Journal Pre-proof

Testing obesity Kuznets curve for Türkiye

Muhammed Benli, Yasin Acar, Semih Bas

PII: S2451-8476(24)00007-1
DOI: https://doi.org/10.1016/j.obmed.2024.100537
Reference: OBMED 100537

To appear in: Obesity Medicine

Received Date: 17 September 2023

Revised Date: 18 January 2024
Accepted Date: 17 February 2024

Please cite this article as: Benli, M., Acar, Y., Bas, S., Testing obesity Kuznets curve for Türkiye, Obesity
Medicine (2024), doi: https://doi.org/10.1016/j.obmed.2024.100537.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.

© 2024 Published by Elsevier Ltd.

Testing obesity Kuznets curve for Türkiye
Muhammed Benlia*, Yasin Acarb, Semih Basc
a Department of Economics, Bilecik Şeyh Edebali University, 11230, Gülümbe, Bilecik, Türkiye.
b Department of Public Finance, Bilecik Şeyh Edebali University, 11230, Gülümbe, Bilecik, Türkiye.
c Department of Medical Documentation and Secretariat, Osmaneli Vocational School, Bilecik Şeyh Edebali University, 11500,

Osmaneli, Bilecik, Türkiye

ABSTRACT

Aim: This paper tests for the existence of Obesity Kuznets Curve (OKC); that is, an inverse U-shape pattern between
income and obesity (as measured by Body Mass Index).
Methods: We employ linear probability model and logistic regression using cross-sectional data obtained from “2019

of
Türkiye Health Research” provided by the Turkish Statistical Institute (TUIK).
Results: Our findings reveal that OKC hypothesis holds for Türkiye such that the likelihood of obesity first increases

ro
as income rises and after reaching a maximum point, it starts to decrease. Furthermore, the study reveals that obesity
differs according to gender, so women are more likely to be obese than men. Our empirical study also reveals that
marriage, education, age, and physical activity are closely associated with the likelihood of being obese in Türkiye.

-p
Conlusions: There is a non-linear relationship between income and obesity. The study also reveals that obesity
differs according to gender and marriage, education, age, and physical activity are closely associated with the
re
likelihood of being obese in Türkiye.
lP

Keywords: Obesity, Income, OKC hypothesis, Logistics regression

1. Introduction
na

The World Health Organization (WHO) defines overweight (body mass index – 𝐵𝑀𝐼 > 25) and obesity (𝐵𝑀𝐼 >
30) as abnormal or excessive fat accumulation that may impair health and estimates that there were about 2 billion
ur

adults (39% of men and 40% of women, about 13% of the world’s adult population) with overweight worldwide; of
those, 650 million were obese (11% of men and 15% of women) in 2016. If current trends continue, it is estimated
that 1 in 7 men and 1 in 5 women will be living with obesity, equating to over 1 billion people worldwide by 2030,
Jo

according to the latest projections by the World Obesity Federation (World Obesity Atlas 2022). Overweight and
obesity, previously seen as a problem only in high-income countries, have been increasing dramatically in low- and
middle-income countries as well, particularly in urban settings.
According to the WHO, worldwide obesity has nearly tripled since 1975 and 2.8 million deaths yearly are
attributable to being overweight or obese, putting the global population at increased risk for diseases including
diabetes, cardiovascular diseases, and certain types of cancer. However, the impact of obesity is not limited just to
public health concerns and has important personal, social, and economic consequences. Obesity-associated chronic
diseases have major effects on the personal budgets of individuals and families and may impose a large economic
burden on national economies by reducing the life expectancy and the quality of human capital as it reduces the labor
force productivity, educational attainment, academic performance, workforce participation, and increasing disability
and health expenditures, and thus, negatively affect the national income of a country, create fiscal pressure, and
increase taxation (Tremmel et al., 2017; OECD, 2019).
While numerous factors contribute to an increase in average population bodyweight, including environmental
factors and genetic propensity, overweight and obesity primarily occur due to the high caloric intake from diet and
low levels of physical activity levels (Withrow and Alter, 2011; Romieu et al., 2017; OECD, 2019). Globally, as
globalization and urbanization have grown in the past few decades, diets and activity patterns have changed
significantly, and there has been a significant decrease in physical activity and increases in sedentary behavior

*
Corresponding author. Department of Economics, Bilecik Şeyh Edebali University, Gülümbe, Bilecik, Türkiye.
E-mail address: [email protected] (M. Benli), [email protected] (Y. Acar), [email protected]
(S. Bas)

1
associated with the developments in technology and shift towards office jobs and more inactive modes of transport,
but there has also been a shift in food consumption patterns towards more energy-dense foods associated with the
increasing availability of calories, particularly for certain nutrients, such as sugar. (Popkin et al., 2012; OECD, 2019;
Shekar and Popkin, 2020). Besides these primary reasons, the socioeconomic and environmental factors which are
also mostly associated with diet and activity patterns such as income and education level, sex, age, ethnicity,
socioeconomic status, living environment, and working conditions also affect overweight and obesity as the
prevalence of obesity varies across countries, sexes, socioeconomic groups, regions, and social classes within
countries (Branca et al., 2007; OECD, 2019). For example, global average data suggest that, in general, the
prevalence of obesity and overweight among women is higher than that of men. On the other hand, in most countries,
individuals with low levels of education and in the lowest income group are more likely to be overweight or obese
than individuals with the highest education and income (OECD, 2019). Therefore, it can be argued that addressing
the issue of obesity is complicated but also necessary given its significant health and economic costs, for designing
well-targeted government policies to combat this epidemic. In this context, it is of vital importance to determine the
link between obesity and its determinants. One of these determinants, which has received a great deal of attention in
the theoretical and empirical literature, but whose effect on obesity requires a thorough analysis, is income.
From the theoretical perspective, at the individual level, the principal mechanism by which income induces
obesity by affecting individuals’ energy intake and expenditure is through its effect on food preferences, dietary

of
habits, physical activity trends, lifestyle, and other behavioral and psychosocial effects. At the individual level, as
incomes rise above the subsistence level, individuals can afford and generally prefer a higher-calorie diet and leisure,

ro
which leads to higher energy intake and less energy expenditure. On the other hand, it can be argued that individuals
with low income generally live in an environment requiring less physical activity (Edwards and Tsouros, 2006;

-p
Heinrich et al., 2007; Armstrong et al., 2018) and are less likely to consume healthful foods and tend to spend a
larger proportion of their income on less healthful foods (low price foods filled with excess calories) that lower the
diet quality (French et al. 2019; Drewnowski and Specter, 2004), and thus, they are more likely to suffer from
re
nutrition-related health problems such as overweight and obesity (Ren et al., 2019). As their incomes rise, they can
access healthier foods and health services and have time for physical activities such as exercising regularly, which
lP

leads to lower energy intake and more energy expenditure (Zhang and Xiang, 2019). Therefore, the theoretical
considerations (and empirical findings) regarding the income effect on obesity and overweight do not provide a
persuasive answer as this effect may go either in a positive or negative direction. Although a large body of research
na

agrees that there is a significant relationship between income and obesity, empirical results appear to be
contradictory, and the relationship between the two remains mostly unclear, given the fact that the association
between them is more complex than it seems (Ameye and Swinnen, 2019; Mathieu‐Bolh, 2022). Besides, recent
ur

studies in the literature have shown that the effect of income on obesity is not uniform across countries and differs
according to the level of economic development and global integration (Zhou, 2021). There are a number of studies
suggesting a positive association between income and obesity in less developed countries as those with higher
Jo

income are likely to be obese in this group of countries, and a negative relationship between the two in developed
countries as obesity is highest among the least well-off groups in such societies (Popkin, 1998; Giskes et al., 2008;
Sallis & Glanz, 2009; Dinsa et al., 2012; Wells et al., 2012; Pampel et al., 2012; Malik et al., 2013; Marteleto et al.,
2017; Zhou, 2019; Ameye and Swinnen, 2019; Zhou, 2021). The impact of economic growth on obesity is mostly
explained by technological development shaping the structure of the economy, consumption patterns, changing
overall price of food consumption, and lifestyle (Philipson and Posner, 1999; Cutler et al., 2003; Lakdawalla and
Philipson, 2009; Minos et al., 2016; Ozyapi, 2019; Zhou, 2021). Furthermore, involvement in globalization may also
condition the relationship between income and obesity by affecting consumption patterns, lifestyle, and cultural
norms and perceptions. Global integration is expected to reduce obesity among more well-off than among poor ones
(Zhou, 2021).
Theoretically, the relationship between individual (or national) income and obesity prevalence could take the
form of inverted-U, which is generally called as obesity Kuznets curve (OKC) hypothesis in the literature
(Lakdawalla and Philipson 2009). According to the hypothesis, at the individual level, obesity first inreases as
incomes rise since higher incomes allow people to afford excess food and nutrients more than they need, reduce
physical activity and induce sedentary behavior, so the income-obesity curve is upward sloping. However, beyond a
certain income threshold, higher income allows individuals to shift their consumption towards healthier food, raises
awareness of health benefits from proper weight and health-related activities, and thus leads to a negative association
between income and obesity, and the income-obesity curve becomes sloping downward (Akee et al., 2013; Grecu
and Rotthoff, 2015; Windarti et al., 2019; Talukdar et al., 2020; Mathieu‐Bolh, 2022). The OKC hypothesis has been
investigated through a variety of econometric methods and data sets in the literature; nevertheless, the number of
empirical studies with a specific focus on the hypothesis, especially in the micro context, is still fairly limited.

2
Windarti et al. (2019) investigate the relationship between economic development and weight-related health statuses
using panel data on 130 countries for the period 1975-2010. Using static panel data models and two-step system
GMM estimation to account for endogeneity, the study confirms the OKC for both genders. The results from
analyses indicate that the weight-related health status deteriorates as incomes increase in low-income countries and
improve as incomes increase in high-income countries. Deuchert et al. (2014), using Demographic and Health
Surveys providing anthropometric measurements (weight and height) for women (age 15–49 years) in 52 countries
for the period 1990-2008, test the main aggregated predictions that health inequality and average body weight are
associated with economic development. The results from the regression analysis imply that the rates of obesity (and
overweight) and the average BMI increase significantly with economic development. The findings also suggest a
pattern of the OKC with a turning point at $2800. Similarly, Grecu and Rotthoff (2015) test the hypothesis of OKC
using state-level data from the USA for the period 1991-2010. Utilizing a differences-in-differences strategy, they
find evidence of an obesity Kuznets curve for white females, peaking at $15192. In another study on the US, Alola
and Bekun (2021) explore the validity of the OKC and the U-shaped relationship between ecological footprint and
obesity for the period 1975-2016. Employing an autoregressive distributed lag (ARDL) model, they find supportive
evidence of the OKC as a higher income per capita leads to an increase in obesity until a certain threshold, with an
estimated turning point of $10391. The empirical evidence also suggests that an increase in ecological footprint
decreases the prevalence of obesity until a minimum threshold; then, the prevalence of obesity becomes severe as the

of
ecological footprint continues to rise.
Relying on longitudinal data from the China Health and Nutrition Survey covering the period 1991–2011,

ro
Clément (2017) attempts to reveal the relationship between income and adult body size in urban China. Using
parametric and semi-parametric methods, the study finds that the relationship between the two is gender-specific and

-p
nonlinear. The empirical findings provided by the parametric estimate show that the relationship between income
and body shape goes from positive to negative for both genders, while the semi-parametric estimate suggests that this
is the case only for women, with a turning point just above the median income. Aydin (2019), using a bounds testing
re
approach within ARDL framework, examines the relationship between economic growth and obesity in the top 20
obese countries for the period 1991-2016. The empirical findings confirm the validity of OKC hypothesis in Türkiye,
lP

Oman, the United Arab Emirates, and Saudi Arabia. On the other hand, Talukdar et al. (2020) systematically
examine the strength and pattern of the relationship between national income and obesity prevalence as well as the
macro-environmental moderators of this relationship using an extensive dataset covering 147 countries over the
na

period 1975-2014. Relying on a Bayesian hierarchical model, they find a positive association between national
obesity prevalence and national income. However, the findings do not support the obesity Kuznets curve hypothesis
as the relationship does not turn negative at higher income levels.
ur

Among the others, geographic distribution, gender (Ogden et al., 2018; Ameye and Swinnen, 2019; Zhou, 2021),
and education level (Monteiro et al., 2001; Ogden et al., 2018; Mosli et al., 2020; Zhou, 2021) may also condition the
relationship between income and obesity. For example, Ameye and Swinnen (2019) identify important gender
Jo

differences in the relationship between the two. Their study shows that females are more obese in low- and middle-
income countries, while this gap disappears in high-income countries. They also show that the geographic
concentration of obesity also changes with income growth. The study finds that obesity is higher in urban areas than
in rural areas in low-income countries, whereas the rural-urban obesity gap is reversed in high-income countries.
Mosli et al. (2020), on the other hand, find that individuals in the highest income bracket with lower levels of
education in Saudi Arabia may be more vulnerable to obesity. Moreover, Monteiro et al. (2001) find that education
does not affect male obesity rates in lower income groups in Brazil, while the obesity rate declines for men with
higher income as the education level rises. Their findings, on the other hand, show that education has a much more
significant effect on obesity for women, regardless of which income group they belong to (Ozyapi, 2019).
In light of these arguments, this study aims to extend the income-obesity analysis in Türkiye with an emphasis on
the OKC hypothesis. Although there are are a number of studies attempting to measure OKC, to the best our
knowledge, none of these studies have focused on Türkiye, particularly in micro context. According to the data from
the Nutrition, Health and Food Consumption Research, individuals’ energy intake provided by carbohydrates and
fat-based foods in 1974 was 64% and 24%, respectively (Köksal, 1977), while, in 2017, these figures were 50.2%
and 34.5%, respectively (Ministry of Health of Türkiye, 2019). During the same period, there was no significant
change in the amount of energy that individuals received from protein-derived foods, which implies that the dietary
habits of individuals in Türkiye switched from carbohydrate-based foods to fat-based foods. The change in eating
habits is also confirmed by the study of Foodservice Monitor conducted by the Association of Out-of-Home
Consumption Suppliers and Ipsos Research Company in 2016. According to the study, due to an increase in urban
population, the rising role of women in work life, and changes in lifestyle, expenditure in restaurants and fast food
places constituted about half of the out-of-home consumption expenditures of individuals in Türkiye. The same

3
period can also be characterized by increasing rates of overweight and obesity. According to the data from Türkiye
Health Survey (2019), the overweight (pre-obesity) and obesity rates showed a significant rise in the last decade
(17.9 percent), reaching 56.1 percent in 2019, compared to 47.6 percent in 2008, and obese individuals accounted for
21.1 percent of the total population in 2019 (15.2 percent in 2008 and 19.6 percent in 2016). Although this increase
is observed in both male and female populations, it is more evident in the female population, with the rate of
overweight and obesity increasing by 20.3 percent in the last decade (45.9 percent in 2008 and 55.2 percent in 2019).
This figure is 15.9 for the male population, reaching from 49.2 percent in 2008 to 57 percent in 2019. According to
the survey, the prevalence of obesity was higher among women, while the prevalence of overweight was higher
among men for all data points. The distribution of individuals’ BMI by gender is summarized for the period 2008-
2019 in Table 1.
Furthermore, according to the results of the WHO European Childhood Obesity Surveillance Initiative (COSI)
research (2016), 24.5% of primary school 2nd-grade children (6-9 years old) in Türkiye were overweight and obese in
2016 (24.9% in boys, 24.2% in girls).

Table 1
Distribution of Individuals’ BMI by Gender (15+ age), 2008-2019

of
Normal weight
Year and sex Total Underweight (%) Pre-obese (%) Obese (%)
(%)

ro
Total 100 4.2 48.2 32.4 15.2

-p
2008 Male 100 2.7 48.1 36.9 12.3
Female 100 5.9 48.2 27.4 18.5
re
Total 100 4.7 45.5 33.0 16.9
lP

2010 Male 100 3.5 46.1 37.3 13.2

Female 100 5.9 44.7 28.4 21.0
na

Total 100 3.9 44.2 34.8 17.2

2012 Male 100 2.7 44.7 39.0 13.7
ur

Female 100 5.1 43.6 30.4 20.9

Total 100 4.2 42.2 33.7 19.9
Jo

2014 Male 100 2.8 43.7 38.2 15.3

Female 100 5.5 40.7 29.3 24.5
Total 100 4.0 42.1 34.3 19.6
2016 Male 100 2.5 43.8 38.6 15.2
Female 100 5.6 40.4 30.1 23.9
Total 100 3.8 40.1 35.0 21.1
2019
Male 100 2.7 40.3 39.7 17.3
Female 100 4.9 40.0 30.4 24.8
Note: Numbers in the table may not add up due to rounding.
Source: Turkish Statistical Institute (2020). Türkiye Health Survey – 2019.

2. Materials and methods

Many empirical studies have used several functional forms to test the Kuznets’ hypothesis. Our strategy is to
estimate Obesity Kuznets Curve (OKC) in its simplest quadratic form which can be formulated as follows:
𝐵𝑀𝐼𝑖 = 𝛽0 + 𝛽1 𝑦𝑖 + 𝛽2 𝑦𝑖2 + 𝛽3 𝑧𝑖 + 𝜀𝑖 (1)

4
where BMI refers to Body Mass Index. In this paper, we also use an indicator variable with the help of BMI,
describing whether a person is obese or not. 𝑦𝑖 refers to personal income level which is hypothesized to follow a
quadratic relationship (𝑦𝑖2 ); 𝑧𝑖 relates to other variables which potentially affect obesity such as sex, age, educational
level, marital status, sport activities; 𝜀𝑖 is an error term. This specification allows testing whether there exists an
inverted-U-shaped relationship between obesity and income such that 𝛽1 >0 and 𝛽2 <0. It is also possible to have 𝛽1 =
𝛽2 = 0 (indicating no relationship exists) and monotonic relationship in case of 𝛽1 > 0 𝑎𝑛𝑑 𝛽2 = 0. The turning
point can easily be obtained by the formula below:
−𝛽1
𝑦∗ =
2𝛽2

This is the level of income where BMI or obesity (whichever the dependent variable in the econometric
specification) stops increasing and begin to decrease.
We use two different approaches to estimate Eq.(1). We estimate the equation using logistic regression and
ordinary least squares (OLS) to report the results as our dependent variable is binary, which takes 1 for those whose
BMI is equal and greater than 30 and 0 otherwise.
Regarding the independent variables, the following socio-economic and demographic factors were included,

of
based on the literature: gender, marital status, income, education, level of physical activity. Table 2 summarizes
descriptive statistics for the sample. The analyses were performed with Obese as the dependent variable.

ro
Table 2
Descriptive Statistics

-p
Variable N % Description
Status of Obesity Obese 13249 21.3 1 if obese, 0 otherwise
re
Gender Female 13249 51.7 1 when female, 0 otherwise
Marital status Married 13249 69 1 when married, 0 otherwise
Income level 0-1500 TL 1833 11.74 Categorical variable (1-6)
lP

1501-2424 TL 3749 24.02

2425-3695 TL 3692 23.65
3696-5784 TL 3654 23.41
na

5785-8912 TL 1759 11.27

8913 TL and over 922 5.91
Age 15-24 2713 17.38 Categorical variable (1-7)
ur

25-34 2975 19.06

35-44 3230 20.69
Jo

45-54 2707 17.34

55-64 2212 14.17
65-74 1249 8.00
75 and over 523 3.35
Education status No school degree 822 5.27 Categorical variable (1-5)
Primary school 5585 35.78
Secondary school 2952 18.91
High school 3224 20.65
University 3026 19.39
Daily Walking time 10-29 minutes 6112 46.13 Categorical variable (1-5)
30-59 minutes 4396 33.18
1-2 hours 1935 14.60
2-3 hours 463 3.49
3+ hours 343 2.59

The grouping given in the income level variable in the Turkish Health Survey (THS) is determined according to
the 5% income groups of the Income and Living Conditions Survey (ILCS) conducted by TurkStat. For the
household income groups in the reference period of the THS, the results of the most recently released ILCS are used.
For the calculation of the ILCS, household disposable net income is defined as the sum of personal annual disposable
income (salaries, wages, wages, entrepreneurial income, pensions, widows' and orphans' pensions, payments to the
elderly, unpaid scholarships, etc.) received by each member of the household. It is calculated by subtracting taxes
paid during the income reference period and regular transfers to other households or individuals from the sum of the

5
sum of annual household income (real estate rental income, unrequited aid to the household, income earned by
individuals under the age of 15, etc.). Households are sorted from smallest to largest according to household
disposable income or equivalent household disposable income of all individuals in the household, and
household/individual groups of 5 percent are formed by dividing into 20 groups, 10 percent by dividing into 10
groups or 20 percent by dividing into 5 groups. The shares of these groups in total income provide information on
income inequality. For the purpose of our study, we combined some of these groups and we conducted the analysis
using these six new groups.
21.3% of respondents are obese. While 51.7% of all respondents are female, 69% of them are married. We use
income as categorical variable from 0-1500 TL to 8913 TL and over. Age, education and daily walking time were
also categorized as shown in Table 1.
For the purpose of the study, we use cross-sectional data obtained from “2019 Türkiye Health Research”
provided by the Turkish Statistical Institute (TUIK). The data were obtained by face-to-face interviews with 23199
persons from 8,166 households living in the Republic of Türkiye between September and December 2019. In the
collection of data, questionnaires including questions for children aged 0-14 were used together with the modules
recommended by the European Union Statistical Office (Eurostat). The four modules used in data collection are
health status, health care use, health determinants, and socio-economic background variables. We exclude
individuals under the age of 15 as the data set does not include the height and weight data for them. There are also

of
some missing data for 105 individuals aged 15 and over. Hence, the data of 13249 individuals aged 15 and over in
total were included in the analyses. The data set provides demographic information such as age, gender, educational

ro
status, and income as well as data on height-weight, and active lifestyles.
Since BMI values for individuals are not directly included in the data set, these values were calculated using the

-p
height and weight data of the individuals. Specifically, BMI value is calculated by dividing the weight in kilograms
by the square of the height in meters (Centers for Disease Control and Prevention, 2022). According to the WHO
(2022), as stated earlier, individuals with a BMI greater than 25 are defined as overweight while individuals with a
re
BMI greater than or equal to 30 are considered as obese.
In this study, using regression analysis, we specifically test the following hypotheses for the case of Türkiye:
lP

H1: There is a nonlinear relationship between obesity and income such that the probability of being obese firstly
increases as income rises and then begins to fall after reaching a peak.
H2: Males and females differ in the rates of obesity. Accordingly, females are more likely to be obese than males.
na

H3: Marital status is a factor affecting being obese.

H4: Older people tend to be more obese than younger people.
H5: More educated people are less likely to be obese.
ur

H6: Daily physical activities reduce the risk of being obese.

3. Results
Jo

Table 3 presents the regression results. Column (1) reports the OLS estimates while column (2) and (3) show the
logit coefficients and odd ratios, respectively. The final column presents confidence intervals at the 95% level. We
observe that there is a non-linear relationship between income and obesity status. The likelihood of being obese
increases as income rises and after reaching a peak point, higher income leads to a fall in the likelihood of being
obese. Our results indicate that BMI increases with income at a decreasing rate. These income variables are jointly
significant and peak at an income category between 3 and 4. This nearly corresponds to an income level between
3145 and 3398 TL in our database. Hence, we argue that individuals earning more than 3398 TL are more likely to
be obese in Türkiye.

Table 3
OLS and Logistic estimates
(1) (2) (3) (4)
VARIABLES Obese Obese Obese 95% CI
OLS Logistic Logistic
(Coefficients) Odd ratios

Income 0.030*** 0.182** 1.20 [1.08-1.46]

(0.011) (0.081)
Income2 -0.004** -0.025** 0.97 [0.94-0.98]
(0.001) (0.012)

6
Female 0.031*** 0.218*** 1.24 [1.13-1.36]
(0.006) (0.047)
Married 0.007 0.099 1.10 [0.98-1.25]
(0.009) (0.063)
Age 25-34 0.064*** 0.897*** 2.45 [1.92-3.07]
(0.012) (0.119)
Age 35-44 0.134*** 1.448*** 4.25 [3.38-5.34]
(0.013) (0.117)
Age 45-54 0.211*** 1.845*** 6.32 [5.05-8.00]
(0.013) (0.118)
Age 55-64 0.214*** 1.859*** 6.41 [5.09-8.12]
(0.01) (0.119)
Age 65-74 0.164*** 1.614*** 5.02 [3.89-6.47]
(0.016) (0.130)
Age 75 and over 0.055** 0.991*** 2.69 [1.88-3.88]
(0.024) (0.185)
Primary School -0.024 -0.156 0.85 [0.70-1.03]

of
(0.016) (0.097)
Secondary School -0.105*** -0.547*** 0.58 [0.46-0.72]

ro
(0.018) (0.114)
High School -0.120*** -0.683*** 0.50 [0.41-0.63]

-p
(0.018) (0.111)
University -0.157*** -1.006*** 0.36 [0.30-0.48]
(0.018) (0.118)
re
Daily walking 30-59min -0.029*** -0.191*** 0.82 [0.74-0.91]
(0.007) (0.051)
lP

Daily walking 1-2 hours -0.055*** -0.390*** 0.67 [0.58-0.78]

(0.010) (0.072)
Daily walking 2-3 hours -0.046** -0.337** 0.71 [0.54-0.93]
na

(0.018) (0.139)
Daily walking 3+ hours -0.046** -0.324** 0.72 [0.53-0.97]
(0.021) (0.156)
ur

Constant 0.123*** -2.581*** 0.07 [0.05-0.10]

(0.024) (0.178)
Jo

Observations 13,249 13,249

R-squared 0.086 0.092
Standard errors in parentheses, Pseudo R-square is shown in logistic regression. Reference categories: Age:15-24, no school
finished, walking 10-29 minutes a day. *** p<0.01, ** p<0.05, * p<0.1

Females tend to be more obese than males. For instance, women are 1.24 times more likely to be obese than men.
We do not find any link between marital status and obesity as its coefficient is not significant neither in linear
probability model (OLS) nor in logistic regression. When a person gets older, the probability of being obese
increases since reference group is people with age of 15-24. For instance, people whose age between 25-34 years are
2.45 times more likely to be obese compared to 15-24 years category. The highest probability of being obese occur
between the ages of 55-64. After peaking at those ages, obesity ratios start to decrease. Individuals with lower
education tend to exhibit higher levels of a greater probability of being obese. The same goes for sedentary or
physically low active individuals who have higher expected levels of obesity than non-sedentary/active individuals.
This means that each additional increase of walking time is associated with a decrease in the odds of being obese. To
put it more clearly, daily 1-2 hours of walking is associated with a %33 decrease in the odds of being obese
compared to walking 10-29 minutes a day†. From all these results, we show that all the hypotheses above hold except
H3 which say that marital status is a factor determining obesity.

†
To illustrate the distribution of BMI across different groups defined by age, and physical activity levels, we also provide the
presentation of our data by including intuitive graphical formats using box plots in the appendix section. These visualizations
provide a clear, immediate understanding of the data and highlight the central tendencies and variances within each group. We

7
Table 4
Logistic estimates for females and males
(1) (2)
Variables Obese Odds Obese Odds
Female Male

Income 0.262** 1.30 0.210* 1.23

(0.108) (0.113)
Income2 -0.0473*** 0.95 -0.0282* 0.97
(0.0165) (0.0166)
Married 0.140* 1.15 0.255** 1.29
(0.0842) (0.107)
Age 25-34 0.889*** 2.43 0.798*** 2.22
(0.166) (0.174)
Age 35-44 1.456*** 4.29 1.267*** 3.55

of
(0.160) (0.177)
Age 45-54 1.930*** 6.88 1.580*** 4.85
(0.161) (0.181)

ro
Age 55-64 2.128*** 8.40 1.391*** 4.02
(0.163) (0.184)

-p
Age 65-74 1.941*** 6.96 1.194*** 3.30
(0.181) (0.197)
re
Age 75 and over 1.456*** 4.29 0.497* 1.64
(0.262) (0.270)
Primary School 0.00401 1.00 -0.299 0.74
lP

(0.114) (0.198)
Secondary School -0.464*** 0.62 -0.579*** 0.56
(0.144) (0.211)
na

High School -0.686*** 0.50 -0.621*** 0.53

(0.138) (0.209)
University -1.191*** 0.30 -0.769*** 0.46
ur

(0.158) (0.214)
Daily walking 30-59min -0.211*** 0.80 -0.128* 0.88
Jo

(0.0713) (0.0759)
Daily walking 1-2 hours -0.388*** 0.68 -0.350*** 0.70
(0.110) (0.0979)
Daily walking 2-3 hours -0.356 0.70 -0.306* 0.73
(0.237) (0.172)
Daily walking 3+ hours -0.194 0.82 -0.327* 0.72
(0.323) (0.179)
Constant -2.584*** 0.07 -2.511*** 0.08
(0.226) (0.281)

Observations 6,676 6,573

Standard errors in parentheses. Reference categories: Age:15-24, no school finished, walking 10-29 minutes a day. *** p<0.01, **
p<0.05, * p<0.1

When we split the sample in terms of gender, the findings still support the OKC for both males and females. The
results suggest evidence of a Kuznets curve with income cut-off points differing across genders. The cut-off points
for females and males are 2215-2424 TL and 3399-3695 TL, respectively. This means that the likelihood of being
obese begins at lower level of income for females than males as income increases. Furthermore, we obtain some
notable results regarding the effect of marital status on obesity rates across genders. For example, being married for

also compare BMI differences between groups using Analysis of Variance (ANOVA) which we provide in the appendix section
to assess the impact of age and physical activity on BMI.

8
males increases the likelihood of obese than it does for females. The odd ratio of 1.29 indicates that the likelihood of
obesity is 29% greater for males who are married while the odd ratio is just 1.15 for females. The effect of aging on
obesity also differs across genders. However, it is important to highlight that the most obese age range for women is
55-64 (odd ratio: 8.40), while for men it is 45-54 (odd ratio: 4.85). Our findings are just about the same for both
genders in regard to education and daily walking time. However, it should be noted that university graduate women
are less likely to be obese than men (0.30<0.46).

4. Discussion

This study tests the validity of the OKC hypothesis in Türkiye controlling for socio-economic factors and
physical activity. To do so, we employ linear probability model and logistic regression benefitting from the Türkiye
Health Research 2019 dataset. Our findings reveal that OKC hypothesis holds for Türkiye such that the likelihood of
obesity first increases as income rises and after reaching a maximum point, it starts to decrease. In other words, there
is a non-linear relationship between income and obesity. This result proves the validity of OKC for Türkiye. These
results are in line with other studies in the literature (Ameye and Swinnen, 2019; Aydin, 2019; Grecu and Rothoff,
2015; Mathieu-Belh, 2021; Windart et al., 2019). We also find that the more educated individuals are less likely to be
obese.

of
Furthermore, the study reveals that obesity differs according to gender, so women are more likely to be obese
than men, which is also supported by other studies (Ameye and Swinnen, 2019; Garawi et al., 2014; Kapoor et al.,

ro
2021; Khan et al., 2021; Jonikas et al., 2016). However, the turning point of OKC in women is lower than in men. In
other words, although women enter the obesity process earlier than men, they exit the obesity process earlier due to

-p
increased income. This might be explained by the fact that women with an increasing income attach more
importance to their health than men do (Windarti et al. 2019), as well as the fact that BMI values have a stronger
effect on body image in women than it has in men (Algars at al. 2009). Moreover, it is stated in the literature that
re
there is a linear relationship between women's BMI values and body dissatisfaction ( Algars et al., 2009; Chithambo
and Huey, 2013; Dunkel et al., 2010; Swami et al., 2015).
lP

Our empirical study reveals that married men were more likely to be obese than married women. This finding is
in line with Liu et al. (2021), Liao et al. (2018), and Teachman (2016). The reason for this situation is thought to be
the emergence of more meal consumption with a regular family life brought by marriage and the adoption of a less
na

active lifestyle (Burke et al., 2004; Mata et al., 2015;). As the level of education increases, the obesity rate decreases.
According to Lawrence (2017), the reason for this is that university graduates exhibit healthier behaviors and have a
better socio-economic status.
ur

Age is also another factor affecting obesity. According to the results obtained in the study, obesity rates are more
in middle-aged people than in higher-aged ones. In other words, obesity rates increase from the 15-24 age range to
the 55-64 age range. Accordingly, the participants with the highest obesity rate are those between 55 and 64. Obesity
Jo

rates begin to decrease for participants 65 years old and over. The low obesity rate in old age may be because the diet
in old age is different from the diet in young and middle age, as well as the changing work and living conditions with
the development of technology. Namely, with the development of technology, a sedentary lifestyle prevails both in
business and social life and positively affects obesity (Woessner et al., 2021). However, health problems may occur
due to nutrition in old age, which may negatively affect obesity, with the transformation of the diet in young and
middle-aged periods into a healthier diet (Granic et al., 2018).
We also examined the relationship between walking time and obesity in this study. The findings reveal that the
probability of being obese in individuals with increased walking time is lower compared to less active individuals.
Physical activity and an active lifestyle are proven methods of treating obesity (Niemiro et al., 2022). The Centers for
Disease Control and Prevention (CDC) treats walking as one of the most effective methods of physical activity
(CDC, 2022b). Our results are supported by Jakicic et al. (2018) and Kim et al. (2017) which found physical activity
has a negative effect on obesity.
The main aim of this study is to test the validity of the OKC hypothesis for Türkiye. Therefore, it is worth
mentioning its importance in terms of health economics. Because obesity is mainly caused by an imbalance between
energy intake and energy expenditure, the importance of income, eating habits, and an active lifestyle comes to the
fore (Wyszyńska et al., 2020). Although there is a strong relationship between eating habits and obesity (Birch and
Fisher, 1998; Branen and Fletcher, 1999; Manios et al., 2004), one of the most important factors affecting nutritional
habits is food consumption and, therefore, income status affecting food preference (Konttinen at al., 2021;
Giacobone at al., 2021). French et al. (2019) show that the amount of out-of-home consumption increases depending
on income rise. Hence the number of calories and energy taken from the consumption of ready-made and processed
foods or fast foods is multiplied, causing obesity (Ameye and Swinnen, 2019; Grecu and Rotthoff, 2015).

9
 Along with the benefits of technological developments to society, the significant restrictions on active lifestyle
causes obesity and health problems associated with obesity (Woessner et al., 2021). At high-income levels,
individuals have become more aware of obesity and the health problems associated with obesity (Butler-Wall, 2016).
In addition, the level of awareness about personal health and external appearance increases in individuals whose
income continues to increase (Zhou, 2019). As a result, individuals can reduce their obesity levels by allocating their
income to healthier food consumption and an active lifestyle (Grecu and Rotthoff, 2015).
The negative externality problem of obesity emerges in terms of health economics (Bhattacharya and Sood, 2007;
McCormick and Stone, 2007). This situation requires public intervention against obesity. In this manner, the
prevention of obesity is crucial in terms of easing the economic burden on both governments and individuals.
Accordingly, obese individuals may have higher health insurance premium payments or standing contribution
payments to reduce negative externality than non-obese individuals (Bernard et al., 2018; Woolford et al., 2013).
Thus, individual efforts will be encouraged to overcome obesity. For example, taxes on high-calorie and sugary
beverages in some countries decrease consumption of these products (Pedraza et al., 2019; Redondo et al., 2018). In
this way, additional taxation can be applied to such products to reduce the consumption of high-calorie and
unhealthy nutrition. However, such deterrent measures alone may not be sufficient to combat obesity. Along with
these measures, particularly to prevent childhood obesity and adult obesity in the following years, compulsory
physical education lessons should be included in the curriculum of schools, and the duration of these lessons should

of
be increased (Wyszyńska et al., 2020). Besides these policies, obesity awareness levels among children and parents
might be increased, and healthy eating and an active lifestyle should be encouraged to combat the obesity prevalence

ro
among the society.
Although cross-sectional data provides insightful views into the interaction between variables at a given time, its

-p
main purpose is to serve as a starting point for more in-depth and dynamic study. The conclusions drawn from our
study can inform hypotheses for future research and can also highlight potential targets for interventions. Future
research could focus on longitudinal studies to better understand the interaction between income and obesity. Such
re
studies could help to verify the existence of the OKC over time and in different settings.
Insights from the OKC can guide the development of health education programs that address dietary habits and
lP

physical activity, tailored to different income levels. Furthermore, understanding at what income levels the risk of
obesity is highest can help public health officials to design targeted interventions for those income groups, which
may be more cost-effective. The OKC theory suggests that economic policies aimed at reducing poverty could also
na

impact obesity rates. This might include efforts to provide better access to healthy foods and promoting jobs that
require physical activity. The findings can inform healthcare providers about the need for more proactive screening
and intervention strategies in populations that are emerging into higher income brackets. The intersection of
ur

economics and health indicated by the OKC implies a need for collaboration between different sectors – such as
healthcare, finance, education, and urban planning – to address the multifaceted determinants of obesity.
Jo

5. Conclusion

Our results indicate that there is a non-linear relationship between income and obesity. The study reveals that
obesity differs according to gender and marriage, education, age, and physical activity are closely associated with the
likelihood of being obese in Türkiye.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could
have appeared to influence the work reported in this paper.

Ethical Approval

This study does not require ethical approval since it involves information freely available in the public domain
and the analysis of the data relies on database upon request.

Acknowledgement

10
 The authors extend their sincere appreciation to the anonymous reviewers whose insightful and constructive
comments significantly contributed to enhancing the quality of this manuscript. We also thank the editor for their
guidance throughout the review process.

CRediT authorship contribution statement

Muhammed Benli: Investigation, Writing – original draft, Writing – review & editing. Yasin Acar:
Conceptualization, Supervision, Methodology, Formal analysis, Writing – original draft. Semih Bas: Visualization,
Data Curation, Writing – original draft.

Appendix

of
ro
-p
re
lP

Fig. 1. Box Plot (BMI – Age Groups)

na

Table 1
ANOVA Analysis (Age Groups)
ur

Source SS df MS F P-value
Between Groups 67690.533 6 11281.756 548.44 0.000
Jo

Within Groups 320943.174 15602 20.571

Total 388633.707 15608 24.9
No. of obs. 15609 R-squared 0.174
Root MSE 4.535 Adj. R-squared 0.174

11
 Fig. 2. Box Plot (BMI – Physical Activity)

of
Table 2

ro
ANOVA Analysis (Physical Activity)
Source SS df MS F P-value

-p
Between Groups 1722.51 4 430.627 18.01 0.000
Within Groups 316659.168 13244 23.61
re
Total 318381.678 13248 24.9
No. of obs. 13249 R-squared 0.005
lP

Root MSE 4.89 Adj. R-squared 0.005

na

References

Akee, R., Simeonova, E., Copeland, W., Angold, A., Costello, E. J., 2013. Young adult obesity and household
ur

income: effects of unconditional cash transfers. American Economic Journal: Applied Economics, 5(2), 1-28.
https://doi.org/10.1257/app.5.2.1
Jo

Algars, M., Santtila, P., Varjonen, M., Witting, K., Johansson, A., Jern, P., Sandnabba, N. K., 2009. The adult body:
how age, gender, and body mass index are related to body image. Journal of Aging and Health, 21(8), 1112–
1132. https://doi.org/10.1177/089826430934802
Alola, A. A., Bekun, F. V., 2021. Obesity Kuznets curve and the reality of eco-income ellipsoids (EIE). The
European Journal of Health Economics, 22(7), 1095-1101. https://doi.org/10.1007/s10198-021-01308-x
Ameye, H., Swinnen, J., 2019. Obesity, income and gender: the changing global relationship. Global Food
Security, 23, 267-281. https://doi.org/10.1016/j.gfs.2019.09.003
Armstrong, S., Wong, C. A., Perrin, E., Page, S., Sibley, L., Skinner, A., 2018. Association of physical activity with
income, race/ethnicity, and sex among adolescents and young adults in the United States: findings from the
National Health and Nutrition Examination Survey, 2007-2016. Jama Pediatrics, 172(8), 732-740.
https://doi.org/10.1001/jamapediatrics.2018.1273
Aydin, M., 2019. The effect of economic growth on obesity for the most obese countries: new evidence from the
obesity Kuznets curve. The European Journal of Health Economics, 20(9), 1349-1358.
https://doi.org/10.1007/s10198-019-01099-2
Bernard, M., Riedel-Heller, S. G., Luck-Sikorski, C., 2019. Weigh more, pay more? Public opinion on varying health
insurance contributions among divergent weight groups. Obese Facts, 12, 509–517.
https://doi.org/10.1159/000502799
Bhattacharya, J., Sood, N., 2007. Health insurance and the obesity externality. Advances in Health Economics and
Health Services Research, 17: 279–318. https://doi.org/10.1016/S0731-2199(06)17011-9
Birch, L.L., Fisher, J.O., 1998. Development of eating behaviours among children and adolescents. Pediatrics,
101(3), 539–549. https://doi.org/10.1542/peds.101.S2.539

12
Branca, F., Nikogosian, H., Lobstein, T. (Eds.). 2007. The challenge of obesity in the WHO European Region and
the strategies for response: summary. World Health Organization.
Branen, L., Fletcher, J., 1999. Comparison of college students’ current eating habits and recollections of their
childhood food practices. Journal of Nutrition Education, 31(6), 304-310. https://doi.org/10.1016/S0022-
3182(99)70483-8
Burke, V., Beilin, L. J., Dunbar, D., Kevan, M., 2004. Changes in health-related behaviours and cardiovascular risk
factors in young adults: associations with living with a partner. Preventive Medicine, 39, 722-730.
https://doi.org/10.1016/j.ypmed.2004.02.038
Butler-Wall K., 2016. Feeling healthy: media, affect, and the governance of health. University of Minnesota Digital
Conservancy. Available at: http://hdl.handle.net/11299/182202
Centers for Disease Control and Prevention, 2022. Defining adult overweight & obesity.
https://www.cdc.gov/obesity/basics/adult-defining.html
Centers for Disease Control and Prevention, 2022b. Physical activity - walking.
https://www.cdc.gov/physicalactivity/walking/index.htm
Chithambo, T. P., Huey, S. J., 2013. Black/white differences in perceived weight and attractiveness among
overweight women. Journal of Obesity, 2013, 1–4. https://doi.org/10.1155/2013/320326
Clement, M., 2017. The income-body-size gradient among Chinese urban adults: a semiparametric analysis. China

of
Economic Review, 44, 253-270. https://doi.org/10.1016/j.chieco.2017.05.003
Cutler, D. M., Glaeser, E. L., Shapiro, J. M., 2003. Why have Americans become more obese?. Journal of Economic

ro
Perspectives, 17(3), 93-118. https://doi.org/10.1257/089533003769204371
Deuchert, E., Cabus, S., Tafreschi, D., 2014. A short note on economic development and socioeconomic inequality

-p
in female body weight. Health economics, 23(7), 861-869. https://doi.org/10.1002/hec.2968
Dinsa, G. D., Goryakin, Y., Fumagalli, E., Suhrcke, M., 2012. Obesity and socioeconomic status in developing
countries: a systematic review. Obesity Reviews, 13(11), 1067-1079. https://doi.org/10.1111/j.1467-
re
789X.2012.01017.x
Drewnowski, A., Specter, S. E., 2004. Poverty and obesity: the role of energy density and energy costs. The
lP

American Journal of Clinical Nutrition, 79(1), 6-16. https://doi.org/10.1093/ajcn/79.1.6

Dunkel, T. M., Davidson, D., Qurashi, S., 2010. Body satisfaction and pressure to be thin in younger and older
muslim and non-muslim women: the role of Western and non-Western dress preferences. Body Image, 7(1),
na

56–65. https://doi.org/10.1016/j.bodyim.2009.10.003
Edwards, P., Tsouros, A. D., 2006. Promoting physical activity and active living in urban environments: the role of
local governments. WHO Regional Office Europe.
ur

French, S. A., Tangney, C. C., Crane, M. M., Wang, Y., Appelhans, B. M., 2019. Nutrition quality of food purchases
varies by household income: the SHoPPER study. BMC Public Health, 19, 231.
https://doi.org/10.1186/s12889-019-6546-2
Jo

Garawi, F., Devries, K., Thorogood, N., Uauy, R., 2014. Global differences between women and men in the
prevalence of obesity: is there an association with gender inequality? European Journal of Clinical Nutrition,
68(10), 1101–1106. https://doi.org/10.1038/ejcn.2014.86
Giacobone, G., Tiscornia, M. V., Guarnieri, L., Castronuovo, L., Mackay, S., Allemandi, L., 2021. Measuring cost
and affordability of current vs. healthy diets in Argentina: an application of linear programming and the
INFORMAS protocol. BMC Public Health, 21(1). https://doi.org/10.1186/s12889-021-10914-6
Giskes, K., Van Lenthe, F. J., Turrell, G., Kamphuis, C. B., Brug, J., Mackenbach, J. P., 2008. Socioeconomic
position at different stages of the life course and its influence on body weight and weight gain in adulthood: a
longitudinal study with 13‐year follow‐up. Obesity, 16(6), 1377-1381. https://doi.org/10.1038/oby.2008.54
Granic, A., Mendonça, N., Hill, T., Jagger, C., Stevenson, E., Mathers, J., Sayer, A., 2018. Nutrition in the very old.
Nutrients, 10(3), 269. https://doi.org/10.3390/nu10030269
Grecu, A. M., Rotthoff, K. W., 2015. Economic growth and obesity: findings of an obesity Kuznets curve. Applied
Economics Letters, 22(7), 539–543. https://doi.org/10.1080/13504851.2014.955251
Heinrich, K. M., Lee, R. E., Suminski, R. R., Regan, G. R., Reese-Smith, J. Y., Howard, H. H.,..., Ahluwalia, J. S.,
2007. Associations between the built environment and physical activity in public housing
residents. International Journal of Behavioral Nutrition and Physical Activity, 4(1), 1-9.
https://doi.org/10.1186/1479-5868-4-56
Jakicic, J. M., Rogers, R. J., Davis, K. K., Collins, K. A., 2018. Role of physical activity and exercise in treating
patients with overweight and obesity. Clinical Chemistry, 64(1): 99–107.
https://doi.org/10.1373/clinchem.2017.272443

13
Jonikas, J. A., Cook, J. A., Razzano, L. A., Steigman, P. J., Hamilton, M. M., Swarbrick, M. A., Santos, A., 2016.
Associations between gender and obesity among adults with mental illnesses in a community health screening
study. Community Mental Health Journal, 52, 406–415. https://doi.org/10.1007/s10597-015-9965-2
Kapoor, N., Arora, S., Kalra, S., 2021. Gender disparities in people living with obesity - an unchartered territory. J.
Mid-life Health, 12(2), 103–7. https://doi.org/10.4103/jmh.jmh_48_21
Khan, S., Nauman, H., Saher, S., Imtiaz, H. A., Bibi, A., Sajid, H., Khan, T. M., Mumtaz, M., Bibi, S., 2021. Gender
difference in obesity prevalence among general population of Lahore, Pakistan. European Journal of Medical
and Health Sciences, 3(3), 55–58. https://doi.org/10.24018/ejmed.2021.3.3.878
Kim, B.Y., Choi, D.H., Jung, C.H., Kang, S.K., Mok, J.O., Kim, C.H., 2017. Obesity and physical activity. Journal
of Obesity & Metabolic Syndrome, 26(1), 15–22. https://doi.org/10.7570/jomes.2017.26.1.15
Köksal O. (1977). Türkiye 1974 beslenme, sağlık ve gıda tüketimi araştırması. Hacettepe University, Ankara.
Konttinen, H., Halmesvaara, O., Fogelholm, M., Saarijärvi, H., Nevalainen, J., Erkkola, M., 2021. Sociodemographic
differences in motives for food selection: results from the LoCard cross-sectional survey. International Journal
of Behavioral Nutrition and Physical Activity, 18, 71. https://doi.org/10.1186/s12966-021-01139-2
Lakdawalla, D., Philipson, T., 2009. The growth of obesity and technological change. Economics & Human Biology,
7(3), 283-293. https://doi.org/10.1016/j.ehb.2009.08.001
Lawrence, E. M., 2017. Why do college graduates behave more healthfully than those who are less educated?

of
Journal of Health and Social Behavior, 58(3), 291–306. https://doi.org/10.1177/0022146517715671
Lee, J., Shin, A., Cho, S., Choi, J.Y., Kang, D., Lee, J.K., 2020. Marital status and the prevalence of obesity in a

ro
Korean population. Obesity Research & Clinical Practice, 14, 217–224.
https://doi.org/10.1016/j.orcp.2020.04.003

-p
Liao, C., Gao, W., Cao, W., Lv, J., Yu, C., Wang, S.,…, Li, L., 2018. Association of educational level and marital
status with obesity: a study of Chinese twins. Twin Research and Human Genetics, 21(2), 126–135.
https://doi.org/10.1017/thg.2018.8
re
Liu, J., Garstka, M. A, Chai, Z., Chen, Y., Lipkova, V., Cooper, M.E., Mokoena, K.K., Wang, Y., Zhang, L., 2021.
Marriage contributes to higher obesity risk in China: findings from the China health and nutrition survey.
lP

Annals of Translational Medicine, 9(7), 564. https://doi.org/10.21037/atm-20-4550

Malik, V.S., Willett, W.C., Hu, F.B., 2013. Global obesity: trends, risk factors and policy implications. Nature
Reviews Endocrinology, 9(1), 13-27. https://doi.org/10.1038/nrendo.2012.199
na

Manios Y., Dimitriou M., Moschonis G., Kocaoglu B., Sur H., Keskin Y. Hayran O., 2004. Cardiovascular disease
risk factors among children of different socioeconomic status in Istanbul, Türkiye: directions for public health
and nutrition policy. Lipids in Health and Disease, 4(3), 11. https://doi.org/10.1186/1476-511X-3-11
ur

Marteleto, L. J., Gama, L. C., Dondero, M., Brown, L. E., 2017. The weight of inequality: socioeconomic status and
adolescent body mass in Brazil. Social Forces, 95(4), 1637-1666.
Jo

https://doi.org/10.1093/sf/sox02810.1093/sf/sox028
Mata, J., Frank, R., Hertwig, R., 2015. Higher body mass index, less exercise, but healthier eating in married adults:
nine representative surveys across Europe. Social Science & Medicine, 138, 119–127.
https://doi.org/10.1016/j.socscimed.2015.06.001
Mathieu‐Bolh, N., 2022. The elusive link between income and obesity. Journal of Economic Surveys, 36, 935-968.
https://doi.org/10.1111/joes.12458
McCormick, B., Stone, I., 2007. Economic costs of obesity and the case for government intervention. Obesity
Reviews, 8(s1): 161–164. https://doi.org/10.1111/j.1467-789x.2007.00337.x
Ministry of Health of Türkiye, 2019. Türkiye beslenme ve sağlık ve araştırması. General Directorate of Public
Health, Ankara.
Minos, D., Butzlaff, I., Demmler, K. M., Rischke, R., 2016. Economic growth, climate change, and obesity. Current
Obesity Reports, 5(4), 441-448. https://doi.org/10.1007/s13679-016-0234-7
Monteiro, C.A., Conde, W.L., Popkin, B.M., 2001. Independent effects of income and education on the risk of
obesity in the Brazilian adult population. The Journal of Nutrition, 131(3), 881S-886S.
https://doi.org/10.1093/jn/131.3.881S
Mosli, H.H., Kutbi, H.A., Alhasan, A.H., Mosli, R.H., 2020. Understanding the interrelationship between education,
income, and obesity among adults in Saudi Arabia. Obesity Facts, 13(1), 77-85.
https://doi.org/10.1159/000505246
Niemiro, G. M., Revane, A., Algotar, A.M., 2022. Exercise and fitness effect on obesity. [Updated 2022 Jun 5]. In:
StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing.

14
OECD, 2019. OECD Health policy studies the heavy burden of obesity the economics of prevention. OECD.
Available at: https://www.oecd.org/health/the-heavy-burden-of-obesity-67450d67-en.htm. Access date:
22.08.2022.
Ogden, C. L., Carroll, M. D., Fakhouri, T. H., Hales, C. M., Fryar, C. D., Li, X., Freedman, D. S., 2018. Prevalence
of obesity among youths by household income and education level of head of household—United States
2011–2014. Morbidity and Mortality Weekly Report, 67(6), 186. https://doi.org/10.15585/mmwr.mm6706a3
Ozyapi, S. 2019. An empirical investigation of obesity & income and education. MPRA Paper 98727.
Pampel, F. C., Denney, J. T., Krueger, P. M., 2012. Obesity, SES, and economic development: a test of the reversal
hypothesis. Social Science & Medicine, 74(7), 1073-1081. https://doi.org/10.1016/j.socscimed.2011.12.028
Pedraza, L. S., Popkin, B. M., Batis, C., Adair, L., Robinson, W. R., Guilkey, D. K. Taillie, L. S., 2019. The caloric
and sugar content of beverages purchased at different store-types changed after the sugary drinks taxation in
Mexico. International Journal of Behavioral Nutrition and Physical Activity, 16, 103.
https://doi.org/10.1186/s12966-019-0872-8
Philipson, T., Posner, R., 1999. The long-run growth in obesity as a function of technological change. National
Bureau of Economic Research Working Paper 7423. Cambridge, MA: National Bureau of Economic
Research, https://doi.org/10.3386/w7423

of
Popkin, B.M., 1998. The nutrition transition and its health implications in lower-income countries. Public Health
Nutrition, 1(1), 5-21. https://doi.org/10.1079/phn19980004
Popkin, B.M., Adair, L.S., Ng, S.W., 2012. Global nutrition transition and the pandemic of obesity in developing

ro
countries. Nutrition Reviews, 70(1), 3-21. https://doi.org/10.1111/j.1753-4887.2011.00456.x
Redondo, M., Hernández-Aguado, I. Lumbreras, B., 2018. The impact of the tax on sweetened beverages: a

-p
systematic review. The American Journal of Clinical Nutrition, 108(3): 548–563.
https://doi.org/10.1093/ajcn/nqy135
re
Ren, Y., Li, H., Wang, X., 2019. Family income and nutrition-related health: evidence from food consumption in
China. Social Science & Medicine, 232, 58-76. https://doi.org/10.1016/j.socscimed.2019.04.016
Romieu, I., Dossus, L., Barquera, S., Blottière, H.M., Franks, P.W., Gunter, M., ..., Willett, W.C., 2017. Energy
lP

balance and obesity: what are the main drivers?. Cancer Causes & Control, 28(3), 247-258.
https://doi.org/10.1007/s10552-017-0869-z
Sallis, J.F., Glanz, K., 2009. Physical activity and food environments: solutions to the obesity epidemic. The Milbank
na

Quarterly, 87(1), 123-154. https://doi.org/10.1111/j.1468-0009.2009.00550.x

Shekar, M., Popkin, B., 2020. Obesity: health and economic consequences of an impending global challenge. World
Bank Publications. Available at: https://www.worldbank.org/en/topic/nutrition/publication/obesity-health-and-
ur

economic-consequences-of-an-impending-global-challenge. Access date: 15.09.2022.

Swami, V., Tran, U.S., Stieger, S., Voracek, M., 2015. Associations Between women’s body image and happiness:
Jo

Results of the YouBeauty.com Body Image Survey (YBIS). Journal of Happiness Studies, 16(3): 705–718.
https://doi.org/10.1007/s10902-014-9530-7
Talukdar, D., Seenivasan, S., Cameron, A.J., Sacks, G., 2020. The association between national income and adult
obesity prevalence: Empirical insights into temporal patterns and moderators of the association using 40 years
of data across 147 countries. Plos One, 15(5), e0232236. https://doi.org/10.1371/journal.pone.0232236
Teachman, J., 2016. Body weight, marital status, and changes in marital status. Journal of Family Issues, 37(1): 74–
96. https://doi.org/10.1177/0192513x13508404
Tremmel, M., Gerdtham, U.G., Nilsson, P.M., Saha, S., 2017. Economic burden of obesity: a systematic literature
review. International Journal of Environmental Research and Public Health, 14(4), 435.
https://doi.org/10.3390/ijerph14040435
Turkish Statistical Institute, 2020. Türkiye Health Survey - 2019. Available at:
https://data.tuik.gov.tr/Bulten/Index?p=Türkiye-Saglik-Arastirmasi-2019-33661. Access date: 12.09.2022.
Türkiye Childhood (Primary School, 2nd Grade Students) Obesity Surveillance Initiative COSI-TUR 2016” Ministry
of Health -General Directorate of Public Health, Ministry of National Education, World Health Organization
Regional Office for Europe, Ministry of Health Publication No: 1126 Ankara 2019.
Wells, J. C., Marphatia, A. A., Cole, T. J., McCoy, D., 2012. Associations of economic and gender inequality with
global obesity prevalence: understanding the female excess. Social Science & Medicine, 75(3), 482-490.
https://doi.org/10.1016/j.socscimed.2012.03.029
WHO, 2022. Obesity and overweight, https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight.
Access date: 22.08.2022.
WHO, (--). Obesity. https://www.who.int/health-topics/obesity#tab=tab_1.

15
Windarti, N., Hlaing, S.W., Kakinaka, M., 2019. Obesity Kuznets curve: international evidence. Public Health, 169:
26–35. https://doi.org/10.1016/j.puhe.2019.01.004
Withrow, D., Alter, D.A., 2011. The economic burden of obesity worldwide: a systematic review of the direct costs
of obesity. Obesity Reviews, 12(2), 131-141. https://doi.org/10.1111/j.1467-789X.2009.00712.x
Woessner, M. N., Tacey, A., Levinger-Limor, A., Parker, A. G., Levinger, P., Levinger, I., 2021. The evolution of
technology and physical inactivity: the good, the bad, and the way forward. Front. Public Health, 9, 655491.
https://doi.org/10.3389/fpubh.2021.655491
Woolford, S. J., Clark, S. J., Butchart, A., Geiger, J. D., Davis, M. M., Fagerlin, A., 2013. To pay or not to pay:
public perception regarding insurance coverage of obesity treatment. Obesity, 21(12), E709–E714.
https://doi.org/10.1002/oby.20387
World Obesity Atlas, 2022. https://www.worldobesity.org/resources/resource-library/world-obesity-atlas-2022.
Access date: 04.10.2022.
Wyszyńska, J., Ring-Dimitriou, S., Thivel, D., Weghuber, D., Hadjipanayis, A., Grossman, Z.,…, Mazur, A., 2020.
Physical activity in the prevention of childhood obesity: the position of the European childhood obesity group
and the European academy of pediatrics. Frontiers in Pediatrics, 8, 535705.
https://doi.org/10.3389/fped.2020.535705
Zhang, S., Xiang, W., 2019. Income gradient in health-related quality of life—the role of social networking

of
time. International Journal for Equity in Health, 18(1), 1-10. https://doi.org/10.1186/s12939-019-0942-1
Zhou, M., 2019. The shifting burden of obesity: changes in the distribution of obesity in China, 2010–

ro
2015. International Sociology, 34(3), 347-367. https://doi.org/10.1177/0268580919832734
Zhou, M., 2021. The shifting income-obesity relationship: conditioning effects from economic development and

-p
globalization. SSM-Population Health, 15, 100849. https://doi.org/10.1016/j.ssmph.2021.100849
re
lP
na
ur
Jo

16
 Testing Obesity Kuznets Curve for Türkiye

Highlights
• There is a non-linear relationship between income and obesity in Türkiye.
• Obesity differs according to gender; women are more likely to be obese than men.
• Marriage, education, age, and physical activity are closely associated with the likelihood of
being obese in Türkiye.

of
ro
-p
re
lP
na
ur
Jo
Declaration of interests

☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.

☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests:

of
ro
-p
re
lP
na
ur
Jo