The New York Times reports on a recent study by Annette Dobson in the American Journal of Preventive Medicine.  They find that “[a]fter adjusting for other variables, the 10-year weight gain for an average 140-pound woman was 20 pounds if she had a baby and a partner, 15 if she had a partner but no baby, and only 11 pounds if she was childless with no partner. The number of women with a baby but no partner was too small to draw statistically significant conclusions.”

The study does a good job of documenting the weight gain, but does not fully explain why this is occurring.  One of my previous studies looks at just this question.  My co-author and I hypothesize that married individuals have less of an incentive to maintain their weight, because they are not in the dating market.  In fact, the study finds evidence that women who begin cohabitating with their mate gain less weight than if they had gotten married.   Since in both cases the women live with their partner, the weight gain cannot be due to simply moving in with one’s significant other.  We find evidence that the probability a couple will separate (based on marital status and other factors), directly affects weight.  Women in more stable relationships gain more weight then women in less stable relationships, likely because the women in less stable relationships know they may soon re-enter the dating market.

However, the effect of the “dating market” on weight is only 2.4 kg (about 5 lb).  Thus, this effect doesn’t cause one to be obese, but can explain some of the weight gain after marriage.  Marriage does have many other salubrious effects, and one should not avoid marriage simply to avoid a minimal weight gain.

Another study by Michaud et al. (2009) has contradicted this finding.  While obese individuals will have shorter life expectancies, they do have higher health care costs in each year in which they live.  Taking into account a variety of trends that affect life expectancy–such as obesity and diabetes–Michaud and co-authors find that “Together, the reduction in smoking and the rise in obesity have increased net public-sector liabilities by $430bn, or approximately 4% of the current debt burden. Larger effects are observed for specific public programs: annual spending is 10% higher in the Medicaid program, and 7% higher for Medicare.”

It seems like it pays for Uncle Sam to feed grandma cantaloupe and not cheesesteaks.

A study by Jay Bhattacharya, Kate Bundorf, Noemi Pace and Neeraj Sood found that health insurance may actually increase body mass.  According to the authors: “We find weak evidence that more generous insurance coverage increases body mass index. We find stronger evidence that being insured increases body mass index and obesity.”

• A study by Jay Bhattacharya, Kate Bundorf, Noemi Pace and Neeraj Sood (2009) “Does Health Insurance Make You Fat,” NBER Working Paper 15163.

• EMR: Making electronic medical records universal will greatly improve health care quality, but the impact on cost will be minor.  Better quality care can reduce iatrogenic injuries and reduce cost, but the cost reduction–if any–will likely be small in magnitude.
• Prevention.  From the CBO: any gains from reducing obesity would be concentrated in the short and intermediate period “because some of the savings will be offset by increased longevity and the cost of disease that are most prevalent during old age.”
• Outcomes Research:  Laszewski claims that “inefficient use of technology is the key driver in health care spending accounting for an estimated 38% to 65% of spending growth.  The problem…with the suggestions that more outcomes research will save us money is that more than twenty years of outstanding outcomes research, Dartmouth for example, has not kept our health care costs under control.”  Outcomes research is important; it is imperative for physicians to prescribe cost effective treatment.  However, I agree with Laszewski that if financial incentives are not aligned to promote physician use of evidence-based medicine, then health outcomes research will have little impact.
• P4P: Laszewski doesn’t like pay-for-performance because in order for it to save money, it must lead to a reduction in physician payment on average.  Another reason why P4P won’t work is that paying individuals to check a diabetic’s A1C level may increase the frequency the physician monitors this metric, but it also may compel the physician to substitute their time away from other necessary medical services.
• Universal Coverage.  Universal coverage should reduce the percentage of individual who go to the emergency room for primary care needs;.  Nevertheless, providing universal health insurance coverage will certainly increase healthcare spending due to the moral hazard problem as well as supplier-induced demand.

A study by Baum (2009) claims that cigarette taxes may improve health, but not by as much as previously thought.  The paper finds that increasing the cigarette tax decreases smoking, but decreased smoking–an appetite suprressant–increases obesity.  Thus Baum finds that the health benefits of cigarette taxes may be overstated.

Disclaimer: Baum does state that “this research in no way concludes that [cigarette taxes] should be decreased to prompt weight loss.”

• Baum, Charles (2009) “The effects of cigarette costs on BMI and obesity,” Health Economics, Volume 18 Issue 1, Pages 3 – 19.

Getting married increases BMI by 0.70 for men and 0.96 for women.  For the average American male and female, this translates into a 4.8 pound increase for men and a 5.6 pound increase for women.  

Getting divorced actually decreases BMI.  The coefficients estimated were -0.27 for men and -0.63 for women; this translates into weight losses of 1.8 pounds for men and 3.7 pounds for women.  

The authors also found that “spouses tend to become more similar in body weight over time, indicating that environmental influences are an important cause of spouse weight similarity.”

• Jeffery RW, Rick AM. (2002) “Cross-sectional and longitudinal associations between body mass index and marriage-related factors.” Obesity Research 2002 Aug;10(8):809-815.

• HT: Marginal Revolution

However, I am not exactly sure what this proves. If you want to eat healthy food (fruits and vegetables), you need to shop more frequently since these “healthy” foods tend to spoil more easily than fatty, non-perishable foods (frozen burritos). Further, invidiuduals who live in denser, urban environoments likely 1) live closer to grocery stores and 2) have a social network that values healthy eating. Beatty even admits that “The exact causal relationship between dietary quality and expenditure dispersion is ambiguous.”

It seems more sensible that the desire to eat healthy foods determines shopping habits rather than the converse.

• Timothy K. M. Beatty (2008) “Expenditure dispersion and dietary quality: evidence from Canada” Health Economics, Volume 17, Issue 9 (p 1001-1014)