Learning Objectives
Discuss current knowledge of direct and indirect costs associated with obesity , including previous studies of variability between industries.
Summarize the new findings on differences in obesity -related direct and absenteeism/disability costs in US industries compared to the healthcare industry .
Identify industries with higher likelihood of high obesity -related costs and discuss the implications for weight management interventions in specific industries.
According to the World Health Organization (WHO), obesity is estimated as a body mass index (BMI) of 30.0 kg/m2 or greater and is divided into three classes: class I (30.0 ≤ BMI ≤ 34.9 kg/m2 ), class II (35.0 ≤ BMI ≤ 39.9 kg/m2 ), and class III (BMI ≥ 40.0 kg/m2 ).1 obesity among adults in the United States (US) has been rising steadily since the 1980s, with the age-standardized prevalence reported to be 39.6% in 2015 to 2016.2 obesity among the working population.3–5
Compared with normal BMI, obesity has been associated with higher direct healthcare costs 6–9 costs , including costs related to disability, absenteeism (absence from work, such as sick leave), and presenteeism (reduction in productivity while at work).10–13 obesity /overweight and its related comorbidities in 2016.14 obesity can be substantial for both affected employees and their employers.
We previously conducted a study evaluating the direct and indirect costs of obesity among employees of eight major US industries and found considerable variability in economic burden according to industry of employment.15 obesity -related costs were observed across industries, in-depth comparisons between industries were not analyzed. Therefore, this follow-up study was conducted to provide a deeper analysis of the industry-specific factors affecting the economic burden of obesity among the working population, with a specific focus on the incremental costs associated with the three classes of obesity among employees of the healthcare industry . Since healthcare workers are typically health-educated and may theoretically be more health-aware regarding the negative effects of chronic diseases like obesity , we hypothesized that the associated costs may be lower in this subpopulation. Therefore, we compared the economic burden of employees in the healthcare industry to that of other industries to identify specific subpopulations that may be at higher risk of obesity -related costs . Additionally, employment industry was explored as a predictor of high healthcare costs . Insights on at-risk employees are needed to implement targeted employer-based obesity interventions that can effectively promote weight management according to the goals and needs of each employee subpopulation.
METHODS
Data Source
This study was conducted using data from the Optum Health Reporting and Insights employer claims database (employer database) from January 1, 2010 to March 31, 2017. The employer database includes administrative claims for over 19.1 million privately insured individuals covered by 84 self-insured Fortune 500 companies in the US for services provided from 1999 through the first quarter of 2017. In addition, for 42 of the 84 companies, some work loss data for employees (approximately 4.4 million lives) are available, including short- and long-term disability claims. Importantly, the employer database contains no information to measure the indirect impact costs of obesity on presenteeism and workers’ compensation, factors considered to be major contributors to total costs .12
Data were de-identified and complied with the Health Insurance Portability and Accountability Act (HIPAA). Therefore, no reviews by an institutional review board were required.
Study Design and Cohorts
A retrospective longitudinal cohort study was conducted. Study design and patient inclusion and exclusion criteria were described in detail previously.15 obesity class I (employees with BMI between 30.0 and 34.9 kg/m2 ); (2) obesity class II (employees with BMI between 35.0 and 39.9 kg/m2 ); (3) obesity class III (employees with BMI of 40 kg/m2 or over); and (4) reference cohort, as a proxy of a normal-weight population, consisting of a randomly selected sample of employees without overweight, obesity , or underweight BMI codes and without overweight or obesity term International Classification of Diseases (ICD) diagnosis codes. Since normal BMI diagnosis codes are not generally used by healthcare providers except in conjunction with another underlying condition, selection of employees with normal BMI codes would have biased the reference cohort towards unhealthier individuals.
Cohorts were further stratified by employees’ industries of employment as reported in the employer database: healthcare; transportation; manufacturing and energy; retail and consumer goods; government, education, and religious services (GERS); technology; finance and insurance; and other (ie, food services, entertainment, and other service industries).
The index date was defined as the first claim with a diagnosis code for BMI on or after January 1, 2010 (obesity cohorts) or a randomly selected date within study eligibility (reference cohort). Baseline characteristics were evaluated in the 12 months pre-index date (baseline period), while outcomes were evaluated from the index date up to the end of health plan eligibility or end of data availability on March 31, 2017, whichever was earliest (observation period).
Study Outcomes
Study outcomes measured during the observation period included direct healthcare costs (among all employees) and medical-related absenteeism and short-term and long-term disability costs (among employees with work loss information). Direct healthcare costs were obtained from claims and included pharmacy costs and medical costs , including hospitalization, emergency department (ED), outpatient, home healthcare, and other (ie, ambulance, dentist, laboratory, and everything not previously identified) costs . Sick leave (absenteeism) costs were calculated from employees’ resource utilization multiplied by the workers’ recorded wages. Each hospitalization accounted for 8 hours of absenteeism from work (1 work day) and each ED, outpatient, and other visit accounted for 4 hours of absenteeism (1/2 a work day). Five-sevenths of the total sick leave hours were used in the calculation to account for weekend visits that did not result in work loss costs .16 costs were calculated from short- and long-term disability data from claims.
To explore the relationship between the type of industry and the incremental economic impact of obesity , direct, medical-related absenteeism, and disability costs were evaluated for each of the employment industries and compared with the costs of the healthcare industry . The healthcare industry was chosen as a focus and used as a reference because of the presumed propensity for its employees to be health-educated and potentially more health-conscious regarding the negative effects of obesity .
Statistical Analysis
All analyses were conducted using SAS Enterprise Guide software Version 7.1 (SAS Institute, Cary, NC) and a two-sided alpha error of 0.05 was used to declare statistical significance. To minimize the potential confounding between employees with obesity and the reference cohort among healthcare employees, the inverse probability weighting (IPW) approach was used to evaluate the impact of obesity based on the propensity score (PS).17 obesity cohorts, and 1/(1–PS) for the reference cohort. Differences in baseline variables between cohorts were assessed using standardized differences (more than or equal to 10% indicating imbalance between cohorts).
Direct healthcare, absenteeism, and short-term and long-term disability costs were reported as mean cost per-patient-per-year (PPPY). Costs were adjusted for inflation using the US consumer price index for medical services from the Bureau of Labor Statistics from the US Department of Labor, and are reported in 2018 US dollars. Similarly, the wages of employees with work loss data were adjusted using the US hourly compensation index and are reported in 2018 US dollars.
Among healthcare employees, direct, medical-related absenteeism, and disability costs were compared for each of the weighted cohorts to the reference cohort using cost differences. Because cost data have positive values and follow a non-normal distribution, non-parametric bootstrap (B = 499) procedures were used to estimate 95% confidence intervals (CI) and P values of cost differences.18 obesity cohorts were compared with employees of each type of industry within each obesity class for cost differences of total direct, medical-related absenteeism, and disability costs . Cost differences were adjusted using linear regression models including baseline covariates of age, sex, geographical region, type of healthcare plan, Quan-CCI, and comorbidities (overall prevalence more than or equal to 5%).
Multivariate generalized linear models (GLM) with a logit link for binary outcomes were used to estimate the association between employment industry and high direct healthcare, medical-related absenteeism, and disability costs in employees with obesity . Employees with yearly direct healthcare costs or medical-related absenteeism and disability costs at the 80th percentile and above were classified as high-cost employees, which is based on previous literature reporting that more than 80% of all healthcare spending is incurred by the top 20% of spenders in the US population.19,20 obesity cohort.
RESULTS
Baseline Characteristics
A total of 39,196 employees were included in the study. The mean age of employees was 43 to 48 years across all cohorts (Table 1 ). The proportion of female employees was 36.3% in the reference cohort, 39.3% in the obesity class I cohort, 47.3% in the class II cohort, and 57.7% in the class III cohort. Included employees had good geographical representation across all four census regions (Table 1 ). Comorbidity burden, as measured by Quan-CCI, was higher in the obesity cohorts than the reference cohort (standardized difference more than or equal to 10%), with a trend of increasing mean Quan-CCI values with increasing obesity class (range of 0.27 to 0.43; Table 1 ).
TABLE 1: Baseline Demographic and Clinical Characteristics of All Employees in the Obesity and Reference Cohorts
Incremental Costs in the Healthcare Industry
Among employees of the healthcare industry , direct healthcare costs were significantly higher in the class III (BMI more than or equal to 40) obesity cohort compared with the reference cohort (Fig. 1 A). The cost differences were $605 (95% CI = –1388; 2365) PPPY between the obesity class I and reference cohorts, $1962 (–178; 3708) PPPY between class II and reference, and $8438 (6027; 10,595) PPPY between class III and reference (P < 0.05).
FIGURE 1: Adjusted direct healthcare, medical-related absenteeism, and disability cost differences for employees of the healthcare industry in the obesity and reference cohorts. ∗ P < 0.05. CI, confidence interval; ED, emergency department; PPPY, per-patient-per-year.
Medical-related absenteeism and disability costs did not significantly differ between the obesity and reference cohorts (Fig. 1 B).
Direct, Medical-Related Absenteeism, and Disability Costs Across Industries
Across each of the eight industries studied, direct healthcare costs increased with increasing BMI: mean direct healthcare costs PPPY across all eight industries were $4853 in the reference cohort, $7152 in the obesity class I cohort, $9703 in the obesity class II cohort, and $19,169 in the obesity class III cohort (Fig. 2 ).
FIGURE 2: Direct healthcare costs for the obesity and reference cohorts among all employees by industry. GERS, government, education, and religious services; PPPY, per-patient-per-year.
As with direct costs , medical-related absenteeism and disability costs varied across each of the eight industries studied, with employees with class III obesity accruing higher expenses than employees in the reference cohort: mean medical-related absenteeism and disability costs PPPY across all eight industries were $1363 in the reference group, $2494 in the obesity class I group, $2475 in the obesity class II group, and $3912 in the obesity class III group (Fig. 3 ).
FIGURE 3: Medical-related absenteeism and disability costs for the obesity and reference cohorts among all employees with work loss coverage by industry. GERS, government, education, and religious services; PPPY, per-patient-per-year.
Incremental Cost Comparisons With the Healthcare Industry
Compared with the healthcare industry , direct healthcare costs by obesity class were higher in several other US industries (Fig. 4 ). Direct healthcare costs among employees with obesity were significantly higher in the GERS industry (adjusted cost difference [95% CI] = $5630 [1506; 9754] PPPY for class III), other industry (ie, food, entertainment, and other services; $3543 [722; 6364] PPPY for class II and $4922 [1325; 8518] PPPY for class III), and technology industry ($4010 [845; 7175] PPPY for class II and $4321 [201; 8,440] PPPY for class III) compared with the healthcare industry (all P < 0.05).
FIGURE 4: Adjusted direct cost differences among all employees with obesity compared with employees of the healthcare industry 1 . ∗ P < 0.05. 1. Healthcare employees in the obesity class I (n = 781), obesity class II (n = 484), and obesity class III (n = 948) cohorts are compared with employees of each type of industry within each obesity class for cost differences of total direct healthcare cost. CI, confidence interval; GERS, government, education, and religious services; PPPY, per-patient-per-year.
Compared with the healthcare industry , medical-related absenteeism and disability costs by obesity class were higher in several other US industries (Fig. 5 ). Medical-related absenteeism and disability costs among employees with obesity were significantly higher in the other industry (ie, food, entertainment, and other services; adjusted cost difference [95% CI] = $1658 [931; 2385] PPPY for class I, $2349 [1381; 3317] PPPY for class II, and $2249 [125; 4374] PPPY for class III), technology industry ($964 [173; 1756] PPPY for class I and $1144 [75; 2213] PPPY for class II), manufacturing and energy industry ($3758 [1309; 6207] PPPY for class III), and finance and insurance industry ($2933 [336; 5531] PPPY for class III) compared with the healthcare industry (all P < 0.05).
FIGURE 5: Adjusted medical-related absenteeism and disability cost differences among all employees with obesity and work loss coverage compared with employees of the healthcare industry 1 . ∗ P < 0.05. 1. Healthcare employees with work loss coverage in the obesity class I (n = 439), obesity class II (n = 297), and obesity class III (n = 597) cohorts are compared with employees of each type of industry within each obesity class for cost differences of total medical-related absenteeism and disability cost. CI, confidence interval; GERS, government, education, and religious services; PPPY, per-patient-per-year.
Predictors of High Costs Compared With the Healthcare Industry
Among all employees with obesity , GERS employees were at significantly higher odds of having high direct healthcare costs at the 80th percentile and above compared with employees of the healthcare industry (odds ratio [OR; 95% CI] = 2.20 [1.87; 2.59]; P < 0.05; Fig. 6 ). Employees of the retail stores and consumer goods and finance and insurance industries were at significantly lower odds of having high direct healthcare costs at the 80th percentile and above compared with employees of the healthcare industry (OR [95% CI] = 0.75 [0.65; 0.87] and 0.69 [0.60; 0.79], respectively; P < 0.05; Fig. 6 ).
FIGURE 6: Industries as predictors of high direct healthcare costs among all employees with obesity . ∗ P < 0.05. CI, confidence interval; GERS, government, education, and religious services; OR, odds ratio; P80, 80th percentile.
Among all employees with obesity and work loss coverage, employees of almost all other industries were at significantly higher odds of having high medical-related absenteeism and disability costs at the 80th percentile and above compared with employees of the healthcare industry (Fig. 7 ). The highest odds were associated with employees with obesity in the technology (OR [95% CI] = 2.01 [1.64; 2.47]; P < 0.05) and finance and insurance (OR [95% CI] = 2.11 [1.72; 2.58]; P < 0.05) industries relative to the healthcare industry (Fig. 7 ).
FIGURE 7: Industries as predictors of high medical-related absenteeism and disability costs among all employees with obesity and work loss coverage. ∗ P < 0.05. CI, confidence interval; GERS, government, education, and religious services; OR, odds ratio; P80, 80th percentile.
DISCUSSION
This retrospective longitudinal cohort study confirmed the variation observed previously in the economic burden of obesity across employment industries in the US,15 obesity -related costs . Relative to the healthcare industry , direct, medical-related absenteeism, and disability costs for employees with class I, II, and III obesity were higher in several other US industries, especially the technology, other (i.e., food services, entertainment, and other service industries), and GERS industries. Moreover, GERS employees were significantly most likely to be among the top 20% of employees with the highest healthcare costs .
Many other industries studied had higher direct healthcare costs compared with the healthcare industry , suggesting the existence of healthcare-specific factors associated with low direct costs . One can hypothesize that those with a formal medical education may feel capable dealing with obesity -related health ailments on their own or may feel shameful for needing external medical support, thus not using healthcare resources and not incurring the associated direct costs . Indeed, Taber et al21 22 23,24 costs of obesity among employees of the healthcare industry ; therefore, additional research is needed to further explore the factors associated with low costs in this population.
Similar to direct costs , medical-related absenteeism and disability costs were also higher in many other industries compared with healthcare, which again, may suggest the existence of factors specific to employment in the healthcare industry associated with low work loss-related costs . While nationally representative evaluations of absenteeism among healthcare workers compared with other employment industries in the US are scarce, several questionnaire-based studies in Europe have found that the rate of sickness-related absenteeism among physicians is low.25–27 28,29 30,31 26–28 32 28 costs observed in healthcare employees in the present study, it is unclear if the low absenteeism rates and related costs would be offset by a higher prevalence of presenteeism.
Among all employees with obesity , GERS employees were significantly more likely to be among the top 20% of patients with the highest direct healthcare costs compared with employees of the healthcare industry . Public sector health plans are typically more generous than private sector plans despite public sector employees contributing less to the higher premium costs .33 costs .34 35 costs in the present study, it is clear that GERS employees with obesity used considerable healthcare resources and incurred the associated costs . Based on these patterns of use, it is reasonable to believe that the implementation of accessible, comprehensive, employer-based resources targeted at weight management would attract comparable usage in employees, thus leading to reductions in obesity and its associated costs .
While real-world evidence is currently lacking, simulation studies have shown potential cost-effectiveness with the implementation of weight management interventions, such as national nutrition standards and sugar reduction policies, as well as diet and exercise programs.36–38 economic burden in the US, with obesity being a main target.39 costs for obesity alone.39 economic burden . Indeed, in Australia, where policy makers have established rigorous modeling strategies to assess the economic effects of obesity prevention programs and policies at a national level, these cost-effectiveness studies have estimated that regulatory and policy-based interventions may produce more health benefits and cost savings in the long term than program-based interventions.40 economic burden of obesity .
Limitations
The present study is subject to several important limitations. First, only medical-related absenteeism and disability costs were reported in this study. Additional indirect costs , in the form of presenteeism and workers’ compensation, were not included, leading to a significant underestimation of the total economic burden associated with obesity . This was due to a lack of available data to model into these analyses, highlighting a need for more robust and comprehensive employee datasets that provide full indirect cost data. Consequently, this underestimation of costs could have been a contributing factor towards the cost differences observed between industries. Second, the cost difference analyses among healthcare employees were adjusted for baseline comorbidities, including any obesity -related comorbidities that may have already developed by the index date, given the chronic nature of obesity and its comorbidities. Therefore, only a partial effect of obesity may have been captured using this conservative approach, resulting in potentially underestimated cost differences between reference and obesity cohorts among healthcare employees. Third, there may have been a selection bias in the obesity cohorts towards unhealthier employees who used more healthcare services, thus leading to an obesity BMI diagnosis code and inclusion in the study. Lastly, the BMIs of employees in the reference cohort were unknown. Employees in the reference cohort had no claims for BMI in obesity or overweight categories, but the potential presence of undiagnosed obesity could have led to an underestimation of the burden of obesity .
CONCLUSIONS
This retrospective study confirms the varying cost burden of obesity in the US according to industry of employment. Employees of specific industries, such as GERS, food/entertainment services, and technology, are at higher risk of incurring high obesity -related healthcare costs and may therefore benefit most from targeted, employer-led weight management approaches that encompass a comprehensive range of diet-, medication-, and surgery-based interventions. Finally, employees of the healthcare industry generally incurred lower obesity -related direct, absenteeism, and disability costs compared with other industries.
Information regarding the industry-specific trends in obesity -related spending gained from this study will contribute towards the implementation and tailoring of employer-based weight management programs according to the specific needs of each industry subpopulation in order to achieve effective and sustained improvements in employee health.
Acknowledgments
Medical writing assistance was provided by Christine Tam, an employee of Groupe d’analyse, Ltée, which received research grants from Novo Nordisk Inc. to conduct the present study.
Previous presentation: Part of the material in this manuscript was presented at the American Occupational Health Conference (AOHC) 2019 on April 28–May 1 in Anaheim, California, USA.
REFERENCES
2. Hales CM, Fryar CD, Carroll MD, Freedman DS, Ogden CL. Trends in
obesity and severe
obesity prevalence in US youth and adults by sex and age, 2007–2008 to 2015–2016.
JAMA 2018; 319:1723–1725.
3. Caban AJ, Lee DJ, Fleming LE, Gomez-Marin O, LeBlanc W, Pitman T.
Obesity in US workers: the National Health Interview Survey, 1986 to 2002.
Am J Public Health 2005; 95:1614–1622.
4. Gu JK, Charles LE, Bang KM, et al. Prevalence of
obesity by occupation among US workers: the National Health Interview Survey 2004–2011.
J Occup Environ Med 2014; 56:516–528.
5. Luckhaupt SE, Cohen MA, Li J, Calvert GM. Prevalence of
obesity among U.S. workers and associations with occupational factors.
Am J Prev Med 2014; 46:237–248.
6. Andreyeva T, Sturm R, Ringel JS. Moderate and severe
obesity have large differences in health care
costs .
Obes Res 2004; 12:1936–1943.
7. Kamble PS, Hayden J, Collins J, et al. Association of
obesity with healthcare resource utilization and
costs in a commercial population.
Curr Med Res Opin 2018; 34:1335–1343.
8. Musich S, MacLeod S, Bhattarai GR, et al. The impact of
obesity on health care utilization and expenditures in a medicare supplement population.
Gerontol Geriatr Med 2016; 2:2333721415622004.
9. Suehs BT, Kamble P, Huang J, et al. Association of
obesity with healthcare utilization and
costs in a Medicare population.
Curr Med Res Opin 2017; 33:2173–2180.
10. Andreyeva T, Luedicke J, Wang YC. State-level estimates of
obesity -attributable
costs of absenteeism.
J Occup Environ Med 2014; 56:1120.
11. Finkelstein EA, DiBonaventura M, Burgess SM, Hale BC. The
costs of
obesity in the workplace.
J Occup Environ Med 2010; 52:971–976.
12. Goettler A, Grosse A, Sonntag D. Productivity loss due to overweight and
obesity : a systematic review of indirect
costs .
BMJ Open 2017; 7:e014632.
13. Kudel I, Huang JC, Ganguly R. Impact of
obesity on work productivity in different US occupations: analysis of the National Health and Wellness Survey 2014 to 2015.
J Occup Environ Med 2018; 60:6–11.
14. Waters H, Graf M. America's
Obesity Crisis. The Health and Economic
Costs of Excess Weight. Santa Monica, California: Milken Institute; 2018.
15. Ramasamy A, Laliberte F, Aktavoukian SA, et al. Direct and indirect cost of
obesity among the privately insured in the united states: a focus on the impact by type of industry.
J Occup Environ Med 2019; 62:877–886.
16. Jensen JT, Lefebvre P, Laliberte F, et al. Cost burden and treatment patterns associated with management of heavy menstrual bleeding.
J Womens Health (Larchmt) 2012; 21:539–547.
17. Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies.
Multivariate Behav Res 2011; 46:399–424.
18. Afifi AA, Kotlerman JB, Ettner SL, Cowan M. Methods for improving regression analysis for skewed continuous or counted responses.
Annu Rev Public Health 2007; 28:95–111.
19. LaMori J, Tandon N, Laliberte F, et al. Predictors of high healthcare resource utilization and liver disease progression among patients with chronic hepatitis C.
J Med Econ 2016; 19:364–373.
20. National Institute for Health Care Management (NIHCM). The concentration of U.S. health care spending; 2017. Available at:
https://www.nihcm.org/categories/concentration-of-us-health-care-spending . Accessed February 8, 2019.
21. Taber JM, Leyva B, Persoskie A. Why do people avoid medical care? A qualitative study using national data.
J Gen Intern Med 2015; 30:290–297.
22. Gross CP, Mead LA, Ford DE, Klag MJ. Physician, heal Thyself? Regular source of care and use of preventive health services among physicians.
Arch Intern Med 2000; 160:3209–3214.
23. Collier R. The “physician personality” and other factors in physician health.
CMAJ 2012; 184:1980.
24. Lemaire JB, Wallace JE. How physicians identify with predetermined personalities and links to perceived performance and wellness outcomes: a cross-sectional study.
BMC Health Serv Res 2014; 14:616.
25. Kivimaki M, Sutinen R, Elovainio M, et al. Sickness absence in hospital physicians: 2 year follow up study on determinants.
Occup Environ Med 2001; 58:361–366.
26. McKevitt C, Morgan M, Dundas R, Holland WW. Sickness absence and ’working through’ illness: a comparison of two professional groups.
J Public Health Med 1997; 19:295–300.
27. Rosvold EO, Bjertness E. Physicians who do not take sick leave: hazardous heroes?
Scand J Public Health 2001; 29:71–75.
28. Jena AB, Meltzer DO, Press VG, Arora VM. Why physicians work when sick.
Arch Intern Med 2012; 172:1107–1108.
29. Rhodes SM, Collins SK. The organizational impact of presenteeism.
Radiol Manage 2015; 37:27–32. quiz 33–24.
30. Zuvekas SH, Cohen JW. Fee-For-Service, while much maligned, remains the dominant payment method for physician visits.
Health Aff (Millwood) 2016; 35:411–414.
31. Merritt Hawkins. Review of Physician and Advanced Practitioner Recruiting Incentives; 2018.
32. Aronsson G, Gustafsson K. Sickness presenteeism: prevalence, attendance-pressure factors, and an outline of a model for research.
J Occup Environ Med 2005; 47:958–966.
33. Zawacki AM, Vistnes JP, Buchmueller TC. Why are employer-sponsored health insurance premiums higher in the public sector than in the private sector?
Monthly Labor Rev 2018; Available at:
https://www.bls.gov/opub/mlr/2018/article/employer-sponsored-health-insurance-premiums.htm . Accessed April 24, 2019.
34. Costrell RM, Dean J. The rising cost of teachers’ health care.
Educ Next 2013; 13:66–72.
35. Mehta N, Ni J, Srinivasan K, Sun B. A dynamic model of health insurance choices and healthcare consumption decisions.
Market Sci 2017; 36:1–23.
36. Amies-Cull B, Briggs ADM, Scarborough P. Estimating the potential impact of the UK government's sugar reduction programme on child and adult health: modelling study.
BMJ 2019; 365:l1417.
37. Forster M, Veerman JL, Barendregt JJ, Vos T. Cost-effectiveness of diet and exercise interventions to reduce overweight and
obesity .
Int J Obes (Lond) 2011; 35:1071–1078.
38. Gortmaker SL, Wang YC, Long MW, et al. Three interventions that reduce childhood
obesity are projected to save more than they cost to implement.
Health Aff (Millwood) 2015; 34:1932–1939.
39. Soler R, Orenstein D, Honeycutt A, et al. Community-based interventions to decrease
obesity and tobacco exposure and reduce health care
costs : outcome estimates from communities putting prevention to work for 2010–2020.
Prev Chronic Dis 2016; 13:E47.
40. Assessing Cost-effectiveness of
Obesity Prevention Policies in Australia. Melbourne, Australia: Deakin Health Economics; 2018.
