OUP user menu

Body weight dissatisfaction in the Icelandic adult population: a normative discontent?

Erna Matthiasdottir, Stefan Hrafn Jonsson, Alfgeir Logi Kristjansson
DOI: http://dx.doi.org/10.1093/eurpub/ckq178 116-121 First published online: 2 December 2010

Abstract

Background: Body weight dissatisfaction is a common problem in many modern societies and it appears to have grown over time. This study examined the prevalence and predictors of body weight dissatisfaction among 18- to 79-year-old Icelanders. The distribution of body weight dissatisfaction according to the WHO body mass index criteria for underweight, normal weight, overweight and obesity was also assessed. Methods: The study was based on 5832 participants in the cross-sectional survey ‘Health and Well-Being of Icelanders’, administered by the Public Health Institute in Iceland in the fall of 2007. Results: Around 43% of adult Icelanders are dissatisfied with their own body weight (50% of females; 35% of males) and 72% believe they need to lose weight (80% of females; 63% of males). Despite being in normal range according to the WHO-based BMI criteria, 64% of females in that category believe they need to lose weight. Multivariate results show that indicators of socio-economic status (SES) are not significantly associated with body weight dissatisfaction among females, while among males, those with university education are more often dissatisfied than those with secondary education (OR = 1.5, P < 0.05; for self-perceived need to lose weight, estimates are OR = 1.4, P < 0.05 for females, and OR = 2.5, P < 0.001 for males). Conclusion: The prevalence of body dissatisfaction is so prominent among Icelandic female adults that it can rightfully be labelled a normative discontent in this population.

Introduction

Body weight dissatisfaction is common in many modern societies and it appears to have grown over time.1–3 An increasingly large discrepancy appears between current body image ideals and the body size of ordinary people.4 The body image favoured in many societies in recent times is very slim5–7 but at the same time the average weight has increased in many countries.8–10 Studies have found some gender disparity in the way body dissatisfaction and body weight dissatisfaction appears in the population with the prevalence of dissatisfaction generally higher among females.5,11,12 Because of how widespread body weight dissatisfaction is among females, it has been labelled ‘a normative discontent’3 in the population. The term ‘normative discontent’ was first described by Rodin et al.3 in 1984 and has become a growing public health concern.11–13 Studies have shown that body- and physical appearance seem to be of more importance to females than males.14 Besides being more dissatisfied with their bodies11–13 females are more likely than males to overestimate their body weight and to be overly concerned about it.15 Even so, body weight dissatisfaction also appears to be a growing problem among males.12,16 Whereas females are more likely to want to be thinner12 males that are dissatisfied with their body weight either want to lose or gain weight and are more concerned about their figure and muscularity.16–18

It has been argued that in modern Western culture, the demand for having a slender body is strong and problems related to negative body image are no longer limited to middle class females as before.6 Recent developments in mass communication technology have resulted in greater overall circulation of messages about body shape ideals through media, regardless of social position or education.7 It has been pointed out that much more work is needed in comparing body satisfaction in people from different social groups, with emphasis on income and class differentials in body image.5 Body weight images may influence people’s quality of life and health behaviours. Studies have reported a positive relationship between negative body image and body dissatisfaction with various psychological, physical and social problems.5,13,19,20

The aim of this study was to identify the prevalence and main predictors of body weight dissatisfaction among 18- to 79-year-old Icelanders, particularly with regard to the normative discontent assertion. Furthermore, the distribution of body weight dissatisfaction in relation to the WHO division of body mass index (BMI) into underweight, normal weight, overweight and obesity was assessed.

Methods

Study population

The study was based on data from ‘the survey on health and well-being of Icelanders in 2007’, which was administered by the Public Health Institute of Iceland. Anonymous questionnaires were mailed to 18- to–79-years-old residents in Iceland based on a cross-sectional stratified random probability sample.21 The sample was stratified on age and residence, with half coming from the capital area and half outside it, but 63.4% of the Icelandic population lived in the capital area in 2007. The sample was stratified by age with equal number of people sampled from each age group because of unequal age distributions in Iceland. When running multiple binominal regression on data from stratified sample one must take into account the survey design and unequal sample weights in estimating the standard errors and other model statistics. For statistical modelling, we use the R software with the ‘survey’ package.22,23 The total study sample was 9807 and a response rate of 60.3% (n = 5918) was received.24 Due to item non-response, this study is based on 5832 responses (53.3% females).

Measures

Body weight dissatisfaction

As the main dependent variable in the study, participants were asked the following question: ‘How satisfied or dissatisfied do you feel with your own body weight?’ Response categories were: very dissatisfied, rather dissatisfied, neither satisfied nor dissatisfied, rather satisfied and very satisfied. The variable was collapsed to form three groups in the cross-tabulations; dissatisfied with body weight, neither satisfied nor dissatisfied and satisfied with body weight. For the multivariate analyses, the variable was collapsed into two categories, very or rather dissatisfied (1) and very or rather satisfied or neither (0).

The need to lose weight

To measure preferences regarding the need to lose weight, we use the question: ‘Do you feel like you need to lose or gain weight?’ Response formats were: lose more than 15 kg, lose 11–15 kg, lose 6–10 kg, lose 1–5 kg, neither lose nor gain weight, gain 1–5 kg, gain 6–10 kg, gain 11–15 kg, gain >15 kg. The number of categories were reduced to three in the cross-tabulations: 1= those that believe they need to lose weight, 2= do not want to change their body weight and 3= those that believe they need to gain weight. In the multivariate modelling groups were reduced to two: 1= want to lose weight and 0= does not need to lose weight or need to gain weight. The reason is that association between the need to gain weight and body weight dissatisfaction is much smaller than with needing to lose weight (analysis not shown).

Age

Age was measured by subtracting self-reported year of birth from 2007 that yields age at last birthday at the end of 2007. Age is recoded into six groups: 18–29, 30–39, 40–49, 50–59, 60–69 and 70–79 year olds, respectively.

BMI

BMI was calculated from self-reported height and weight by dividing weight in kilograms by height in squared metres (kg/m2). The BMI scores were grouped according to the WHO classifications into underweight (<18.5), normal range (18.5–24.9), overweight (25.0–29.9) and obesity (≥30.0). In order to allow for curvilinear relationship between BMI and the dependent variables in the multivariate analyses, we centred BMI by subtracting 18.5 from its original values and then squared the difference. Both the centred and the squared values were entered in the multivariate models.

Socio-economic status

Socio-economic status (SES) was assessed with several variables. First, people were asked to report the highest educational degree they have completed. Response categories were classified into secondary school education (reference category in multivariate analyses), junior college level and university level education. Four dummy coded variables were extracted from a multiple response question about labour force participation; paid employment, being self-employed, unemployment and disability. The reference category is based on all other labour force variables put together, including; working at home, retired and attending school only. As a simple estimate of income, we used a single item measure of economic affluence hardship where people were asked how easy or difficult it has been for them and their family (if applicable) to pay the bills during the last 12 months. The response categories range from very easy, rather easy, neither, rather difficult and very difficult.

Data analysis

Data analysis and statistical tests were carried out using both SPSS 17.022 and R.25 Cross-tabulations were performed in SPSS 17.0 to report frequencies and proportions and Pearson’s Chi-square significance tests were run in R to identify differences between proportions. Because of unequal weights the complex survey design module in R was used. Unequal weights are derived from the sample design that was stratified on age and residency.22,23 Multivariate analyses are carried out in SPSS 17.0 using binary logistic regression with odds ratios.

Results

Average height of study participants was 172.8 cm (SD: 9.0) and the average weight was 81.2 kg (SD: 16.8). The average BMI in the study sample was 27.1 (SD: 4.9). BMI values ranged from 15.2 to 75.4 (data not shown). About 1% of the study participants were identified as underweight (BMI < 18.5), 38.9% in normal range (BMI 18.5–24.9), 39.9% as overweight (BMI 25–29.9) and 20.1% obese (BMI ≥ 30). Females over normal range were 53.7%, with just over 32% being overweight and 21.4% obese. Males over normal range were 66.4%, with about 47% being overweight and 19% identified as obese. Although more males than females were over normal range, or 66.4% compared to 53.7% [χ2 (8) = 97 629, P < 0.01], more females than males were categorized as obese, or 21.4% against 19.0% [χ2 (1) = 5089, P < 0.05].

The results reveal that 42.7% of 18- to 79-year-old Icelanders are dissatisfied with their body weight and over 71% think they need to lose weight. Table 1 shows weight satisfaction and dissatisfaction by gender, age, BMI, educational level and perceived need to lose weight. Among females, 50.1%, reported to be dissatisfied with their body weight and 35.1% of males, which is a significant gender difference [χ2 (2) = 127.912, P < 0.01]. Body weight dissatisfaction is more common among females than males in all age groups studied. In fact, the lowest prevalence of body weight dissatisfaction among females (39.9% in the 70- to 79-year-old group) is similar to the highest prevalence among males (40.6% among 50- to 59-year olds). Table 1 also shows that the prevalence of body weight dissatisfaction increases in the population with age up to ∼60 years, but decreases thereafter [χ2 (10) = 64.752, P < 0.01]. Body weight dissatisfaction is greatest among those in the obese category, almost 77% among males and 87.5% for females. This prevalence is also over 64% among overweight females but only 37% among males in that category. Weight dissatisfaction does not seem to vary with education among females, but a slight upper trend was observed among males with higher educational attainment revealing higher prevalence of body weight dissatisfaction.

View this table:
Table 1

Body satisfaction and dissatisfaction by gender and selected characteristics

MalesFemales
Satisfied (%)Neither (%)Dissatisfied (%)Sample naSatisfied (%)Neither (%)Dissatisfied (%)Sample na
Total47.517.435.1270236.413.550.13086
Age, years (%)
    18–2948.721.629.732835.712.951.5463
    30–3946.715.537.838531.612.256.2490
    40–4945.017.837.343436.813.749.6548
    50–5945.413.940.650337.313.349.4534
    60–6949.315.435.354637.115.747.1520
    70–7954.418.926.750645.714.539.9531
BMI
    Underweight58.52.738.81161.211.627.238
    Normal weight77.713.88.578263.514.022.51213
    Overweight41.721.237.1132120.015.964.11040
    Obese9.213.976.95503.98.687.5699
Education
    Secondary school48.020.531.598034.915.050.01450
    Junior college48.717.234.1119935.313.051.6862
    University44.412.942.745740.111.748.2713
Need
    Need to lose weight26.121.452.5170323.715.361.02423
    Neither63.019.717.323041.911.446.782
    Need to gain weight93.96.00.275794.75.00.3559
  • a: Sample n are unweighted sample size for group defined by sex and variables on the leftmost column. The proportions in the table are all based on weighted counts. Sex distribution for ages 18- to 79-year olds in Iceland in 2007 are about equal (50.4% males, 49.6% females), which allows the reader to take a simple average of males and females to get the overall proportions

Table 2 shows that just over 80% of females and close to 63% of males believe they need to lose weight [χ2 (1) = 5089, P < 0.05]. The table also shows the need to lose weight among males and females in line with the standardized WHO-based BMI groups. About four in five males in the overweight group and nearly all in the obese group report that they need to lose weight. This prevalence is even higher among females with 94.5% in the overweight group needing to lose weight and nearly all in the obese category. Furthermore, almost 64% of females in normal BMI range want to lose weight but only ∼ 19% of males in normal range want to lose weight.

View this table:
Table 2

The need to lose weight by BMI categories

Under- weight (%)Normal weight (%)Over weight (%)Obese (%)All (%)
Males
    Lose weight0.018.880.497.262.8
    Neither40.147.417.51.724.5
    Gain weight59.933.82.11.112.7
Females
    Lose weight11.763.894.598.780.4
    Neither53.033.04.60.717.3
    Gain weight35.33.30.80.62.4

Multivariate findings are shown in table 3. The need to lose weight is by far the strongest predictor for body weight dissatisfaction among both males and females (OR: 5.58 and 6.46, respectively). Age has a slight inverse relationship with body weight dissatisfaction for both sexes but the SES indicators do not show much impact, apart from the university education indicator among males (OR: 1.45). Body weight dissatisfaction increases with BMI levels but the squared BMI variable shows this relationship to be slightly curvilinear among both males and females. The need to lose weight is most strongly predicted by increased levels of BMI among both males and females (OR: 3.07 and 2.33, respectively) but as with body weight dissatisfaction this is observed as a slightly curvilinear relationship. To a small extent, age also seems to be a defending factor for the need to lose weight. The SES indicators do not reveal any impact on the need to lose weight apart from the measure for university education which is particularly strong among males (OR: 2.46) but less so for females (OR: 1.44).

View this table:
Table 3

ORs for body weight dissatisfaction and the need to lose weight predicted by age, SES indicators and BMI

Body weight dissatisfactionNeed to lose weight
B95% CI (+/−)exp(B)B95% CI (+/−)exp(B)
Males
    (Intercept)−4.4641.1750.012***−5.5211.1230.004***
    Age−0.0130.0080.987**−0.0080.0090.992
    Need to lose weight1.7190.3215.579***
    Junior college education0.0590.2701.061−0.0030.3060.997
    University education0.3720.3241.451*0.9000.3852.459***
    Paid work−0.0900.1990.914−0.0950.2330.909
    Self-employed−0.0820.1910.9210.0310.2471.031
    Unemployed0.0000.0001.0000.0000.0001.000
    Disability0.1340.2661.143−0.2060.3610.814
    Problems paying bills0.0800.1131.083−0.0850.1310.918
    BMI centred at 18.50.3810.1181.464***1.1200.1363.065***
    (BMI-18.5)2−0.0070.0040.993***−0.0230.0040.977***
Females
    (Intercept)−3.1780.8050.042***−0.1210.7920.886
    Age−0.0290.0080.972***−0.0690.0120.933***
    Need to lose weight1.8660.4276.462***
    Junior college education−0.0690.2580.9330.1940.3311.215
    University education−0.1490.2670.8620.3650.3361.440*
    Paid work−0.0950.1690.9100.0410.2091.041
    Self-employed0.1840.2531.2020.0110.2991.011
    Unemployed0.0000.0001.0000.0000.0001.000
    Disability−0.1060.2710.8990.2530.3321.287
    Problems paying bills0.0800.0941.083−0.0580.1220.944
    BMI centred at 18.50.4230.0701.527***0.8460.0982.330***
    (BMI-18.5)2−0.0070.0030.993***−0.0130.0020.987***
  • *P < 0.05; **P < 0.01; ***P < 0.001

Discussion

Findings indicate that a large proportion of the Icelandic adult population is dissatisfied with their body weight. More females than males are dissatisfied with their body weight in all age groups even though the proportion of males with BMI over normal range is close to 13% higher. It is particularly noteworthy that among overweight participants 64% of females are dissatisfied with their bodyweight but only 37% of males. More females than males also feel they need to lose weight. This fact is strikingly revealed in the normal BMI category where close to 64% of females but only 19% of males claim that they need to lose weight. In short, neither body weight dissatisfaction nor the need to lose weight is as pervasive among Icelandic males as they are with females independent of weight status.

The prevalence of body weight dissatisfaction is so prominent among Icelandic females that it can rightfully be labelled ‘a normative discontent’ in the population. These findings are in concordance with a number of studies from other countries.3,11–13 Such high levels of discontent is of public health concern because research has shown negative body weight dissatisfaction to be related to various physical-, psychological- and social risk factors.5,13,19,20 However, some studies, e.g. Heinberg and colleagues, suggest the opposite that some degree of worry about body image may be an important predictor of health behaviours.13 Studies have also shown that body weight dissatisfaction can be a predictive factor for successful weight loss attempts.26 On the other hand, it has been shown that chronic body weight dissatisfaction plays a role in the aetiology of eating pathology.27 And researchers have shown that attempts to lose weight has the tendency to lead to increase in weight, counter to intended goals (yo-yo dieting).5 It has also been shown that long-term maintenance of weight loss has not been effective28,29 and up to 90–95% adults that lose weight will gain weight again.28

In the current literature on normative discontent, there is no precise definition for when it is proper to label body- and body weight dissatisfaction as such but it may seem appropriate to set the frame of reference at including ≥50% (the majority) of a population. According to this assertion, the term normative discontent does apply to Icelandic females but not to males.

The study results also show that females’ dissatisfaction with their own body weight remains rather stable throughout the lifespan. The prevalence of body weight dissatisfaction among females does not seem to decline until the oldest group aged 70–79 years. Still there are close to 40% dissatisfied with their body weight in that group. Hence, a normative discontent with own body weight exists in the Icelandic female population more or less throughout the lifespan. Studies from other countries have reported similar age-related patterns.12,30,31

Existing studies of body weight dissatisfaction in relation to respondent's social characteristics have shown mixed results.5,32 Some have shown females to be more dissatisfied with their own body weight with increased levels of education and higher social grade.33,34 Others have shown the opposite34 and still others have indicated that these factors do not have much impact.35 More research is needed in this regard. Our results indicate that SES does not have strong relations with body weight dissatisfaction in Iceland and neither does it with the perceived need to lose weight. The exception is for university education that predicted greater levels of dissatisfaction with body weight and the need to lose weight, particularly for males. This indicates that in Iceland body weight dissatisfaction is more or less independent of social status. Female dissatisfaction seems to be a social norm.

Even though the results of the study show a large proportion of the Icelandic population being over normal BMI range, the division between overweight and obesity should be taken into account. Much larger proportion is overweight than obese. In that context, it is important to bear in mind that the WHO criteria for overweight is still debated and some researchers believe there not to be enough scientific evidence for the assumption that health risk increases at BMI = 25.36–38 Furthermore, greater mortality rates have not been directly related to being overweight.39 Finally, some studies have shown that it is possible to improve the health and wellbeing of overweight or even obese people by improving their ways of living by bringing them towards healthier lifestyles without direct implications to their weight status.37,39,40

Implications for practice and future research

These findings may be of use for Icelandic health policy regarding weight loss. Since studies have shown that attempts to lose weight are usually not successful in the long run, it might be worthwhile to search for new approaches where the aim should be an overall improvement of health and well-being. Among other things, this may be done by emphasizing a healthy lifestyle, good diet, modest exercise and improvement in self- and body image regardless of size or body weight.

Our results highlight the need for further research in this area. Body image as represented by satisfaction and dissatisfaction of body weight may contribute to the overall health and well-being of individuals. Therefore, it is important to identify those groups of people who may be at risk for developing a negative body weight image. We may want to investigate further who particularly are dissatisfied with their body weight and believe they need to lose weight. It would also be beneficial to design a prospective longitudinal study to assess the social- and behavioural trajectories that may lead into body weight dissatisfaction and if body weight dissatisfaction changes and/or develops with age.

Study limitations and strengths

Several limitations of this study should be noted. First, the study is based on a cross-sectional data, which does not provide causal evidence. Second, the response rate was 60.3% which is rather low. Third, single measure variables were used to measure body weight satisfaction and the perceived need to lose or gain weight. Preferably, pre-tested measure instruments should be used in population studies of body weight satisfaction and related factors. Due to this limitation, we may have over- or underestimated the true prevalence of these variables in the Icelandic population. Fourth, height and weight that were used to calculate BMI were based on self-reports. We are, therefore, unable to rule out underreporting in this regard. Furthermore, the BMI measure does not take into account the muscle-to-fat ratio which may be one of the reasons why males are so much less inclined than females to be dissatisfied with their body weight and think they need to lose weight in our study. Despite these limitations, the study also has a number of strengths. The assessment is based on a large number of participants and the sample is a random probability one that reflects the Icelandic nation. Such a comparative study has not been carried out in Iceland before.

Conclusions

Findings indicate that a large proportion of the Icelandic population is dissatisfied with their body weight. The prevalence of body weight dissatisfaction in the female population by far exceeds the number of overweight and/or obese females and is independent of socio-economic or demographic factors. Body weight dissatisfaction is so prominent among Icelandic females that it can rightfully be labelled ‘a normative discontent’ in the population. Public health policy in this area should aim to improve general population health by emphasizing a healthy lifestyle, good and nutritional diet and modest exercise along with improvements in self- and body image.

Conflicts of interest: None declared.

Key points

  • A large proportion of the Icelandic population is dissatisfied with their body weight.

  • The prevalence of body weight dissatisfaction is greater among females than males at all age groups even though the proportion of males with BMI over normal range is 13% higher than for females. More females than males think they need to lose weight.

  • About 64% of females in normal weight range feel they need to lose weight. The prevalence of body weight dissatisfaction is so prominent among females in Iceland that it can rightfully be labelled ‘a normative discontent’.

  • Males with higher educational status were more likely than other males to be dissatisfied with their body weight but body weight dissatisfaction was not predicted by any socio-economic variables among females in our multivariate analyses. This further underlines the normative discontent assertion.

References

View Abstract