overweight/obese patients who perceive themselves as overweight at baseline respond appropriately to this knowledge by losing weight during 4 years' follow-up.
2. Methods
2.1. Patients
All patients in the present study were recruited to the Fremantle Diabetes Study (FDS), a longitudinal observational study of a representative sample of patients from a postcode-defined urban Australian community of 120,097 people (Bruce et al., 2000, Davis et al., 2000). The study protocol was approved by the Fremantle Hospital Human Rights Committee and all subjects gave informed consent before participation. Descriptions of recruitment, sample characteristics including classification of diabetes type, and details of nonrecruited patients have been published elsewhere (Davis et al., 2000). Of 2258 diabetic patients identified during the study registration period between 1993 and 1996, 1426 (63%) were recruited to the FDS and 1294 had type 2 diabetes. Eligible patients who declined participation were a mean of 1.4 years older than participants, but the proportion with type 2 diabetes and the distribution of treatment modalities were not significantly different (Davis et al., 2000).
2.2. Clinical methods
The present study analysed cross-sectional and longitudinal data from the detailed baseline and annual FDS assessments (Davis et al., 2000). Comprehensive questionnaires were administered to all participants, and self-perception of weight was measured using a single-item question: “Do you consider yourself overweight?” Patients were also asked (i) whether they had ever been informed that they were overweight and, if so, by whom, and (ii) whether they weighed themselves regularly. Each participant's body mass index (BMI) was calculated from height and weight measured as part of a detailed physical examination conducted by trained study staff. BMI was categorised by standard clinical definitions as normal (<25.0 kg/m2), overweight (≥25.0 and <30.0 kg/m2), or obese (≥30.0 kg/m2) (World Health Organisation, 2000). Waist circumference measurements were also obtained and categorised for abdominal overweight/obesity using currently accepted standards for Europids (males ≥94 cm; females ≥80 cm) (Han, Vanleer, Seidell, & Lean, 1995). Diabetes knowledge was assessed from 15 standard multiple-choice questions about diabetes and its management (Bruce, Davis, Cull, & Davis, 2003) and depression status from responses to a validated questionnaire (Bruce, Davis, Starkstein, & Davis, 2005). Weight measurements at baseline and their fourth annual visit were used to determine change in weight over a 4-year period.
2.3. Statistical analysis
The computer package SPSS for Windows (version 14.0; SPSS, Inc., Chicago, IL, USA) was used for statistical analysis. Data are presented as proportions, means (±S.D.), or, in the case of variables which did not conform to a normal distribution, median [interquartile range]. For independent samples, two-way comparisons for proportions were by Fisher's Exact Test, for normally distributed variables by Student's t test, and for non-normally distributed variables by Mann–Whitney U test. Multiple comparisons for proportions were by Fisher's Exact Test or chi-squared test, for normally distributed variables by one-way ANOVA, and for non-normally distributed variables by Kruskal–Wallis H test. Multiple logistic regression (forward conditional modelling using P<.05 for entry and P>.10 for removal) was used to identify factors relating to (i) self-perception of weight after adjustment for potential confounders, and (ii) weight loss in overweight and obese patients between baseline and fourth visit. Clinically plausible variables with univariate P<.20 were considered for entry into the multivariate models. The level of significance was taken as a P value <.05.
3. Results
3.1. Baseline patient characteristics
Of the 1294 FDS patients with type 2 diabetes, 1272 (98.8%) both responded to the question relating to self-perception of overweight and had a valid BMI measurement. These subjects comprised the present sample. Approximately half (51.0%) were females, their mean age was 64.0±11.2 years, and median diabetes duration was 4.0 [1.0–9.0] years. Patient details are summarised by BMI category in Table 1.
Table 1. Baseline characteristics of 1272 type 2 diabetic patients responding to questions on self-perception of overweight by weight status defined by BMI
| Normal (<25.0 kg/m2) | Overweight (25.0–29.9 kg/m2) | Obese (≥30.0 kg/m2) | P value | |
|---|---|---|---|---|
| Number (%) | 230 (18.2%) | 508 (40.0%) | 534 (41.8%) | |
| Age (years) | 66.3±11.4 | 65.7±10.1 | 61.4±11.6 | <.001 |
| Sex (% male) | 51.7 | 55.9 | 41.2 | <.001 |
| Diabetes duration (years) | 4.0 [1.0–9.3] | 4.0 [0.9–10.0] | 3.5 [1.0–8.0] | .42 |
| Diabetes knowledge score (0–15) | 9 [6–11] | 9 [5–11] | 9 [6–11] | .53 |
| Body mass index (kg/m2) | 22.7±1.9 | 27.6±1.4 | 34.5±4.4 | <.001 |
| Abdominal overweight/obesity (%) | 43.6 | 91.1 | 99.8 | <.001 |
| Self-perception of overweight (%) | 11.7 | 52.8 | 83.7 | <.001 |
| Informed overweight (%)a | 9.1 | 47.3 | 82.9 | <.001 |
 By GP | 4.3 | 34.9 | 70.5 | <.001 |
| By specialist | 1.3 | 5.1 | 13.1 | <.001 |
| By diabetes educator | 1.7 | 16.8 | 29.8 | <.001 |
| By dietitian | 1.7 | 5.9 | 12.8 | <.001 |
| By family/friends | 0.4 | 2.8 | 4.1 | .020 |
| Regular self-weighing (%) | 64.8 | 68.1 | 64.4 | .42 |
| Times/monthb | 4 [2–10] | 4 [2–12] | 4 [2–12] | .23 |
| Given exercise programme (%) | 56.1 | 57.2 | 62.3 | .15 |
| Any exercise in past 2 weeks (%) | 76.1 | 75.4 | 67.5 | .006 |
| Visited dietitian in past year (%) | 17.4 | 16.9 | 19.1 | .64 |
| Smoking status (%) | ||||
| Never | 42.4 | 41.8 | 48.6 | .09 |
| Ex- | 38.9 | 43.6 | 37.3 | |
| Current | 18.8 | 14.6 | 14.1 | |
| Alcohol (standard drinks/day) | 0 [0–0.8] | 0 [0–0.8] | 0 [0–0.3] | .08 |
| Married/de facto relationship (%) | 68.7 | 66.9 | 63.6 | .32 |
| Educated beyond primary level (%) | 79.9 | 70.6 | 74.9 | .025 |
| Not fluent in English (%) | 12.2 | 17.5 | 14.4 | .14 |
| Ethnic background (%) | ||||
| Anglo-Celt | 70.9 | 60.8 | 62.0 | <.001 |
| Southern European | 9.1 | 19.9 | 21.2 | |
| Other European | 6.1 | 10.8 | 7.7 | |
| Asian | 8.3 | 2.6 | 1.3 | |
| Mixed/other | 4.3 | 4.3 | 6.6 | |
| Indigenous Australian | 1.3 | 1.6 | 1.3 | |
| Household income (AUD; %) | ||||
| ≤$12,000 | 32.0 | 29.6 | 28.7 | .029 |
| $12,001–25,000 | 53.6 | 54.7 | 48.8 | |
| >$25,000 | 14.4 | 15.7 | 22.5 | |
| Depressed (%) | 28.8 | 30.1 | 33.8 | .28 |
aPatient may have been informed he/she is overweight by more than one source. |
bIn those who weighed themselves regularly. |
3.2. Baseline weight self-perception
The majority of obese subjects identified themselves as being overweight compared with just over half of the participants in the overweight category (Table 1). Males were significantly less likely than females to view themselves as overweight (52.6% vs. 63.8%; P<.001). In particular, fewer overweight males considered themselves overweight compared with their female counterparts (49.3% vs. 57.1%; P=.09), while more normal-weight females than males perceived themselves as overweight (16.2% vs. 7.6%; P=.06).
Participants self-identified as overweight were more likely to have been informed they were overweight compared with those who did not consider themselves overweight (77.0% vs. 25.0%; P<.001). Informants comprised mainly the patient's general practitioner (GP) (80.1%), diabetes educator (35.3%), dietitian (14.5%), and/or medical specialist (14.1%), while only 5.3% of informants were family members or friends. In patients who were overweight/obese, 19.8% neither considered themselves overweight nor had been informed they were. Of the overweight/obese patients who had been informed they were overweight, 82.6% considered themselves so compared with 42.5% who had not been told (P<.001).
3.3. Variables associated with baseline weight self-perception
Multiple logistic regression analyses showed that, for obese patients, being informed that they were overweight increased the odds of overweight awareness nearly fourfold (see Table 2). Better diabetes knowledge and educational attainment beyond primary school were also independently associated with higher odds of correct self-perception of overweight. For overweight patients, the number of different informant sources was important in increasing the odds of correct self-perception (nearly threefold for one informant and sevenfold for two or more informants compared with not being told). Abdominal obesity increased the odds of self-perception threefold, while overweight males and those who did not speak English fluently had less self-awareness. Overweight patients with higher household incomes were significantly more likely to be aware of their overweight status than those with low incomes (Table 2). Age, diabetes duration, marital status, ethnic background, and presence of depression were not independent associates of self-perception of being obese or overweight.
Table 2. Variables independently associated with self-perception of obesity and overweight in multiple logistic regression analysis
| Variable | Odds ratio (95% CI) |
|---|---|
| Obese (n=534) | |
| Informed overweight | 3.85 (2.22–6.69) |
| Diabetes knowledge score | 1.16 (1.07–1.25) |
| Education beyond primary level | 1.92 (1.10–3.34) |
| Overweight (n=508) | |
| Male | 0.61 (0.40–0.94) |
| Number of informant sources | |
| 0 | 1 |
| 1 | 2.79 (1.77–4.42) |
| ≥2 | 7.45 (3.82–14.56) |
| Abdominally obese | 3.38 (1.51–7.59) |
| Not fluent in English | 0.26 (0.14–0.47) |
| Income (AUD$) | |
| ≤12,000 | 1 |
| 12,001–25,000 | 1.83 (1.15–2.92) |
| >25,000 | 4.36 (2.18–8.75) |
3.4. Baseline variables associated with weight-loss after 4 years' follow-up
Of the 1042 type 2 participants who were overweight or obese at baseline, 518 (49.7%) had ≥5 annual FDS assessments (baseline plus at least four annual reviews). This subset of patients was used to examine factors predicting weight loss subsequent to the baseline FDS assessment. After 4.4±0.6 years' follow-up, 298 (57.5%) had lost a mean of 5.5±4.3 (range 1 to 30) kg (see Table 3). In bivariate statistics, age was positively associated with weight loss at 4 years, while cerebrovascular disease and a dietitian visit within the last year were negatively associated (Table 3). In multiple logistic regression analysis, older age [odds ratio (95% confidence interval): 1.69 (1.35–2.12) for an increase of 10 years] and higher BMI [1.07 (1.02–1.11) for an increase of 1 kg/m2] at baseline were independently associated with increased odds of weight loss after 4 years' follow-up, while a dietitian visit during the previous year [0.46 (0.29–0.75)] and the presence of cerebrovascular disease [0.31 (0.13–0.72)] and/or neuropathy [0.56 (0.36–0.87)] were associated with reduced odds. After adjustment for these independent determinants, neither correct self-perception of overweight status at baseline nor being informed of being overweight by their GP influenced whether or not a patient subsequently lost weight (P≥.22). For those whose weight remained stable or who gained weight during follow-up, the average weight gain was 3.9±4.0 (range 0 to 20) kg (see Table 3).
Table 3. Baseline characteristics of 518 overweight/obese patients who attended at least five annual assessments categorised by weight loss status after 4 years' follow-up
| Weight loss | No weight loss | P value | |
|---|---|---|---|
| Number (%) | 298 (57.5%) | 220 (42.5%) | |
| Age (years) | 63.8±9.7 | 61.1±8.8 | .002 |
| Sex (% male) | 51.0 | 55.5 | .33 |
| Diabetes duration (years) | 3.0 [0.9–7.0] | 3.0 [0.5–7.0] | .53 |
| Treatment [diet/oral agents (OGLM)/insulin±OGLM; %] | 32.7/60.6/6.7 | 33.3/55.3/11.4 | .15 |
| Fasting plasma glucose (mmol/l) | 8.5 [7.1–10.4] | 8.2 [6.6–10.7] | .31 |
| HbA1c (%) | 7.3 [6.5–8.6] | 7.2 [6.2–8.5] | .27 |
| Diabetes knowledge score (0–15) | 10 [6–11] | 10 [7–12] | .31 |
| Body mass index (kg/m2) | 31.2±5.3 | 30.6±3.8 | .12 |
| Abdominal overweight/obesity (%) | 94.9 | 94.1 | .70 |
| Self-perception of overweight (%) | 72.1 | 71.8 | 1.00 |
| Informed overweighta (%): | 66.3 | 73.2 | .10 |
| By GP | 52.9 | 59.5 | .15 |
| By specialist | 10.8 | 10.9 | 1.00 |
| By diabetes educator | 24.9 | 30.0 | .23 |
| By dietitian | 8.1 | 10.5 | .36 |
| By family/friends | 4.7 | 3.2 | .50 |
| Regular self-weigh (%) | 64.8 | 68.1 | .42 |
| Times/monthb | 4 [2–12] | 4 [3–20] | .27 |
| Given exercise programme (%) | 62.3 | 66.8 | .31 |
| Any exercise in past 2 weeks (%) | 77.1 | 79.0 | .67 |
| Visited dietitian in past year (%) | 13.1 | 23.2 | .003 |
| Diabetes education/advice in past year (%) | 37.6 | 43.6 | .18 |
| Smoking status (never/ex-/current; %) | 46.6/41.6/11.7 | 43.2/43.2/13.6 | .68 |
| Alcohol (standard drinks/day) | 0 [0–0.8] | 0 [0–0.8] | .81 |
| Married/de facto relationship (%) | 71.8 | 70.0 | .70 |
| Educated beyond primary level (%) | 73.0 | 77.3 | .31 |
| Not fluent in English (%) | 14.4 | 13.2 | .70 |
| Ethnic background (%) | |||
| Anglo-Celt | 65.8 | 63.2 | .67 |
| Southern European | 20.1 | 20.0 | |
| Other European | 7.7 | 11.8 | |
| Asian | 2.0 | 1.8 | |
| Mixed/other | 4.0 | 3.2 | |
| Indigenous Australian | 0.3 | 0 | |
| Household income (AUD; %) | |||
| ≤$12,000 | 26.0 | 23.8 | .52 |
| $12,001–25,000 | 53.5 | 51.4 | |
| >$25,000 | 20.5 | 24.8 | |
| Depressed (%) | 28.8 | 32.1 | .33 |
| Coronary heart disease (%) | 28.8 | 32.6 | .38 |
| Cerebrovascular disease (%) | 3.4 | 7.7 | .044 |
| Peripheral arterial disease (%) | 22.2 | 21.7 | .91 |
| Retinopathy (%) | 13.2 | 11.5 | .59 |
| Neuropathy (%) | 22.2 | 27.9 | .14 |
| Microalbuminuria (or worse) (%) | 37.0 | 39.6 | .58 |
| Self-reported cancer (ever; %) | 11.1 | 10.9 | 1.00 |
aPatient may have been informed he/she is overweight by more than one source. |
bIn those who weighed themselves regularly. |
4. Discussion
Although self-perception of body weight paralleled BMI in the present study, there were still a large proportion of patients (31.4%) who did not recognise their overweight or obesity. While this percentage was not as large as that found in the general Australian population (38.8%) (Donath, 2000), it is still of concern given the importance of body weight in type 2 diabetes management. Abdominal obesity proved more influential in the correct identification of body weight than BMI, which is an encouraging finding given the influence of visceral fat on the development of diabetes and other cardiovascular risk factors (Abate et al., 1996).
Baseline self-perception of weight was influenced by several socio-demographic factors after adjusting for BMI. The existing strong evidence from general population and diabetes (Blokstra et al., 1999, McTigue et al., 2006, Paeratakul et al., 2002) studies that women are more likely to overestimate their weight than males was confirmed in the present study, particularly in the overweight group. This is likely to reflect social and media pressures that reinforce the desirability of a lean female body habitus in developed countries. The importance of socioeconomic factors (both household income and educational attainment) in determining the accuracy of weight self-perception was highlighted in the present study. English proficiency was a further barrier to correct self-identification of body weight. Because 18.5% of our sample was of Southern European ethnic background, limited English-speaking ability could reflect linguistic barriers to weight-related health messages, but it could also be a surrogate for cultural differences in the way overweight is viewed. Indeed, frequent misperception of weight has been reported in the general Spanish population (Gutiérrez-Fisac, López García, Rodriguez-Artalejo, Banegas Banegas, & Guallar-Castillón, 2002).
Self-identification of overweight was independently associated with the number of sources who had informed the patient of their weight status. These sources were mainly health care professionals, especially GPs who usually have greater contact with patients than specialist physicians, nurses, and allied health personnel. The relatively minor role played by friends and family underlines the importance of addressing weight status in a clinical setting. Although there was no relationship between the number of informant sources and self-perception of weight status in the obese group, it is encouraging that approaching three-quarters of these patients had been informed of their obesity by their GP. In a US study, only 42% of obese patients were instructed to lose weight by their primary care physician (Galuska, Will, Serdula, & Ford, 1999).
In contrast to the results of a previous study of the views of diabetic patients on healthy body weight (McTigue et al., 2006), we found that diabetes knowledge was independently associated with an increased odds of obese patients being aware they were overweight. Although this suggests that current educational initiatives in an urban Australian setting are addressing weight management issues in type 2 diabetes, the lack of such an association in the overweight group may mean that metabolic and vascular consequences of a BMI in the overweight range are not being adequately emphasised.
In the present study, neither accurate body weight self-perception nor being informed of overweight by the GP positively influenced weight loss over the next 4 years. Paradoxically, a visit to a dietitian during the year before baseline assessment reduced the odds of subsequent weight loss. This may simply reflect the fact that a selected, refractory group, i.e., those patients who had been consistently unable to lose weight, was referred to a dietitian who was also unable to effect a significant change in lifestyle. Such patients are likely to require more than the knowledge of weight status and support from health care professionals to lose weight and may be candidates for pharmacotherapy and perhaps surgical intervention. Consistent with our findings, exercise levels did not increase in proportion to appropriate advice given by health care professionals in a US study of type 2 diabetic patients (Morrato, Hill, Wyatt, Ghushchyan, & Sullivan, 2006).
Patient barriers exist in initiating behavioural change. A patient's perception of the ease or difficulty of undertaking a behaviour (perceived behavioural control) has been identified under the Theory of Planned Behaviour framework to explain a majority of the variance in physical activity and healthy eating intentions in a study of those at risk of type 2 diabetes (White, Terry, Troup, & Rempel, 2007). This indicates that patients need information on methods of weight loss in addition to the intention to undertake the lifestyle changes. Furthermore, fewer costs and control factors such as availability or discipline influenced individuals with type 2 diabetes in being physically active (White et al., 2007). Interventions must address these potential barriers to allow patients to initiate changes to lifestyle factors which affect weight loss.
The present study had limitations. We relied on self-reported sources of information relating to informants of overweight status which is subject to recall bias. In addition, we did not identify whether lifestyle recommendations or advice was provided by GPs and, if so, whether they were appropriate. Interestingly, studies in both the US and Australia have identified that patients view GPs as having a role in assisting with weight loss (Morrato et al., 2006, Tan et al., 2006), but, in the Australian study (Tan et al., 2006), only 46% of patients felt GPs could provide enough time to supply effective weight loss advice.
Health care professionals can play a valuable role in promoting correct self-identification of body weight by diabetic patients in the community. Nevertheless, diabetes education programmes should recognise the impact of gender and socio-demographic variables such as educational level, English fluency, and income on accurate weight self-perception, in addition to providing help with weight loss beyond just the recognition of the issue.
Acknowledgments
We thank the patients for their participation, and FDS and Fremantle Hospital staff for their assistance and cooperation.
References
- . Relationship of generalized and regional adiposity to insulin sensitivity in men with NIDDM. Diabetes. 1996;45:1684–1693
- . Standards of medical care in diabetes — 2006. Diabetes Care. 2006;29:S4–S42
- . Perception of weight status and dieting behaviour in Dutch men and women. International Journal of Obesity and Related Metabolic Disorders. 1999;23:7–17
- . Diabetes education and knowledge in patients with type 2 diabetes from the community: the Fremantle Diabetes Study. Journal of Diabetes and its Complications. 2003;17:82–89
- . Glycemic control in older subjects with type 2 diabetes mellitus in the Fremantle Diabetes Study. Journal of the American Geriatrics Society. 2000;48:1449–1453
- . A prospective study of depression and mortality in patients with type 2 diabetes: the Fremantle Diabetes Study. Diabetologia. 2005;48:2532–2539
- . Self-perception of weight appropriateness in the United States. American Journal of Preventive Medicine. 2003;24:332–339
- . Autoantibodies to glutamic acid decarboxylase in diabetic patients from a multi-ethnic Australian community: the Fremantle Diabetes Study. Diabetic Medicine. 2000;17:667–674
- . Who's overweight? Comparison of the medical definition and community views. Medical Journal of Australia. 2000;172:375–377
- . Are health care professionals advising obese patients to lose weight?. Journal of the American Medical Association. 1999;282:1576–1578
- . Self-perception of being overweight in Spanish adults. European Journal of Clinical Nutrition. 2002;56:866–872
- . Waist circumference action levels in the identification of cardiovascular risk-factors — prevalence study in a random sample. BMJ. 1995;311:1401–1405
- . Dietary treatment and long-term weight loss and maintenance in type 2 diabetes. Obesity Research. 2001;9:S348–S353
- . Perception of “healthy” body weight by patients with diabetes. Diabetes Care. 2006;29:695–697
- . National standards for diabetes self-management education. Diabetes Care. 2006;29(Suppl 1):S78–S85
- . Are health care professionals advising patients with diabetes or at risk for developing diabetes to exercise more?. Diabetes Care. 2006;29:543–548
- . Sex, race/ethnicity, socioeconomic status, and BMI in relation to self-perception of overweight. Obesity Research. 2002;10:345–350
- . Weight management in general practice: what do patients want?. Medical Journal of Australia. 2006;185:73–75
- . Behavioral, normative and control beliefs underlying low-fat dietary and regular physical activity behaviors for adults diagnosed with type 2 diabetes and/or cardiovascular disease. Psychology, Health & Medicine. 2007;12:485–494
- . Obesity: preventing and managing a global epidemic. A report of a WHO consultation. Geneva: WHO; 2000;
☆ The Fremantle Diabetes Study was funded by the Raine Foundation, University of Western Australia.
PII: S1056-8727(09)00127-5
doi:10.1016/j.jdiacomp.2009.11.001
© 2011 Elsevier Inc. All rights reserved.
Volume 25, Issue 1 , Pages 1-6, January 2011
