Determining strategies to combat obesity: analysis of G7 countries with WENSLO and EDAS methods

Abstract

Proposing the optimal strategies to combat obesity for the health systems is a challenging task as there are many intertwined factors influencing it. Thus, huge responsibility is bestowed on the decision makers to identify the most suitable strategies to minimize obesity. This study is designed around the question: “What are the relative importance levels of the key factors to be considered in the fight against obesity, and how do G7 countries rank in terms of their performance in relation to these factors?” The study aims to address the main factors influencing obesity and to effectively prioritize them. In thus paper, two multi-criteria decision making (MCDM) methods, i.e. weights by envelope and slope (WENSLO) and evaluation based on distance from average solution (EDAS) are integrated in picture fuzzy environment to propose the most appropriate strategies to combat obesity in the G7 countries. The corresponding weights of the considered strategies are evaluated using WENSLO, while they are ranked based on EDAS. For selection of the experts to provide their judgments, the Hartigan-Wong algorithm is adopted. The findings from the study reveals that the most important criterion that the decision makers should consider while proposing strategies to prevent obesity is nutrition habits (w = 330), followed by physical activity and lifestyle (w = 193), and socioeconomic factors (w = 128). Based on the derived results, it is thought that educational policies regarding nutritional habits of the individuals can be an important step in preventing obesity. In addition, it is also important to implement policies to increase physical activity. Among the considered G7 countries, Japan appears as the most successful one and can be benchmarked for the remaining countries struggling with this issue.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12889-026-28625-1.

Introduction

Obesity can be defined as a health condition due to excessive fat accumulation in the body at a level negatively affecting the health [1]. The World Health Organization (WHO) evaluates obesity through calculation of body mass index (BMI). A scale is determined according to the value obtained by dividing an individual’s weight by the square of his/her height. This value is treated as normal if it is between 18.5 and 24.9; overweight between 25 and 29.9; and obese if it is 30 and above. It is possible to mention many factors responsible for obesity [2]. Preference for high-calorie and fatty foods, low daily energy burn, family history of obesity or psychological factors, such as stress and depression significantly affect obesity. However, obesity is a risk factor for many chronic diseases, leading to cardiovascular diseases, type 2 diabetes, and some types of cancer [3].

It is a well known fact that obesity negatively affects the health system as well as the individuals who are obese. Since obesity is one of the main causes of many chronic diseases, it also results in increased health expenditures [4]. Indirectly, it prevents the individuals from participating in the labor force. Since obese individuals need special care and support, home care and rehabilitative services gain importance [5]. Due to serious health problems caused by obesity, there is also an increase in the demand for efficient health services. This situation causes more consumption of the scarce resources and an increase in workload of the healthcare professionals. For the obese individuals to benefit from the health services, specialized medical devices and equipment, such as specially designed beds and stretchers are required. Therefore, an increase in the rate of obese individuals has detrimental effects on the health system [6].

Many different procedures can be followed to prevent the individuals from becoming obese. Healthily regulating eating habits is one of the important factors [7]. In addition, daily exercise programs and reducing sedentary behaviors significantly contribute to energy burning [8]. Furthermore, the socio-economic power of the individuals affects their access to obesity-preventive practices. Significant results can be achieved by improving social support for the individuals with low socio-economic status [9]. Psychosocial factors are also important triggers for obesity. Expanding stress management programs and providing importance to rehabilitative services, such as therapy can be another major step. Apart from these, strategies, like increasing walking and sports areas, and reducing air pollution stand out in the fight against obesity. It is also observed that development of technology negatively affects obesity, provoking sedentary lifestyles [10]. Besides, policies, such as promoting healthy eating and obesity awareness programs are the other actions to be deployed for improving the process [11].

Obesity is a significant public health problem resulting from numerous environmental, individual, psychological, socioeconomic, and health system-related factors. While the existing literature includes studies that examine and evaluate the effects of these factors individually, research that addresses these factors holistically and systematically assesses their importance is limited. In particular, decision-support-based studies that guide policymakers, given limited resources such as budget, manpower, and time, to prioritize intervention in specific factors are insufficient. Furthermore, the vast majority of previous research defines obesity or describes the current situation in various countries. However, studies do not address these two situations comparatively. This creates a significant research gap, both in determining the order of importance of key obesity indicators and in evaluating countries’ performance in combating obesity within the same analytical framework. The motivation for this paper is to address the complexity of factors affecting obesity and effectively prioritize them. This paper thereby aims to provide policymakers with the effectiveness of the prioritized strategies to combat obesity.

To achieve the said goal, two multi-criteria decision making tools, i.e. weights by envelope and slope (WENSLO) and evaluation based on distance from average solution (EDAS) are hybridized in picture fuzzy environment to propose the most effective strategies to prevent obesity in the G7 countries (Canada, France, Germany, Italy, Japan, the United Kingdom and the United States). WENSLO, being an objective weighting method, estimates the relative importance of the considered strategies, based on the subjective judgments of several policymakers. On the other hand, EDAS analyzes those strategies and ranks them considering their distances from the average solution. Finally, the uncertainty involved in the decision making process is effectively with the deployment of picture fuzzy numbers. As an innovative approach, the Hartigan-Wong algorithm is employed in this paper to ease out the expert selection process. People having sufficient expertise in the relevant field are chosen and clustered based on this algorithm.

Literature review

Nutrition habits

Nutrition habits that include components such as consumption of high-calorie processed foods, unhealthy food-based nutrition and frequent fast-food consumption are important triggers of obesity [12]. Consumption of such foods has usually been preferred due to their low cost and high calorie content [13]. Increased consumption of sugar-sweetened beverages and fast-food has also constituted the primary dietary risk for obesity [14]. Marlow et al. [15], in their study examining modifiable risk factors for obesity, identified a strong association between carbohydrate consumption and skipping meals, and obesity. Thus, encouraging healthy eating practices has been identified to be crucial to halting the obesity epidemic.

Physical activity and lifestyle

The risk of obesity is substantially raised by a lifestyle involving hazardous habits and insufficient physical activity. Development of technology, corporate business structure and spread of urban life have also led to sedentary life and weight gain [16]. Haas et al. [17] stated that having an active lifestyle had been associated with less obesity; therefore, doing any sport would be more important than the amount, type and intensity. Prolonged sitting and inactivity can also lead to fat accumulation around the waist and hips. At this point, lifestyle change and transition to an active life can prevent the risk of overweight that may occur in the following years [18]. Promoting physical activity would play an important role in the fight against obesity by supporting the weight control.

Socio-economic factors

Low socioeconomic status limits access to healthy foods and encourages individuals to consume high-calorie and nutrient-poor foods. The impact of socioeconomic factors on obesity is concentrated in areas such as access to healthy foods and the nutritional value consumed, especially for low-income groups. Low-income individuals are mostly orientated towards lower-cost and high-calorie foods. Overweight, especially in developing age individuals, is more common among individuals with poorer socioeconomic conditions [19] (Nogueira-de-Almeida et al., 2024). In addition, the impact of socioeconomic status on access to health services is also very important. Individuals living in areas with low socioeconomic status have difficulty in accessing regular health check-ups and diet programs, as well as shopping for healthy food. Lofton et al. [20] stated that individuals exposed to inequality cannot access adequate health services and this plays a role in increasing obesity rates. In another study.

Technology usage

The role of sedentary activities in daily life has grown along with the use of technology. The increase in screen time both reduces physical activity and accelerates weight gain. Digital devices and screen time constitute the majority of the time that children and young people are awake [21] (Zerón-Rugerio et al., 2024). Doruk et al. [22] reported that the increase in the daily consumption of technological devices such as phones, tablets and televisions has a negative effect on the body mass index values of developing individuals. In addition, the consumption of high-calorie foods such as junk food during the time spent in front of the television and computer also triggers obesity. The continuous use of these technological devices has potential negative effects such as stress and anxiety, sleep problems, asocial behaviour and eating problems [23] (Oniz et al., 2023). Poor sleep quality and stress also lead to overweight and obesity tendencies among adults [24] (Dakanalis et al., 2024).

Health systems and policies

Health systems and policies play a critical role in preventing and managing obesity. Strong health systems can reduce the risk of obesity by providing resources that support individuals to lead healthy lives [25] (Mytton et al., 2024). It has been discussed for years that obesity is not only the responsibility of the individual but also a social problem that concerns the public and institutions [26] (Mehta, 2023). At this point, health policies can take protective measures against obesity with strategies that promote healthy eating and physical activity. The meta-analysis of Pineda et al. [27] reveals that regulations such as taxation of unhealthy foods and beverages encourage healthy living habits and reduce obesity rates.

Multi criteria decision making methods

Multi-criteria decision-making methods allow for the simultaneous evaluation of multiple criteria. Many studies using this method have been conducted in the literature in recent years. It is particularly preferred in areas requiring strategic policies. In the health sector, these methods are frequently used to develop strategies for addressing not only physical problems such as obesity but also psychological issues [28, 29]. Furthermore, they are among the preferred methods for current issues such as energy use and technology evaluation. Through these methods, expert opinions are integrated with quantitative data, providing policymakers with a scientific decision-making framework for directing limited resources to more priority areas [30, 31].

The set of criteria obtained as a result of the literature review is as shown in Table 1.

Table 1.

Criteria set

Criteria	Code	Explanation	Source
Nutrition habits	NH	It is a way of life expressing how often, in what quantity and which type of food and beverages the individuals consume.	[12, 14]
Physical activity and lifestyle	PAL	It is a set of daily habits, like sleep patterns and stress management, as well as all kinds of movements that increase muscle and energy utilization of the individuals.	[16, 18, 27]
Socio-economic factors	SF	It refers to the social and economic status of the individuals, including income level, education level, occupation and living conditions.	[32, 33]
Psychosocial factors	PF	It covers several factors, such as individuals’ psychological state, stress level, social support networks and life satisfaction.	[34, 35]
Environmental factors	EF	The physical environment in which the individuals live includes elements, like traffic, air pollution and walkable areas.	[36, 37]
Medical conditions and medication use	MCMU	It covers those individuals with hormonal imbalances, metabolic diseases or medication that may cause weight gain.	[38, 39]
Technology usage	TU	It refers to the time the individuals spend in front of digital tools and screens in daily life.	[40, 41]
Health systems and policies	HSP	It includes strategies, programs and legislation that regulate the community access to services promoting healthy living.	[42, 43]

Research gap and novelty of the paper

It can be noted from the literature review that there are many factors affecting obesity, making it difficult to treat each factor individually with the available resources to combat obesity. Therefore, the most important indicators need to be identified. However, only a limited number of studies have conducted prioritized analysis, indicating a strong need for such studies. Considering these issues, this paper aims to address the main factors affecting obesity and make effective prioritization among them. The aim presented is to fill a gap in the literature.

Obesity is not merely an individual health issue. It is a global problem that triggers chronic diseases and leads to increased healthcare expenditures. The main motivation of this study is to scientifically reveal which strategies policymakers fighting obesity should focus on to achieve more accurate results. It is to analyze the performance of G7 countries, which guide global health policies, in combating obesity and to provide an optimal roadmap for other countries. This article presents several key innovations. The WENSLO and EDAS methods have been integrated with Picture Fuzzy sets and used for the first time to determine strategies for combating obesity. The Hartigan-Wong algorithm used in expert selection has added a new dimension to the fight against obesity. Finally, the G7 analysis provides an important roadmap for policymakers and developing/underdeveloped countries.

Methods

It has already been mentioned that to cope with obesity, the most important factors affecting it need to be identified. In this paper, it is aimed to prioritize those factors in terms of public health and evaluate the G7 countries. The status of those countries is evaluated based on the criteria considered, the most important criterion is determined using WENSLO method [44] and the G7 countries are finally ranked employing EDAS method [45]. Unlike classical methods such as AHP, the WENSLO approach contributes to a more precise analysis of the uncertainty in expert opinions through the envelope and slope approach. By offering decision-makers the opportunity to evaluate within more flexible ranges rather than exact numbers, it reduces the burden on expert. Unlike approaches such as Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) and VlseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR), the EDAS approach measures the acceptability of alternatives based on the average performance of all options. This approach ensures a more balanced decision-making process. The WENSLO and EDAS methods have been integrated with Picture Fuzzy numbers to facilitate decision-making under uncertainty. Picture Fuzzy numbers consider the state of neutrality in addition to membership, non-membership, and indecision in Intuitionistic Fuzzy numbers. Therefore, it can be said that more accurate results are obtained.

Why G7 countries?

Lifestyle-related health issues such as obesity yield significant findings when examined in G7 countries, which share similar levels of economic development and industrialized society characteristics. G7 countries have the capacity to shape global health policies. Identifying best practices in these countries provides a roadmap for other nations in the fight against obesity.

The WENSLO and EDAS methods were used to evaluate the differences and similarities among G7 countries. However, the evaluation of these methods requires the opinions of experts meeting specific criteria. Experts examined the G7 countries within the framework of socio-economic, cultural, and technological factors and provided their opinions. These opinions were then analyzed using the WENSLO and EDAS methods. This revealed which criteria were prioritized and which countries were taking more appropriate steps.

Expert selection

In the proposed approach, expert opinions will play a significant role, and therefore, selecting a group of individuals knowledgeable about obesity in terms of public health and obesity status in G7 countries will be a critical task. Opinions will be gathered based on linguistic expressions. Since ambiguity and vagueness of linguistic expressions are a concern in the decision-making process, a promising uncertainty model, the fuzzy set model, has also been included in this study. Certain criteria are sought when selecting the expert group needed to conduct these analyses. These criteria are as follows:

• Having expertise in obesity, public health, health policy, health management, nutrition, endocrinology, or related fields.

• Having at least 10 years of professional experience in the relevant field.

• Having worked in a managerial or decision-making position in a health institution, public institution, university, ministry, or related organization.

• Having participated in policies, programs, or projects related to the prevention, treatment, or management of obesity.

• Having conducted scientific publications, research, projects, or academic studies on the subject.

The mathematical details of all the approaches considered in this paper are provided in the subsequent sub-sections.

Data collection

Face-to-face meetings were requested with experts. Meetings were scheduled with those who agreed, and data was collected. Online meetings were scheduled with experts who could not attend the face-to-face meetings. A total of 8 experts were interviewed, and their opinions were obtained. The analysis was conducted using the opinions obtained from these 8 experts.

Hartigan-Wong algorithm

The Hartigan-Wong algorithm [46–48] is one of the K-means algorithms that assigns data in a dataset to the most appropriate cluster for a given number of clusters. Its formulation is based on the principle that the data is assigned to the clusters iteratively according to their distances to the updated cluster centers. The steps involved in this algorithm for selecting an expert team are presented as below:

At first, a dataset containing features of all the experts under consideration is framed. Assuming the number of experts as e and number of features as n, the dataset is thereby defined in the following Eq. 

1

Here, represents j. feature as age, experience for i. expert.;  e represents the total number of experts; and  n represents the number of features considered. Since the measurement unit of each feature of the experts is dissimilar to each other, the dataset is now normalized using Eq. (2).

2

where is the normalized value of element of the above dataset, is the arithmetic mean of j^th column and is the standard deviation of j^th column of the dataset. Afterwards, data are randomly selected for the number of clusters determined using Eq. (3) and their values are determined as the initial cluster centers.

3

In the next step, the Euclidean distances of each data to the centers of clusters are calculated employing Eq. (4).

4

The expression is used to measure the geometric distance of the standardized expert number  i to the cluster center number  t, symbolized by . Here,  k represents the number of clusters, and j represents the specific feature being compared. Each data is then assigned to the cluster to which it is closest, based on its distances to the cluster centers. After each data is assigned to a cluster, the cluster centers are updated. The distances of each data to the updated cluster centers are recalculated and the memberships are updated based on their proximity to the new cluster centers. This process continues until the maximum number of iterations is reached or until there is no change in the cluster membership.

After the iteration process is complete, the average of the elements of K clusters is computed. In other words, the average values of the experts’ features are calculated for each cluster. Finally, based on these averages, the clusters representing the expert teams are evaluated. The cluster with the highest averages is selected as the most successful expert team.

Picture fuzzy-WENSLO

Picture fuzzy sets allow for the simultaneous evaluation of degrees of membership (), neutral membership (), and non-membership (). This enables a more comprehensive modeling of uncertainty in decision-making processes. For two picture fuzzy numbers A and B, the addition operation is performed using the formula . The multiplication operation used to weight an opinion is defined by the formula ; >0. To ensure the consistency of picture fuzzy theory, the constraint is considered.

WENSLO is one of the recently developed objective methods for computing priority weights of criteria in a decision-making problem. It is based on calculation of the ratio between envelope of each criterion and slope of each criterion. According to WENSLO method, when the slope value of a criterion with a high envelope value is low, its priority increases. In this paper, WENSLO method is integrated with picture fuzzy numbers to take care of the subjective opinions of the experts [44]. At the beginning, the corresponding decision matrix is developed, accumulating the experts’ opinions and transforming them into corresponding picture fuzzy numbers. Subsequently, the average of each expert’s evaluation for any alternative with respect to a specific criterion is derived based on Eqs. (5)-(6).

5

6

where and v denote the degrees of positive, neutral and negative memberships, respectively. Let the number of alternatives be m and number of criteria be n, the fuzzy decision matrix can now be using Eq. (7).

7

where denotes the average value of i^th alternative against j^th criterion. Next, the elements of the fuzzy decision matrix are defuzzified applying the following Eq. 

8

where and refer to the degrees of positive, neutral and negative memberships of j^th criterion of i^th alternative in the fuzzy decision matrix, respectively. In the next step, the elements of the normalized decision matrix are calculated using Eq. (9).

9

Thereafter, the class intervals for each criterion are obtained using Eq. (10) according to Sturges’ rule.

10

The value calculated at this stage represents the class interval used for systematically grouping the data. After calculating the class interval, graduation criteria are created for each criterion. Accordingly, the criterion slopes are defined employing Eq. (11).

11

The criterion envelope value is subsequently determined using the total Euclidean distance and the criterion envelope is computed using Eq. (12).

12

The value, which forms the basis of the WENSLO method, represents the numerical slope of each obesity strategy, while the symbol represents the envelope value, which represents the distribution range of data. The ratio, which determines the final significance level, is the fundamental coefficient obtained by dividing these two parameters and determines the priority order of the strategy. In the next step, the envelope-slope ratio is estimated based on the following Eq. 

13

Finally, the priority values of the considered criteria are evaluated applying Eq. (14).

14

Picture fuzzy-EDAS

EDAS method evaluates the alternatives based on their distances from the average solution [45]. To effectively cater with the subjectivity in the decision making process, EDAS is also combined with picture fuzzy numbers. Following the same initial steps of picture fuzzy-WENSLO method, the corresponding fuzzy decision matrix is also constructed in this integrated method, as shown in Eq. (15).

15

where represents the average value of i^th alternative with respect to j^th criterion. The elements of the fuzzy decision matrix are subsequently defuzzified and the average value of all the criteria is computed using Eq. (16).

16

Depending on the type of the criterion (benefit or cost), the negative and positive distances of the corresponding criterion from the average value are estimated using Eqs. (17) and (18), respectively.

17

18

where and represent the positive and negative distances from the average value for i^th alternative against of j^h criterion in the decision matrix. Subsequently, the weighted sums of these distances are evaluated employing Eqs. (19)-(20).

19

20

where denotes the priority weight of j^h criterion. The weighted sums are then normalized using the following two equations:

21

22

Finally, the assessment scores of the considered are computed with the help of Eq. (23).

23

The score reached in the final stage represents the final assessment score of each G7 country based on all criteria in the fight against obesity. The higher this value, the more successful and strategically effective the country’s obesity policies are.

Results

Selecting the expert team using Hartigan-Wong algorithm

In this paper, eight experts from both the academic and healthcare fields having knowledge on obesity are considered, and the relevant information with respect to their age and experience are accumulated, as shown in Table 2.

Table 2.

Information about the considered experts

Expert	Age	Global experience	Teaching experience	Total experience
E1	50	20	17	21
E2	45	12	9	19
E3	43	12	8	19
E4	45	17	9	21
E5	47	15	11	24
E6	48	16	11	24
E7	46	14	10	23
E8	43	12	7	15

Global experience refers to the expert’s overall competence not only locally but also in accordance with international standards and within different healthcare system. Teaching experience, on the other hand, represents the expert’s mastery of the obesity literature not only in practice but also theoretically and methodologically. Total experience refers to the sum of the expert’s clinical practice and academic processes in the sector. Now, based on the procedural steps of Hartigan-Wong algorithm, all the elements of Table 2 are normalized. The corresponding normalized decision matrix with respect to the expert’s age and experience is provided in Table 3.

Table 3.

Normalized decision matrix

Expert	Age	Global experience	Teaching experience	Total experience
E1	1.71	1.83	2.21	-1.36
E2	-0.36	-0.96	-0.41	-0.27
E3	-1.19	-0.96	-0.74	-0.27
E4	-0.36	0.79	-0.41	0.27
E5	0.47	0.09	0.25	1.09
E6	0.88	0.44	0.25	1.09
E7	0.05	-0.26	-0.08	0.82
E8	-1.19	-0.96	-1.06	-1.36

The optimal number of expert teams based on the normalized set is calculated using Eq. Equation (3). Since, in this problem, the number of experts is eight, hence, the number of expert teams is identified as two. Based on the adopted approach, the distances of the experts to the two centers are randomly determined applying Eq. (4), leading to formation of two expert clusters, i.e. the first cluster consisting of five experts (E1, E4, E5, E6 and E7) and the second cluster containing three experts (E2, E3 and E8). The memberships of the experts are visualized in Fig. 1.

Fig. 1.

Formation of expert clusters

In the next step, the center values of the developed clusters are calculated. In other words, the average values of the considered variables for the experts in the teams are estimated, as shown in Table 4. The comparison of the averages according to the demographic information of the teams is presented in Fig. 2. After examining the information provided in Table 4; Fig. 2, it can be noticed that the first team of experts is more successful than the second one. Therefore, the opinions are sought from the five experts in the first team, and the subsequent analysis is carried out considering their judgments.

Table 4.

Average values of the cluster variables

Cluster	Age	Global experience	Teaching experience	Total experience
1	0.548	0.576	0.441	0.63
2	-0.914	-0.96	-0.736	-1.05

Fig. 2.

Mean values of the expert teams

Prioritization of criteria using PF-WENSLO

Keeping in mind the basic objective of this paper, the experts from the first team evaluate the G7 countries with respect to each criterion. The experts’ opinions are presented in the appendix, Table A1. The expert evaluations are converted into corresponding picture fuzzy numbers. Subsequently, the average of the picture fuzzy evaluations is calculated using Eqs. (5) and (6). The developed fuzzy decision is depicted in Table 5.

Table 5.

The developed fuzzy decision matrix

G7 countries	NH			PAL			SF			PF			EF			MCMU			TU			HSP
United States	0.30	0.50	0.20	0.41	0.42	0.17	0.50	0.37	0.13	0.45	0.41	0.13	0.64	0.30	0.00	0.65	0.25	0.10	0.47	0.38	0.15	0.60	0.30	0.10
United Kingdom	0.65	0.25	0.10	0.62	0.28	0.10	0.63	0.26	0.10	0.61	0.29	0.10	0.84	0.13	0.00	0.58	0.32	0.10	0.56	0.34	0.10	0.64	0.26	0.10
Germany	0.66	0.24	0.10	0.48	0.38	0.13	0.81	0.17	0.00	0.62	0.30	0.00	0.84	0.14	0.00	0.67	0.23	0.10	0.48	0.39	0.13	0.70	0.20	0.10
Italy	0.62	0.28	0.10	0.71	0.22	0	0.63	0.26	0.10	0.64	0.26	0.10	0.70	0.20	0.10	0.56	0.34	0.10	0.64	0.26	0.10	0.64	0.26	0.10
France	0.60	0.30	0.10	0.61	0.29	0.10	0.63	0.26	0.10	0.64	0.26	0.10	0.60	0.30	0.10	0.55	0.35	0.10	0.62	0.28	0.10	0.70	0.20	0.10
Japan	0.76	0.17	0.00	0.65	0.29	0.00	0.58	0.29	0.13	0.70	0.23	0.00	0.70	0.20	0.10	0.54	0.36	0.10	0.67	0.26	0.00	0.76	0.17	0.00
Canada	0.47	0.42	0.11	0.51	0.36	0.13	0.50	0.37	0.13	0.48	0.39	0.13	0.84	0.14	0.00	0.54	0.36	0.10	0.51	0.36	0.13	0.64	0.26	0.10

The elements of the fuzzy decision-making matrix are defuzzified with Eq. (8)and the decision-making matrix is constructed. Then, the values are normalized using Eq. (9). The normalized decision matrix is given in Table 6.

Table 6.

The normalized decision matrix

Normalized	NH	PAL	SF	PF	EF	MCMU	TU	HSP
United States	0.105	0.120	0.128	0.123	0.134	0.151	0.130	0.134
United Kingdom	0.151	0.149	0.145	0.143	0.156	0.142	0.142	0.139
Germany	0.153	0.130	0.169	0.150	0.154	0.154	0.130	0.146
Italy	0.148	0.165	0.145	0.148	0.138	0.140	0.153	0.139
France	0.145	0.146	0.145	0.148	0.126	0.138	0.150	0.146
Japan	0.170	0.157	0.138	0.160	0.138	0.137	0.161	0.157
Canada	0.127	0.133	0.128	0.127	0.154	0.137	0.134	0.139

In the next step, the class intervals are obtained by Eq. (10). The size of class intervals of the criteria is exhibited in Table 7.

Table 7.

The size of the class intervals

NH	PAL	SF	PF	EF	MCMU	TU	HSP
0.017	0.012	0.011	0.010	0.008	0.004	0.008	0.006

Table 8 present the class range values for each criterion with the mathematical formula specified by Eq. (10). These values are used as primary input in determining the slope value and criterion weights used, as required by the steps of the WENSLO method. In addition, graduation of criteria or positioning of alternatives is illustrated in Table 8.

Table 8.

The graduation of criteria or positioning of alternatives

	NH	PAL	SF	PF	EF	MCMU	TU	HSP
0	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
1	0.017	0.012	0.011	0.010	0.008	0.004	0.008	0.006
2	0.034	0.023	0.021	0.019	0.015	0.009	0.016	0.012
3	0.051	0.035	0.032	0.029	0.023	0.013	0.024	0.018
4	0.068	0.047	0.043	0.039	0.031	0.017	0.033	0.025
5	0.085	0.058	0.054	0.048	0.039	0.022	0.041	0.031
6	0.102	0.070	0.064	0.058	0.046	0.026	0.049	0.037

Table 9 shows the performance levels defined for each criterion. This value forms the basis of the slope calculations. Then, the slope of the criteria is calculated with the help of Eq. (11). The slope results of the criteria are displayed in Table 9.

Table 9.

The slope of criteria

	NH	PAL	SF	PF	EF	MCMU	TU	HSP
Slope Value	9.78	14.25	15.56	17.30	21.56	38.30	20.46	27.10

The envelopes of the criteria computed by Eq. (12) are presented in Table 10.

Table 10.

The envelope of criteria

	NH	PAL	SF	PF	EF	MCMU	TU	HSP
Envelope	0.178	0.152	0.110	0.104	0.095	0.050	0.102	0.068

Later, the envelope-slope ratio of the criteria is estimated by Eq. (13). The results are shared in Table 11.

Table 11.

The Envelope-Slope Ratio

NH	PAL	SF	PF	EF	MCMU	TU	HSP
0.018	0.011	0.007	0.006	0.004	0.001	0.005	0.002

Finally, after the priority values of the criteria are obtained with Eq. (14), the priorities of the criteria are summarized in Table 12.

Table 12.

The priorities of the criteria

Criteria	PV
NH	0.330
PAL	0.193
SF	0.128
PF	0.109
EF	0.080
MCMU	0.024
TU	0.090
HSP	0.045

According to the analysis results, the top priority criterion is nutrition habits. To compare the analysis results, the analyses are repeated with the Entropy method. The priority order of the criteria according to the two methods is visualized in Fig. 3.

Fig. 3.

The comparative results

Ranking of the G7 countries using PF-EDAS

The fuzzy decision matrix for ranking the countries shown in Eq. (15) is given in Table 6. After the values are defuzzified, the average values of the criteria are calculated with Eq. (16). The average values of the criteria are given in Table 13.

Table 13.

The average values

	NH	PAL	SF	PF	EF	MCMU	TU	HSP
AV	0.723	0.720	0.743	0.736	0.830	0.724	0.713	0.783

Pos and Neg values are calculated using Eqs. (17) and (18). The eight criteria are benefits. Then, weighted sums are obtained with Eqs. (19) and (20). The sums are presented in Table 14.

Table 14.

The weighted sums

United States	SPos	SNeg
	0.001	0.161
United Kingdom	0.037	0.002
Germany	0.063	0.025
Italy	0.054	0.005
France	0.021	0.010
Japan	0.111	0.008
Canada	0.007	0.082

The weighted sums are normalized by Eqs. (21) and (22). The normalized weighted sums are shared in Table 15.

Table 15.

The normalized weighted sums

United States	NrmSPos	NrmSNeg
	0.013	0.000
United Kingdom	0.334	0.988
Germany	0.566	0.846
Italy	0.485	0.972
France	0.189	0.937
Japan	1.000	0.951
Canada	0.059	0.490

Finally, AS values are obtained with the help of Eq. (23). AS values of G7 countries are given in Table 16.

Table 16.

The AS values

G7 Countries	Values
United States	0.006
United Kingdom	0.661
Germany	0.706
Italy	0.728
France	0.563
Japan	0.976
Canada	0.275

As can be seen from the results in Table 16, the most successful country is Japan. To test the consistency of the results, an analysis was conducted according to 16 different scenarios. The results of the 16 different scenario analyses are presented in Fig. 4. The results show consistency across each scenario.

Fig. 4.

Country performance rankings

As shown in Fig. 4, Japan is the most successful G7 country in combating obesity.

Discussion

The analysis revealed that the most important factor affecting the process of minimizing obesity is dietary habits. Another result obtained is that Japan is the country that carries out the most effective activities in the fight against obesity among the G7 countries. In Japan, the Metabo Law and the Shukuiku Law play a significant role in the fight against obesity. The Metabo Law requires annual waist measurements for company employees aged 45–74. Individuals exceeding 85 cm (men) and 90 cm (women) are provided with counseling services. Furthermore, the Shukuiku Law mandates a diet free of frozen or processed foods in schools. It also aims to instill portion control awareness in children from an early age. It is known that obesity generally occurs in the case of excessive weight gain as a result of excessive consumption of high-calorie, low-nutrient and high-fat foods. Considering this situation, it is seen that proper and regular eating habits are critical in the fight against obesity [42]. Minami et al. [49] conducted a study in Japan investigating the relationship between dietary habits during adolescence and gestational diabetes in later periods. In the results of the study, it is stated that avoiding snacking between meals and at night during adolescence reduces gestational diabetes. Cui et al. [50] examined the relationship between unhealthy eating habits and obesity in men aged 20–39 years in Japan. Accordingly, it is stated that an unhealthy diet triggers obesity. Azuma et al. [51] tried to determine the link between irregular breakfast eating and obesity with an observational study on 254 people in Japan. In this study, it is said that men who have irregular breakfast have a higher risk of obesity.

According to the results of the study, the second most important factor affecting the solution of obesity problem is physical activity and lifestyle. Preferring a sedentary lifestyle in daily life leads to weight gain and obesity by reducing calorie expenditure. An active lifestyle and sports habits, on the other hand, not only reduce the risk of obesity but also contribute to the treatment of many diseases such as heart health. Yamashiro et al. [52] conducted a study investigating the relationship between morbid obesity and physical activity and lifestyle habits by surveying 15.122 university students in Japan between 2011 and 2019. In the results of the study, it is stated that sleep patterns and exercise habits affect morbid obesity. Fujiwara et al. [53] conducted a study investigating the prevalence of childhood obesity in Japan. This study suggests that obesity can be minimized by promoting healthy eating and physical activity through school policies and curricula. Hamasaki [54] conducted a study examining the role of physical activity in the management of obesity by meta-analysis method. Accordingly, it is stated that structured exercise programs, especially combinations of aerobic exercise and resistance training, improve the body composition of obese individuals. Tataka et al. [55] conducted a study investigating whether there are differences between obese and non-obese individuals in terms of physical activity and diet. The study included 143 men between 65 and 75. In the study, it is said that there is no difference between the two groups in terms of physical activity and nutritional profiles.

Conclusion

Obesity is one of the most common health problems today. In the coming years, it is predicted that the problem of obesity will become more widespread. This problem affects society and states as well as the individual. Because both obesity itself and other diseases caused by obesity bring a significant burden to the health system. In addition, health problems experienced by obese individuals reduce labor productivity. Therefore, action should be taken to avoid these negative effects caused by obesity. However, there are as many factors that cause obesity as there are in minimizing it. It is not possible to intervene in all of them. For this reason, it is necessary to know the importance of the factors in order to determine which action will be taken among the relevant factors. The aim of this study is to identify the factors that are effective in minimizing obesity and to determine their importance ranking. In this framework, 8 criteria were determined with the support of the literature, the importance ranking of the determined criteria was analyzed with the Fuzzy WENSLO method and the country rankings were analyzed with the Fuzzy EDAS method.

The results of the analysis show that nutrition habits are the most important factor in minimizing obesity. Physical activity and lifestyle were found to be the second, socioeconomic factors, the third. Among the G7 countries, the best countries in the fight against obesity were found to be Japan, Italy, Germany, United Kingdom, France, Canada and United States, respectively. In this framework, focusing on the first three factors contribute to achieving significant success in the fight against obesity. However, this struggle should be supported by national policies as well as individual policies. Especially for children in pre-school and primary education age to gain healthy eating habits, adopt an active lifestyle and participate in sports activities, policies should be implemented at the national level. In this way, the obesity problem will be prevented before it occurs. In other words, it is recommended that obesity should first be combated at the point of preventive health services. In addition, actions should be taken to treat existing obese individuals and to bring these individuals into society. In addition, in order to raise public awareness, awareness should be raised by providing training for healthy eating habits and other factors in cooperation with local governments and health institutions.

Limitations and future study suggestions

Although this study provides a significant roadmap for combating obesity, it has several limitations. The analysis was specifically conducted on G7 countries, which share similar economic characteristics. Since the dynamics of developing countries may differ, this should be noted as an important limitation. Additionally, country groups could be determined based on geographic proximity in addition to economic similarity. Indeed, geography is one of the key factors influencing nutritional and cultural habits. Furthermore, the weighting analysis relies on expert judgments. Although the disadvantage of this situation was attempted to be minimized using Picture Fuzzy numbers, the analysis results are ultimately shaped by the subjective perspectives of the expert team.

Future studies are recommended to be conducted on different country groups, such as developing and underdeveloped countries. Furthermore, new analyses considering different fuzzy number sets and different MCDM approaches are needed for future research. In addition, conducting different analyses in geographically similar regions with similar dietary habits (such as Mediterranean countries) may yield different results. The opinions provided by the experts take into account the current time period and needs. It is highly probable that societal needs and dynamics will change in the future. Therefore, it is important to repeat the studies at specific time intervals.

Research implications

From a research perspective, our findings offer various important implications for the academic literature. This study enables a holistic approach to be taken to strategies for combating obesity. The literature is enriched by the set of criteria that influence the fight against obesity, which is an important contribution. The results obtained show that the theoretical focus has shifted from individual behaviour to systemic policy frameworks. Additionally, the study contributes to the development of priority strategies and policies for the effective management of scarce health resources, which is of critical importance in this area.

Practical implications

The fact that nutritional habits are the most important factor in combating obesity shows that public health budgets should primarily be directed toward programs that increase nutritional literacy. In addition, tax regulations are needed to facilitate access to healthy food. The physical activity criterion is the second most important factor affecting the struggle against obesity. Accordingly, urban planners and health managers need to work together to increase the number of walking paths. Regulations that support an active life in workplaces will provide a positive effect in reducing the burden of obesity. Finally, the nutritional policies and preventive health service models of Japan, which was found to be the most successful country in combating obesity, should be examined and efforts should be carried out for their integration. The Metabo Law and the Shukuiku Law, which played a significant role in Japan’s success, should be applied through cultural adaptation.

Authors’ contributions

Y.G. was the principal investigator of the study. Y.G. and S.A. conceptualized and conducted the study, performed data analysis, and authored the article. S.K. worked on data collection and literature review. Y.U. and S.E. contributed to all phases of the project, including study design and analysis. S.C. supervised the project and provided significant corrections and feedback to the article. All authors have read and approved the final version of the article.

Funding

Open access funding was provided by Istanbul Medipol University.

Data availability

Data can be shared upon request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the ethical principles of the Declaration of Helsinki. Since the research is based on expert evaluations within a fuzzy multi-criteria decision-making framework and does not involve clinical interventions, identifiable personal data, or human/animal tissue, formal ethics committee approval was deemed unnecessary according to the institutional guidelines of İstanbul Medipol University. Informed consent was obtained from all individual experts prior to their participation. All participants were fully informed about the study’s objective, and their involvement was entirely voluntary and anonymous.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.