Volume 17, Issue 2

Both tobacco use and obesity are among the most important and costly health challenges faced in developed countries. Unfortunately, they may be inversely linked. While policy interventions that have placed limits on tobacco use have increased substantially over time, one unintended consequence may be to increase obesity rates. Issues of selection and unobserved heterogeneity make it difficult to empirically assess the relationship between the two health outcomes. Additionally, there may be heterogeneous policy effects by cessation cause – smoking bans or medical treatments or tobacco prices. This paper focuses on the effects of a rapidly expanding policy by using within-individual differences in exposure to workplace smoking bans to estimate the impact of smoking cessation on weight gain using a large study of over 5000 White and Black respondents followed since 1986. Findings suggest that individuals affected by the smoking bans gained more weight in the short-term than suggested by OLS estimates.

We analyzed the effect of oral nutritional supplement (ONS) use on 30-day readmission rates, length of stay (LOS), and episode costs in hospitalized Medicare patients (≥65), and subsets of patients diagnosed with acute myocardial infarction (AMI), congestive heart failure (CHF) or pneumonia (PNA). Propensity-score matching and instrumental variables were used to analyze ONS and non-ONS episodes from the Premier Research Database (2000–2010). ONS use was associated with reductions in probability of 30-day readmission by 12.0% in AMI and 10.1% in CHF. LOS decreases of 10.9% in AMI, 14.2% in CHF, and 8.5% in PNA were associated with ONS, as were decreases in episode costs in AMI, CHF and PNA of 5.1%, 7.8% and 10.6%, respectively. The effect on LOS and episode cost was greatest for the Any Diagnosis population, with decreases of 16.0% and 15.8%, respectively. ONS use in hospitalized Medicare patients ≥65 is associated with improved outcomes and decreased healthcare costs, and is therefore relevant to providers seeking an inexpensive, evidence-based approach for meeting Affordable Care Act quality targets.

Many state Medicaid programs have implemented policies designed to reduce spending on prescription drugs by restricting access to branded products. For patients with major depressive disorder, formulary restrictions could severely limit access to antidepressant therapies and disrupt care. We linked data on patient outcomes and spending from 24 state Medicaid programs to information on formulary restrictions from 2001 to 2008. Outcomes included frequency of MDD-related hospitalizations and ER visits per patient and total healthcare spending. We estimated the effect of the policies on patient outcomes and spending using a difference-and-difference approach. We found that restricting access to antidepressants increased the probability of an MDD-related hospitalization by 1.7 percentage points (16.6%). Furthermore, we found no evidence that these restrictions resulted in any net savings for Medicaid.

Bans on retail tobacco displays, of the type proposed by New York’s Mayor Bloomberg in March 2013, have been operative in several economies since 2001. Despite an enormous number of studies in public health journals using attitudinal data, we can find no econometric event studies of the type normally used in Economics. This paper attempts to fill that gap by using data from 13 cross sections of the annual Canadian Tobacco Use Monitoring Surveys. These data afford an ideal opportunity to study events of this type given that each of Canada’s 10 provinces implemented display bans at various points between 2003 and 2009. Accordingly, we use difference-in-difference methods to study three behaviors following the introduction of bans: participation in smoking, the intensity of smoking and quit intentions. A critical element of the study concerns the treatment of contraband tobacco. Our estimates provide very little support for the hypothesis that behaviors changed following the bans.