The Economics of Obesity: Volume 17

This volume is part of an annual series entitled Advances in Health Economics and Health Services Research. We express our gratitude to the series editors Björn Lindgren and Michael Grossman for inviting us to edit this volume.

Five papers in the volume use economic models to predict obesity and related behaviours. Two of the papers are theoretical. Liqun Liu, Andrew J. Rettenmaier, and Thomas R. Saving analyse the importance of food quality for bodyweight. Bodyweight is treated as a variable of choice – the individual derives utility from health, food consumption and consumption of a composite good. Bodyweight is assumed to decrease health whenever it differs from its physiologically optimal level. Their model implies that much of further income growth will be used to improve food quality rather than increase caloric intake.

This paper follows previous economic studies on bodyweight by looking at bodyweight as individuals’ choices in response to changes in income and food prices. However, it goes one step further to add another relevant choice under individual control: food quality. It shows that the upward trend of bodyweight caused by economic growth and technological innovations may well be self-limiting in the sense that the bodyweight growth in the future is likely to be slower or reversed. In particular, it finds that much of further income growth will be used for improving food quality rather than increasing caloric intake. Moreover, further technological innovations that focus on lowering the price of high-quality (healthy) food relative to that of low-quality food would encourage substituting food quality for food quantity (calories).

We examine the extent to which food prices and restaurant outlet density are associated with adolescent fruit and vegetable consumption, body mass index (BMI), and the probability of overweight. We use repeated cross-sections of individual-level data on adolescents from the Monitoring the Future Surveys from 1997 to 2003 combined with fast food and fruit and vegetable prices obtained from the American Chamber of Commerce Researchers Association and fast food and full-service restaurant outlet density measures obtained from Dun & Bradstreet. The results suggest that the price of a fast food meal is an important determinant of adolescents’ body weight and eating habits: a 10% increase in the price of a fast food meal leads to a 3.0% increase in the probability of frequent fruit and vegetable consumption, a 0.4% decrease in BMI, and a 5.9% decrease in probability of overweight. The price of fruits and vegetables and restaurant outlet density are less important determinants, although these variables typically have the expected sign and are often statistically associated with our outcome measures. Despite these findings, changes in all observed economic and socio-demographic characteristics together only explain roughly one-quarter of the change in mean BMI and one-fifth of the change in overweight over the 1997–2003 sampling period.

This paper presents a behavioral economics model with bounded rationality to describe an individual's food consumption choices that lead to weight gain and dieting. Using a physiological relationship determining calories needed to maintain weight, we simulate the food consumption choices of a representative female over a 30-year period. Results show an individual will periodically choose to diet, but that diet will reduce weight only temporarily. Recurrence of weight gain leads to cyclical dieting, which reduces the trend rate of weight increase. Dieting frequency is shown to depend on decision period length, dieting costs, and habit persistence.

In this study, we examined the association between girls’ participation in high school sports and the physical activity, weight, body mass, and body composition of adolescent females during the 1970s when girls’ sports participation was dramatically increasing as a result of Title IX. We found that increases in girls’ participation in high school sports, a proxy for expanded athletic opportunities for adolescent females, were associated with an increase in physical activity and an improvement in weight and body mass among girls. In contrast, adolescent boys experienced a decline in physical activity and an increase in weight and body mass during the period when girls’ athletic opportunities were expanding. Taken together, these results strongly suggest that Title IX and the increase in athletic opportunities among adolescent females it engendered had a beneficial effect on the health of adolescent girls.

The entrance of economics into the literature on obesity and diabetes has been instrumental in showing how people respond to incentives when maximizing their health. In this paper some of the roles that prices and policies have played in the surging obesity and diabetes rates across the world are addressed. The paper focuses on the possible role that prices of foods with high glycemic indexes play in determining blood sugar levels, and addresses the recent concern with high fructose corn syrup and genetically modified goods across the world. The possible links and implications suggest that future research in the area is urgently needed.

Context. The most advanced and fastest growing form of bariatric surgery is laparoscopic gastric bypass. Very little is known about population-based 180-day laparoscopic bypass costs, complication rates, readmission rates, and post-operative care.

Objective. To examine the 6-month costs and outcomes of laparoscopic vs. open bariatric bypass surgery using a national population-based sample.

Design. We use the 1998–2003 Nationwide Inpatient Sample to examine national trends in the rate of laparoscopic bypass. To examine post-operative outcomes, we examine insurance claims for 2,384 bariatric bypass surgeries, at 308 hospitals, among a population of 5.6 million non-elderly people covered by large employers across 49 states in 2001 and 2002. Multivariate logit regression analysis is performed to risk-adjust outcomes.

Main Outcome Measures. 180-day outcomes: 12 complications specific to bariatric surgery and 44 general post-operative conditions, readmission rates, ER rates, and expenditures following bariatric surgery.

Results. Between 1998 and 2003, the national percentage of bariatric bypass surgeries that were laparoscopic grew from 1.5 to 17.1%. There was no significant difference in in-hospital mortality between laparoscopy and open surgery. With the 2001–2002 claims data, we find that of the patients having bypass surgery, men had 48% lower odds of having laparoscopy and that high bariatric volume hospitals were close to four times more likely to use laparoscopy. Laparoscopic bypass, compared with open bypass, had 34% lower odds of a complication during the initial surgical stay, 27% lower odds of a 30-day complication, but no statistically significant difference in 180-day complications. Laparoscopy had 49% higher odds of having the general 44 post-operative conditions, with 45% higher odds of a readmission and 54% higher odds of an ER visit. However, overall, laparoscopy resulted in a 23% lower number of hospital days and 9% lower 180-day expenditures.

Conclusion. The laparoscopic cost-savings during the less invasive initial surgery stay outweigh the increase in post-discharge utilization. Further cost-savings will only emerge from laparoscopy only if its late post-operative complications are reduced. More cost-savings will also emerge as more physicians switch to the use of laparoscopy for bypass surgery.

The doubling of obesity in the U.S. over the last 25 years has led policymakers and physicians to encourage weight loss, but few methods of weight loss are effective. One promising avenue is pharmacotherapy. However, little is known about the use of anti-obesity drugs. This paper describes the market for anti-obesity drugs and studies the utilization of anti-obesity drugs using data from the Medical Expenditure Panel Survey for 1996–2002, a period that is interesting because it covers the introduction of three, and the withdrawal of two, anti-obesity drugs from the market.

Our results point to wide sociodemographic disparities in anti-obesity drug use. Women are almost 200% more likely than men to use anti-obesity drugs. Hispanics and African-Americans are only 39% as likely as Whites to use them. Those with prescription drug coverage are 46% more likely to use anti-obesity drugs.

We also find that the vast majority of subjects who are approved to take these drugs are not taking them, and a significant number who are not approved to take the drugs are taking them. We find strong evidence that the well-publicized 1997 withdrawal of fenfluramine and dexfenfluramine had a chilling effect on the overall market for anti-obesity drugs. We find little difference in observed characteristics between those who took the withdrawn drugs and those who took the other anti-obesity drugs in the market.

This paper examines the associations between obesity, employment status and wages for several European countries. Our results provide weak evidence that obese workers are more likely to be unemployed or tend to be more segregated in self-employment jobs than their non-obese counterparts. We also find difficult to detect statistically significant relationships between obesity and wages. As previously reported in the literature, the associations between obesity, unemployment and wages seem to be different for men and women. Moreover, heterogeneity is also found across countries. Such heterogeneity can be somewhat explained by some labor market institutions, such as collective bargaining coverage and employer-provided health insurance.

This paper brings a European perspective to the mainly U.S.-based literature on the relationship between obesity and labour-market outcomes. Using micro-data on workers aged 50 and over from the newly developed SHARE database, the effects of obesity on employment, hours worked, and wages across 10 European countries were analysed. Pooling all countries, the results showed that being obese was associated with a significantly lower probability of being employed for both women and men. Moreover, the results showed that obese European women earned 10% less than their non-obese counterparts. For men, however, the effect was smaller in size and insignificant. Taking health status into account, obese women still earned 9% less. No significant effect of obesity on hours worked was obtained, however. Regressions by country-group revealed that the effects of obesity differed across Europe. For instance, the effect of obesity on employment was greatest for men in southern and central Europe, while women in central Europe faced the greatest wage penalty. The results in this study suggest that the ongoing rise in the prevalence of obesity in Europe may have a non-negligible effect on the European labour market.

Previous research provides evidence of a negative effect of body mass on women's economic outcomes. We extend this research by using a much older sample of individuals from the Panel Study of Income Dynamics and by using a body mass measure that is lagged by 15 years instead of the traditional 7 years. One of the main contributions of this paper is a replication of previous research findings given our differing samples and measures. We compare OLS estimates with sibling fixed effects estimates and find that obesity is associated with an 18% reduction in women's wages, a 25% reduction in women's family income, and a 16% reduction in women's probability of marriage. These effects are robust – they persist much longer than previously understood and they persist across the life course, affecting older women as well as younger women.

If rational individuals pay the full costs of their decisions about food intake and exercise, economists, policy makers, and public health officials should treat the obesity epidemic as a matter of indifference. In this paper, we show that, as long as insurance premiums are not risk rated for obesity, health insurance coverage systematically shields those covered from the full costs of physical inactivity and overeating. Since the obese consume significantly more medical resources than the non-obese, but pay the same health insurance premiums, they impose a negative externality on normal weight individuals in their insurance pool.

To estimate the size of this externality, we develop a model of weight loss and health insurance under two regimes – (1) underwriting on weight is allowed and (2) underwriting on weight is not allowed. We show that under regime (1), there is no obesity externality. Under regime (2), where there is an obesity externality, all plan participants face inefficient incentives to undertake unpleasant dieting and exercise. These reduced incentives lead to inefficient increases in bodyweight, and reduced social welfare.

Using data on medical expenditures and bodyweight from the National Health and Interview Survey and the Medical Expenditure Panel Survey, we estimate that, in a health plan with a coinsurance rate of 17.5%, the obesity externality imposes a welfare cost of about $150 per capita. Our results also indicate that the welfare loss can be reduced by technological change that lowers the pecuniary and non-pecuniary costs of losing weight, and also by increasing the coinsurance rate.

Objective. To quantify the association between obesity and somatic hospital costs and number of overall somatic hospital contacts – number of inpatient admissions, number of outpatient visits, and number of emergency department visits – based on anthropometric measurements of waist circumference (WC) and information from The National Patient Registry and The Danish Case-Mix System (DRG).

Participants. The study population consisted of two sub-samples from the Inter99 study at Research Centre for Prevention and Health in 1999–2001. One sub-sample used WC as an indicator for obesity (n=5,151), whereas the other used BMI as an indicator for obesity (n=4,048). Using WC, obesity was defined as WC > 102cm for men and > 88cm for women. Normal weight was defined as circumference < 94cm for men and < 80cm for women. Using BMI, obesity was defined as BMI > 30kg/m2, whereas individuals with BMI=18.5–24.9kg/m2 were defined as normal weight. Individuals with BMI < 18.5kg/m2 were excluded from both sub-samples.

Design. We undertook a 3-year retrospective study of the relationship between obesity and use of hospital resources. Data on hospital contacts and costs were obtained from The National Patient Registry and DRG. Analyses were performed using two-part models and Poisson regression. Outcome variables were costs and hospital contacts.

Results. This study has demonstrated that obese individuals have a greater use of hospital services and greater hospital costs compared with normal weight individuals. When using WC as an indicator for obesity, mean hospital costs were 33.8% greater among obese women and 45.3% greater among obese men in a 3-year period but the differences were not significant. When using BMI to measure obesity, obese men had significantly greater costs (57.5%) than normal weight men.

Furthermore, obese men and women (indicated byWC) had an increased number of hospital contacts compared with normal weight individuals (rate ratio 1.32, 95% CI 1.21–1.43 for men and 1.20, 95% CI 1.11–1.28 for women) including inpatient admissions, outpatient visits, and emergency department visits. The same trends were seen when obesity was indicated by BMI.

A five category self-reported health indicator together with the self-reported prevalence of diabetes and heart disease for older Canadians, are examined using data from five cohorts of men and women from the 2001 Canadian Community Health Survey. Consistent with other studies we find that smoking and dietary behaviors are highly correlated with general self-reported health, diabetes, and heart disease. Individual standardized weight, the body mass index, was negatively associated with health outcomes for all age groups, but became less important with age as socioeconomic variables became more important. Socioeconomic variables explained more of the variation in health outcomes than the combined effects of tobacco use and excessive weight problems. In addition, there is compelling evidence that obesity could overtake smoking as the leading cause of health problems in Canada.