Background

PLoS ONE

plos

plosone

PLOS ONE

1932-6203

Public Library of Science

San Francisco, CA USA

10.1371/journal.pone.0247307

PONE-D-20-36222

Research Article

Biology and life sciences

Physiology

Physiological parameters

Body weight

Body mass index

Biology and life sciences

Physiology

Physiological parameters

Body weight

Obesity

Biology and life sciences

Physiology

Physiological parameters

Body weight

Obesity

Childhood obesity

Biology and life sciences

Physiology

Physiological parameters

Body weight

Social sciences

Economics

Health economics

Medicine and health sciences

Health care

Health economics

Biology and life sciences

Physiology

Physiological parameters

Body weight

Overweight

Medicine and health sciences

Pediatrics

People and places

Population groupings

Age groups

Adults

Association of body mass index with health care expenditures in the United States by age and sex

BMI-related health care expenditures in the US

https://orcid.org/0000-0002-4007-2207

Ward

Zachary J.

Conceptualization Data curation Formal analysis Methodology Software Writing – original draft Writing – review & editing ¹ *

https://orcid.org/0000-0003-3098-3119

Bleich

Sara N.

Conceptualization Writing – review & editing ² Long

Michael W.

Conceptualization Writing – review & editing ³ Gortmaker

Steven L.

Conceptualization Writing – review & editing ⁴

Center for Health Decision Science, Harvard T. H. Chan School of Public Health, Boston, MA, United States of America

Department of Health Policy and Management, Harvard T. H. Chan School of Public Health, Boston, MA, United States of America

Department of Prevention and Community Health, Milken Institute School of Public Health, George Washington University, Washington, DC, United States of America

Department of Social and Behavioral Sciences, Harvard T. H. Chan School of Public Health, Boston, MA, United States of America

Siegel

Robert

Editor

Cincinnati Children’s, UNITED STATES

The authors have declared that no competing interests exist.

* E-mail: zward@hsph.harvard.edu

24 3 2021

2021

16 3

e0247307

17 11 2020 4 2 2021

2021

Ward et al

This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Background

Estimates of health care costs associated with excess weight are needed to inform the development of cost-effective obesity prevention efforts. However, commonly used cost estimates are not sensitive to changes in weight across the entire body mass index (BMI) distribution as they are often based on discrete BMI categories.

Methods

We estimated continuous BMI-related health care expenditures using data from the Medical Expenditure Panel Survey (MEPS) 2011–2016 for 175,726 respondents. We adjusted BMI for self-report bias using data from the National Health and Nutrition Examination Survey (NHANES) 2011–2016, and controlled for potential confounding between BMI and medical expenditures using a two-part model. Costs are reported in $US 2019.

Results

We found a J-shaped curve of medical expenditures by BMI, with higher costs for females and the lowest expenditures occurring at a BMI of 20.5 for adult females and 23.5 for adult males. Over 30 units of BMI, each one-unit BMI increase was associated with an additional cost of $253 (95% CI $167-$347) per person. Among adults, obesity was associated with $1,861 (95% CI $1,656-$2,053) excess annual medical costs per person, accounting for $172.74 billion (95% CI $153.70-$190.61) of annual expenditures. Severe obesity was associated with excess costs of $3,097 (95% CI $2,777-$3,413) per adult. Among children, obesity was associated with $116 (95% CI $14-$201) excess costs per person and $1.32 billion (95% CI $0.16-$2.29) of medical spending, with severe obesity associated with $310 (95% CI $124-$474) excess costs per child.

Conclusions

Higher health care costs are associated with excess body weight across a broad range of ages and BMI levels, and are especially high for people with severe obesity. These findings highlight the importance of promoting a healthy weight for the entire population while also targeting efforts to prevent extreme weight gain over the life course.

The JPB Foundation

Grant No. 1085

http://dx.doi.org/10.13039/100000002

National Institutes of Health

R01HL146625

Centers for Disease Control and Prevention

U48DP006376

All of the authors received support from The JPB Foundation (Grant No. 1085). ZJW, SNB, and SLG were supported by the National Institutes of Health (Grant No. R01HL146625). SNB and SLG were supported by the Centers for Disease Control and Prevention (CDC) (Grant No. U48DP006376). This work is solely the responsibility of the authors and does not represent official views of the CDC or other agencies. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data Availability

Data from the Medical Expenditure Panel Survey (MEPS) are available at: https://www.meps.ahrq.gov/mepsweb/, and data from the National Health and Nutrition Examination Survey (NHANES) are available at: http://www.cdc.gov/nchs/nhanes.htm. Estimates of age/sex/BMI-specific annual medical costs are available at: https://doi.org/10.7910/DVN/872OW1.

Introduction

Seven out of ten adults and three out of ten children in the United States currently have overweight or obesity [1, 2], and the prevalence continues to rise, with half of US adults projected to have obesity by 2030 [3], and nearly 60% of today’s children predicted to have obesity by age 35 [4]. Excess body weight is associated with a wide array of comorbidities and premature mortality [5], and higher health care costs [6–13], which are expected to increase as population body mass index (BMI) continues to rise in the United States [13].

Accurate estimates of excess weight-related health care costs are necessary to evaluate the cost-effectiveness of policies and programs aimed at helping to reverse the obesity epidemic and promote a healthy weight across a range of ages and BMI levels [14, 15]. However, commonly used cost estimates are not sensitive to changes in weight across the entire BMI distribution as they are often based on discrete categories, such as binary classification (obesity vs non-obesity) or BMI categories (e.g. moderate vs severe obesity) [12, 13]. Using discrete categories likely underestimates the health care cost impact of changes in population BMI, as only changes in weight that cross specific category thresholds are accounted for, therefore ignoring changes within categories at all other parts of the BMI distribution. In contrast, estimating continuous BMI-related costs provides a more accurate and flexible approach as it reflects the entire BMI distribution and does not rely on specific category thresholds.

To address key gaps in the literature, in this paper we estimate continuous BMI-specific medical costs by age and sex, and provide updated estimates of the medical costs attributable to obesity using recent data.

Methods Data

We used publicly available, de-identified data from the Medical Expenditure Panel Survey (MEPS) 2011–2016. We harmonized variable definitions across years and adjusted total expenditures to $US 2019 using the Personal Consumption Expenditures (Health) index [16]. After excluding pregnant women and respondents with missing variables of interest, our pooled dataset contained 175,726 respondents– 139,143 adults (aged 20 and older) and 36,583 children (aged 6 to 19). BMI was not available for children younger than 6. See S1 File, Section 1 for details on dataset harmonization, exclusion criteria, and respondent characteristics.

Adjustment for self-report bias

We adjusted reported BMI in MEPS to correct for self-report bias that leads to underestimates of obesity prevalence [17, 18]. We used a semi-parametric method [3] to adjust the distribution of self-reported BMI to match nationally-representative, measured data from the National Health and Nutrition Examination Survey (NHANES). Specifically, we used cubic splines to estimate the magnitude of self-report bias by BMI quantile and adjust the self-reported BMI in MEPS by age group. Using this approach our sex-specific distributions of adjusted BMI were statistically similar to NHANES (p>0.05). See S1 File, Section 2 for details.

Expenditure standardization

To adjust for potential confounding of the relationship between BMI and medical expenditures, we standardized respondents’ expenditures to be representative of a synthetic, ‘average’ population, thus controlling for the effects of other salient factors (e.g. smoking, insurance coverage, etc.). Using a well-established approach [7, 9, 11], we fit a two-part regression model to predict medical expenditures. The first part of the model fit a logistic regression to predict the probability of non-zero expenditures. The second part of the model fit a linear regression of the log expenditure given positive expenditure. The two parts were then multiplied together to yield the full model.

Similar to previous analyses [9, 11], we controlled for the following variables: BMI (continuous), year (continuous), geographical region (Northeast, Midwest, South, West), age (continuous), sex, race/ethnicity (White, Black, Hispanic, American Indian/Alaska Native, Asian/Native Hawaiian/Pacific Islander, multiple races), marital status (married, widowed, divorced, separated, never married), education (less than high school, some high school, GED or high school diploma, some college, college graduate, graduate school, unknown), smoking status (yes/no), poverty level (continuous), and insurance coverage (private, TRICARE, Medicare, Medicaid, other public A, other public B, none). We fit separate models for children and adults, and did not control for education or marital status when adjusting expenditures for children. Continuous variables were modeled as cubic polynomials for greater flexibility, and were standardized to reduce multicollinearity and improve numerical stability [19].

We used ridge regression [20, 21] to help guard against over-fitting to extreme values that can occur given the highly skewed nature of health expenditures [22]. Using the fitted two-part model we then adjusted each respondent’s probability and level of total expenditure to be representative of a standardized individual. We adjusted for all variables except BMI (and age when fitting bivariate models–see below), thus controlling expenditures for other salient factors. See S1 File, Section 3 for details.

Predicted expenditures by BMI

We then used generalized additive models (GAMs) [23] to estimate the relationship between log BMI and the log of the standardized expenditures using the same two-part model described above. We fit two different types of models: a univariate model (cubic smoothing splines) with log BMI as a continuous predictor, and a bivariate model [24] with both age and log BMI as continuous predictors to capture the interaction between age and BMI (see S1 File, Section 4.1 for details). We predicted adult expenditures for BMI values between 10 and 80. For children we predicted expenditures for BMI z-scores between -3 and +3. Although we fitted our model using BMI for both children and adults, we present the univariate predicted expenditures for children in terms of BMI z-score for visual ease, as the overweight and obesity thresholds are defined by BMI z-score for children. With these predicted expenditures we also estimated costs by binary obesity status: non-obesity vs obesity, and by BMI category: underweight, normal weight, overweight, moderate obesity, severe obesity (see S1 File, Section 4.2 for category definitions).

To estimate population-level excess costs we assumed that all respondents would instead follow the BMI distribution observed in the reference category (i.e. weight category with lowest costs) and re-estimated the total costs. Excess costs were then calculated as the difference between the current predicted costs and the predicted costs for the reference weight population. We scaled the per-person excess costs to the population-level using BMI category prevalence estimates from NHANES 2011–2016 and 2019 population estimates of the civilian, non-institutionalized population [25].

We also estimated how categorical expenditures change with age by fitting smooth splines to the predicted costs within each BMI category by age. Because age is top-coded at 85 years old in MEPS, we predicted costs from ages 6 to 85.

Model uncertainty

To estimate confidence intervals for all results we bootstrapped the MEPS dataset 1,000 times, taking into account the complex survey structure, and re-estimated all models described above. We calculated 95% confidence intervals (CIs) as the 2.5 and 97.5 percentiles of the bootstrapped results (see S1 File, Section 5 for details). All analyses were performed in R (version 3.6.1).

Results Adults

We found a J-shaped curve of medical expenditures by BMI (Fig 1A), with higher costs in general for females and the lowest expenditures occurring at a BMI of 20.5 for females and 23.5 for males. Above a BMI of 30, predicted costs continued to increase linearly, with each one-unit increase in BMI associated with an additional cost of $253 (95% CI $167-$347) per person on average. Obesity was associated with $1,861 (95% CI $1,656-$2,053) excess annual medical costs per person, accounting for nearly $173 billion (95% CI $153.70-$190.61) of annual spending in the US (Table 1). Most of these costs are from individuals with severe obesity, who have excess annual costs of over $3,000 (95% CI $2,777-$3,413). We also found that having overweight is associated with over $600 per person in excess costs (95% CI $503-$756), contributing to $50 billion (95% CI $40.64-$61.12) in medical spending per year.

10.1371/journal.pone.0247307.g001

Fig 1 Estimated BMI-related medical expenditures, children and adults.

Estimated expenditures are controlled for potential confounding variables. Shaded areas represent 95% confidence intervals.

10.1371/journal.pone.0247307.t001

Table 1 Total and excess annual medical expenditures by BMI category ($US 2019).

Obesity Status (BMI range)	Total Cost Per Person^a (95% CI)	Excess Cost Per Person^b (95% CI)	Excess Cost (Billions)–Population-level^c (95% CI)
Children (6–19)^d
Non-Obesity (BMI < 95%_ile)	1,871 (1,826–1,918)	Reference	Reference
Obesity (BMI ≥ 95%_ile)	1,987 (1,893–2,072)	116 (14–201)	1.32 (0.16–2.29)
BMI Category
Underweight (BMI < 5%_ile)	1,913 (1,806–2,039)	41 (-59-161)	0.09 (-0.13–0.35)
Normal weight (5%_ile ≤ BMI < 85%_ile)	1,873 (1,821–1,925)	Reference	Reference
Overweight (85%_ile ≤ BMI < 95%_ile)	1,852 (1,794–1,916)	-21 (-84-45)	-0.21 (-0.83–0.44)
Moderate Obesity (95%_ile ≤ BMI < 120% x 95%_ile)	1,882 (1,803–1,957)	9 (-83-96)	0.07 (-0.61–0.70)
Severe Obesity (BMI ≥ 120% x 95%_ile)	2,183 (2,013–2,327)	310 (124–474)	1.27 (0.51–1.94)
Adults (20+)
Non-Obesity (BMI < 30)	4,525 (4,450–4,616)	Reference	Reference
Obesity (BMI ≥ 30)	6,385 (6,221–6,558)	1,861 (1,656–2,053)	172.74 (153.70–190.61)
BMI Category
Underweight (BMI < 18.5)	4,419 (3,970–4,921)	228 (-201-721)	0.85 (-0.75–2.68)
Normal weight (18.5 ≤ BMI < 25)	4,191 (4,092–4,306)	Reference	Reference
Overweight (25 ≤ BMI < 30)	4,812 (4,716–4,936)	621 (503–756)	50.19 (40.64–61.12)
Moderate Obesity (30 ≤ BMI < 35)	5,672 (5,548–5,808)	1,480 (1,305–1,650)	77.03 (67.91–85.83)
Severe Obesity (BMI ≥ 35)	7,288 (7,002–7,594)	3,097 (2,777–3,413)	126.39 (113.35–139.29)

^a Mean of predicted costs for respondents in each BMI category, controlling for age, sex, and other covariates in the two-part model.

^b Excess costs were estimated by assuming that all respondents would instead follow the BMI distribution observed in the reference category, then calculating the difference between the current predicted costs and the predicted costs for the reference weight population.

^c Population-level costs were estimated by scaling the per-person excess costs using BMI category prevalence estimates from NHANES 2011–2016 and 2019 population estimates of the civilian, non-institutionalized population.

^d %_ile = percentile.

Children

We found a very shallow J-shaped curve of medical expenditures by BMI z-score for boys, meaning that we observed higher expenditures for boys with the lowest BMI z-scores and only a slight increase in expenditures as z-scores increased from the lowest expenditure level. Expenditures for girls were lower and did not exhibit increased costs at low BMI z-scores (Fig 1B). For both boys and girls, we found that medical expenditures only increased substantially over the 99^th percentile of BMI. Among children, obesity is associated with $116 (95% CI $14-$201) excess annual medical costs per person and $1.32 billion (95% CI $0.16-$2.29) of medical spending (Table 1). By BMI category we find increased costs for children with severe obesity of over $300 per person a year (95% CI $124-$474).

Age-specific

For predictions by BMI and age (Fig 1C) we found a similar J-shaped relationship of expenditures by BMI at all ages, with increasing expenditures by age. The highest predicted expenditures are for individuals with severe obesity between 60–70 years of age.

Our predictions of expenditure by obesity status and BMI category revealed increasing costs in all groups by age, with differential cost increases for obesity, especially for severe obesity (Fig 2). We found that severe obesity is associated with increased costs at all ages. BMI-specific medical costs by age and sex are available in a public repository (https://doi.org/10.7910/DVN/872OW1).

10.1371/journal.pone.0247307.g002

Fig 2 Estimated age-specific medical expenditures by BMI category.

Estimated expenditures are controlled for potential confounding variables. Shaded areas represent 95% confidence intervals.

Discussion

Using a continuous costing approach, we provide updated national and individual-level estimates of the annual direct health care costs associated with excess weight for children and adults in the US. We found a J-shaped curved of medical expenditures by BMI for adults, consistent with the epidemiologic data on BMI-related mortality [5]. The lowest predicted medical costs for adults occurred at a BMI between 20 and 24 for all ages. We find that among adults, obesity is associated with over $1,800 excess annual medical costs per person, accounting for over $170 billion of annual spending in the US. This figure rises to over $200 billion if excess costs from overweight (over $600 per person) are included, highlighting the large economic impact of overweight and obesity in the US.

For children, we find that across most of the range of BMI z-scores there was no association with health care expenditures, with large increases in costs occurring only above the 99^th percentile of BMI. Overall, we find that childhood obesity is associated with over $100 per child with obesity and over $1 billion of excess medical costs. Health care expenditures for children with severe obesity were increased by $300 per person per year. Although children’s obesity costs are a relatively small contributor to excess medical spending (less than 1% of all obesity-related medical expenditures), preventing childhood obesity may help to avert future health care costs given that excess body weight during childhood is a strong predictor of excess weight during adulthood [4].

We find that obesity-related costs increase with age, starting around age 30. This is similar to findings by the Global Burden of Disease and Global BMI Mortality Collaboration that report increased relative risks of obesity-related morbidity and mortality starting at ages 25–29 and 35+, respectively [5, 26]. Thus, our findings of little excess cost at younger ages is consistent with the epidemiological evidence that obesity-related disease mostly occurs later in life.

However, even at younger ages we do find increased costs associated with severe obesity, highlighting the importance of preventing extreme weight gain at all ages. The high costs at higher levels of BMI is especially concerning given that the prevalence of severe obesity among adults is projected to increase further and become the most common BMI category for some subgroups [3].

Our results suggest that obesity-related excess costs increase with age until about age 65, at which point the gap between obesity and non-obesity begins to narrow. This is partly due to increasing costs among those with normal weight as a result of ageing. However, it is also driven by a flattening out of costs in the severe obesity group; the costs for this group plateau while they continue to increase for all other BMI categories over this age range. This observed flattening may be due to selection bias (i.e. informative censoring) caused by higher mortality among individuals with severe obesity. We see that excess costs peak at progressively later ages for moderate obesity and overweight compared to severe obesity, adding support to the idea of mortality-induced censoring. Models to estimate the cost-effectiveness of obesity interventions among adults thus need to take into account effects on mortality as well [14].

Our estimates of obesity-related excess medical costs are similar to but higher than previous estimates by Finkelstein [6] and Wang [11], but lower than estimates by Cawley [10] using an instrumental variable (IV) approach (see S1 File, Section 6 for details of comparisons to previous estimates). Our overall results are also similar to a meta-analysis of 12 studies which estimated that obesity-attributable medical costs were $1,901 (95% CI $1,239-$2,582) per person in 2014 $US, accounting for $150 billion at the national level [12]. Consistent with previous findings [27], we found that overall per capita spending was higher for adult women than adult men, but lower for girls than boys. Also similar to past research [13], we found higher excess obesity-related costs for adult women, but no difference for boys and girls.

Limitations

While we control for a broad range of covariates to estimate the relationship between BMI and medical expenditures, there may be residual confounding due to unobserved variables, such as physical activity. In addition, our estimates are based on the cross-sectional association between BMI and medical expenditures. Large-scale longitudinal data tracking changes in individual-level BMI and expenditures over time would help more firmly establish the causal relationship between BMI and medical costs.

Also, we pooled all MEPS data from 2011–2016 to improve the stability of our estimates, so we could not examine trends within this period. Lastly, we only considered direct medical costs of obesity in this study. Including the indirect costs of obesity, such as lost wages due to obesity-related illness or disability or loss of future earnings due to premature death [28], would provide more comprehensive estimates of the economic impact of obesity.

Conclusions

We found that health care expenditures are higher for people with excess weight across a wide range of ages and BMI levels, with especially high costs for people with severe obesity. Obesity-related medical costs are higher for adult females, and increase with age for all adults, with the highest estimated costs occurring for 60–70 year olds. Although childhood obesity contributes a small proportion of total obesity-related medical costs, because excess weight in childhood is a strong predictor of adult obesity, policies to prevent excess weight gain at all ages are needed to mitigate the health and economic impact of the obesity epidemic, which accounts for over $170 billion in excess medical costs per year in the United States. These findings highlight the importance of promoting healthy weight across the entire BMI distribution, and provide policy makers and practitioners with more accurate estimates of the health care cost impact of excess weight by age, sex, and continuous BMI.

Supporting information

S1 File Supplemental appendix.

Additional methodological details and results.

(PDF)

References 1

Fryar CD, Carroll MD, Ogden CL. Prevalence of overweight, obesity, and extreme obesity among adults aged 20 and over: United States, 1960–1962 through 2011–2014. National Center for Health Statistics Data, Health E-Stats, July 2016 [cited 2020 October 22]. Available from: https://www.cdc.gov/nchs/data/hestat/obesity_adult_13_14/obesity_adult_13_14.htm.

Fryar CD, Carroll MD, Ogden CL. Prevalence of overweight and obesity among children and adolescents aged 2–19 years: United States, 1963–1965 through 2013–2014. National Center for Health Statistics Data, Health E-Stats, July 2016 [cited 2020 October 22]. Available from: https://www.cdc.gov/nchs/data/hestat/obesity_child_13_14/obesity_child_13_14.htm.

Ward

, Bleich

, Cradock

, Barrett

, Giles

, Flax

, et al. Projected US State-Level Prevalence of Adult Obesity and Severe Obesity. N Engl J Med. 2019;381(25):2440–2450. doi: 10.1056/NEJMsa1909301

31851800

Ward

, Long

, Resch

, Giles

, Cradock

, Gortmaker

. Simulation of Growth Trajectories of Childhood Obesity into Adulthood. N Engl J Med. 2017;377(22):2145–2153. doi: 10.1056/NEJMoa1703860

29171811

Global BMI Mortality Collaboration. Body-mass index and all-cause mortality: individual-participant-data meta-analysis of 239 prospective studies in four continents. Lancet. 2016;S0140-6736(16)30175-1. doi: 10.1016/S0140-6736(16)30175-1

27423262

Finkelstein

, Brown

. Why does the private sector underinvest in obesity prevention and treatment? N C Med J 2006; 67: 310–312. 17066665

Trasande

, Liu

, Fryer

, Weitzman

. Effects of childhood obesity on hospital care and costs, 1999–2005. Health Aff (Millwood). 2009;28(4):w751–60. doi: 10.1377/hlthaff.28.4.w751

19589800

Tsai

, Williamson

, Glick

. Direct medical cost of overweight and obesity in the USA: a quantitative systematic review. Obes Rev. 2011;12(1):50–61. doi: 10.1111/j.1467-789X.2009.00708.x

20059703

Trogdon

, Finkelstein

, Feagan

, Cohen

. State- and payer-specific estimates of annual medical expenditures attributable to obesity. Obesity (Silver Spring). 2012;20(1):214–20. doi: 10.1038/oby.2011.169

21681222

Cawley

, Meyerhoefer

. The medical care costs of obesity: An instrumental variables approach. J Health Econ 2012; 31: 219–230. doi: 10.1016/j.jhealeco.2011.10.003

22094013

Wang

, Pamplin

, Long

, Ward

, Gortmaker

, Andreyeva

. Severe Obesity In Adults Cost State Medicaid Programs Nearly $8 Billion In 2013. Health Aff (Milwood). 2015;34(11):1923–31. doi: 10.1377/hlthaff.2015.0633

26526251

Kim

, Basu

. Estimating the Medical Care Costs of Obesity in the United States: Systematic Review, Meta-Analysis, and Empirical Analysis. Value Health. 2016;19(5):602–13. doi: 10.1016/j.jval.2016.02.008

27565277

Kent

, Fusco

, Gray

, Jebb

, Cairns

, Mihaylova

. Body mass index and healthcare costs: a systematic literature review of individual participant data studies. Obes Rev. 2017;18(8):869–879. doi: 10.1111/obr.12560

28544197

Gortmaker

, Wang

, Long

, Giles

, Ward

, Barrett

, et al. Three Interventions That Reduce Childhood Obesity Are Projected To Save More Than They Cost To Implement. Health Aff (Millwood). 2015;34(11):1932–9. doi: 10.1377/hlthaff.2015.0631

26526252

Cradock

, Barrett

, Kenney

, Giles

, Ward

, Long

, et al. Using cost-effectiveness analysis to prioritize policy and programmatic approaches to physical activity promotion and obesity prevention in childhood. Prev Med. 2017;95 Suppl:S17–S27. doi: 10.1016/j.ypmed.2016.10.017

27773710

Medical Expenditure Panel Survey. Using appropriate price indices for analysis of health care expenditures or income across multiple years [Internet]. Rockville (MD): Agency for Healthcare Research and Quality; 2020 June [cited 2020 Aug 24]. Available from: https://meps.ahrq.gov/about_meps/Price_Index.shtml.

Connor Gorber

, Tremblay

, Moher

, Gorber

. A comparison of direct vs. self-report measures for assessing height, weight and body mass index: a systematic review. Obes Rev. 2007;8(4):307–326. doi: 10.1111/j.1467-789X.2007.00347.x

17578381

Ward

, Long

, Resch

, Gortmaker

, Cradock

, Giles

, et al. Redrawing the US Obesity Landscape: Bias-Corrected Estimates of State-Specific Adult Obesity Prevalence. PLoS One. 2016;11(3):e0150735. doi: 10.1371/journal.pone.0150735

26954566

Bring

. How to Standardize Regression Coefficients. Am Stat. 1994;48(3):209–213.

Hoerl

, Kennard

. Ridge regression: biased estimation for nonorthogonal problems. Technometrics. 1970;12(1):55–67.

Cessie

, van Houwelingen

. Ridge estimators in logistic regression. Appl Stat. 1992;41(1):191–201.

Mitchell E. Concentration of Health Expenditures in the U.S. Noninstitutionalized Population, 2014 [Internet]. Rockville (MD): Agency for Healthcare Research and Quality; 2016 Nov [cited 2018 June 6]. (Statistical Brief #497). Available from: http://meps.ahrq.gov/mepsweb/data_files/publications/st497/stat497.pdf.

Hastie

, Tibshirani

. Generalized Additive Models. London: Chapman and Hall; 1990.

Wood

. Low rank scale invariant tensor product smooths for generalized additive mixed models. Biometrics. 2006;62(4):1025–1036. doi: 10.1111/j.1541-0420.2006.00574.x

17156276

United States Census Bureau. Annual Estimates of the Resident Population by Single Year of Age and Sex for the United States: April 1, 2010 to July 1, 2019 [Internet]. 2020 June [cited 2020 Aug 24]. Available from: https://www2.census.gov/programs-surveys/popest/tables/2010-2019/national/asrh/nc-est2019-syasexn.xlsx

GBD 2013 Risk Factors Collaborators, Forouzanfar

, Alexander

, et al. Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks in 188 countries, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2015;386(10010):2287–323. doi: 10.1016/S0140-6736(15)00128-2

26364544

Lassman

, Hartman

, Washington

, Andrews

, Catlin

. US Health Spending Trends by Age and Gender: Selected Years 2002–10. Health Aff (Millwood). 2014;33(5):815–22. doi: 10.1377/hlthaff.2013.1224

24799579

Spieker

, Pyzocha

. Economic Impact of Obesity. Prim Care. 2016;43(1):83–95. doi: 10.1016/j.pop.2015.08.013

26896202

10.1371/journal.pone.0247307.r001

Decision Letter 0

Siegel

Robert

Academic Editor

2021

Robert Siegel

Submission Version

18 Dec 2020

PONE-D-20-36222

Association of body mass index with health care expenditures in the United States by age and sex

PLOS ONE

Dear Dr. Ward,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jan 29 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Robert Siegel

Academic Editor

PLOS ONE

Additional Editor Comments:

Please address all the reviewer comments and suggestions.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified (1) whether consent was informed and (2) what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.

If you are reporting a retrospective study of medical records or archived samples, please ensure that you have discussed whether all data were fully anonymized before you accessed them and/or whether the IRB or ethics committee waived the requirement for informed consent. If patients provided informed written consent to have data from their medical records used in research, please include this information.

Once you have amended this/these statement(s) in the Methods section of the manuscript, please add the same text to the “Ethics Statement” field of the submission form (via “Edit Submission”).

For additional information about PLOS ONE ethical requirements for human subjects research, please refer to http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research.

3. You have not indicated whether ethical approval was waived or necessary for your study. We understand that the framework for ethical oversight requirements for studies of this type may differ depending on the setting and we would appreciate some further clarification regarding your research. Could you please provide further details on ethical oversight of your study. Please clarify whether or why your study is exempt from the need for approval.

4. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This is a very carefully constructed analysis of the cost of continuously, in addition to categorically, measured BMI considering both age and sex with which I can find no fault in method.

I have two suggestions for improvement in the paper. The first is to explain why you undertake all the corrections for self-report bias relative to NHANES rather than simply using NHANES. It seems like NHANES may not be sufficiently representative, but it is unclear as they cover the same time period.

The second in the development of the conclusions. The paper set out to “To address key gaps in the literature, in this paper we estimate continuous BMI-specific medical costs by age and sex, and provide updated estimates of the medical costs attributable to obesity using recent data.” Please link back to this in the conclusions—note the higher costs of females, the comparatively low cost of children, and the high costs of 60-70 year olds. Linking back the billions of dollars in excess costs as a motivation (including opportunity cost) as an immediate call to policy action could be emphasized.

Reviewer #2: I appreciate this very relevant and important topic.

The data and information as presented for the obesity-related costs by BMI units as well as by BMI percentile categories is very clear. There is limited data out there for pediatrics so greatly appreciate the inclusion of pediatrics from ages 6-19 as well as your ability to control for all the many variables listed and the use of separate models for children vs adults.

I did have a question about the use of z-scores in children vs the use of actual BMI values like in the adults as the use of z- scores in the literature has become somewhat controversial.

You may want to consider reviewing this article below from 2017 that is similar in nature (but nto the same) to your study.

The Additional Costs and Health Effects of a Patient Having Overweight or Obesity: A Computational Model

Saeideh Fallah-Fini1,2, Atif Adam1, Lawrence J. Cheskin1, Sarah M. Bartsch1, and Bruce Y. Lee

Obesity (2017) 25, 1809-1815. doi:10.1002/oby.21965

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

10.1371/journal.pone.0247307.r002

Author response to Decision Letter 0

Submission Version

22 Jan 2021

Additional Editor Comments:

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

We have revised the manuscript formatting and file naming to meet the style requirements.

For additional information about PLOS ONE ethical requirements for human subjects research, please refer to http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research.

We used public use samples of de-identified datasets which are freely available online. We have amended the Methods section of the manuscript to clarify this point.

Line 24: “We used publicly available, de-identified data from the Medical Expenditure Panel Survey (MEPS) 2011-2016.”

This work does not involve human subjects, and is thus IRB exempt. We used public use samples of de-identified datasets which are freely available online.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

We have revised the cover letter to provide this information. We have also added our estimates of age/sex/BMI-specific medical expenditures to a data repository, which will be made publicly available if the analysis is accepted for publication.

Lines 132-133:

“BMI-specific medical costs by age and sex are available in a public repository (https://doi.org/10.7910/DVN/872OW1).”

Reviewers' comments:

Reviewer #1:

This is a very carefully constructed analysis of the cost of continuously, in addition to categorically, measured BMI considering both age and sex with which I can find no fault in method.

Thank you for your supportive comments.

Thank you for this clarifying question. NHANES is generally considered the gold-standard for nationally-representative, measured BMI. However, NHANES does not have information on medical costs – the outcome of interest in this analysis. Although MEPS does have individual-level data on both medical costs and BMI, the BMI data are self-reported, which is well-known to cause bias – generally underestimating BMI for adults, and often overestimating it for young children. Because NHANES and MEPS are both designed to be nationally-representative samples, we can use information on the BMI distribution from NHANES to adjust MEPS data for self-reporting bias, using the corrected individual-level BMI to more accurately estimate the relationship between BMI and medical expenditures.

Thank you for this suggestion. We have highlighted these findings in the conclusions.

Lines 194-203:

“We found that health care expenditures are higher for people with excess weight across a wide range of ages and BMI levels, with especially high costs for people with severe obesity. Obesity-related medical costs are higher for adult females, and increase with age for all adults, with the highest estimated costs occurring for 60-70 year olds. Although childhood obesity contributes a small proportion of total obesity-related medical costs, because excess weight in childhood is a strong predictor of adult obesity, policies to prevent excess weight gain at all ages are needed to mitigate the health and economic impact of the obesity epidemic, which accounts for over $170 billion in excess medical costs per year in the United States. These findings highlight the importance of promoting healthy weight across the entire BMI distribution, and provide policy makers and practitioners with more accurate estimates of the health care cost impact of excess weight by age, sex, and continuous BMI.”

Reviewer #2:

I appreciate this very relevant and important topic. The data and information as presented for the obesity-related costs by BMI units as well as by BMI percentile categories is very clear. There is limited data out there for pediatrics so greatly appreciate the inclusion of pediatrics from ages 6-19 as well as your ability to control for all the many variables listed and the use of separate models for children vs adults.

Thank you for your supportive comments. We agree that data for children are often not available, and felt it was important to include them in this analysis

I did have a question about the use of z-scores in children vs the use of actual BMI values like in the adults as the use of z- scores in the literature has become somewhat controversial.

Thank you for the opportunity to clarify this point. We did use the actual BMI values when estimating the models, but use the corresponding z-scores to display the predicted expenditures in the figure. We have clarified this in the manuscript.

Lines 68-72:

“We predicted adult expenditures for BMI values between 10 and 80. For children we predicted expenditures for BMI z-scores between -3 and +3. Although we fitted our model using BMI for both children and adults, we present the univariate predicted expenditures for children in terms of BMI z-score for visual ease, as the overweight and obesity thresholds are defined by BMI z-score for children.”

You may want to consider reviewing this article below from 2017 that is similar in nature (but nto the same) to your study.

The Additional Costs and Health Effects of a Patient Having Overweight or Obesity: A Computational Model Saeideh Fallah-Fini1,2, Atif Adam1, Lawrence J. Cheskin1, Sarah M. Bartsch1, and Bruce Y. Lee Obesity (2017) 25, 1809-1815. doi:10.1002/oby.21965

Thank you for this suggestion. The referenced paper estimates incremental costs over an adult’s lifetime using a Markov model, with the direct medical costs of overweight/obesity vs normal weight one of the inputs to the model, based on data from MEPS. However it is not clear that the authors adjusted for self-report bias or potential confounding when analyzing the MEPS data. Our current analysis aims to provide such cost estimates which could be used in simulation models such as this.

Attachment

Submitted filename: Response to reviewers.docx

10.1371/journal.pone.0247307.r003

Decision Letter 1

Siegel

Robert

Academic Editor

2021

Robert Siegel

Submission Version

5 Feb 2021

Association of body mass index with health care expenditures in the United States by age and sex

PONE-D-20-36222R1

Dear Dr. Ward,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Robert Siegel

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

You have a successfully addressed all the reviewer concerns

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

**********

6. Review Comments to the Author

Reviewer #1: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

10.1371/journal.pone.0247307.r004

Acceptance letter

Siegel

Robert

Academic Editor

2021

Robert Siegel

2 Mar 2021

PONE-D-20-36222R1

Association of body mass index with health care expenditures in the United States by age and sex

Dear Dr. Ward:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Robert Siegel

Academic Editor

PLOS ONE