|
Session Title: Using Multiple Sources of Evidence to Evaluate Policy Changes: Examining the Impact of the 2007 Smoke-Free Arizona Act
|
|
Panel Session 621 to be held in Panzacola Section F3 on Friday, Nov 13, 4:30 PM to 6:00 PM
|
|
Sponsored by the Health Evaluation TIG
|
| Chair(s): |
| Michele Walsh, University of Arizona, mwalsh@u.arizona.edu
|
| Abstract:
Adequately evaluating public policy changes requires several important elements, including an understanding of the context in which the policy is applied; a general effect theory of what the desired outcomes are, and how change should occur; and multiple sources of data to determine whether the policy was actually implemented, and what the outcomes were. This session outlines how multiple sources of evidence can be used to evaluate the impact of larger-scale public policies, addressing both the initial context of the policy and examining outcomes. The evaluation of the adoption of a statewide smoking ban will be used to illustrate the described approach.
|
|
Understanding the Context of Policy: Setting the Stage For Evaluation
|
| Michele Walsh, University of Arizona, mwalsh@u.arizona.edu
|
|
Policy change is generally a long, drawn-out process that occurs in a highly-charged environment involving multiple stakeholders, with sometimes-conflicting agendas. Developing an evaluation of the outcome of such a process requires that multiple perspectives are taken into account in defining what outcomes to examine and what sources of data should be used. This paper sets the stage for the methodological approaches to follow by outlining the contextual background of a statewide smoking ban and describing how that affected some of the methods chosen.
|
|
|
Evaluating New Policies: General Principles of Multi-method Approaches - Exemplified With Arizona's 2007 Smoke-free Law
|
| Frederic Malter, University of Arizona, fmalter@email.arizona.edu
|
|
New policies usually have a multitude of effects. An evaluation of new policies begins with careful examinations of the changes to be expected in the social, political and psychological reality of people affected by it. The necessity of a general effect (causal) theory becomes evident when evaluators try to answer questions around why a given change should occur. In the case of Arizona's 2007 smoking ban, a causal model was developed that specified anticipated effects in several domains, such as compliance, reductions in air pollution, changes in smoking behavior and impacts on hospitality's business revenue. Estimates of anticipated effects were counterbalanced against a careful assessment of the share of the population already covered by smoke-free provisions before the state-wide ban to obtain realistic projections. The presentation will give an overview of the multiple data sources that were identified and the data-analytic methods chosen to best answer questions about time-dependent changes.
| |
|
The Effect of the Smoke-Free Arizona Act on Cessation and Other Behaviors
|
| John Daws, University of Arizona, johndaws@email.arizona.edu
|
|
The campaign to pass Arizona's state-wide smoking ban focused on the reduction of second-hand smoke. As a matter of policy, the campaign did not directly promote smoking cessation. Nonetheless, it would be a desired indirect outcome of the ban that some smokers decide to quit. Calls to the state's telephone cessation service were used as a proxy for intentions to quit in the general population. We examined how call volumes responded to ban-related health messages, to the ban's passage in November 2006, and to its implementation seven months later. We used regression analysis to isolate the effects of the ban from the effects of other events, such as the cigarette-tax increase in December 2006. Other measures of public response to the ban included (a) reduced particle pollution inside bars and restaurants, (b) public-opinion surveys of smokers and non-smokers both before and after implementation, and (c) public complaints about non-compliant locations.
| |
|
Public Health and Economic Impact of the New Smoke-free Law
|
| Patricia Herman, University of Arizona, pherman@email.arizona.edu
|
|
One of the main arguments for the statewide smoking ban was evidence of the detrimental health effect of secondhand smoke (SHS) on those exposed, and many smoking ban opponents cited the potential for it to cause serious economic losses to establishment owners. In order to measure the impact on health of the ban we compared statewide hospital admission data before and after the ban for diagnoses known to be associated with SHS exposure. In order to better imply causation we also looked at admissions for diagnoses not likely to be associated with SHS exposure and we split the analysis between counties with and without previous smoking bans. To determine the economic impacts we analyzed three types of data: self-report patronage of smoking and non-smoking establishments before and after the ban, statewide taxable monthly sales for bars and restaurants, and statewide monthly employment in the food services and drinking places sector.
| |