Abstract
Objectives
This study was designed to investigate factors affecting the discordance between body image and body mass index amongst Korean adults aged 19–39 years.
Methods
Data (N = 59,361) from the 2014 Korean Community Health Survey was analyzed using descriptive analysis, chi-square test, and logistic regression analysis.
Results
To examine the factors affecting body image discordance as observed in 43.1% of participants, the group was subdivided into underestimation and overestimation. There were 36.0% of participants that were body image discordant underestimators and 7.1% were overestimators. Multivariate logistic regression analysis showed that the underestimators tended to be men, graduated from high school, married, current/ex-smokers, ex-drinkers, had between 5 to 9 hours sleep (inclusive), had fair to good self-rated health, and demonstrated healthy weight control behavior, relative to the reference group. In comparison, overestimators tended to be in the 19–29 year group and had signs of depression. The OR of individuals who were men, married, and had healthy weight control behavior was significantly lower in the overestimators group.
Conclusion
These findings show that tailored interventions to promote accurate body awareness should be based on the type of body image discordance. In addition, it is necessary to develop and implement a program of body image improvement that considers the factors affecting body weight discordance.
Keywords: adult, body image, body mass index, overweight
Introduction
Obesity is a risk factor for chronic diseases and a major cause of death [1,2]. It has been attributed to mental and social disorders such as sleep disorders, depression, and low self-esteem [3,4]. The World Health Organization (WHO) classifies obesity as a disorder, and many countries are implementing diverse health policies to prevent it. Nevertheless, the number of obese people is continuously growing globally. In 2016, approximately 13% of the world’s adult population was obese [5]. In the populations of countries that form the Organization for Economic Cooperation and Development, 19% of the adult population from these countries was obese [6].
Being underweight means weighing 10% to 20% less than normal body weight, which corresponds to a body mass index (BMI) of < 18.5 kg/m2 in Asian countries [2,7]. The percentage of underweight adults in South Korea increased from 3.9% in 2010, to 4.7% in 2014 [8]. Being underweight can lead to serious health issues, such as deterioration in body function, decreased bone density, and weak immunity. Underweight women of child-bearing age, can also experience irregular menstrual periods, poor fetus growth, and premature labor [9]. To prevent these problems and improve an individual’s quality of life, maintaining an adequate weight is critical and can be achieved through appropriate weight control behaviors. Generally this is initiated by a subjective self-perception of body shape (i.e., body image) as opposed to objective physical indices, such as body weight or waist size [10].
The discrepancy between body image and BMI is called body image discordance (BID). Perceiving one’s body as slimmer than the actual BMI is categorized as underestimation, whilst perceiving one’s body as heavier than the actual BMI is categorized as overestimation. BID leads to body image dissatisfaction and negatively affects mental health, such as lowering self-esteem and increasing depression [11,12].
A previous study reported that people who overestimate typically put more effort into losing weight than those with body image concordance [13]. Women, in particular, engage in unnecessary weight control behaviors by overestimating their body shapes [14]. Whilst 30.9% of the South Korean population is overweight, 63.5% of the total population is trying to lose weight [8], this implies that BID is becoming more serious. Since BID can negatively affect psychosocial issues, it is necessary to elicit appropriate health-promoting behaviors by promoting accurate body image [15].
Adulthood is the time period during which physical maturation is reached. Our body requires different calorific and nutrient content compared to the growth period of youth, therefore, it is crucial to form healthy eating habits from childhood and to manage health. However, a busy lifestyle is often an issue resulting in a lack of physical activity. Furthermore, Westernized diets and the development of convenience foods may hinder the formation of healthy eating habits [16]. Healthy diet habits, and health management in adulthood is important for middle-aged and elderly individuals, therefore, it is imperative to have an appropriate body image and a good health management habit.
In recent years, the prevalence rate of metabolic syndrome for Koreans in their 40s ranged from 19.9% to 30.8%, showing a sharp increase at this age [17]. This highlights the significance of health management during young adulthood. Therefore, this study investigated BID amongst adults aged under 40 years and established baseline data to improve health during adulthood.
The WHO BMI-based criterion to determine obesity amongst Asians was recently modified from an overweight category BMI ≥ 25 kg/m2, to BMI ≥ 23 kg/m2 [18]. When this new criterion was applied to Koreans, people who were previously classified as being in a normal weight group, were reclassified as being in a high-risk group for chronic diseases; therefore, it was necessary to study BID using the new criterion. Most studies on BID that were conducted in South Korea did not apply this BMI criterion for Asians [10,11]; therefore, there is a limited understanding of the factors affecting BID among young Korean adults. Consequently, this study sought to identify the specific factors affecting adults aged 19–39 years to develop an intervention program that encourages appropriate health-promoting behaviors and an accurate body image to improve individuals’ quality of life.
Materials and Methods
1. Design
A descriptive study design was used for secondary data analysis.
2. Data collection and participants
Since 2008, the Korean Community Health Survey has been conducted annually in 253 community clinics throughout Korea by the Korean Centers for Disease Control and Prevention. To maintain the representativeness of sample households, an average of 900 samples was extracted and selected using a multi-phase probability sampling method. Data were collected after a 1:1 interview by a trained surveyor. There were 228,712 participants in 2014, including 60,194 adults aged 19–39 years. Data were analyzed from 59,361 participants after excluding 833 adults who had insufficient data (e.g., weight, height) to confirm BID (Figure 1).
Figure 1.
Selection process of the study
KCHS = Korean Community Health Survey; BMI = body mass index.
3. Research variables
3.1. General characteristics
We examined participants’ general characteristics including gender, age, education level, marital status, and employment status, which were divided as follows: age = “19–29 years,” or “30–39 years;” education level = “≤ high school education” or “≥ college education;” marital status = “married” or “single;” and employment status = “employed” or “unemployed.”
3.2. Health behaviors
We examined participants’ health behaviors including smoking status alcohol consumption, frequency of breakfast consumption, sleep duration, perceived stress, depression, perceived health condition, physical activity, weight control behavior, and obesity, which were divided as follows: smoking = “non-smoker,” “ex-smoker,” and “currently smoking;” alcohol consumption = “non-drinker,” “ex-drinker,” and “currently drinking”; breakfast consumption = “≤ 4 times a week” and “over 5 times a week;” sleep duration = “≤ 5 hours,” “6–8 hours,” and “over 9 hours;” and stress perception = a non-perceived stress group who reported “rarely feeling stressed” in daily life, and a perceived stress group who reported “frequently feeling stressed.” For depression, participants who felt sad or depressed for more than 2 consecutive weeks in the last year to the point that it interfered with their daily life were categorized as the “depression” group. For perceived health status, those who reported “good” or “very good” were categorized as the “good health” group; those who reported “fair” were categorized as the “average health” group; and those who reported “poor” or “very poor” were categorized as the “poor health” group. Regarding physical activity level, questions on intense physical activity, moderate level physical activity, and walking were used. Intense physical activity comprised of exercising over 30 minutes/day for at least 5 days/week; moderate level physical activity comprised of exercising over 20 minutes/day for at least 3 days/week; walking comprised of walking over 30 minutes/day for at least 5 days/week. Participants engaged in one of the 3 physical activities were categorized as the “physically active” group. Regarding weight control behaviors, anyone who tried to lose, maintain, or gain weight in the last year was categorized as the “weight control behavior” group. For obesity, the height (m) and weight (kg) of each respondent was used to calculate BMI: < 18.5 kg/m2 = “underweight”, 18.5 kg/m2–22.9 kg/m2 = “normal weight”, and ≥ 23.0 kg/m2 = “obese”.
For body image, “very thin” was categorized as “slender.” “Slightly thin,” “average,” and “slightly overweight” were categorized as “average.” “Extremely overweight” was categorized as “obese.” Based on obesity determination and body image, the cases of “underweight-slender,” “normal-average,” and “obese-obese” were categorized as body image concordance. The remainders were categorized as BID. The BID group was sub-divided based on the obesity determination and body image: “normal-slender,” “obesity-slender,” and “obesity-average” were categorized as underestimation; “underweight-average,” “underweight-obese,” and “normal-obese” were categorized as overestimation.
4. Statistical analysis
Data were analyzed considering the complex sampling design. To estimate the population, individual weight was applied. The general characteristics and health behavior-related characteristics were analyzed using frequency analysis. BID discrepancies based on participants’ general characteristics were analyzed using a chi-square test. To identify factors affecting BID, multivariable logistic regression analysis was performed. Standard error (SE), odds ratio (OR), 95% confidence interval (CI), and relative frequency (%) were all used to examine the data. Significance level for all tests was set at α = 0.05
5. Ethical considerations
The Institutional Review Board (IRB) of the Korean Centers for Disease Control and Prevention approved the Korean Community Health Survey (2014-08EXP-09-4C-A). The data were disclosed to the public except personal information of participants. This study was also approved by a university IRB (HYI-17-026-1).
Results
1. The general and health behavior-related characteristics according to the type of BID
Table 1 shows the characteristics of the study population. Of the population, 56.9% had body image concordance, 43.1% had discordance (36.0% for underestimation, 7.1% for overestimation). For underestimators, 52.7% were men, 41.7% were in their 30s, and 85.0% were obese. For overestimators, 1.1% were men, 5.2% were in their 30s, and 81.1% were underweight.
Table 1.
Participants’ general and health behavior-related characteristics per body image discordance group (N = 59,361).
| Characteristics | Total | Discordance between body mass index and body image | |||||||
|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||
| Concordance | Discordance | ||||||||
|
| |||||||||
| Under-estimation | Over-estimation | ||||||||
|
| |||||||||
| N | % (SE) | n | % (SE) | n | % (SE) | n | % (SE) | ||
|
| |||||||||
| 59,361 | - | 33,864 | 56.9 (0.2) | 21,262 | 36.0 (0.2) | 4,235 | 7.1 (0.1) | ||
| Gender | Men | 28,100 | 52.1 (0.2) | 12,559 | 46.2 (0.3) | 15,243 | 52.7 (0.3) | 298 | 1.1 (0.1) |
| Women | 31,261 | 47.9 (0.2) | 21,305 | 68.4 (0.3) | 6,019 | 17.8 (0.2) | 3,937 | 13.8 (0.2) | |
|
| |||||||||
| Age (y) | 19–29 | 25,313 | 48.0 (0.2) | 15,554 | 61.0 (0.3) | 7,317 | 29.8 (0.3) | 2,442 | 9.2 (0.2) |
| 30–39 | 34,048 | 52.0 (0.2) | 18,310 | 53.0 (0.3) | 13,945 | 41.7 (0.3) | 1,793 | 5.2 (0.1) | |
|
| |||||||||
| Education level | ≤ High school | 14,696 | 21.6 (0.2) | 8,012 | 54.1 (0.5) | 5,789 | 39.6 (0.5) | 895 | 6.3 (0.2) |
| ≥ College | 44,665 | 78.4 (0.2) | 25,852 | 57.6 (0.3) | 15,473 | 35.0 (0.3) | 3,340 | 7.4 (0.1) | |
|
| |||||||||
| Marital status | Married | 29,401 | 44.4 (0.2) | 16,002 | 53.5 (0.3) | 11,727 | 40.7 (0.3) | 1,672 | 5.7 (0.2) |
| Single | 29,960 | 55.6 (0.2) | 17,862 | 59.5 (0.3) | 9,535 | 32.2 (0.3) | 2,563 | 8.3 (0.2) | |
|
| |||||||||
| Employment status | Employed | 38,760 | 64.6 (0.2) | 20,793 | 53.6 (0.3) | 15,541 | 40.1 (0.3) | 2,426 | 6.3 (0.1) |
| Unemployed | 20,601 | 35.4 (0.2) | 13,071 | 62.8 (0.4) | 5,721 | 28.4 (0.4) | 1,809 | 8.8 (0.2) | |
|
| |||||||||
| Smoking status | Currently smoking | 14,380 | 25.3 (0.2) | 6,469 | 45.8 (0.5) | 7,616 | 52.1 (0.5) | 295 | 2.1 (0.1) |
| Ex-smoker | 5,091 | 9.2 (0.2) | 2,237 | 44.7 (0.8) | 2,740 | 53.0 (0.8) | 114 | 2.3 (0.2) | |
| Non-smoker | 39,890 | 65.5 (0.2) | 25,158 | 62.8 (0.3) | 10,906 | 27.3 (0.3) | 3,826 | 9.8 (0.2) | |
|
| |||||||||
| Alcohol consumption | Currently drinking | 50,818 | 87.3 (0.2) | 28,747 | 56.5 (0.2) | 18,683 | 36.8 (0.2) | 3,388 | 6.7 (0.1) |
| Ex-drinker | 4,824 | 7.5 (0.2) | 2,872 | 58.4 (0.8) | 1,484 | 31.3 (0.8) | 468 | 10.3 (0.5) | |
| Non-drinker | 3,719 | 5.2 (0.2) | 2,245 | 60.6 (1.0) | 1,095 | 29.1 (0.9) | 379 | 10.3 (0.6) | |
|
| |||||||||
| Breakfast consumption (d/wk) | ≤ 4 | 29,789 | 52.1 (0.2) | 17,010 | 56.8 (0.3) | 10,605 | 36.0 (0.3) | 2,174 | 7.2 (0.2) |
| ≥ 5 | 29,572 | 47.9 (0.2) | 16,854 | 56.9 (0.3) | 10,657 | 36.0 (0.3) | 2,061 | 7.1 (0.2) | |
|
| |||||||||
| Sleep duration (h/d) | ≤ 5 | 6,719 | 12.2 (0.2) | 3,701 | 55.0 (0.7) | 2,586 | 38.7 (0.7) | 432 | 6.3 (0.3) |
| 6–8 | 49,721 | 83.2 (0.2) | 28,311 | 56.7 (0.2) | 17,880 | 36.1 (0.2) | 3,530 | 7.2 (0.1) | |
| ≥ 9 | 2,921 | 4.6 (0.2) | 1,852 | 64.1 (1.0) | 796 | 26.5 (0.9) | 273 | 9.4 (0.6) | |
|
| |||||||||
| Perception of stress | Yes | 18,783 | 32.4 (0.2) | 10,628 | 56.3 (0.4) | 6,820 | 36.6 (0.4) | 1,335 | 7.1 (0.2) |
| No | 40,578 | 67.6 (0.2) | 23,236 | 57.1 (0.3) | 14,442 | 35.7 (0.3) | 2,900 | 7.2 (0.1) | |
|
| |||||||||
| Perception of depression | Yes | 3,540 | 6.1 (0.2) | 2,164 | 59.7 (0.9) | 1,054 | 30.5 (0.9) | 322 | 9.7 (0.6) |
| No | 55,821 | 93.9 (0.2) | 31,700 | 56.7 (0.2) | 20,208 | 36.3 (0.2) | 3,913 | 7.0 (0.1) | |
|
| |||||||||
| Perceived health status | Good | 32,060 | 54.4 (0.2) | 18,317 | 57.1 (0.3) | 11,584 | 36.2 (0.3) | 2,159 | 6.7 (0.2) |
| Average | 24,233 | 40.6 (0.2) | 13,723 | 56.1 (0.4) | 8,683 | 36.2 (0.3) | 1,827 | 7.7 (0.2) | |
| Poor | 3,068 | 5.0 (0.2) | 1,824 | 60.6 (1.0) | 995 | 31.2 (0.9) | 249 | 8.2 (0.6) | |
|
| |||||||||
| Physical activity | Yes | 23,803 | 43.8 (0.2) | 13,576 | 56.9 (0.4) | 8,544 | 36.1 (0.4) | 1,683 | 7.0 (0.2) |
| No | 35,558 | 56.2 (0.2) | 20,288 | 56.8 (0.3) | 12,718 | 35.9 (0.3) | 2,552 | 7.3 (0.2) | |
|
| |||||||||
| Weight control behavior | Yes | 42,077 | 72.0 (0.2) | 23,553 | 55.5 (0.3) | 16,140 | 38.7 (0.3) | 2,384 | 5.8 (0.1) |
| No | 17,284 | 28.0 (0.2) | 10,311 | 60.3 (0.4) | 5,122 | 29.0 (0.4) | 1,851 | 10.7 (0.3) | |
|
| |||||||||
| Body mass index (kg/m2) | Underweight | 5,046 | 8.6 (0.1) | 937 | 18.9 (0.6) | 0 | 0 | 4,109 | 81.1 (0.6) |
| Normal weight | 30,031 | 50.7 (0.2) | 29,175 | 96.9 (0.1) | 730 | 2.7 (0.1) | 126 | 0.4 (0.0) | |
| Obese | 24,284 | 40.7 (0.2) | 3,752 | 15.0 (0.3) | 20,532 | 85.0 (0.3) | 0 | 0 | |
|
| |||||||||
| Body image | Slender | 1,680 | 3.0 (0.1) | 937 | 54.2 (1.4) | 743 | 45.8 (1.4) | 0 | 0 |
| Average | 53,798 | 90.7 (0.1) | 29,175 | 54.2 (0.2) | 20,519 | 38.2 (0.2) | 4,104 | 7.7 (0.1) | |
| Obese | 3,883 | 6.3 (0.1) | 3,752 | 96.8 (0.3) | 0 | 0 | 131 | 3.2 (0.3) | |
All percentages were weighted to represent the total population of the 2014 Community Health Survey.
SE = standard error.
2. BID based on general and health behavior-related characteristics
For the underestimators, there were statistically significant differences depending on the following variables: gender (x2 = 6,133.89, p < 0.001), age (x2 = 725.52, p < 0.001), education level (x2 = 80.45, p < 0.001), marital status (x2 = 381.21, p < 0.001), employment status (x2 = 720.40, p < 0.001), smoking status (x2 = 2,894.53, p < 0.001), alcohol consumption (x2 = 80.81, p < 0.001), sleep duration (x2 = 113.91, p < 0.001), perceived depression (x2 = 35.82, p < 0.001), perceived health condition (x2 = 29.05, p < 0.001), and weight control behavior (x2 = 339.03, p < 0.001).
For overestimators, there were statistically significant differences depending on the following variables: gender (x2 = 1,898.81, p < 0.001), age (x2 = 164.87, p < 0.001), education level (x2 = 5.04 p = 0.040), marital status (x2 = 61.21, p < 0.001), employment status (x2 = 27.31, p < 0.001), smoking status (x2 = 596.29, p < 0.001), alcohol consumption (x2 = 78.09, p < 0.001), sleep duration (x2 = 9.30, p = 0.025), perceived depression (x2 = 21.71, p < 0.001), perceived health condition (x2 = 22.29, p < 0.001), and weight control behavior [x2 = 267.66, p < 0.001 (Table 2)].
Table 2.
The associations between concordance and discordance of body image and general and health behavior-related characteristics (N = 59,361).
| Characteristics | Concordance (n = 33,864) | Under-estimation (n = 21,262) | χ2 | p | Concordance (n = 33,864) | Over-estimation (n = 4,235) | χ2 | p |
|---|---|---|---|---|---|---|---|---|
|
|
|
|||||||
| %(SE) | %(SE) | %(SE) | %(SE) | |||||
| Gender | ||||||||
| Men | 45.2(0.3) | 54.8(0.3) | 6,133.89 | < 0.001 | 97.7(0.1) | 2.3(0.1) | 1,898.81 | < 0.001 |
| Women | 78.0(0.3) | 22.0(0.3) | 83.2(0.3) | 16.8(0.3) | ||||
|
| ||||||||
| Age (y) | ||||||||
| 19–29 | 67.2(0.3) | 32.8(0.3) | 725.52 | < 0.001 | 86.8(0.3) | 13.2(0.3) | 164.87 | < 0.001 |
| 30–39 | 56.0(0.3) | 44.0(0.3) | 91.0(0.2) | 9.0(0.2) | ||||
|
| ||||||||
| Educational level | ||||||||
| ≤ High school | 57.7(0.5) | 42.3(0.5) | 80.45 | < 0.001 | 89.5(0.4) | 10.5(0.4) | 5.04 | 0.040 |
| ≥ College | 62.2(0.3) | 37.8(0.3) | 88.6(0.2) | 11.4(0.2) | ||||
|
| ||||||||
| Marital status | ||||||||
| Married | 56.8(0.3) | 43.2(0.3) | 381.21 | < 0.001 | 90.3(0.3) | 9.7(0.3) | 61.21 | < 0.001 |
| Single | 64.9(0.3) | 35.1(0.3) | 87.8(0.3) | 12.2(0.3) | ||||
|
| ||||||||
| Employment status | ||||||||
| Employed | 57.2(0.3) | 42.8(0.3) | 720.40 | < 0.001 | 89.5(0.2) | 10.5(0.2) | 27.31 | < 0.001 |
| Unemployed | 68.9(0.4) | 31.1(0.4) | 87.8(0.3) | 12.2(0.3) | ||||
|
| ||||||||
| Smoking status | ||||||||
| Currently smoking | 46.7(0.5) | 53.3(0.5) | 2,894.53 | < 0.001 | 95.6(0.3) | 4.4(0.3) | 596.29 | < 0.001 |
| Ex-smoker | 45.7(0.8) | 54.3(0.8) | 95.2(0.5) | 4.8(0.5) | ||||
| Non-smoker | 69.7(0.3) | 30.3(0.3) | 86.5(0.2) | 13.5(0.2) | ||||
|
| ||||||||
| Alcohol consumption | ||||||||
| Currently drinking | 60.6(0.3) | 39.4(0.3) | 80.81 | < 0.001 | 89.4(0.2) | 10.6(0.2) | 78.09 | < 0.001 |
| Ex-drinker | 65.1(0.8) | 34.9(0.8) | 85.0(0.7) | 15.0(0.7) | ||||
| Non-drinker | 67.5(1.0) | 32.5(1.0) | 85.5(0.9) | 14.5(0.9) | ||||
|
| ||||||||
| Breakfast consumption (d/wk) | ||||||||
| ≤4 | 61.2(0.3) | 38.8(0.3) | 0.06 | 0.825 | 88.7(0.3) | 11.3(0.3) | 0.39 | 0.581 |
| ≥5 | 61.3(0.3) | 38.7(0.3) | 88.9(0.3) | 11.1(0.3) | ||||
|
| ||||||||
| Sleep duration (h/d) | ||||||||
| ≤5 | 58.7(0.7) | 41.3(0.7) | 113.91 | < 0.001 | 89.8(0.5) | 10.2(0.5) | 9.30 | 0.025 |
| 6–8 | 61.1(0.3) | 38.9(0.3) | 88.8(0.2) | 11.2(0.2) | ||||
| ≥9 | 70.8(1.0) | 29.2(1.0) | 87.2(0.8) | 12.8(0.8) | ||||
|
| ||||||||
| Perception of stress | ||||||||
| Yes | 60.6(0.4) | 39.4(0.4) | 4.63 | 0.061 | 88.9(0.3) | 11.1(0.3) | 0.05 | 0.845 |
| No | 61.6(0.3) | 38.4(0.3) | 88.8(0.2) | 11.2(0.2) | ||||
|
| ||||||||
| Perception of depression | ||||||||
| Yes | 66.2(0.9) | 33.8(0.9) | 35.82 | < 0.001 | 86.0(0.8) | 14.0(0.8) | 21.71 | < 0.001 |
| No | 60.9(0.2) | 39.1(0.2) | 89.0(0.2) | 11.0(0.2) | ||||
|
| ||||||||
| Perceived health status | ||||||||
| Good | 61.2(0.3) | 38.8(0.3) | 29.05 | < 0.001 | 89.5(0.2) | 10.5(0.2) | 22.29 | < 0.001 |
| Average | 60.8(0.4) | 39.2(0.4) | 88.0(0.3) | 12.0(0.3) | ||||
| Poor | 66.1(1.0) | 33.9(1.0) | 88.1(0.8) | 11.9(0.8) | ||||
|
| ||||||||
| Physical activity | ||||||||
| Yes | 61.2(0.4) | 38.8(0.4) | 0.03 | 0.878 | 89.1(0.3) | 10.9(0.3) | 1.98 | 0.207 |
| No | 61.3(0.3) | 38.7(0.3) | 88.6(0.2) | 11.4(0.2) | ||||
|
| ||||||||
| Weight control behavior | ||||||||
| Yes | 58.9(0.3) | 41.1(0.3) | 339.03 | < 0.001 | 90.6(0.2) | 9.4(0.2) | 267.66 | < 0.001 |
| No | 67.5(0.4) | 32.5(0.4) | 84.9(0.4) | 15.1(0.4) | ||||
SE =standard error.
3. Factors affecting BID
To identify the factors affecting BID, multivariable logistic regression analysis was conducted after univariable analysis for calibrating the variables. The adjusted ORs for BID relative to concordance are shown in Table 3 and 4. For underestimators, participants who were men (OR 5.26, 95% CI 4.98–5.57), had graduated from high school (OR 1.21, 95% CI 1.15–1.28), were married (OR 1.59, 95% CI 1.50–1.68), were current (OR 1.13, 95% CI 1.07–1.20) or ex-smokers (OR 1.15, 95% CI 1.06–1.24), were ex-drinkers (OR 1.25, 95% CI 1.10–1.42), had ≤ 5 (OR 1.15, 95% CI 1.01–1.30) or 6–8 hours of sleep (OR 1.13, 95% CI 1.01–1.26), had good (OR 1.25, 95% CI 1.12–1.38) or fair self-rated health (OR 1.32, 95% CI 1.18–1.46), and had healthy weight control behavior (OR 2.05, 95% CI 1.95–2.16) relative to their reference groups, had significantly higher ORs. Participants who were aged 19–29 years (OR 0.74, 95% CI 0.70–0.78) and were physically active (OR 0.96, 95% CI 0.92–0.99) relative to their reference groups had significantly lower ORs (Table 3). For overestimators, participants who were aged 19–29 years (OR 1.48, 95% CI 1.33–1.63) and had depression (OR 1.21, 95% CI 1.05–1.39) relative to their reference groups had significantly higher ORs. The participants who were men (OR 0.09, 95% CI 0.07–0.10), were married (OR 0.67, 95% CI 0.60–0.74), and had healthy weight control behavior (OR 0.41, 95% CI 0.38–0.44) relative to their reference groups, had significantly lower ORs (Table 4).
Table 3.
Multivariable logistic regression analysis of factors for underestimation (N = 55,126).
| Characteristics | Unadjusted | Adjusted† |
|---|---|---|
|
| ||
| OR (95% CI) | OR (95% CI) | |
| Gender | ||
| Men | 4.40 (4.21–4.59)* | 5.26 (4.98–5.57)* |
| Women | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Age (y) | ||
| 19–29 | 0.62 (0.60–0.65)* | 0.74 (0.70–0.78)* |
| 30–39 | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Education level | ||
| ≤ High school | 1.21 (1.15–1.26)* | 1.21 (1.15–1.28)* |
| ≥ College | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Marital status | ||
| Married | 1.41 (1.36–1.46)* | 1.59 (1.50–1.68)* |
| Single | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Employment status | ||
| Employed | 1.66 (1.59–1.73)* | 1.00 (0.95–1.06) |
| Unemployed | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Smoking status | ||
| Currently smoking | 2.62 (2.50–2.74)* | 1.13 (1.07–1.20)* |
| Ex-smoker | 2.73 (2.55–2.92)* | 1.15 (1.06–1.24)* |
| Non-smoker | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Alcohol consumption | ||
| Currently drinking | 1.36 (1.24–1.48)* | 0.97 (0.88–1.08) |
| Ex-drinker | 1.12 (1.00–1.25) | 1.25 (1.10–1.42)* |
| Non-drinker | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Breakfast consumption (d/w) | ||
| ≤ 4 | 1.00 (1.00–1.04) | 0.96 (0.92–1.01) |
| ≥ 5 | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Sleep duration (h/d) | ||
| ≤ 5 | 1.71 (1.53–1.90)* | 1.15 (1.01–1.30)* |
| 6–8 | 1.54 (1.40–1.70)* | 1.13 (1.01–1.26)* |
| ≥ 9 | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Perception of stress | ||
| Yes | 1.04 (1.00–1.09) | 1.02 (0.97–1.07) |
| No | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Perception of depression | ||
| Yes | 0.80 (0.73–0.87)* | 0.94 (0.85–1.03) |
| No | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Perceived health status | ||
| Good | 1.24 (1.13–1.35)* | 1.25 (1.12–1.38)* |
| Average | 1.26 (1.15–1.38)* | 1.32 (1.18–1.46)* |
| Poor | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Physical activity | ||
| Yes | 1.00 (0.96–1.04) | 0.96 (0.92–0.99)* |
| No | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Weight control behavior | ||
| Yes | 1.45 (1.39–1.52)* | 2.05 (1.95–2.16)* |
| No | 1.00 (reference) | 1.00 (reference) |
p<0.05.
Adjusted for gender, age, education level, marital status, employment status, smoking status, alcohol consumption, sleep duration, depression, perceived health status, and weight control behavior.
CI = confidence interval; OR = odds ratio.
Table 4.
Multivariable logistic regression analysis of factors for overestimation (N = 38,099).
| Characteristics | Unadjusted | Adjusted† |
|---|---|---|
|
| ||
| OR (95% CI) | OR (95% CI) | |
| Gender | ||
| Men | 0.12 (0.10–0.13)* | 0.09 (0.07–0.10)* |
| Women | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Age (y) | ||
| 19–29 | 1.53 (1.42–1.65)* | 1.48 (1.33–1.63)* |
| 30–39 | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Education level | ||
| ≤ High school | 0.91 (0.83–0.99)* | 1.00 (0.91–1.11) |
| ≥ College | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Marital status | ||
| Married | 0.77 (0.71–0.83)* | 0.67 (0.60–0.74)* |
| Single | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Employment status | ||
| Employed | 0.84 (0.78–0.91)* | 1.03 (0.96–1.12) |
| Unemployed | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Smoking status | ||
| Currently smoking | 0.29 (0.26–0.33)* | 1.15 (0.97–1.36) |
| Ex-smoker | 0.32 (0.26–0.41)* | 0.83 (0.66–1.05) |
| Non-smoker | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Alcohol consumption | ||
| Currently drinking | 0.70 (0.61–0.81)* | 0.88 (0.75–1.02) |
| Ex-drinker | 1.04 (0.87–1.24) | 1.04 (0.86–1.25) |
| Non-drinker | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Breakfast consumption (d/wk) | ||
| ≤ 4 | 1.02 (0.95–1.10) | 1.03 (0.96–1.11) |
| ≥ 5 | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Sleep duration (h/d) | ||
| ≤ 5 | 0.78 (0.65–0.93)* | 0.97 (0.80–1.17) |
| 6–8 | 0.86 (0.74–0.99)* | 1.02 (0.87–1.21) |
| ≥ 9 | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Perception of stress | ||
| Yes | 0.99 (0.92–1.07) | 0.93 (0.84–1.01) |
| No | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Perception of depression | ||
| Yes | 1.32 (1.16–1.51)* | 1.21 (1.05–1.39)* |
| No | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Perceived health status | ||
| Good | 0.87 (0.74–1.01) | 0.91 (0.77–1.07) |
| Average | 1.01 (0.87–1.18) | 0.99 (0.84–1.17) |
| Poor | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Physical activity | ||
| Yes | 0.96 (0.89–1.03) | 1.02 (0.95–1.10) |
| No | 1.00 (reference) | 1.00 (reference) |
|
| ||
| Weight control behavior | ||
| Yes | 0.58 (0.54–0.63)* | 0.41 (0.38–0.44)* |
| No | 1.00 (reference) | 1.00 (reference) |
p< 0.05.
Adjusted for gender, age, education level, marital status, employment status, smoking status, alcohol consumption, sleep duration, depression, and weight control behavior.
CI =confidence interval; OR =odds ratio.
Discussion
This study showed that the rate of concordance between BMI and body image was 56.9%. Many previous studies on body image assessed the degree of concordance based on BMI guidelines published by the WHO. Therefore, it is difficult to directly compare them to this study’s results. Although our results showed a lower percentage than a study reporting 74.9% body image concordance among adults aged 19–64 years [11], this study had similar results to a study that showed 56.4% body image concordance among women aged 20–30 years [14]. Since this study used the BMI criterion for Asians, these results may more accurately reflect body image concordance and provide a clearer understanding of the factors influencing BID.
Regarding both BID types, the results support a previous finding showing that men tend to underestimate and women tend to overestimate their body shapes [10]. This can be due to differences in the “ideal” body shapes that Korean society has for men and women. While women view a slim, frail, and “Westernized” body shape as ideal, men view a well-built body as ideal. Perhaps it is a social phenomenon that women wish to be slimmer and men wish to be more muscular. Moreover, the greater the BID, the more inappropriate weight control methods women try: disproportionate diets, such as fasting and meal replacements, or drugs, such as antiemetics and diuretics [19]. In contrast, men have a higher tendency than women to underestimate their weight, which often leads to the negligence of weight control behavior [10].
Gender, age, marital status, and weight control behavior all significantly affected both BID types. Participants in their 20s underestimated their body shape less and overestimated more than those in their 30s. This finding matches the claim that younger adults in their 20s and 30s tend to overestimate their body shapes [14,20]; it also supports the assertion that teenagers have more distorted body images than young adults [21]. Moreover, it is consistent with the results that 45% of female and 25% of male college students overestimate their body shapes [22].
Generally, men tend to underestimate their body shapes; however, men aged in their 20s, unlike other male adults, typically control their weight to maintain a good-looking body shape [23]. This appears to reflect the phenomenon that both women and men try to maintain a socially desirable body shape. Therefore, it is necessary to implement an intervention program for young adults in their 20s to develop healthy body images.
Married participants underestimated their body shapes more than single participants, which was consistent with a previous study [14]. Perhaps ideal body images displayed in mass media and advertising target young adults. In comparison, people who are married and have children are less pressured to maintain socially desirable body shapes. Additionally, married people may prefer to maintain an ideal weight for medical and health-related reasons as opposed to social reasons [24,25].
While those who utilize weight control behavior were found to underestimate their body shapes more than their non-control counterparts, it appears that various weight control behaviors improve individuals’ self-respect and self-image [26]. For this reason, even obese participants have a high likelihood of underestimating their body weight and possibly consider it ideal.
Furthermore, alcohol consumption, smoking, sleep duration, perceived health condition, and physical activity, significantly affected weight underestimation. Participants with a drinking history underestimated their body image, perhaps due to weight loss from alcohol withdrawal. In contrast, participants who once smoked or were currently smoking had a higher likelihood of underestimating their body image compared to non-smokers. There is certainly a common notion that smoking is effective for weight loss; however, it is necessary to investigate the correlation between alcohol consumption, smoking, and weight discordance in a follow-up study.
Participants who perceived their health condition as “average” or “good” were more likely to underestimate their body image compared to the “poor” health condition group. Since this could be due to a bias that society views being underweight as healthy, it is necessary to promote the understanding of what a “normal” healthy body shape looks like and to establish a proper understanding of how an individual’s body shape changes with the state of health.
Physically active participants were more likely to underestimate body image compared to physically inactive participants. This is consistent with research showing that people who walk and exercise moderately or regularly have a lower OR of underestimation [11,27]. Physical activity is encouraged as a measure to improve body image as they aim to improve health and fitness, which contributes to becoming slim [11].
The participants in this study were not asked to explain the purpose of their physical activities (e.g., physical fitness or weight control). Most participants who underestimated their body shapes in this study were obese (85.0%). If the purpose of their physical activity was to lose weight, participants who underestimated their BMI could have believed that they had normal BMI, thus making it unnecessary to lose weight. Due to difficulty in clarifying a causal relationship, a study determining the relationship between physical activity motivation and body image is required.
Perceived depression also significantly affected the overestimation of body image. These results are consistent with the finding that physical dissatisfaction and depression are closely related [4,28], and that self-perceived obesity amplifies depression [29]. In particular, depression can lead to problems, such as eating disorders, which can lower self-esteem and further increase depression [30]. Therefore, for those who overestimate their body shapes, it is necessary to check for depression and implement an intervention to reduce depression; in doing so, it is possible to promote an appropriate body image.
Underestimation of body image is highly likely to lead to being overweight or obese. In contrast, overestimation can promote unhealthy weight control behaviors, decrease self-esteem, and increase depression or eating disorders [11]. Self-perception of body shape requires not only individual effort, but also a simultaneous social and cultural approach [30]. Public health education on weight control and obesity prevention should be implemented among adults. Furthermore, accurate information about normal weight and body shape should be portrayed in mass media to improve an individual’s self-perceived body image.
Appropriate body image during adulthood is closely related to health management behaviors, such as physical activity and eating habits, and psychosocial wellbeing. It forms the basis of healthy living during later adulthood. It has been suggested that development and implementation of a body image improvement program would help to improve BID, taking into account factors such as gender and age. Moreover, a follow-up study should be conducted to survey the specific factors pertaining to BID, such as health management behaviors, eating habits, and socio-psychological factors, and to examine related factors influencing BID among adults aged over40 years.
This study had a key limitation. Since a secondary data analysis was conducted, psychosocial factors of body image, such as self-esteem and self-efficacy, were not addressed. In a follow-up study, these should be included, along with further investigation into the differences in BID in male and female adults to enable the development of a tailored intervention program. Despite this limitation, this study is significant in that it used raw data from a nationally representative, standard survey and identified the important factors affecting young adults’ BID.
Footnotes
Conflicts of Interest
All authors have no conflicts of interest to declare.
References
- 1.Ladhani M, Craig JC, Irving M, et al. Obesity and the risk of cardiovascular and all-cause mortality in chronic kidney disease: a systematic review and meta-analysis. Nephrol Dial Transplant. 2016;32(3):439–49. doi: 10.1093/ndt/gfw075. [DOI] [PubMed] [Google Scholar]
- 2.Yoo S, Kim H, Cho HI. Heterogeneity in obesity status and cardiovascular risks in multiethnic Asian female immigrants in South Korea. Asia Pac J Public Health. 2015;27(2):NP448–56. doi: 10.1177/1010539512459751. [DOI] [PubMed] [Google Scholar]
- 3.Son Y-J, Kim G. The relationship between obesity, self-esteem and depressive symptoms of adult women in Korea. J Korean Med Sci. 2012;21(2):89–98. [Google Scholar]
- 4.Preiss K, Brennan L, Clarke D. A systematic review of variables associated with the relationship between obesity and depression. Obes Rev. 2013;14(11):906–18. doi: 10.1111/obr.12052. [DOI] [PubMed] [Google Scholar]
- 5.World Health Organization [Internet] Fact sheet: obesity and overweight. 2016 [cited 2018 Jan 20]. Available from: http://www.who.int/mediacentre/factsheets/fs311/en.
- 6.OECD. Health at a Glance; OECD indicators. Paris (France): OECD Publishing; 2017. Overweight and obesity among adults; pp. 80–1. [Google Scholar]
- 7.World Health Organization [Internet] The Asia-Pacific perspective: redefining obesity and its treatment. [cited 2017 Jan 23]. Available from: http://apps.who.int/iris/handle/10665/206936.
- 8.Ministry of Health and Welfare [Internet] Korea Health Statistics 2014: Korea National Health and Nutrition Examination Survey (KNHANES VI-2) [cited 2016 Dec 30]. Available from: https://knhanes.cdc.go.kr/knhanes/index.do.
- 9.Kodama H. Problems of underweight in young females and pregnant women in Japan. Japan Med Assoc J. 2010;53(5):285–9. [Google Scholar]
- 10.Hwang JK, Kim HK, Kong HS, et al. The self-assessment of obesity in Korean adults. J Korean Soc Study Obes. 2002;11(4):349–55. [Google Scholar]
- 11.Chun IA, Ryu SY, Park J, et al. The Associations between discordance of body image and physical activities among adults aged 19 to 64 years: based on the data from 2010 Community Health Survey. Korean J Obes. 2014;23(4):274–80. doi: 10.7570/kjo.2014.23.4.274. [DOI] [Google Scholar]
- 12.Kwak H-K, Lee MY, Kim MJ. Comparisons of body image perception, health related lifestyle and dietary behavior based on the self-rated health of university students in Seoul. Korean J Community Nutr. 2011;16(6):672–82. doi: 10.5720/kjcn.2011.16.6.672. [DOI] [Google Scholar]
- 13.Fan M, Jin Y, Khubchandani J. Overweight misperception among adolescents in the United States. J Pediatr Nurs. 2014;29(6):536–46. doi: 10.1016/j.pedn.2014.07.009. [DOI] [PubMed] [Google Scholar]
- 14.Park S, Hyun T, Lee H. Weight control behaviors, health-related quality of life and nutritional status by overestimation of body image among young Korean females: data from the fifth Korea National Health and Nutrition Examination Survey, 2010–2011. Korean J Community Nutr. 2015;20(5):362–74. doi: 10.5720/kjcn.2015.20.5.362. [DOI] [Google Scholar]
- 15.Werch CC, Bian H, Moore MJ, et al. Brief multiple behavior interventions in a college student health care clinic. J Adolesc Health. 2007;41(6):577–85. doi: 10.1016/j.jadohealth.2007.06.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kang M, Joung H, Lim JH, et al. Secular trend in dietary patterns in a Korean adult population, using the 1998, 2001, and 2005 Korean national health and nutrition examination survey. Korean J Nutr. 2011;44(2):152–61. doi: 10.4163/kjn.2011.44.2.152. [DOI] [Google Scholar]
- 17.Park KR, Cho YC. Prevalence rates of risk factors of metabolic syndrome, and its related with obesity indices among the health checkup examinees. J Korea Acad Ind Coop Soc. 2016;17(3):153–62. [Google Scholar]
- 18.Shiwaku K, Anuurad E, Enkhmaa B, et al. Appropriate BMI for Asian populations. Lancet. 2004;363(9414):1077. doi: 10.1016/S0140-6736(04)15856-X. [DOI] [PubMed] [Google Scholar]
- 19.Neumark-Sztainer D, Paxton SJ, Hannan PJ, et al. Does body satisfaction matter? Five-year longitudinal associations between body satisfaction and health behaviors in adolescent females and males. J Adolesc Health. 2006;39(2):244–51. doi: 10.1016/j.jadohealth.2005.12.001. [DOI] [PubMed] [Google Scholar]
- 20.Boo S. Misperception of body weight and associated factors. Nurs Health Sci. 2014;16(4):468–75. doi: 10.1111/nhs.12154. [DOI] [PubMed] [Google Scholar]
- 21.Rozin P, Trachtenberg S, Cohen AB. Stability of body image and body image dissatisfaction in American college students over about the last 15 years. Appetite. 2001;37(3):245–8. doi: 10.1006/appe.2001.0426. [DOI] [PubMed] [Google Scholar]
- 22.Wardle J, Haase AM, Steptoe A. Body image and weight control in young adults: international comparisons in university students from 22 countries. Int J Obes. 2006;30(4):644–51. doi: 10.1038/sj.ijo.0803050. [DOI] [PubMed] [Google Scholar]
- 23.Hankey CR, Leslie WS, Lean MEJ. Why lose weight? Reasons for seeking weight loss by overweight but otherwise healthy men. Int J Obes. 2002;26(6):880–2. doi: 10.1038/sj.ijo.0801999. [DOI] [PubMed] [Google Scholar]
- 24.Sobal J, Rauschenbach B, Frongillo EA. Marital status changes and body weight changes: a US longitudinal analysis. Soc Sci Med. 2003;56(7):1543–55. doi: 10.1016/S0277-9536(02)00155-7. [DOI] [PubMed] [Google Scholar]
- 25.Shin NM, Park E, Choi J, et al. Self-portrait of obese and overweight Korean women based on lifetime phase. Korean J Adult Nurs. 2016;28(6):701–8. doi: 10.7475/kjan.2016.28.6.701. [DOI] [Google Scholar]
- 26.McAuley E, Blissmer B, Katula J, et al. Physical activity, self-esteem, and self-efficacy relationships in older adults: a randomized controlled trial. Ann Behav Med. 2000;22(2):131–9. doi: 10.1007/BF02895777. [DOI] [PubMed] [Google Scholar]
- 27.Kruger J, Lee CD, Ainsworth BE, et al. Body size satisfaction and physical activity levels among men and women. Obesity. 2008;16(8):1976–9. doi: 10.1038/oby.2008.311. [DOI] [PubMed] [Google Scholar]
- 28.You S, Shin K. Body dissatisfaction and mental health outcomes among Korean college students. Psychol Rep. 2016;118(3):714–24. doi: 10.1177/0033294116648634. [DOI] [PubMed] [Google Scholar]
- 29.Haff DR. Racial/ethnic differences in weight perceptions and weight control behaviors among adolescent females. Youth Soc. 2009;41(2):278–301. doi: 10.1177/0044118X08328006. [DOI] [Google Scholar]
- 30.Jeon HO. Relations of body perception, anxiety, psychological flexibility and abnormal eating attitudes of college students. J Korea Acad Ind Coop Soc. 2014;15(6):3675–84. [Google Scholar]