Annual Health Expenditures by Body Mass Index Categories and the Economic Impact of Obesity in Türkiye: A Retrospective Modeling Study

Abnormal or excessive fat accumulation that may impair health is defined as overweight and obesity [1]. The worldwide prevalence of overweight and obesity has increased gradually over the years, as reported to have reached a threefold increase within the 40 years since 1975 [1].

According to the World Health Organization (WHO) European Regional Obesity Report in 2022, Türkiye has become the country with the highest rate of obesity among the European countries, with a rising prevalence of obesity (>30%) starting from the age 20 and reaching 50% in females and >30% in males in the 45–74 age group [2]. Similarly, the 2023 World Obesity Atlas by the World Obesity Federation indicated a very high (55% by 2035) prevalence of obesity in adults along with a high (2.3%) annual increase in adult obesity (2020–2035) in Türkiye [3].

Turkish population-based diabetes, hypertension, and obesity surveys (TURDEP I and II) conducted with a 12-year interval reported an increase in the prevalence of obesity in the Turkish adult population from 22.3% in 1998 (30% in females, 13% in males) to 35% in 2010 (44% in females, 27% in males), and the rate of increase in obesity was particularly higher for males (107%) than for females (34%) [4, 5]. According to 1997–1998 data from the TEKHARF study on the cardiac disease and risk factors in the Turkish adult population and 1999–2000 data from the TOHTA study on obesity and hypertension in Türkiye, the prevalence of adult obesity was 28.6% (38.8% in females and 18.7% in males) and 19.4% (24.6% in females, 14.4% in males), respectively [6].

Data from Türkiye Nutrition and Health Survey (TNHS) in 2010 revealed similar rates of adult obesity (30.3% overall, 41% in females, 20.5% in males) in the TURDEP II study, while the class III obesity rate was 2.9% (5.3% in females and 0.7% in males) [7]. The TNHS-2017 study revealed the prevalence of overweightness (overall: 34%; males: 39.9%; females: 27.6%) and obesity (overall: 31.5% [class I+II: 27.8%, class III: 3.7]; males: 24.6% [class I+II: 32.7%, class III: 6.4%]; females: 39.1% [class I+II: 32.7%, class III: 6.4%]) in individuals aged ≥15 years [8]. Similarly, the National Household Health Survey in 2017, which was conducted via the WHO-stepwise methodology and in the same period with TNHS 2017, reported the prevalence of overweight and obesity in individuals aged ≥15 years to be 35.6% (41.2% in males, 30.1% in females) and 28.8% (21.8% in males, 35.9% in females), respectively [9].

In a meta-analysis of 12 randomized controlled trials (in 106,533 adults, within the 2005–2016 period) with data on body mass index (BMI) measurement, obesity prevalence was found to be 33.2% in females and 18.2% in males, while the meta-regression analysis indicated the age among the inclusion criteria but not the year of the study to be an important factor likely to explain varied obesity prevalence across different studies [10]. Accordingly, the standardized obesity prevalence was calculated using the crude age- and gender-specific obesity rates in the TURDEP II study and the Turkish Statistical Institute (TurkStat) yearly reports on the address-based population registration system (ABPRS). When TURDEP II data on obesity were standardized with respect to 2010–2022 ABPRS population reports, the prevalence of obesity in individuals aged ≥20 years in our country is estimated to be 29.8% (36.7% in females, 22.7% in males) in 2010, increasing to 31.5% (38.9% in females, 23.8% in males) in 2022 and expected to reach 34.9% (43.2% in females, 26.5% in males) by 2045 [10]. Hence, obesity remains a significant health issue in Türkiye, where two-thirds of the adult population is overweight and more than one-third of the overweight individuals are obesity [11, 12].

Obesity appears to underlie almost a hundred diseases or health conditions in eleven different disease groups, mainly the cardiovascular diseases, diabetes, hypertension, certain cancers, and musculoskeletal disorders. Obesity and the related increase in morbidity and mortality are considered responsible for 4.7 million deaths each year worldwide and for decreasing the life expectancy of individuals in Organization for Economic Cooperation and Development (OECD) countries by 3 years [13, 14].

Obesity remains a public health issue on a global scale, while it also adversely affects the world economies [13]. Estimation of economic impact of obesity, particularly in resource-limited countries, may help in drawing policymakers’ attention and enabling an improved recognition of the fact that the most economically beneficial solution is to take immediate action [14]. The economic burden of consequences related to the high BMI was reported to account for 13% of total health expenditures in 2019 [15]. Obesity-related health expenditures constitute 2–7% of overall health expenditures in the developed countries, approximately 21% of annual health expenditures in the USA and 2–8% of annual health expenditures in European countries [16, 17]. In OECD countries, 8.4% of the health budget is estimated to be spent on treating the consequences of overweight and obesity in the next 30 years [13]. In Türkiye, the economic impact of overweight and obesity has not yet been properly investigated at a national level.

This study aimed to determine the economic impact of obesity on Türkiye. In addition, in order to evaluate the effectiveness of interventions regarding obesity risk factors, the impact of obesity on healthcare resource utilization and annual health expenditure in anthropometric terms, and the potential economic results of changes in BMI distribution, as a determinant of health expenditure, were investigated.

This study focused on two different analyses. In the first step, annual healthcare resource utilization and annual health expenditure were analyzed based on BMI categories of adults in Türkiye. In the second step, current and future economic impacts of obesity in adults were estimated using the data on obesity, as well as the 17 diseases and 15 malignancies related to obesity. The estimated economic impacts were adjusted according to 2021 purchasing power parities (PPPs) US dollar values (2.782 TRY = 1 PPP) [18].

BMI is widely used for the classification of obesity in adults, as a simple height-weight index. BMI (kg/m²) is calculated by dividing the adults’ weight in kilograms by their height in meters squared [1, 19]. WHO defines a BMI value between 25 kg/m² and 29.9 kg/m² in adults as overweight, and a value equal to or greater than 30 kg/m² as obesity, while obesity is further classified in three subgroups including class I (30.0–34.9 kg/m²), class II (35.0–39.9 kg/m²) and class III (≥40.0 kg/m²) obesity [1, 19]. In this study, BMI groups were classified in accordance with these definitions [1, 19].

The study was approved by the Ethics Committee of Baskent University (Data of Approval: 28.09.2022, Protocol No: KA22/394). Permission was obtained from the Turkish Statistical Institute for the use of micro dataset in our study (Permission No.: 789). The need for informed consent was waived by the Baskent University Ethics Committee.

TurkStat Türkiye Health Survey (THS) raw datasets, which were obtained through computer-assisted personal interviewing (CAPI) with 29,761 individuals living in 10,028 households, using the modules recommended by the statistical office of the European Union (Eurostat) in addition to the questions related to children in the 0–14 years age group, were subjected to secondary analysis [11, 20, 21]. By combining the relevant categories of datasets related to 2014–2019 years, a new dataset comprising the annual healthcare resource utilization and the diagnoses for chronic diseases among 72,880 persons was generated and transferred to IBM SPSS Statistics for Windows, version 26.0 (IBM Corp., Armonk, NY). Individuals aged ≤18 years and those with BMI <18.5 kg/m² and short stature (dwarfism) were excluded from the dataset. Data of persons without any health insurance such as the Social Security Institution (SSI), which is the only national reimbursement institution in Türkiye, or private health insurance or health insurance funds were not included in the analysis. Following the data cleansing, a total of 42,006 individuals were subjected to the first-step analyses (Fig. 1).

In order to estimate the economic impact of obesity from the societal perspective, a cost of illness model was developed using Microsoft Excel (Microsoft Corp. Released 2021. Redmond, WA) program (Fig. 2). In this study, the costs of accessing healthcare such as transportation, accommodation and meals, care provided by family, and other out-of-pocket expenses were considered. Travel costs were calculated based on the average transport cost (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000542821), using the average number of inpatient and outpatient treatments and obesity prevalence [22, 23].

By estimating the average daily wage of caregivers, for a hospitalized person or a person with obesity, cost of services received for wage payment for the hospitalization days and home-care were estimated. Average cost was calculated using the average wage payments of caregivers in different provinces across Türkiye. Accordingly, average wage payment for a paid daytime caregiver working 6 days a week and 8 h a day was found to be PPP 1,617.5 for January 2022. Paid caregiver service is received by 3.7% of individuals with obesity. In case of care provided by family members, taking into account that no extra wages are paid, the cost of the loss of workforce (online suppl. Table 2) for these persons was calculated separately within the indirect costs, according to labor statistics in the employed group [24].

Indirect costs are disease-related workforce loss, the loss caused by disability or early retirement and societal costs that are caused by premature deaths. In the calculation of indirect costs, Human-Capital Approach (HCA) method was utilized. HCA is the value of lost productivity due to premature death and disability of the person caused by her/his disease [22, 25]. Indirect costs or lost productivity are business earnings lost due to adverse health outcomes. For a disease that leads to premature death, indirect costs are the loss of potential wages and profits. In this study, impacts that depend on obesity-related premature deaths were calculated based on the current value of the annual economic impact of the death year (DY) (discount rate: 10.2%), the value of life year (VLY) and total number of persons to be still alive (people death year [PDY]) (online suppl. Table 3) [24, 26‒28].

The disease, death, side effects and time passed while receiving treatment result in a decrease in production and productivity. For employees with obesity, economic impacts on workplaces were estimated based on the absenteeism or decrease in productivity. Indirect costs also include lost productivity and earnings of the families and family members who take care of the patients. Besides the workplace, these were calculated as the impacts of absenteeism due to obesity-related conditions, through average daily wage and employed population with obesity, excess days of absence, in accordance with the obesity prevalence (online suppl. Table 2) [24]. Costs related to early retirement due to obesity and related diseases or disability caused by chronic diseases were obtained via multiplying the difference between the average and the realized retirement age by the monthly wage, tax and social security contributions (online suppl. Table 4) [29‒31].

Obesity data from SSI 2009–2014 demand data repository and 2015–2019 bariatric surgery data were used for projections of distribution of health services presented in the 2019 Health Statistics Report of Ministry of Health (MoH). Besides the 2004 dataset of National Burden of Disease and Cost Effectiveness (NBD-CE) Study, Republic of Türkiye Public Health Institution (PHIT), Chronic Diseases and Risk Factors Survey (CDRFS) in 2013 and the results of its extension study, namely, MoH General Directorate of Public Health Türkiye Non-Communicable Diseases and Risk Factors Cohort Study (NCD Türkiye in 2021, and MoH 2016 and 2017 National Cancer Statistics served as the national data sources used in projection calculations [28, 32‒40].

Analyses were subjected to 30-year projection. By using TurkStat Türkiye Demographic Information for patients diagnosed and treated with obesity, information regarding the individuals diagnosed with obesity in the 2009–2014 data repository of SSI and those with bariatric surgery in 2015–2019 data repository, projection for the year 2021 was performed. The total number of patients diagnosed with obesity (ICD10 E66.x) in 2021 was estimated to be 2,211,865.

In patients with obesity, comorbidity and complication costs were calculated in two different groups of diseases (prevalent and incident cases). If data on the disease prevalence were found in the data repository of SSI for reimbursement demands or TurkStat data, these prevalence values were used primarily (online suppl. Tables 5 and 6) [9, 11, 20, 21, 32, 36, 39].

In all direct medical cost calculations, related sections of Legal Notice of the SSI on the Implementations of Health (Health Implementation Notification/Communique – HIC) were taken into consideration [41]. Outpatient costs in obesity treatment were calculated over one calendar year. Emergency admissions for outpatient clinics were calculated in their own cost item. Outpatient costs related to treatment of side effects, comorbidity and complications were also calculated separately under the related cost items. Costs of laboratory and imaging tests regarding side effects, comorbidities and complications were calculated separately in the related cost item.

For the outpatient treatments, health institutions that are capable of providing treatment for obesity and related conditions were selected from the List for Classification of Healthcare Providers in Outpatient Treatment by taking into consideration the distribution of hospitals and related specialties listed in the Public Hospitals Institution 2017 report [35]; and “outpatient application fee” for the related outpatient clinics were calculated as weighted average according to the distribution of MoH hospitals (73.9%), private hospitals (13.8%), university hospitals (8.8%), and private medical centers (3.5%) [38]. For the services specified by the Legal Notice and conducted in tertiary care health institutions, a calculated 10% extra was added to cost items [41]. Laboratory and imaging tests and diagnostic procedures on the HIC Annex 2/B procedure score lists were included in the cost calculation with their scores. Laboratory and imaging test costs were added to the costs for inpatients with non-package (HIC Annex 2/C) hospitalization, while for outpatient admissions, the tests and medical supply not on the HIC Annex 2/A-2 outpatient additional procedures list were not added to the cost.

The annual hospitalization rate and length of hospital stay (LOS, average days) of patients with obesity were retrieved from TurkStat data. Hospitalizations for bariatric surgery and approximately 15,000 bariatric surgery operations (50.0%) presumed to have SSI coverage were included in the costs from the SSI perspective. Nearly 10.0% of patients with obesity receive psychotherapy and 9.0% receive physiotherapy, costs of which were evaluated under the interventions cost item. Health services and diagnostic or therapeutic procedures on the HIC Annex 2/B (per service) and Annex 2/C (package) procedure score lists were included in cost calculation.

Except for the conditions specific to the cause of hospitalization, daily hospitalization cost was calculated as PPP 66.6 for the medical wards and PPP 74.1 for the surgical wards. The healthcare resource utilization that may realize within hospitalization for a week was calculated and this cost was converted to daily hospitalization cost (per diem).

The HIC Annex 2/C package procedure list scores were also used for intensive care unit hospitalizations, and the costs for the first and the last day of intensive care unit stay were calculated as PPP 151.5, while costs of in between days were calculated as PPP 225.6 per day, with consideration of each level of care to be equal. For cases where only secondary or tertiary intensive care services were used, the daily cost was calculated as PPP 290.9. In the calculations, the utility percentage of university hospitals (8.3%) was considered with inclusion of 50% extra in accordance with this percentage [35, 38, 41].

Drug costs were calculated using the Turkish Medicines and Medical Devices Agency (TMMDA) and HIC Annex 4 list. Drugs were classified according to active ingredients, all forms of all products with the active ingredient were included in the analysis and their arithmetic means were reflected to the calculations. For the prices, valid public discounts were taken into account [41, 42]. The costs related to medical equipment which are not included in the procedure scores and payments of which are performed separately by SSI as per the HIC Annex 3 list, were included in the cost calculations, by taking arithmetic mean of prices in case of more than one product and by using the listed price in case of one product, along with the addition of 8% tax rate.

For the annual number of outpatient treatments, inpatient treatments and health service admissions of the patients, mean, standard deviation (SD), and 95% confidence interval (95% CI) values were calculated. The differences were evaluated with Pearson chi-square test, Fisher’s exact test and one-way analysis of variance (ANOVA). The impact of BMI on the health expenditures was analyzed via Generalized Linear Model (GLM) with gamma distribution and log link function [43]. Odds ratios (ORs) adjusted for age and sex, and 95% CIs were calculated according to GLM.

The current value of money and the discount rate that can be used for Türkiye in the next 10-year period were calculated. For the development of chronic diseases, the identified rates were converted to yearly estimations and then included in the population. By calculating the relative risks with 95% CI values, the population attributable fraction (PAF) was calculated. These methods were also explained in the related sections.

Mean age of participants was 46.5 ± 16.3 years. Overall, 36.5% (n = 15,336) of participants had normal BMI, while 38.7% (n = 16,272) were overweight, and 24.8% (n = 10,398) were with obesity according to BMI. The obesity classification revealed class I, class II and class III obesity in 73.0% (n = 7,569), 21.0% (n = 2,165) and 6.0% (n = 664) of those with obesity, respectively. Prevalence of obesity was 19.6% (n = 3,703) for males and 29.0% (n = 6,695) for females (Table 1).

Mean age was higher in individuals with obesity than in those with normal BMI and overweightness. For each obesity class, mean age was above 50 years, and the highest percentage of patients was in the 45–54 years age group (25.2%, 27.0%, and 30.1% for classes I, II, and III, respectively) (Table 1).

Charlson score was found to increase with advanced BMI, from mean ± SD score of 0.4 ± 0.9 in normal BMI to 0.8 ± 1.1 in class I obesity, 1.0 ± 1.3 in class II obesity and 1.3 ± 1.4 in class III obesity (p < 0.001). The most common comorbidities in individuals with obesity were the spinal diseases of lumbar (45%) and cervical (31%) regions, hypertension (36%), diabetes (23%), and arthrosis (18%) (Table 1). In individuals with obesity vs. those with normal BMI, both the smoking (16–18% vs. 25.4%, p < 0.001) and the regular (>3 times per month within the last 12 months) alcohol consumption (2.1–4.2% vs. 6.6%, p < 0.001) rates were significantly lower (Table 1).

Rate of utilization of outpatient and inpatient health services was found to increase with increasing BMI in the overall population and in those with SSI-based health insurance coverage. In the overall population, outpatient health service utilization rate within the last year was 66.2% with an average 5.8 ± 9.2 per person days in the outpatient care. In individuals with normal BMI, outpatient health service utilization rate within the last year was 63.4% with an average 5.3 ± 8.8 per person days in the outpatient care. In overweight individuals, these values were 66.4% and 5.7 ± 9.0 days, while in those with obesity, the values were 68.7% and 6.4 ± 8.9 days in class I obesity, 73.8% and 7.4 ± 11.9 days in class II obesity and 74.2% and 8.5 ± 12.3 days in class III obesity, respectively (p < 0.001) (Table 2).

Overall, the physician examination rate within the last 1 year was 88.8% including 4.7 ± 6.2 visits in the primary care and 6.6 ± 9.7 specialist visits. The physical examination rates and the number of primary care and specialist visits were 86.6%, 4.3 ± 5.7 and 6.2 ± 9.2, respectively, in the normal BMI group; 89.0%, 4.6 ± 6.2 and 6.5 ± 9.6, respectively, in the overweight group; 91.2%, 5.2 ± 6.8 and 7.1 ± 9.8, respectively, in class I obesity; 91.7%, 5.5 ± 7.1 and 7.9 ± 10.9, respectively, in class II obesity and 93.7%, 5.8 ± 7.9 and 9.3 ± 14.0, respectively, in class III obesity (Table 2).

In the overall population, inpatient admission rate within the last 1 year was 12.3% and average LOS was 0.8 ± 5.3 days. The inpatient admission rates and LOS were 11.1% and 0.8 ± 5.3 days in the normal BMI group; 12.3% and 0.7 ± 4.6 days in the overweight group; 13.4% and 0.9 ± 5.2 days in the class I obesity, 15.9% and 1.3 ± 7.0 days in the class II obesity, and 17.2% and 1.7 ± 8.0 days in the class III obesity (Table 2).

Average annual health expenditure in the normal BMI group was calculated as PPP 950.7 (95% CI: 928.8–972.6). In other BMI subgroups, average annual health expenditure was calculated via the age- and gender-adjusted weighted GLM. The average annual health expenditure was estimated to be PPP 1,148.6 (OR = 1.040; 95% CI: 1.018–1.064; p < 0.001) in the overweight group. In the obesity group (OR = 1.243; 95% CI: 1.206–1.281; p < 0.001), average annual health expenditure was PPP 1,436.6 (OR = 1.159; 95% CI 1.128–1.192; p < 0.001) in class I obesity, PPP 1,817.2 (OR = 1.378; 95% CI: 1.318–1.440; p < 0.001) in class II obesity, and PPP 2,240.9 (OR = 1.686; 95% CI: 1.564–1.818; p < 0.001) in class III obesity (Table 2).

In females, the average annual health expenditure was PPP 1,017.8 (95% CI: 989.0–1,046.5) in the normal BMI group, PPP 1,348.1 (95% CI: 1,313.1–1,383.0) in the overweight group, PPP 1,647.4 (95% CI: 1,597.0–1,697.8) in class I obesity, PPP 1,973.7 (95% CI: 1,874.9–2,072.5) in class II obesity and PPP 2,284.8 (95% CI: 2,113.7–2,456.0) in class III obesity. In males, the annual average health expenditure in the normal BMI, overweight, class I obesity, class II obesity and class III obesity groups were PPP 865.9 (95% CI: 832.2–899.7), PPP 961.9 (95% CI: 931.2–992.5), PPP 1,122.1 (95% CI: 1,067.4–1,176.8), PPP 1,369.5 (95% CI: 1,230.8–1,508.2), and PPP 2,004.4 (95% CI: 1,420.3–2,588.5), respectively (Table 2).

The health insurance coverage was based on SSI in 96.3% (n = 40,458) of participants, and on private health insurance and/or funds in 3.7% (n = 1,548). Annual utilization of health services and annual health expenditures per patient from the health insurance perspective were calculated based on BMI and gender, using the costs of health services for the year 2021 (Table 3).

The average annual health expenditure by health insurance coverage in females revealed the expenditures in the class I, II, and III obesity groups to be PPP 1,623.2 (95% CI: 1,573.8–1,672.6), PPP 1,950.4 (95% CI: 1,853.4–2,047.4), and PPP 2,256.6 (95% CI: 2,087.0–2,426.3), respectively, in those with SSI-based insurance; and to be PPP 2,827.8 (95% CI: 2,149.5–3,506.2), PPP 3,239.1 (95% CI: 1,777.4–4,700.8), and PPP 3,836.5 (95% CI: 1,435.1–6,237.9), respectively, in those with private health insurance. Among males, the average annual health expenditure in the class I, II, and III obesity groups were PPP 1,108.6 (95% CI: 1,053.5–1,163.8), PPP 1,332.6 (95% CI: 1,192.1–1,473.0), and PPP 2,059.1 (95% CI: 1,460.8–2,657.4), respectively, in those with SSI-based insurance and were PPP 1,470.1 (95% CI: 1,111.9–1,828.2) for class I obesity and PPP 2,274.7 (95% CI: 1,496.9–3,052.5) for class II obesity in those with private health insurance. There were no males with class III obesity and private insurance (Table 3).

The use of BMI categorization in planning of health services devoted to obesity prevention and in identification and monitoring of cost-effective interventions is considered not sensitive enough to understand the effect of alterations in different distributions of BMI values. For this reason, a linear regression analysis as adjusted for age and gender was used to evaluate the effect of alterations in BMI values on the health expenditures. This analysis revealed the association of 1 kg/m² change in BMI with average PPP 55.4 (95% CI: PPP 52.4–58.5; p < 0.001) change in the annual health expenditure.

For the year 2021, among the 2,211,865 patients diagnosed and treated with obesity, 77.6% (n = 1,715,390) were females and 22.4% (n = 496,475) were males. Overall, 21.5% (n = 475,685) of patients were in the age group of 35–44 (online suppl. Table 7).

In patients receiving obesity treatment, the average per patient annual costs were calculated to be PPP 68.1 for the outpatient visits and laboratory and imaging tests, to be PPP 1,582.7 for the hospitalization, and to be PPP 37.4 for the interventions. The average annual cost per patient for prescribed drugs and medical equipment was estimated to be PPP 19.8. After the weighting of costs regarding the comorbidities and complications of obesity, the average total cost per patient was found to be PPP 2,526.5 (Table 4).

In patients with obesity, the main components of co-payment were considered to include the insurance contributions for physician examination and drug treatment services, costs of dietitians, popular diet applications and food supplements not reimbursed by insurance and traditional and complementary therapies such as acupuncture and mesotherapy. Average annual cost per patient for the co-payment was estimated to be PPP 1,546.3 (Table 4).

The annual direct medical cost was calculated to be PPP 9,366,034,469 in total from the insurance perspective and to be PPP 12,786,334,605 when considered together with the co-payment. Average annual cost per patient related to direct medical cost item was PPP 5,780.8 (Table 4).

In healthcare services provided outside the city of residence, except for days of the hospitalization, around 8.9% of accommodation and food costs are met by the insurance institution. Average annual food cost per patient was determined as PPP 235.2 and accommodation cost as PPP 652.8. Average annual travel cost per patient and per diem reimbursement of the security institution is PPP 86.8.

Average annual cost of waged caregivers per patient was estimated to be PPP 708.5. Direct non-medical cost was estimated to be PPP 14,566,867,737 overall, including travel (74.4%), caregiver (10.8%), food and accommodation (14.8%) costs. Average annual direct non-medical cost per patient was estimated to be PPP 6,585.8 (Table 5).

Mean age of patients with obesity is 44.8 years which refers to 35.6 years of life expectancy according to TurkStat life table [28]. In patients with obesity, mortality rate was calculated based on the SSI data repository and TurkStat cause of death statistics for Türkiye (online suppl. Table 3) [24, 26‒28]. As a result of premature death, the total cost of financial loss of families was PPP 7,920,838,544 and total cost of public losses caused by taxes and contributions was PPP 3,184,798,220. The average annual cost per patient of premature death was estimated to be PPP 5,020.9 (Table 6).

Due to obesity and related diseases and required treatments, income losses occur at the individual and public levels. The total public cost of sick leaves and reports was PPP 5,249,756,940 which corresponds to average annual cost per patient of PPP 2,373.5. Total cost of sick leaves for the employed patients and for an employed family-member caregiver was calculated as PPP 3,473,061,898. Together with this cost, average workforce loss per patient was estimated to be PPP 3,943.6 (Table 6).

Total cost of early retirement pension paid by the insurance institution was determined as PPP 11,858,598,999. When evaluated together with the tax and contribution loss of the public due to early/disabled retirement (PPP 6,995,702,287), the average per patient cost of early/disability retirement was estimated to be PPP 8,524.2 (online suppl. Table 4) [29‒31]. With respect to financial aids, total annual cost of disability pension was PPP 820,098,451 and average per patient annual cost of disability pension was estimated to be PPP 370.8 (Table 6).

In this cost of illness study, average per patient annual costs were estimated to be PPP 5,780.8 for the direct medical cost, PPP 6,585.8 for the direct non-medical cost and PPP 17,859.5 for the indirect cost, and the total per patient annual cost of illness for obesity was estimated to be PPP 30,226.1. When weighted according to gender, total economic burden of the obesity in females was calculated to be PPP 48,701,167,638, and the average annual cost per patient was PPP 28,390.7. In male patients with obesity, total economic burden of disease was calculated as PPP 18,154,890,044 and average annual cost per patient was PPP 36,567.6.

Direct cost of illness estimates for obesity based on 2.2 million obese people registered on SSI registration system is equivalent to 8.4% of total health expenditure of Türkiye and 14.7% of health expenditure of SSI. For the year 2021, total annual economic burden of obesity for Türkiye was estimated to be PPP 66.9 billion, which corresponds to 2.6% of the gross national product of Türkiye.

Using the trends related to overweightness, obesity and the diagnosed and treated obesity conditions between 2009–2021 in Türkiye based on our data, the obesity rate in the next 30 years and the impact of ±5% change in this rate were investigated. Based on the current value of money, it was predicted that a 5% decrease in the rate of obesity would provide a cost saving of PPP 88 billion in 30 years, while a 5% increase in the rate of obesity would cause a cost increment of PPP 153 billion. The current rise of obesity has a course that seems to catch 5% increase in the rate in 30 years. This shows that the economic burden attributable to obesity may increase by almost 50% within 30 years (Table 7).

Obesity is defined as the most important public health issue of the 21st century. Manageability of a health issue depends also on its measurability. Obesity is often perceived as an individual problem, neglecting the involvement of societal factors. In this study, the economic burden of obesity in Türkiye was estimated for the year 2021, to determine the societal burden of obesity and provide every stakeholder involved in the subject with a picture of their economic shares. In this regard, the cost of illness related to diagnosed obesity was evaluated along with the impact of obesity on the healthcare resource utilization and annual health expenditures in anthropometric terms, and for the first time in Türkiye, the national economic burden of obesity was determined along with estimation of potential cost savings related to prevention of obesity via the projections.

Almost 2/3 of adults in Türkiye are overweight or obese [5, 8]. As the degree of obesity increases, the percentage of females in the group also increases. Given the high rates of obesity in the 45–64 years age group, as this group gets older, it is likely that prevalence of obesity in older adults (age ≥65) would continue to increase in the future.

Utilization of health services as well as the number of utilizations were found to be higher in individuals with obesity than in those with normal BMI. In Türkiye, the number of annual per person physician visits was estimated to be 9.8 in 2019 including 3.5 primary care visits and 6.3 specialist visits [44]. In our study, the number of primary care and specialist visits were found to be 4.3 and 6.2 in the normal BMI group, and to increase by 154% in the obesity group including 5.2 and 7.1 visits, respectively, in class I obesity, 5.5 and 7.9 in class II obesity, and 5.8 and 9.3 in class III obesity. The outpatient treatment days were increased from 5.3 to 8.5, while per person LOS in hospitalizations was prolonged by 213%. The annual health expenditures in case of obesity relative to normal BMI was found to increase by 117% in class I obesity, by 139% in class II obesity and by 169% in class III obesity.

Chronic diseases are more prevalent in individuals with obesity. For the direct medical costs, the costs related to complications were 1.5 times greater regarding obesity. The prevalence of diabetes, dyslipidemia, and hypertension increases more than 3 times, with higher degrees of obesity. This increase is more than 2 times for the liver diseases and arthrosis and 2 times for coronary artery disease, asthma, and COPD. Accordingly, the complications have become the major cost driver accounting for 60% of the direct medical costs. Direct medical costs are expected to represent the cost items that do not decrease the economic growth directly but cause the shift of resources from the fields that would reinforce the economy to the health services. For obesity, these costs were identified to be derived mostly by the burden of obesity-related complications.

The high rate of health services utilization in obesity is the main cause of transportation costs related to access to these services. Consequently, the largest component of non-medical direct costs has become the transportation expenditures. In Türkiye, annual transportation cost caused by obesity and related conditions was estimated to be around PPP 10.8 billion, which corresponds to annual transportation cost of PPP 144 per capita for the overall population.

There is increasing evidence regarding the employees with obesity to have higher rates of absenteeism and presenteeism. Previous studies reported the likelihood of individuals with obesity to be less productive than the non-obese individuals and the tendency of individuals with obesity to go to work (presenteeism) even if they do not have proper physical conditions for working. In the current study, a 20% lower rate of employment was noted in individuals with obesity compared to those without obesity, as taken into account in calculating the “loss of productivity.” In 2021, 17.2 million days of incapacity payment occurred due to obesity and related conditions, which accounts for nearly 36% of annual incapacity payments made by SSI in the overall population. In Türkiye, one-third of health-related loss of workforce in 1 year and 16% of disability pensions for chronic disease occur due to obesity and related conditions.

Obesity-related health expenditures are PPP 12.8 billion per year and PPP 9.4 billion of this is covered by the health insurances. Out-of-pocket payments of patients comprise one-third of the cost with major contribution of the health services and treatments not covered by the insurance. Obesity is responsible for 8.4% of total health services expenditures and 15% of national reimbursement institution expenditures in Türkiye. If no measures are taken to cease this health issue, an additional expenditure of PPP 4.7 billion will be required up to the year 2050. Obesity-related expenditures constitute 2–7% of health expenditures in developed countries. In our study, these expenditures were found to be consistent with the 30-year budget burden of obesity and related conditions foreseen for the OECD countries [13], which is also lower than rates reported in the USA (21%) and higher than the upper bound of the range (2–7%) declared in the European region [16, 17].

Türkiye spends PPP 66.9 billion per year for the treatment of conditions related with obesity. Obesity comprised the 2.6% of the gross domestic product (GDP) in 2021. Medical costs of individuals with obesity are PPP 616.5 higher than those with normal body weight. If nothing changes in the current situation, an obesity-related cost of PPP 2.5 trillion is expected in Türkiye in the next 30 years. Globally, obesity continues to be both a disease burden and an economic burden. Therefore, there are many studies in the literature supporting our findings. Cawley and Meyerhoefer [16] estimated that the annual medical care costs of obesity in the USA were 147 billion annually. Gortmaker et al. [45] presented a global perspective on the economic burden of obesity, estimating that obesity could cost up to 2.8% of global GDP. Wang et al. [46] estimated the health and economic burden of projected obesity trends in the USA and the UK, finding that obesity could cost up to USD 344 billion annually in the USA and GBP 24 billion annually in the UK.

In light of these results, it is evident that obesity poses a significant burden not only on healthcare systems but also on the economy. These findings are supported by various studies available in the literature.

D’Errico et al. [47] revealed that the overall cost attributed to obesity in Italy reached EUR 13.34 billion in 2020, with direct costs estimated at EUR 7.89 billion. Cardiovascular diseases accounted for the highest impact on costs at EUR 6.66 billion, followed by diabetes (EUR 0.65 billion), cancer (EUR 0.33 billion), and bariatric surgery (EUR 0.24 billion). Indirect costs totaled EUR 5.45 billion, with similar contributions from absenteeism (EUR 2.62 billion) and presenteeism (EUR 2.83 billion). The study emphasizes the considerable direct and indirect costs associated with obesity, highlighting the critical need for cost-effective preventive programs to address this public health threat in Italy [47].

Chen et al.’s [48] article underscores the insufficiency of pharmacoeconomic analyses for obesity treatments in China. Current studies have predominantly focused on the health system perspective, often omitting productivity loss. The article notes the scarcity of such analyses, citing total direct medical cost estimates for overweight and obesity at USD 62.6 billion. Additionally, the article highlights that existing pharmacoeconomic analyses in China have primarily centered on evaluating the cost effectiveness of bariatric surgeries and comprehensive medical interventions, particularly in blood glucose management. It also emphasizes variations in methodologies and discrepancies in study quality among current analyses.

The results of Hoogendoorn et al.’s (2023) study underscore the substantial economic burden of obesity across five European countries. Base-case analyses revealed varying total lifetime healthcare costs for obese individuals aged 40 with a BMI of 35 kg/m², ranging from EUR 75,376 in Greece to EUR 343,354 in the Netherlands, while the life expectancies varied from 37.9 years in Germany to 39.7 years in Spain. The analysis demonstrated that a one-unit decrease in BMI led to gains in life expectancy, ranging from 0.65 to 0.68 years, with changes in total healthcare costs varying from EUR −1,563 to EUR +4,832. These results emphasize the significance of considering both obesity and non-obesity-related healthcare costs in decision-making concerning the implementation of preventive interventions [49].

These studies suggest that obesity is a significant economic burden both nationally and globally. However, the results may vary due to the use of different methods to estimate the economic burden of obesity.

Certain limitations to this study should be considered. First, since the pediatric general population was excluded from the study, the economic burden could be estimated only for the adult population. The overall economic burden is expected to be much higher when the cost of obesity in pediatric population is also considered in the calculations. Second, given the lack of available data or local publications regarding the cost of presenteeism, which is defined as the loss in productivity related to illness, the presenteeism was neglected. The impact of obesity on workforce cost needs to be further investigated in terms of presenteeism in Türkiye. In this regard, it can be stated that costs regarding workforce were underestimated in this study. Third, it is known that the information on BMI in the SSI database is very limited and that obesity diagnosis is often not entered into the system. Fourth, obesity in the THSs is analyzed based on BMI values calculated using the self-reported height and weight measurements by the participants. However, studies conducted by OECD have shown that the prevalence of obesity is higher when BMI is calculated according to actual height and weight measurements [50]. Fifth, while the total number of patients diagnosed with obesity (ICD 10 E66.x) in 2021 is estimated to be 2,211,865, this seems to be a very low estimate for prevalent obesity cases in our country (approximately 30% prevalence and slightly over 0.4 million new cases with 2.3% annual increase rate for adult obesity), which is probably due to the fact that physicians do not enter obesity diagnosis codes into the SSI system, except for cases of 2nd and 3rd degree obesity that require medication or bariatric surgery. Accordingly, while our analysis indicates that obesity is responsible for 8.4% of total health expenditures and 15% of the contribution made by the reimbursement system, these amounts are calculated for 2.2 million obese individuals despite the fact that there are actually around 18 million obese individuals living in Türkiye, emphasizing that the burden of obesity is largely underestimated. In addition, since this study aimed to estimate the economic impact of obesity from the societal perspective, “out-of-pocket” expenses and private sector health expenditure were not considered in the cost analysis.

Obesity has both an important disease burden and an important economic burden. Besides being a burden itself, along with the accompanying comorbidities and complications obesity increases the health expenditures, causes the loss of workforce and the unintended consequences such as early retirement and premature death [51‒55]. Obesity is not an individual issue but rather a societal issue, necessitating an intervention with contribution of all related fields besides the field of health. It is obvious that this issue cannot be resolved with the interventions limited to health field per se. Obesity management requires a multi-dimensional approach involving psycho-socio-cultural, educational, political, economic, and communicational aspects of the disease. A great portion of total national cost of obesity is comprised by indirect costs, which forms a rationale for nationwide interventions toward prevention and reduction of obesity.

Obesity is a disease that has a progressive course, including an increase in the body weight at an early stage and then the emergence of problems related to increased body weight and the development of complications leading to life-threatening diseases. Prevention is far cheaper than the treatment. Early recognition and prompt management of the condition at the time of initial body weight increase is easier and cheaper. There is a need for implementation of emergency action plans to enable optimal utilization of the national health resources, by putting the improvement of health and preventive medicine services on the agenda.

The study was approved by the Ethics Committee of Baskent University (Data of Approval: 28.09.2022, Protocol No: KA22/394). Permission was obtained from the Turkish Statistical Institute for the use of micro dataset in our study (Permission No.: 789). The need for informed consent was waived by the Baskent University Ethics Committee.

L.T. has received honoraria for lectures or consultancy from Abbott, Amgen, AstraZeneca, Bayer, Daiichi Sankyo, Lilly, MSD, Novartis, Novo Nordisk, Sanofi, Pfizer, and Ultragenyx. V.D.Y. reports honoraria from Eli Lilly for providing a single advisory activity and from Novo Nordisk for providing educational sessions and attending advisory boards. A.S. reports honoraria for advisory board attendance from AstraZeneca, Novo Nordisk, Lilly, MSD, Novartis, Eczacıbasi, Roche Diagnostics, Daiichi Sankyo, and Sanovel, for principal investigator/director role in clinical trials from Novo Nordisk, Sanofi, and Novartis and fees for serving as a speaker from Bilim Ilac, Novo Nordisk, AstraZeneca, and Medtronic. B.G. and E.S.Y. are employees of Novo Nordisk Turkey. The other authors have no conflicts of interest to declare.

Novo Nordisk Turkey provided funding for the study. Novo Nordisk Turkey had no role in study design, selection and collection of literature data, decision to publish, or preparation of the manuscript.

E.O. and S.M. contributed to conception/design of the research. S.M., H.A., D.G.Y., T.K., M.S., A.S., L.T., B.G., E.S.Y., and V.D.Y. contributed to acquisition, analysis, and interpretation of the data and carried out investigation and project administration. S.M. and EO drafted the manuscript. S.M., D.G.Y., A.S., and V.D.Y critically revised the manuscript. All authors read and approved the final manuscript and agree to be fully accountable for ensuring the integrity and accuracy of the work.

Ergun Oksuz is deceased on January 18, 2023.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.