Obesity in Ukraine: Understanding and Addressing the Rural Divide

Why does obesity matter?

For many, obesity may not seem as harmful as other public health issues, but it actually is. The Ministry of Health reports that almost 60% of adults in Ukraine are overweight, with nearly 25% of them being obese (WHO STEPS 2019). The rate is among the highest in Eastern Europe.

Obesity has devastating health consequences, ranging from cancer to mental health disorders. In terms of numbers, overweight and obesity are responsible for about USD 1 trillion in annual healthcare costs worldwide (Diamantis et al. 2022). The breakdown of these costs is shown in Figure 1.

Why is the urban/rural divide important?

Since social resources are limited, we need to tackle each problem in the most efficient way. Therefore, we need to identify population groups at higher risks of obesity and related problems and address them in the first place. We hypothesize that rural residents constitute such a group as they are more prone to obesity compared to urban dwellers. This hypothesis is supported by evidence from both Europe and the US. Thus, rural residents had 35.8% greater odds of obesity than urban residents in the US (Wen et al., 2018) and 32% greater odds in Poland (Stos et al., 2022). Testing this hypothesis, we obtained similar results: 24.6% lower odds of obesity for urban residents.

Note that odds and probability are not the same. Probability says how often something happens (e.g., 25% chance). Odds is the ratio of happening vs. not happening (e.g., 1/3 odds). In this article, we report probabilities for easier interpretation.

Data and method

The data source for this article is the WHO STEPS survey on non-communicable diseases, conducted in Ukraine in 2019. The target population was male and female Ukrainians aged 18 to 69 years, from both urban and rural areas, residing in the country. A total of 4,409 respondents participated in the face-to-face survey, the sample was representative for government-controlled areas. Respondents answered a number of questions on their health, lifestyle and demography, and their blood sugar and cholesterol were measured with Cardiocheck Plus devices.

Our dependent variable is obesity: a person was considered obese if their Body Mass Index (BMI, calculated as weight/height²) was 30 kg/m² or higher. It is worth noting that other alternatives to BMI – such as Waist Circumference (WC), Waist-to-Hip Ratio (WHR), and Waist-to-Height Ratio (WHtR) – were also evaluated. However, since these measures are strongly correlated, we opted to use BMI, as it is more standard and allows for comparisons with other similar studies.

We estimate the impact on this binary variable (obese=1) of socioeconomic variables – from basic factors like gender and age to more detailed measures such as fruit, vegetable, and processed food consumption. To do this, we apply various specifications and econometric methods: linear LPM and non-linear Logit. All the models (i.e., all combinations of specifications and methods) tell the same consistent story. Therefore, to save time and space, we present the results of only one of them.

What determines obesity? Estimation results

We find that living in an urban area reduces the probability of obesity by 4.3 p.p. compared to rural residents (figure 2). Gender, age and smoking also significantly impact the probability of being obese. Other variables, including eating processed food, turned out to be insignificant (see Annex).

We include both age and age squared into the regression equation to account for non-linearity of age effect. We also include the interaction term Female*Age to capture the possibility that the effect of gender on obesity may become more pronounced with age.

Indeed, one additional year of age increases the obesity risk by 2.4 p.p. but the effect diminishes at older ages. Thus, for example, for a 30-year-old male an additional year increases the probability of obesity by 1.2 p.p., while for a 50-year-old male this probability increases by only 0.4 p.p. A turning point for men is 60 years: after that each additional year decreases the probability of obesity. 

For females obesity risk is higher: thus, for a 30-year-old woman an additional year would increase obesity risk by 1.6 p.p., and for a 50-year-old woman – by 0.8 p.p. A turning point for women in 70 years.

Note that we don’t analyze Female coefficient alone because due to inclusion of Female*Age variable, it becomes just a baseline shift (a coefficient of being female at age 0). We must interpret gender differences only in combination with age.

Finally, smokers are 8.0 p.p. Less likely to be obese, probably due to appetite-suppressing effect of nicotine.

Estimation results only for rural dwellers are similar to results for the entire sample, but the value of the coefficients is higher. For example, an additional year of age increases the probability of obesity by 2.4 p.p. in the total sample and by 3.5 p.p. in the rural sample. Turning points for age in the rural sample are 58 years for men and 65 years for women.  Women start with a lower obesity risk (the baseline effect of being Female (when Age = 0) is -0.137), but it increases more rapidly with age (Female*Age is positive and significant).

An interesting finding is that Lives Alone becomes significant: rural individuals who live alone have a 6.3 percentage point higher obesity risk, which is substantial. This could be because rural individuals living alone are less likely to seek preventive medical care, either due to lower healthcare availability or reduced motivation. As a result, obesity-related risks may remain unaddressed for longer periods, as there may be no one to encourage or motivate these individuals.

What can we do about rural obesity?

We propose a simple framework. First, we need to admit the problem and raise awareness about it, with campaigns tailored to the rural context. The message can be delivered by local authorities and trusted leaders (e.g., teachers, doctors, religious figures) for better reach. We can also focus on public events and community workshops rather than internet-based campaigns in areas with digital gaps.

Since the best treatment for the disease is its prevention, we suggest several measures to treat obesity. First, teachers should encourage healthy habits in students. Second, family doctors in rural areas should be trained to track obesity issues and integrate nutrition counselling into routine visits. Third, local authorities should invest in infrastructure like sporting grounds, bike lanes, walking paths, and community centers that offer some physical activity, e.g. dancing.

However, even with the best preventive measures, people will still face obesity. In such cases, access to treatment is essential. Solutions include mobile clinics for remote areas and telehealth services, while advanced care can be provided by specialized centers. 

References

1. Diamantis, D. V., K. Karatzi, P. Kantaras, S. Liatis, V. Iotova, Y. Bazdraska, T. Tankova, G. Cardon, K. Wikström, I. Rurik, E. Antal, A. M. Ayala-Marín, N. G. Legarre, K. Makrilakis, and Y. Manios. 2022. “Prevalence and Socioeconomic Correlates of Adult Obesity in Europe: The Feel4Diabetes Study.” International Journal of Environmental Research and Public Health 19, no. 19 (October): 12572. 
2. Okunogbe, A., R. Nugent, G. Spencer, J. Powis, J. Ralston, and J. Wilding. 2022. “Economic Impacts of Overweight and Obesity: Current and Future Estimates for 161 Countries.” BMJ Global Health 7 (April): e009773. 
3. Stival, C., A. Lugo, A. Odone, P. A. van den Brandt, E. Fernandez, O. Tigova, J. B. Soriano, M. José López, S. Scaglioni, and S. Gallus. TackSHS Project Investigators. 2022. “Prevalence and Correlates of Overweight and Obesity in 12 European Countries in 2017–2018.” Obesity Facts 15, no. 5: 655–665. 
4. Stoś, K., E. Rychlik, A. Woźniak, M. Ołtarzewski, M. Jankowski, M. Gujski, and G. Juszczyk. 2022. “Prevalence and Sociodemographic Factors Associated with Overweight and Obesity Among Adults in Poland: A 2019/2020 Nationwide Cross-Sectional Survey.” International Journal of Environmental Research and Public Health 19, no. 3 (January): 1502.
5. Traczyk, I., A. Kucharska, B. I. Sińska, M. Panczyk, L. Wronka, F. Raciborski, D. Szostak-Węgierek, and B. Samoliński. 2023. “Every Second Adult Inhabitant of Poland (Aged 18–64) is Overweight—Results of Representative Cross-Sectional Studies Conducted in 2017–2020.” Annals of Agricultural and Environmental Medicine 30, no. 2 (June): 322–330.
6. World Health Organization. Regional Office for Europe. 2020. Risk Factors for Noncommunicable Diseases in Ukraine in 2019..
7. Zatońska, K., P. Psikus, A. Basiak-Rasała, Z. Stępnicka, D. Gaweł-Dąbrowska, M. Wołyniec, J. Gibka, A. Szuba, and K. Połtyn-Zaradna. 2021. “Obesity and Chosen Non-Communicable Diseases in PURE Poland Cohort Study.” International Journal of Environmental Research and Public Health 18, no. 5 (March): 2701.
8. Bloomberg News. “Novo Nordisk Overtakes SAP as Europe’s Most Valuable Company.” Bloomberg, June 13, 2025. 

Annex. Estimation results