Comparison of Weight Control Attempts among Middle-Aged and Older Women with Obesity: Findings Based on the Seventh Korea National Health and Nutrition Examination Survey (KNHANES VII-3)

Clinical investigation

Trained interviewers and medical staff performed the interviews and health examinations. Current medical conditions were defined as those that were diagnosed and/or treated by a doctor. Hypertension (HTN) was identified through blood pressure measurement and an interview questionnaire, while diabetes mellitus (DM) and hyperlipidemia were identified through blood sample analysis in a fasting state and questionnaire. The self-administered questionnaire covered smoking status, alcohol use, activity limitations, and aerobic exercise. A current smoker was defined as one who had smoked cigarettes daily or on some days. A drinker was defined as an adult consuming at least one drink per month. Aerobic exercise was defined as performing moderate physical activity for ≥ 150 min or vigorous physical activity for ≥ 75 min per week (Piercy et al., 2018).

Anthropometric and laboratory measurements

Health examinations were conducted in a mobile examination center and included physical examinations and anthropometric measurements. Height and weight were measured to the nearest 0.1 cm and 0.1 kg, respectively. Body mass index (BMI) was calculated as weight (kg) divided by height squared (m²). Results were classified based on BMI categories appropriate for Koreans: underweight (BMI < 18.5 kg/m²), normal weight (18.5–22.9 kg/m²), overweight (23–24.9 kg/m²), and obese [Class I (25-30 kg/m²), Class II (30-35 kg/m²), and Class III (> 35 kg/m²)] (Seo, Lee, Kim, Kang, Kang, Kim, & Committee of Clinical Practice Guidelines, Korean Society for the Study of Obesity, 2019). Waist circumference (WC) was recorded midway between the lower margin of the final rib and upper margin of the iliac crest to the closest 0.1 cm. Abdominal obesity was defined as WC ≥ 85 cm or higher (Seo et al., 2019). The laboratory data quality control program monitored the laboratory performance to ensure that the data met the required standards of accuracy (Kweon et al., 2014).

Dietary intake

Trained interviewers performed computer-assisted personal interviews to assess dietary intake. A 24-hour recall dietary questionnaire based on typical Korean food items was used (Shim, Oh, & Kim, 2014). Its open-ended response structure allowed the respondents to provide a comprehensive and detailed report of all foods. Food models, pictures, and other visual aids were used to help respondents judge and describe the portion size and improve accuracy. All food products ingested by the study participants during the previous 24 hours, together with dietary information (time, location, type of food, amount, and cooking method), were investigated. The amount of total energy and nutrients consumed was calculated using a food composition table published by the Rural Development Administration of Korea, which has acceptable validity and reliability (Rural Development Administration, National Institute of Agricultural Sciences, 2018).

Questionnaire on weight control

Intentional weight control attempts were assessed by participant self-reports (or interviews for participants with low literacy) that involved the following questions and choices: “Have you tried to control your weight in the past year” ([1] to lose weight, [2] to maintain weight, [3] to gain weight, [4] never tried to control weight)?” Thereafter, participants who answered [1] or [2] to the above question were asked to choose what they did for weight control and were provided with multiple choices: (1) exercise, (2) fasting for more than 24 hours, (3) diet control, (4) skipping meals, (5) taking an unprescribed weight-loss drug, (6) taking a prescribed weight-loss drug, (7) taking oriental medicine, (8) taking health functional food, (9) one food diet, (10) etc. Body image perception and 1-year body weight changes were also assessed.

Statistical methods

Weighted complex sampling analysis was used for all data analyses. Generalized linear regression for continuous variables and the chi-square test for categorical variables were used to compare general characteristics between age groups. To identify the factors associated with weight control attempts, we performed a univariate logistic regression analysis in each age group. Based on the analysis, household income, education level, chronic diseases, and body image were significant factors. Thus, we performed a multiple logistic regression analysis to confirm whether household income, educational level, and self-perception of body image were still associated with weight control attempts after adjusting for age, BMI, WC, and chronic diseases. Additionally, to compare the odds ratio (OR) and 95% confidence interval (CI) of weight control attempts in the 60s or 70s group versus the 50s group, multiple logistic regression analysis of the complex sample was performed. Statistical analyses were performed using SPSS 23.0 (IBM Corp., Armonk, NY). In all analyses, P < 0.05 was considered statistically significant.

Demographic and clinical characteristics of the study subjects

A total of 1,571 women participated in the study (mean age, 62.1 years). Women in their 70s had a lower household income and educational level, higher rate of chronic diseases, were less likely to live with their spouse and consume alcohol and were less likely to be current smokers or be physically active than women in their 50s or 60s. In contrast, women in their 70s had significantly higher BMI and WC than those in their 50s or 60s. Regarding dietary intake, older women had lower intakes than younger women. However, the amount of carbohydrate intake did not differ across age groups (Table 1).

Table 1.

Demographic characteristics of all women aged ≥ 50 years

Obesity status of study participants

Figure 2A shows the obesity status of the participants. Approximately 57% of women were overweight or obese, and only 40.3% had normal body weight. The obesity rate was significantly higher in the 70s group than in the 60s or 50s groups. Approximately one-third of the total women were classified as having abdominal obesity (Figure 2 B). The 70s group had the highest abdominal obesity rate (46.5%). Compared with the 50s group, the 60s and 70s groups had a higher risk of abdominal obesity.

Figure 2.

Distribution of obesity status defined based on body mass index (A) and waist circumference (B) by age groups. Underweight, body mass index < 18.5 kg/m2; normal, between 18.5 and 22.9 kg/m2; and overweight, between 23 and 24.9 kg/m2. Class I obesity, between 25 and 29.9 kg/m2; Class II obesity, between 30 and 34.9 kg/m2; and Class III obesity, ≥ 35 kg/m2. Abdominal obesity, waist circumference ≥ 85 cm

Characteristics of women with obesity by age group

Among women with obesity, the 70s group had a lower household income and educational level, a higher rate of chronic diseases, and was less likely to be physically active and live with their spouse than the 50s or 60s groups. The 70s group had lower dietary intake than the 50s or 60s groups; however, the amount of carbohydrate intake did not differ across the age groups (Table 2). Women with obesity (n=533) did not have higher intake levels of calories (p = 0.251), carbohydrates (p = 0.592), proteins (p = 0.152), and fats (p = 0.081) than overall women (n=1,571).

Table 2.

Demographic characteristics of women with obesity aged ≥ 50 years

Weight control attempts in women with obesity according to the different age categories

Table 3 shows the attempts and methods used by obese women to intentionally control weight in the past year by age group. More than 14% of obese women did not consider themselves overweight or obese. More than 70% of the women answered that they had tried to reduce or maintain weight for the past year. However, the older age groups were less likely to attempt intentional weight control than the younger age groups. In particular, 41% of women with obesity in their 70s did not attempt to manage their body weight. Despite intentions for weight reduction, body weight remained unchanged during the previous year in most obese women.

Table 3.

Weight control attempts and methods used by women with obesity aged ≥ 50 years

Methods of maintaining or reducing weight in women with obesity according to age group

Among the women who tried to maintain or lose weight, reducing the amount of food was the most commonly practiced weight control strategy across all age groups, followed by exercise and skipping meals (Table 3). Dietary supplements and oriental herbal medicines were used more frequently for weight management in the 50s group than in the 60s and 70s groups. In contrast, anti-obesity medications prescribed by doctors were less frequently used for weight control than dietary supplements.

Factors associated with weight control attempts in women with obesity

Age, household income, education level, DM, hyperlipidemia, and self-perception of body image were factors significantly associated with weight control attempts in women with obesity (Table 4). However, each age group had different factors. For instance, in the 50s group, women with a lower household income, educational level, and hyperlipidemia were less likely to use weight control strategies than women with a higher household income and educational level, and without hyperlipidemia. Moreover, in the 70s group, women with obesity and a lower educational level, DM, or inappropriate self-body image were less likely to use weight control strategies than women with a higher educational level, no DM, or appropriate self-body image. In the multiple logistic regression analysis adjusted for age, BMI, WC, and chronic diseases, household income, educational level, and self-perception of body image were still significantly associated with weight control attempts in the 50s group. However, in the 70s group, only lower educational level remained a significant factor associated with weight control attempts (Table 5).

Table 4.

Factors associated with the weight control attempts of women with obesity

Table 5.

Odds ratio (95% confidence interval) for the factors associated with weight control attempts in multiple logistic regression analysis

Comparison of the prevalence of weight control attempts among age groups

The 60s and 70s groups were less likely to try weight control than the 50s group after adjusting for age, BMI, and WC (Table 6). In the multiple logistic regression model, including the presence of chronic diseases, weight control attempts were significantly fewer in the 70s group than in the 50s group. However, after additional adjustments for household income, educational level, and body image, the difference between the age groups was insignificant.

Table 6.

Comparison of the prevalence of weight control attempts among age groups

Increasing prevalence of obesity with age

The high rate of weight control attempts in our study subjects indicates national overweight and obesity trends. During the past 10 years, the obesity rate in South Korea has consistently increased, mostly in older women (obesity rate, > 40%) (Jang et al., 2019). Similarly, our study found that the obesity rate, defined based on BMI or WC, increased with age. Older individuals experience normal physiological alterations in body composition, such as increased total body fat (BF) content, decreased lean body mass (LBM), and more centrally distributed fat (Han et al., 2011).

Meanwhile, several recent studies have described an “obesity paradox,” where older adults with higher BMI are at lower risk of mortality from HTN, CVD, and Heart failure (HF) than their healthy weight counterparts (Dorner & Rieder, 2012; Flegal, Kit, Orpana, & Graubard, 2013). Multiple possible mechanisms proposed include a greater likelihood of receiving optimal medications due to earlier presentation of obesity; benefits of higher LBM, such as greater muscle mass, muscular strength, and higher metabolic reserves; and cardioprotective metabolic effects of BF (Chang, Asch, & Werner, 2010). Researchers have tried to explain this phenomenon by the lack of discriminatory power of BMI to differentiate between BF and LBM (Romero-Corral et al., 2006). However, others have demonstrated that even when using BF instead of BMI as the criterion for obesity, this association is consistently observed (Lavie, De Schutter, Patel, Artham, & Milani, 2011). Hence, fundamental questions remain (Flegal, Graubard, Williamson, & Gail, 2005; Flegal et al., 2013; Heiat, Vaccarino, & Krumholz, 2001). Nonetheless, obesity in older adults has a negative impact on functional status and quality of life (Apovian et al., 2002). Obesity is closely associated with physical weakness and slow movement (Lopez-Garcia et al., 2003). Therefore, clinical improvement with weight management in older adults should be clear, and physicians need to determine the risk–benefit profile of weight reduction before initiating obesity treatment. The importance of healthy weight should be emphasized to maintain LBM, muscular strength, and cardiorespiratory fitness.

High rate of weight control attempts that decrease with increasing age

The majority of obese women (71.7%) in our study tried to lose or maintain weight in the past year. The literature on weight control suggests that older people are encouraged to control weight due to health risks and medical conditions, while younger adults are motivated to control weight to improve their self-image (Faucher & Mobley, 2010). Despite the clinically required objectives of weight control, previous studies on Korean adults aged > 19 years have reported that older adults are less likely to modify their weight control behaviors than younger adults (Hwang, Ryu, & Park, 2015). Similarly, our study participants showed a decrease in weight control attempts with increasing age. The highest rate of weight control was observed in the 50s group, while the lowest rate was observed in the 70s group, even after adjusting for age, BMI, WC, and chronic diseases. The significant differences observed between age groups seem to be closely associated with socioeconomic status, such as household income and educational level, influencing health-related behaviors. Our findings imply that women in their 50s need more specific and varied ways to maintain a healthy body weight. Women in the 70s with obesity or overweight require motivation to lose body weight.

Different factors associated with weight control attempts between age groups

In the 50s group, body image was mostly associated with weight control attempts. It is rational that women with an accurate perception of body weight tend to perform more weight control measures than those who underperceive their weight (Hwang et al., 2015), as evidenced by the lower rate of weight control attempts in the older age group, where more women had an underperception of their body weight than those in the 50s group. By developing accurate weight perception, obese individuals can be motivated to engage in weight management programs (Kim, Kim, Cho, & Cho, 2008). Additionally, individuals with obesity who under-perceive their weight try less to lose weight and practice unhealthy methods for weight control, such as fasting, using laxatives, or performing induced vomiting (Jones, Grilo, Masheb, & White, 2010). Thus, accurate weight perception is necessary for developing healthy and effective weight control behaviors.

Household income was associated with weight control attempts in the 50s age group. This indicates that economic status affects weight control behaviors in middle-aged women, although they have greater household income than older women. Similarly, previous studies have reported that individuals with higher socioeconomic or educational levels are more likely to manage their weight (Wardle & Griffith, 2001). This is probably because they are knowledgeable regarding their health status and have ample access to weight management services, collectively facilitating the adaptation of good weight control behaviors (Morland, Wing, Diez Roux, & Poole, 2002). Considering that women in their 50s with obesity in the mid-high to highest quartile of household income have a greater likelihood of attempting weight control, social support for the maintenance of healthy weight in middle-aged women with low economic status should be addressed to reduce health inequality.

Lower educational level was associated with fewer weight control attempts in both the 50s and 70s groups. Particularly, women in their 70s with obesity who had an educational level higher than middle school were 5-times more likely to try appropriate weight control strategies than women with middle school or lower educational levels. Consistent with our observations, studies have reported that educational level in older adults is closely associated with preventive health behaviors, and women with low educational levels are more likely to have a higher prevalence of chronic diseases and lower quality of life than women with high educational levels (Korea Institute for Health and Social Affairs, 2017; Liu, Liu, Li, & Chen, 2015). In particular, a study comprising 3,080 women with obesity showed that low educational status is a contributing factor to the development of obesity and a higher risk for obesity-related diseases (Tanyolaç, Çikim, Azezli, & Orhan, 2008). The educational level is lower in women aged > 70 years than in other population groups (Korea Institute for Health and Social Affairs, 2017), causing this population to have low “health literacy.” (Davis, Michielutte, Askov, Williams, & Weiss, 1998) Hence, healthcare providers should offer sufficient and targeted education on healthy weight maintenance and develop better public health education materials and practical programs for older women. The lower rate of weight control attempts in the 70s group than that in the other age groups could be related to a lower education level; even after multiple logistic regression analysis adjusting for other factors, lower educational level remained a significant factor associated with weight control attempts in the 70s groups. Therefore, education for the age at which obesity leads to many metabolic and degenerative diseases is needed. Understanding the sources of information on health and health care among older women can play an important role in raising awareness regarding the risk of obesity and improving weight control behaviors. Besides healthcare providers, older adults often obtain health information from television programs, newspapers, and family or friends (Chaudhuri, Le, White, Thompson, & Demiris, 2013). Thus, it is necessary to proactively guide older adults to reliable sources of valid health information and help them avoid misinformation.

Personal weight control strategies

In this study, the methods used by most obese women trying to lose or maintain weight were associated with decreased energy intake and increased energy expenditure, consistent with evidence-based guidelines for obesity management (Seo et al., 2019). Reducing the size of meals was the most commonly practiced weight control strategy, and this practice was more prevalent in older individuals than in younger individuals. Women in their 60s and 70s more frequently chose to reduce their meal size than to perform exercise. This is probably because adjusting meal size may be easier for them than to perform exercise or because women aged > 60 years are likely to have chronic conditions, including osteoarthritis and activity limitation. However, there was no difference when comparing the mean nutritional intake of women with obesity and total women aged >50 years. Considering that women with obesity were consuming fewer calories than the estimated average requirement based on the dietary reference intakes for Koreans (1,800 kcal for women aged 50–64 years, 1,600 kcal for women aged 65+ years) (Ministry of Health and Welfare, 2015), reducing or skipping meals for weight control is not a strategy for older women with obesity and should be discouraged.

Many older women try to adjust their meal plan and rely on meal size. Therefore, physicians should discuss dietary habits with these women and determine whether they have any nutritional deficiencies before starting treatment. According to a report by the Nutritional Screening Initiative, 58.7% of adults aged 65 years or older are at risk of nutritional deficiency and need either dietary attention or improvement (Jang et al., 2019). Malnutrition in older adults can occur across the range of BMIs and is also observed in overweight or obese older adults (Winter, Flanagan, McNaughton, & Nowson, 2013). Thus, physicians should consider the potential risks of energy restriction when managing older obese individuals. Improving dietary quality by increasing the consumption of nutrient-dense and healthy foods or restricting the intake of salty and micronutrient-poor foods is considered appropriate. A healthy diet that includes high-fiber foods, such as fruits, vegetables, nuts, and whole grains, will help increase satiety and reduce excessive caloric intake in older individuals (Han et al., 2011).

Exercise was another commonly practiced method of weight control among the older women in our study. A geriatric fact sheet by the Korean National Statistics reported that older individuals spent over 8 hours sitting or lying down each day, and 51.9% exercised in insufficient amounts (Jang et al., 2019). In this report, the most common exercise method was walking (35.4%), followed by other aerobic exercises (34.4%), and only 8.4% of older women practiced resistance exercise regularly. Considering these reports and our findings, physicians need to emphasize the importance of exercise in controlling weight in older people, offer more age-appropriate physical activity and exercise options based on their chronic conditions or physical strength, and educate them to perform exercises properly.

In addition to diet and exercise, there are multiple pharmacological agents that can be used for weight loss. Interestingly, in our study, anti-obesity medications prescribed by physicians were less frequently used by women with obesity than dietary supplements, suggesting a low medication usage rate in South Korea, where several anti-obesity medications for long-term use are now available (Tak & Lee, 2021). This is possibly attributable to the fact that doctors were hesitant to prescribe anti-obesity medications to middle-aged and older people, or women with obesity were reluctant to see a doctor and get prescriptions to lose weight (Thomas, Mauer, Shukla, Rathi, & Aronne, 2016).

Limitations and strengths

This study had limitations. First, since this was a cross-sectional study, it was difficult to assign causality to determine causality between factors. Second, reporting bias cannot be ruled out, as data collection was performed through self-reported questionnaires addressing health-related factors and dietary intake. However, this study has several strengths. One of the major strengths was the large-scale, nationally representative sampling of Korean women aged ≥ 50 years. Moreover, we used the most reliable available data obtained using a variety of health-related questions and tests, which allowed us to control for various potential confounding factors associated with weight control.