Cover Image

HARDBACK
$49.95



View/Hide Left Panel

J
The Role of Public Policies in Reducing Smoking Prevalence: Results from the SimSmoke Tobacco Policy Simulation Model

David T. Levy

Senior Scientist

Pacific Institute for Research and Evaluation

Professor, University of Baltimore

INTRODUCTION

The computer simulation model, known as SimSmoke, is a model that has been developed to examine the effect of tobacco control policies for the United States (Levy et al. 2002; Levy et al. 2000a). SimSmoke projects smoking prevalence over time and estimates the effect of tobacco control policies on those rates. The purpose of this appendix is to describe and to provide predictions from that model.

The development of the model and results from it have been published in a series of papers examining different types of policies (Levy et al. 2000a; Levy et al. 2000b; Levy and Friend 2001; Levy and Friend 2002a; Levy and Friend 2002b; Levy et al. 2001a; Levy et al. 2001b). Other papers have considered the future impact of the policies (Levy et al. 2003; Levy et al. 2005b). In addition, a group of papers has validated the model. Levy and colleagues (2004a; 2004c; 2004d) found that the model predicted smoking prevalence rates well for the United States over the time period 1993–2003, with most of the changes over that period due to prices changes. These studies also show that models for the states of California and Arizona predicts smoking prevalence relatively well after comprehensive programs were developed, and that an important part of the changes was explained by the media campaigns or comprehensive programs implemented in the states (Levy et al. 2004a; Levy et al. 2004c; Levy et al. 2004d).

This appendix considers the effects of individual policies and a combination of different policies on smoking prevalence as explained by the model and using effect sizes developed in conjunction with the Institute of Medicine’s (IOM) Committee on Reducing Tobacco Use. Specifically, we estimate how much smoking rates may be changed by additional policies, including tax changes, clean air laws, media or comprehensive campaigns, school education programs, cessation treatment policies, and youth access enforcement. We also consider the effect of abandoning some of the policies currently in place as well as the effect of policies on smoking rates in total and by age groups.

METHODOLOGY

Basic Model

The SimSmoke simulation model begins with the number of smokers, never-smokers, and ex-smokers by age and gender for the United States in the baseline year. In developing the model, we chose a best year for which there were data to develop the necessary smoking meas-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation J The Role of Public Policies in Reducing Smoking Prevalence: Results from the SimSmoke Tobacco Policy Simulation Model David T. Levy Senior Scientist Pacific Institute for Research and Evaluation Professor, University of Baltimore INTRODUCTION The computer simulation model, known as SimSmoke, is a model that has been developed to examine the effect of tobacco control policies for the United States (Levy et al. 2002; Levy et al. 2000a). SimSmoke projects smoking prevalence over time and estimates the effect of tobacco control policies on those rates. The purpose of this appendix is to describe and to provide predictions from that model. The development of the model and results from it have been published in a series of papers examining different types of policies (Levy et al. 2000a; Levy et al. 2000b; Levy and Friend 2001; Levy and Friend 2002a; Levy and Friend 2002b; Levy et al. 2001a; Levy et al. 2001b). Other papers have considered the future impact of the policies (Levy et al. 2003; Levy et al. 2005b). In addition, a group of papers has validated the model. Levy and colleagues (2004a; 2004c; 2004d) found that the model predicted smoking prevalence rates well for the United States over the time period 1993–2003, with most of the changes over that period due to prices changes. These studies also show that models for the states of California and Arizona predicts smoking prevalence relatively well after comprehensive programs were developed, and that an important part of the changes was explained by the media campaigns or comprehensive programs implemented in the states (Levy et al. 2004a; Levy et al. 2004c; Levy et al. 2004d). This appendix considers the effects of individual policies and a combination of different policies on smoking prevalence as explained by the model and using effect sizes developed in conjunction with the Institute of Medicine’s (IOM) Committee on Reducing Tobacco Use. Specifically, we estimate how much smoking rates may be changed by additional policies, including tax changes, clean air laws, media or comprehensive campaigns, school education programs, cessation treatment policies, and youth access enforcement. We also consider the effect of abandoning some of the policies currently in place as well as the effect of policies on smoking rates in total and by age groups. METHODOLOGY Basic Model The SimSmoke simulation model begins with the number of smokers, never-smokers, and ex-smokers by age and gender for the United States in the baseline year. In developing the model, we chose a best year for which there were data to develop the necessary smoking meas-

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation ures. We chose the year 2002 as our baseline year, which also had the advantage that there were no large changes in policies in recent years (since the large 1998–1999 price changes). The basic SimSmoke model involves a population model, a smoking model, and policy modules. Following a discrete first-order Markov process, the entire population evolves through birth and death rates, and the number of smokers, never-smokers, and ex-smokers evolves through initiation, cessation, and relapse rates. Tobacco control policies change initiation and cessation rates through individual policy modules. Consequently, smoking rates over time depend on tobacco control policies and prior smoking patterns. The version of the model used in this report is built on an Excel platform. This section presents a brief description of that model and a discussion of future policy scenarios. The data sources are summarized in Table J-1. A mathematical formulation and further description of the model can be found at http://cisnet.flexkb.net/mp/pub/cisnet_lung_ pire_profile.pdf and referenced papers. Population Model SimSmoke is built first on a demographic model. The population, distinguished by age, starts in the year 2002. The population evolves over time with fertility (leading to births) and some portion of the population dying at each age. We do not consider immigration or changes in racial or ethnic composition for the purposes of this model. Mathematically, the total population (Pop) is distinguished by time period (T) and age (A) (and is further distinguished in the model by gender and racial ethnic group). Mortality rates (MR) are distinguished by age and gender. The number of newborns depends on first-year death rates and fertility rates (Fert) of females by age, with equal birth rates for males and females. Births through the first year (age 0) for each gender are: PopT,0 = 0.5*(1 − MR0)* ΣA (PopT,A,1 * FertA), where T = 1 … 20; A = 14 … 49 After the first year, the population evolves as: PopT,A = PopT-1,A-1 * (1 − MortRateA) Population data are obtained from the 2000 Census of Population, and projected forward to 2002. Fertility rates are from the U.S. Census Vital Rate Inputs Tables for the year 2002. Mortality rates are from the 2001 Multiple Cause-of-Death File compiled from death certificates, by the National Center for Health Statistics (NCHSU; www.nchs.gov). The file includes information on all deaths in the United States in 2001. Smoking Model SimSmoke next divides the population in the base year into: (1) never-smokers, (2) smokers and (3) 16 categories of ex-smokers (n = 1 … 16+), corresponding to years since last smoking. After the base year, individuals are classified as never-smokers (designated by NS) from birth until they initiate smoking or die, according to: NeversmokersT,A = NeversmokersT-1,A-1 * (1 − MortRateA,NS)*(1 − Initiation RateA) Through age 24, the number of smokers (designated by S) is tracked as:

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation SmokersT,A = SmokersT-1,A-1*(1 − MortRateA,S) + NeversmokersT-1,A-1*(1 − MortRateA,ns)*Initiation RateA Once a smoker, individuals continue in that category until they quit, die, or reenter the group through relapse. After age 24, smokers are tracked as: SmokersT,A = SmokersT-1,A-1*(1 − MortRateT.A,S)*(1 − Cessation RateA) + Σ16N = 1 ExsmokersT-1,A-1,N*(1 − MortRateT,A,N)*(Relapse RateA,N) First year ex-smokers are determined by the first-year cessation rate applied to surviving smokers in the previous year. Individuals who have been ex-smokers for n = 2 … 15 years, are defined as: Ex-smokersT,A,N = Ex-smokersT-1,A-1,N-1*(1 − MortRateA,N)*(1-Relapse RateA,N-1) For those who have ceased smoking for more than 15 years, we add to the above equation the ex-smokers from the previous year who have quit for more than 15 years and did not die or relapse in the previous year. In the model, smokers are defined as individuals who are currently smoking (either daily or on some days) and have smoked more than 100 cigarettes in their lifetime. Due to empirical challenges in measuring initiation and quitting and to ensure the stability and internal consistency of the model, initiation rates at each age are a measured net of quitting. Specifically, net initiation is measured as the difference between the smoking rate at a given age and that same rate at the previous age. Because the duration of smoking is not considered, we do not track the specific year when individuals initiate in this population-level model. Since smoking rates typically level off by age 24 (DHHS 1994), initiation in the model occurs until age 24. Cessation is tracked from age 24, since the relative risks of mortality from smoking are not discernible for those who quit smoking before that age (DHHS 1990; DHHS 2004). Cessation rates in the first year are distinguished by age, but relapse rates in later years are only distinguished by years since quitting. Ex-smokers are defined as those over the age of 24 who were not smoking at the time of the survey. In SimSmoke, ex-smokers are broken down into six categories, categorized by year since quitting through 15 years and then aggregated at >15 years. Never-smokers are those who have not smoked 100 cigarettes in their lifetime or have smoked 100 cigarettes in their lifetime, but are less than the age of 24 and are not currently smoking. The primary source of baseline data on smoking habits by age and gender is the Tobacco Use Supplement (TUS) of the Current Population Survey (CPS), a sample of approximately 475,000 respondents conducted in September 2001, January 2002, and May 2002. Data are obtained by single age from ages 15 to 24 years, and then by 10 year age groups through age 90. Smoking rates are multiplied by the relevant 2002 population to determine the number of smokers and exsmokers by demographic group. In the model, we assign the value for the age bracket to the midpoint age of the bracket, and interpolate between that bracket and the midpoint value in the previous age bracket. Smoking may begin before age 15, but the TUS only asks individuals age 15 and older about their smoking status. For those below age 15, we use data from the 1993 TAPS (Teenage Attitudes and

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation Practices) survey. To maintain comparability, we scaled those data by the ratio of the TAPS 15– 17-year-old smoking rate divided by the U.S. 15–17-year-old smoking rate. Policy Modules In separate policy modules, we examine the effect of tax changes, clean indoor air laws, mass media policies, school education policies, cessation treatment policies, and strategies to reduce youth access to cigarettes. The original policy parameters in the model used to generate the predicted effects are based on thorough reviews of the literature and the advice of an expert panel. These parameters have been reviewed by the IOM committee and are either accepted or modified, as described below. In the case of cessation treatment and school education policies, significant changes have been made in the structure of the original SimSmoke policy modules. The effects of policies are calculated in percentage terms relative to the initial rates [PR = (Post-policy Rate − Initial Rate)/Initial Rate], where PR < 0. For most policies, the greatest effect is generally in the first few years in which the policy is in effect. These are modeled as a permanent additive effect on smoking prevalence, that is, SmokersT,A * (1 + PRI,T,A) for policy (I) at time period (T) that may vary by age (A). While the effect may be spread over several years, we model the effects as occurring in the first year that the policy is in effect. If the policy is maintained, the effects of the policy are maintained through modification of the initiation rates. The percentage reduction is applied throughout the years (T) during which the policy is in effect to the initiation rate [as Initiation RateA *(1 + PR)]. The percentage effects of the policy are also enhanced over time through increases in the first year cessation rate [as Cessation RateA *(1 − PR)]. First-year quit rates continue to be elevated for each of the policies (except youth access policies), because policies reduce the quantity smoked per smoker and quitting is more likely among those who smoke less (Hughes 2000; Hymnowitz et al. 1991; Hymowitz et al. 1997). We assume that relapse rates are unaffected by the policy, except insofar as the amount of relapse increases in proportion to any added cessation. Unless otherwise indicated, the same proportionate effect of a policy is applied to the prevalence, initiation, and cessation rates when a new policy is implemented and maintained. When a long-standing policy is reversed, it is assumed that only initiation and cessation rates are affected (i.e., the effects of a policy are asymmetric in terms of implementation of a proactive policy and the scaling back of that policy). We expect that those who have quit and maintained cessation over a reasonable period of time are unlikely to relapse when the policy is abandoned, although future initiation rates will be higher and cessation rates will be lower. Policy effects may also vary by age. For example, some policies are directed at and are expected primarily to affect youth. When more than one policy is in effect, the percentage reductions are multiplicatively applied, that is, (1 + PCI)*(1 + PCI) for policies (I), which implies that the relative effect is independent of other policies but the absolute effect is smaller when another policy is in effect. Some specific synergies are built into the model as described below. We track the effects of policies from the year 2006 forward. Because the model begins in the year 2002, we track the effect of policies through 2005. Since the CPS and TUS data are collected between September of the preceding year and May of the current year, we consider the estimates as representing smoking rates in the midpoint month (January), and policy data are matched to their levels on January 1 of the particular year.

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation Taxes In the tax module (Levy et al. 2000b), price increases are modeled as age-specific, constant proportional, effects on prevalence, initiation, and cessation rates (Levy et al. 2000b). Based on economic theory, cigarette use is determined by changes in the retail price relative to the prices of other goods, as measured by the participation (i.e., prevalence) and demand elasticity (i.e., the percentage change in consumption from a 1 percent increase in price). Based on the studies that distinguish by age, the simulation model assigns a price elasticity of −0.6 for individuals below age 18, −0.3 for those ages 18 to 24, −0.2 for those ages 25 to 34, and −0.1 for those age 35 and above. Based on recent evidence (Farrelly and Bray 1998), these elasticity estimates have been lowered since our earlier work (Levy et al. 2000b). These parameters have been accepted by the IOM panel. For the period 2002–2005, prices are averaged over states with weights based on tobacco sales and are adjusted for inflation using the Bureau of Labor Statistics (BLS) Consumer Price Index. Data on retail prices and taxes were obtained for 2002 and 2003 from (Orzechowski and Walker 2003) and for 2004 and 2005 from www.tobaccofreekids.org/research/factsheets/pdf/0212.pdf. The retail price is measured by a price index that includes generic cigarettes weighted by their proportionate sales. Inflation-adjusted prices increased slightly from $3.75 to $4.20 between 2002 and 2005, and the average state tax in 2005 was $1.23. From 2005, we assume that cigarette prices relative to inflation stay constant (i.e., we assume that taxes adjust upward to reflect general price inflation). To model the effect of additional tax changes, prices change by the amount of change in the average state plus the federal tax on cigarettes, based on studies reported in Jha and Chaloupka (2000). Clean Air Laws The clean air policy module examines the effect of three types of laws: work site, restaurant, and other public places (Levy et al. 2001b). The module predicts an 11 percent reduction in prevalence rates with all policies fully implemented and with strong enforcement and media publicity. Work site laws have the largest effect, 7 percent, with restaurant and bars laws producing a 2 percent effect, and laws covering other places a 1 percent effect. Work site bans without high compliance have two-thirds of the effect of a total ban with high compliance, and partial work site and restaurant bans have one-third the effect of total bans. Media publicity and enforcement yield an added 0.5 percent effect each for work sites and restaurants. Based on differences in labor participation rates and on the effect on workers who smoke, females experience 80 percent of the effect compared to males, and effects increase between ages 26 to 39 but decrease at older ages. These parameters have been accepted by the IOM panel. The effects of newly implemented clean air laws depend on the extent of clean air laws already in place and the extent of private work site restrictions already implemented. By January 2005, 11 states had adopted smoke-free restaurant laws (California, Connecticut, Delaware, Florida, Idaho, Maine, Massachusetts, New York, Rhode Island, Utah, and Vermont) and 10 states (California, Delaware, Florida, Massachusetts, Maryland, New York, Oregon, Rhode Island, South Dakota, and Washington) had adopted stricter work laws that required smoke-free or separately ventilated areas for smoking. We also consider the effect of partial bans and the percentage of firms that currently have strict bans. In addition to taking into account the extent of policies already in effect, the model considers changes over the tracking period (2002–2005). We estimate that 72 percent of work sites already had strict bans by 2005 (up from 67 percent in 2002) and that 36 percent of restaurants and 31

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation percent of other public places were covered in 2005. We estimate that enforcement and publicity were at half the maximum level from 2002 to 2005. Mass Media The mass media policy module is based on a model of the effect of media campaign expenditures on smoking prevalence (Levy and Friend 2001). Media expenditures must be high enough for messages to reach potential smokers and quitters a sufficient number of times, but after a threshold, additional expenditures show diminishing returns. The effects of media campaigns also depend on other policies that are currently in place. In particular, many states have comprehensive programs as well as local programs and cessation treatment programs. These can also be accompanied by tax increases. These other programs create added publicity, which reinforces the messages of the media campaign, thus having more potential to change attitudes toward smoking. The model distinguishes policies aimed at the entire population and those aimed primarily at youth. The early California, Massachusetts, and Arizona (after the first year) campaigns directed their efforts to all ages. In Massachusetts, where price stayed constant, there was a 6 percent reduction in prevalence with no price change and similar effects are implied for California after netting out prices (CDC 1996; Farrelly et al. 2003; Friend and Levy 2002). It is estimated that across states and over time, tobacco control expenditures at high levels (including an intensive media campaign) would reduce per capita tobacco consumption (which includes prevalence and quantity smoked per smoker changes) by 8 percent, and a recent meta-analysis (Snyder et al. 2004) found that media campaigns (most of which were generally part of a more comprehensive tobacco program) yielded a 5 percent reduction in smoking prevalence. Studies generally have not been able to distinguish the effect of media campaigns from other aspects of comprehensive programs. Using the formal model presented by Levy and colleagues (2001a) that provides the relationship between per capita expenditures and reductions in smoking prevalence, SimSmoke predicts that a highly publicized mass media campaign (publicized heavily on television and other media) directed at all smokers yields a 6 percent reduction in smoking prevalence, which increases over time to as much as 7 percent (Levy et al. 2001a). A low-publicity campaign (publicized only sporadically) has 20 percent the effect of a highly publicized campaign. In the absence of other policies, the effects are halved. These parameters have been accepted by the IOM panel. Media and comprehensive campaigns in Florida and Arizona in the first year and the American Legacy Foundation campaign since 2000 have been directed at youth. A recent study (Tauras et al. 2005) obtained results for youth that were broadly consistent with those found for the effect of adult campaigns, and a recent study (Farrelly et al. 2005) indicated a 7 percent reduction in smoking prevalence associated with the American Legacy Foundation campaign (22 percent of the overall 36 percent decline in youth smoking prevalence). We estimate that youth-oriented campaigns lead to a 6 percent reduction in youth prevalence. We do not consider the effect on smokers ages 18–24 and on those age 25 and above, due to the lack of studies. To incorporate the effect of past media campaigns, state per capita expenditures in 2002 were used to calculate the implied annual reductions in smoking rates by state. The annual reductions were then weighted by the number of smokers in a state, with separate estimates for campaigns directed at youth and all ages. Between 1993 and 1999, Massachusetts—followed by Utah, Arizona, Florida, and Oregon—implemented campaigns. California had a media campaign prior to 1993. Since 1999, Alaska, Maine, Idaho, Indiana, Maryland, Minnesota, Mississippi, New York,

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation New Jersey, and Vermont have added campaigns, but many were directed primarily at youth, and some were conducted at a low level. Since January 2002, Arkansas, Hawaii, and Delaware have implemented campaigns, but many states (including California, Colorado, Massachusetts, Minnesota, New Jersey, and Oregon) reduced campaign expenditures due to fiscal constraints (www.Tobaccofreekids.org/reports/settlements/2004/trends.pdf and www.slati.lungusa.org/reports/SLATI2004MidTermReport.pdf). Due to difficulties in obtaining measures in recent years, media campaigns are considered only for 2002, but media expenditures since then have decreased in some states. Using our mass media model, which relates per capita expenditures to reductions in smoking rates (Levy and Friend 2001), we estimate a 1.5 percent reduction from campaigns implemented in 2002. For youth campaigns, we include the American Legacy Foundation national truth­® campaign directed at youth. We estimate that current youth campaigns reduced smoking prevalence by 5 percent for the years 2002 through 2005. School Education Programs School education policies are added to the model for this report. They consist of well-tested programs applied through middle and high school. These programs have been shown to be more effective when accompanied by sustained media campaigns directed at youth. The effect sizes are based primarily on studies using the 30-day prevalence measure of smoking, which are assumed to ultimately lead to reductions in established smoking. Since the model is in terms of established smokers, the effect of school-based programs has been developed in terms of their ultimate effect on initiation rates into established smoking. Based on the review in this report of school programs (Flay, Appendix D), it is estimated that sustained school programs alone reduce smoking rates by 10 percent and by 20 percent if accompanied by a sustained media campaign. The 10 percent incremental effect of media campaigns reflects the synergies from implementing the campaign in conjunction with the educational programs and is thus higher than the effect of a youth campaign alone (as described above). These effects are modeled as across-the-board reductions in initiation rates at all ages through age 24 applied to males and females. Because there is a lag between the programs and their ultimate effect on initiation, we assume that the program affects initiation rates of youth through age 15 in the first year that the program is in effect, through age 16 the second year that the program is in effect, and one additional age each for each year of implementation through age 24, the last age of initiation. Because current educational programs are generally not implemented in a consistent manner (in other words, using well-tested formats continuously applied throughout middle and high school), it is assumed that they have no measurable effect. There have, however, been youth campaigns in effect, through the American Legacy Foundation campaign and various state campaigns, as described above. The education policy would, therefore, have the entire effect described above, but a concurrent media campaign effect would only reflect the difference between the current campaigns and the additional effects from having the campaigns in conjunction with the educational campaign. Cessation Treatment Policies In a previously published version of the cessation treatment policy module, SimSmoke considers the effects of mandated brief interventions delivered by health care providers to encourage patients to quit smoking, and complete financial coverage of cessation treatments with the

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation smoker having the flexibility to choose from the array of treatment options (Levy and Friend 2002a; Levy and Friend 2002b). Physicians receive training, their practices are monitored, and the financial coverage is well publicized. In that version, cessation policies only affect first-year quit rates. They increased the quit rate by 28 percent, which translates into a 1.4 percent decrease in smoking prevalence in the first 2 years and a 5 percent decreasing after 20 years. The cessation treatment module has been revised for the purposes of this report to consider a more all-inclusive policy. In particular, the module considers the effect of quitlines that are well publicized (e.g., through a media campaign) and that encourage follow-up with multiple sessions. In addition, the quitline is accompanied by a “free NRT” (nicotine replacement therapy) program that enables quitline callers to obtain NRT for a specified period of time. The module has also been modified to allow for a direct prevalence effect as well a continuous effect on the future first-year quit rate as long as the program is in operation. The effect on future one-year quit rates is halved to reflect the greater use of treatments and effectiveness of interventions in the first year of the program. Parameters in the new module have been developed in cooperation with the IOM committee. In the revised module, we set the quit success rate of those who complete a quit attempt at 6 percent, which is consistent with an overall quit rate of 4 percent with about 45 percent of smokers making a quit attempt. We continue to assume that behavioral or pharmacotherapy use doubles quit rates, and their combined use quadruples quit rates (Fiore et al. 2000). Proactive quitlines with follow-up double the quit success rate of those making a quit attempt (Zhu et al. 2002). We estimate that quitlines alone with high media publicity attract 1 percent of smokers and, when free NRT is added, attract 6 percent of smokers (Metzger et al. 2005; Miller et al. 2005; West et al. 2005), of whom 30 percent are new quit attempts. Through the complete coverage of effective cessation treatments, an additional 4 percent of smokers use cessation treatment alone, 2 percent use behavioral treatment alone, and 3 percent use combined pharmacotherapy and behavioral treatment. We estimate that 50 percent of those who use treatments as a result of the policy would not otherwise have made a quit attempt. In addition, brief interventions increase quit attempts by 20 percent and further increase new treatment use (through quitlines and other financial access) by 10 percent. With the combined policies, quit attempts in the first year increase by 40 percent (from 45 to 63 percent of smokers) and average quit success (per quit attempt) increases by 28 percent (from 8.9 to 11.4 percent). As a result of all the policies, the prevalence of smokers is reduced by 3.4 percent in the first-year, and future first-year quit rates increase by about 20 percent. From 2002 forward, the module takes into account the level of treatment coverage and health care involvement. By 2003, 36 Medicaid programs covered some counseling or medication for all Medicaid recipients, but only New Jersey and Oregon offered comprehensive coverage and Medicare did not provide coverage (CDC 2004). Measures of insurance coverage by private payers are more limited (Levy and Friend 2002b). A study of managed care organizations (McPhillips-Tangum et al. 2002) found that 59 percent of plans had some type of pharmacotherapy coverage and 86 percent had some kind of behavioral coverage, but a study of employer coverage (www.cdc.gov/tobacco/educational materials/essation/ page1.html) found that only 24 percent of employers provided any type of cessation treatment coverage. In 2002, 14.6 percent of adults were not covered at any time in last year, 71 percent were covered by private insurance, 11 percent by Medicaid, and 13.5 percent by Medicare (www.ferrer.bls.census.gov/macro/032002/health/h02_001.htm). We estimate that less than 20 percent of the population is covered for pharmacotherapy and for behavioral therapy, and these

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation benefits are not well publicized. We estimate that about 40 states have quitlines—but these quitlines are generally not widely publicized and do not provide free pharmacotherapy (www.cdc.gov/tobacco/quit/Quitlines/Appendix.pdf)—and that about 50 percent of smokers were receiving brief interventions. Youth Access The youth access module considers the effect of self-service and vending machine bans, and three components of retail compliance (compliance checks, penalties, and merchant awareness or community mobilization). The module also takes into account that, as retail sales to youth are reduced, youth switch to non-retail sources such as theft, older peers, and parents. The model considers three levels of enforcement: (1) strongly enforced and publicized (compliance checks are conducted four times per year per outlet, penalties are potent and enforced, and there is heavy publicity and community involvement); (2) well enforced (compliance checks are conducted regularly, penalties are potent, and publicity and merchant training are included, but there is little community support); and (3) weakly enforced (compliance checks are conducted sporadically, penalties are weak, and there is little merchant awareness along with minimal community participation). With a strongly enforced and well-publicized program, we estimate a 20 percent reduction in youth smoking prevalence and future initiation for 16–17 year olds when all policies are in full force, with a 30 percent reduction for those ages 10–15 years (Levy et al. 2001a; Levy and Friend 2000). The well-enforced and weakly enforced policies, respectively, yield 50 percent and 10 percent of the effect. These policies work through the prevalence and initiation rate, but do not affect cessation. These parameters have been accepted by the IOM panel. Data from the Substance Abuse and Mental Health Administration website (http://prevention.samhsa.gov/tobacco/01synartable.asp) indicate that noncompliance is about 15 percent, but these figures may be overstated because they affect future funding. Based on current compliance rates and programs in effect across states, we estimate that states on average have a low enforcement policy. Prediction of Status Quo Trends and the Effect of Tobacco Control Policies The model provides a prediction of smoking prevalence from the year 2002 through 2005, taking into account changes in policies during that time period. The model will be used to project smoking rates in future years beginning in the year 2006. We will consider the smoking prevalence rate over a 20-year time horizon ending in 2025. We examine rates for the adult population (ages 18 and above), as well as breakdowns by age. First, we present a status quo scenario. This scenario incorporates policies in the year 2002 and changes in policy between 2002 and 2005, and then holds policies constant at their 2005 levels to project changes in smoking prevalence in the absence of any policy changes. We then consider the effect of policy changes on smoking rates in future years. Policy changes are made in the year 2006 and maintained in all future years. Their incremental effect will depend on the level of policies in effect in 2005. The effects of policies are presented relative to the status quo level in the same year, that is (Policy RateT − Status Quo RateT)/ Status Quo RateT. Worst Case Scenario Since 1998, large price increases, new clean air laws, and other tobacco control policies have been associated with a reduction in smoking prevalence of about 20 percent (Levy et al. 2005a).

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation However, this reduction might be reversed if policy changes are not maintained. We consider a reverse in some of the more prominent policies, especially those relating to the settlement funds (including those to the American Legacy Foundation), which we call a worst-case scenario. This scenario typifies the possibilities if tobacco control regresses, as it has done in some states. While states are not expected to reduce taxes, cigarette manufacturers might be expected to reduce price to gain back some of the customers that they have lost in recent years due to price increases (Levy et al. 2005a), especially if public pressure is reduced. For example, after the settlement, average cigarette prices increased about $0.80 net of tax increases. That price increase might be reversed once the settlement is no longer an issue, since there would no longer be an incentive to raise prices to reduce youth consumption, and thereby, reduce the size of settlement payouts. With the settlement abandoned, prices might be expected to decrease. We separately consider price reductions of $0.40 and $0.80. As part of the worst-case scenario, we also assume that current taxes do not adjust to future inflation, suggesting a slight erosion of taxes over time. Clean air laws are not expected to revert, but compliance with those laws might be lower as less attention is focused on tobacco control, especially if media campaigns are abandoned. We consider the effect of reduced publicity and enforcement surrounding the laws. In recent years, media campaigns have been abandoned in some states, such as Massachusetts, and faced large cutbacks in others, such as California. The American Legacy Foundation campaign may also be abandoned. We consider the effect of reductions in those campaigns from their current level to no funding, both for youth and for adults. Education policies are currently at levels where they are of minimal effectiveness. Consequently, no change is expected under the worst-case scenario. In addition, no change is expected for youth access policies. These policies are not currently conducted at levels that are expected to have large effects on smoking prevalence, especially adult prevalence, and large cutbacks are not expected. For cessation, we consider only the elimination of the quitlines. It is not expected that Medicaid or other coverage is likely to revert. We then consider the reversion of all policies: a reduction in price of $0.80, a reduction in enforcement and clean air laws, a reduction in adult media campaigns, a reduction in youth media campaigns, and a reduction in cessation treatment programs. Future Policies Finally, we consider the effect of strengthening current policies to what might be viewed as the desired set of policies recommended by the IOM committee. We consider changes in the following policies, individually and in combination: We consider tax increases of $1.00, $2.00, and $3.00. We assume that these taxes are indexed to inflation, so that their value is maintained over time. We consider a clean air policy that bans smoking at all work sites—which includes bars, restaurants, and grocery stores—plus increased compliance through publicity from other policies (especially media policies regarding secondhand smoke). We consider an intensive media campaign as part of a more comprehensive strategy (at levels recommended by the Centers for Disease Control and Prevention), directed at adults and youth in all states. We will consider a comprehensive cessation treatment policy with all of the policies described above (full coverage of pharmacotherapy and behavioral therapy, training and mandated

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation tobacco brief interventions, and multi-session quitlines with free NRT). We further assume that the policies are well-publicized. School education policies consist of well-tested programs applied through middle and high school. We include a media campaign as part of the policy. The youth access policy is assumed to be conducted at a high enforcement level, with high penalties, and be well-publicized. The policy is part of the comprehensive campaign, implying a high degree of community mobilization. RESULTS We present the effect of varying levels of tobacco control policies in isolation and together through a comprehensive tobacco control strategy. The estimates of smoking prevalence under the status quo and varying policy scenarios are shown for the adult population (18 years of age and above) in Tables J-2 and J-3. The Status Quo Scenario The model begins in 2002 with policy levels and changes in policy inputted into the model through 2005. The smoking prevalence is estimated as 21.7 percent in 2002, falling to 20.6 percent in 2004. Part of this decline is due to long-run trends, including policies implemented before 2002, and part is due to policies implemented between January 2002 and January 2005. The average price increased about 14 percent, and several states implemented clean air laws. New policies are implemented and maintained from 2006 through 2025. Their effect on smoking prevalence is presented relative to the status quo, in which tobacco control policies remain unchanged from their 2005 levels. In the status quo scenario, adult smoking prevalence is projected to decline from the 2005 level of 20.6 percent to the 2010 level of 19.3 percent. This absolute reduction of 1.3 percentage points represents a 6.3 percent decline relative to the 2005 level. Kept at 2005 policy levels, smoking rates are projected to fall to 15.5 percent by 2025. This drop represents a 24.7 percent decrease relative to the 2005 level. At least some of the reduction in smoking prevalence is explained by stricter public policies implemented prior to 2005, including the increase in prices since 1998, more stringent smoking restrictions in work and public places, and better information about the effects of smoking (DHHS 2000). The largest reductions are among the 35–64-year-old age groups, due to higher cessation rates among those groups and reduced initiation at earlier ages. Worst-Case Scenario In the worst-case scenario, we first look at the effect of decreasing the average tax rate. A $0.40 decrease without taxes indexed to inflation will lead to a projected 1.6 percent relative increase in adult smoking prevalence within 5 years compared to the status quo, and by the year 2025 it will lead to a 5 percent relative increase. A $0.80 decrease in average tax price is projected to have an even greater effect, causing a 3 percent relative increase within 5 years compared to the status quo, rising steadily to a 7.6 percent relative increase in adult smoking prevalence by the year 2025. Taking away enforcement and publicity of clean air laws has a smaller effect than a tax decrease. This reduction in clean air laws is projected to cause a 0.2 percent relative rise in smoking prevalence compared to status quo within 5 years, and a 0.5 percent relative rise after 20 years. Reductions in media coverage lead to slightly larger increases in smoking prevalence

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation compared to the reduction in clean air laws. Abandoning adult media campaigns is projected to cause a 0.3 percent relative rise in smoking prevalence compared to status quo within 5 years and an 0.8 percent relative rise after 20 years. Abandoning youth media campaigns is projected to cause a 0.1 percent relative rise in smoking prevalence compared to status quo within 5 years and a 0.3 percent relative rise after 20 years. Reductions in cessation treatment policies rise 0.2 percent after 5 years and have a 2.0 percent relative rise after 20 years. Finally, we consider the elimination of all policies: a reduction in price of $0.80, a reduction in enforcement and clean air laws, a reduction adult media campaigns, and a reduction in cessation treatment programs. After 5 years, these reductions are projected to lead to a 3.5 percent increase in smoking prevalence relative to the status quo. The smoking prevalence is projected to increase steadily relative to the status quo. After 20 years, smoking prevalence is projected to be 17.1 percent compared to 15.5 percent under the status quo, or a 10.4 percent relative increase. Relative to the status quo, most of the increases are among the younger age groups due to the greater effect of price increases on those age groups (especially below age 35). Taxes Of the tobacco control policies, SimSmoke attributes the most pronounced effect on smoking prevalence trends between 1993 and 2003 to taxes (Levy et al. 2004e). However, the same absolute increase in taxes or price has a smaller percentage effect at the higher prices found in 2005 than in earlier years, since prices are now at a higher rate and the changes represent smaller relative increases. An increase in the average tax rate of $0.50 from the 2005 level is projected to result in an absolute decline of 0.5 percent in adult smoking prevalence compared to the status quo between 2005 and 2010, which represents a 2.4 percent relative drop. This decrease is projected to continue steadily, reaching a 0.7 percent decline compared to the status quo by the year 2025, which represents a 4.4 percent relative drop. An increase in the average tax rate of $1.00 is projected to result in a 0.9 percent absolute (a 4.4 percent relative) reduction compared to status quo within the first 5 years, rising to a 1 percent reduction (6.8 percent relative to status quo) by 2025. An increase in the average tax rate of $2.00 is projected to result in a 1.5 percent reduction (7.7 percent relative) compared to status quo by the year 2010 and a decrease of 1.8 percent (11.8 percent relative) by 2025. Finally, a $3.00 average tax increase is projected to result in a 2.0 percent reduction in adult smoking prevalence compared to status quo in the first 5 years, which represents a 10.3 percent relative reduction. The smoking prevalence is projected to have a 2.4 percent absolute (15.8 percent relative) decline compared to the status quo by the year 2025. The largest effects of the price increases are on those at younger ages, particularly those below age 18. Consequently, the growth in effect over time is primarily because youth are more responsive to price increases than adults. We also assume that taxes increase with the rate of inflation over time, but some of the effect on smoking prevalence dissipates over time if the perunit taxes are not indexed to inflation (Levy et al. 2000b). Clean Air Policies Clean air policies have a similar, although smaller, effect on smoking prevalence compared to tax policies. The model predicts the effects derived from implementing a total smoking ban in workplaces, restaurants, and public places supported by both publicity and enforcement of the ban. By 2010, these policies lead to a 3.4 percent relative drop in the smoking rate compared to the status quo. By 2025, the model predicts a 4.2 percent drop compared to the status quo, the increased effect due primarily to higher cessation rates. The largest effects are among those in

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation the 25–64-year-old age groups, particularly 35–44-year-old groups, due to the more pronounced effect of work site laws on this group (particularly among males). Mass Media We examine a media campaign directed at all smokers implemented at a high intensity, used in conjuction with other programs, and maintained over time. The decline of 1.1 percent in adult smoking after 5 years compared to status quo translates to a 5.9 percent relative decrease. The effect increases steadily to a 7.3 percent reduction relative to the status quo by 2025. Media campaigns initially have a greater effect on younger smokers, but have greater effects on older smokers in later years. School Education Policies We look at the effects of a sustained school program combined with a media campaign directed at youth. There is a very small projected decline in adult smoking prevalence after 5 years, which is to be expected considering this policy is directed at youth. After 10 years, this policy is project to result in an absolute decline in adult smoking prevalence of 0.5 percent compared to status quo, which is a relative decline of 2.9 percent. By the year 2025 there will be a projected 0.9 percent absolute or 5.9 percent relative drop in smoking prevalence compared to the status quo. These programs only directly affect youth, but their effects spread to lower prevalence rates at older ages over time. Cessation Policies A policy of mandated brief interventions delivered by health care providers, along with full financial coverage of cessation treatments and well-publicized quitlines with free NRT, have smaller effects in the earlier years of the projection, but their impact grows over time through increased cessation rates, which affect those greater than age 24 (Levy and Friend 2002b). The combined cessation policies are projected to reduce adult smoking prevalence by an absolute value of 1.1 percent by 2010 or, in other words, a 5.8 percent relative improvement over the status quo scenario. This effect grows to an 11.2 percent reduction relative to the status quo by 2025. Youth Access Policies We look at a policy of strict control of youth access (bans on access to self-service and vending machines in addition to strict retail compliance checks, penalties for noncompliance, and a high level of publicity). Initially, smoking rates of youth are reduced by about 25 percent. Not surprisingly, adult smoking rates (of which youth are included only in later years) decline by a small amount (1.1 percent) relative to the status quo by 2010, with a greater relative decline of 5.1 percent by 2025 as a large portion of youth affected by the policies become older, replacing those cohort with higher initiation rates. Best-Case Scenario: A Comprehensive Set of Policies The final cases consider a combination of policies representing a tax increase of $1.00, $2.00, and $3.00, along with work site, restaurant, and public place smoking bans with publicity and enforcement; a high-intensity media campaign; comprehensive cessation policies; and strict youth access policies. With a $1.00 tax and the other policies, the smoking rate is projected to

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation fall by 19.7 percent below the status quo level by 2005. Maintaining this policy is projected to reduce the smoking rate 34.0 percent relative to the status quo by 2025. With a $2.00 tax and the other policies, the smoking rate is projected to fall to 14.9 percent by 2010, which is 22.5 percent below the status quo level of 19.3 percent in relative terms. Maintaining this policy is projected to reduce the smoking rate to 9.7 percent compared to a status quo level of 15.5 percent by 2025, which is 37.6 percent lower relative to the status quo. With a $3.00 tax and the other policies, the smoking rate is projected to fall to 14.5 percent by 2010, which is 24.7 percent below the status quo level in relative terms. Maintaining this policy is projected to reduce the smoking rate to 9.3 percent by 2025, which is a 40.3 percent reduction relative to the status quo. Of the policies in the comprehensive package, media campaigns, clean air laws, and taxes have the greatest effect in 2010, but cessation treatment, education, and youth access policies play a greater role by 2025. Some policies have a greater impact on adult smoking prevalence and others on youth prevalence. Overall, the largest effects are on youth through the effects of price, youth access policies, and education programs. The effects of a comprehensive policy strategy are shown as the SimSmoke screen in Figure J-1. CONCLUSIONS From the current smoking prevalence of about 20.6 percent, the SimSmoke model projects a reduction in smoking rates to 19.3 percent by 2010, if policies are maintained at their 2005 levels. The decline occurs due to the aging of older cohorts and the impetus from policies in years through 2005. This rate is substantially above the Healthy People 2010 target of 12 percent. By the year 2025, the smoking rate is projected to fall to 15.5 percent in the absence of policy change. However, if policies regress (the worst-case scenario), the model predicts that the smoking rate would be at 17.1 percent, about 10 percent higher than the status quo scenario. We considered a package of policies as suggested in this report. With a cumulative set of policies (with taxes increased $2.00), we predict that smoking prevalence will fall to about 15 percent by 2010, which is 23 percent below the status quo level of 19 percent in relative terms, and to about 10 percent by 2025, or a 40 percent decrease relative to the status quo. The cumulative impact of the comprehensive set of policies over a 20-year period provides strong encouragement for implementing the policy objectives set out in this report. In summary, the SimSmoke model suggests that policies can have a large impact on smoking rates. With the implementation of strong policies, we project a smoking rate of about 15 percent in 2010. Evidence from California, which has had policies in line with these goals, suggests that this projection is attainable (Gilpin et al. 2003). Maintaining policies at high levels could reduce the smoking prevalence in the United States to about 10 percent by 2025. Although later, these levels are in line with Healthy People 2010 targets.

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation TABLE J-1 Data Used in SimSmoke Variable Source Specifications I. Population model A. Population 2002 Current Population Survey (CPS) Breakdowns by age and gender B. Fertility rates U.S. Census Vital Rate Inputs Tables, 2000 Breakdowns by age C. Mortality rates 2001 Multiple Cause-of-Death File, NCHS Breakdowns by age and gender: total deaths II. Smoking model A. Baseline prevalence rates for current and ex-smokers Tobacco Use Supplement of the CPS (1992—93) for age 15+, and 1993 Teenage Attitudes and Practices Survey (TAPS) for <age 15 Based on 100+ cigarettes lifetime and distinction between current and previous smokers. Breakdowns by smoking experience (<1, 1–2, 3–5, 6–10, 11–14, 15+ years), by age and gender B. Initiation rates Change in smoking rates between contiguous age groups Breakdowns by age and gender C. First-year quit rates Calculated from cessation module with adjusters for demographic group based on the CPS Breakdowns by age and gender D. Relapse rates (DHHS 1989) McWhorter et al. 1990; U.S. DHHS 1990; Gilpin et al. 1997), COMMIT data Breakdowns by age E. Relative death risks of smokers and ex-smokers Cancer Prevention Study II (see Thun et al. 2001) Breakdowns by age and gender III. Policy modules A. Taxes Tobacco Institute, Tobaccofreekids.org, www.bls.gov/cpi/home.htm Prices and taxes for 2002—05 B. Clean air laws www2.cdc.gov/nccdphp/osh/state/report_index.asp and slati.lungusa.org/searchform.asp (National Cancer Institute 2000) Different types of laws and their stringency C. Media and other educational campaigns CDC and various state websites: (Farrelly Matthew C et al. 2003; Wakefield and Chaloupka 2000) Expenditures per capita and audience

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation D. Youth access CDC, SAMHSA, (Levy et al. 2001a) Enforcement checks, penalties, community campaigns, self-service, and vending machine bans NOTE: CDC = Centers for Disease Control and Prevention; NCHS = National Center for Health Statistics; SAMHSA = Substance Abuse and Mental Health Administration.

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation TABLE J-2 Projected Adult Smoking Prevalence (%) from 2003 to 2008, with Projection Through 2025 under Status Quo and Worst-Case Policy Scenariosa YEAR 2002 2005 2006 2010 2015 2020 2025 Prevalence (%) Status quo 21.7 20.6 20.3 19.3 18.1 16.9 15.5 $0.40 price reduction 21.7 20.6 20.4 19.6 18.6 17.6 16.3 $0.80 price reduction 21.7 20.6 20.6 19.9 18.9 18.0 16.7 Clean air reduction 21.7 20.6 20.3 19.3 18.1 17.0 15.6 Media reduction 21.7 20.6 20.3 19.4 18.2 17.0 15.7 Youth media reduction 21.7 20.6 20.3 19.3 18.3 17.2 15.8 Cessation reduction 21.7 20.6 20.3 19.3 18.1 17.0 15.6   21.7 20.6 20.6 20.0 19.2 18.4 17.1 % Change in Prevalence from Status Quob Status quo               $0.40 price reduction     0.7 1.6 2.8 4.0 5.0 $0.80 price reduction     1.4 3.0 4.8 6.4 7.6 Clean air reduction     0.1 0.2 0.3 0.4 0.5 Media reduction     0.2 0.3 0.5 0.7 0.8 Youth media reduction     0.0 0.1 0.2 0.3 0.3 Cessation reduction     0.0 0.2 1.0 1.6 2.0       1.6 3.8 6.4 8.8 10.4 SOURCE: SimSmoke model. a Policies are implemented and maintained from year 2005 forward. b Percent changes calculated relative to the status quo rate at (Policy Rate - Status Quo Rate)/Status Quo Rate.

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation TABLE J-3 Projected Adult Smoking Prevalence (%) from 2003 to 2005, with Projections Through 2025 Under Status Quo and Best-Case Policy Scenariosa YEAR 2002 2005 2006 2010 2015 2020 2025 Prevalence (%) Status quo 21.7 20.6 20.3 19.3 18.1 16.9 15.5 $0.50 tax increase 21.7 20.6 19.9 18.8 17.6 16.4 14.9 $1.00 tax increase 21.7 20.6 19.6 18.4 17.1 15.9 14.5 $2.00 tax increase 21.7 20.6 19.1 17.8 16.4 15.1 13.7 $3.00 tax increase 21.7 20.6 18.6 17.3 15.8 14.5 13.1 Clean air laws 21.7 20.6 19.7 18.6 17.4 16.3 14.9 Media campaign 21.7 20.6 19.2 18.1 16.9 15.8 14.4 Cessation treatment 21.7 20.6 19.7 18.2 16.7 15.3 13.8 Education programs 21.7 20.6 20.3 19.2 17.6 16.1 14.6 Youth access policies 21.7 20.6 20.3 19.1 17.6 16.2 14.7 All policies with $1.00 tax 21.7 20.6 17.4 15.5 13.4 11.8 10.2 All policies with $2.00 tax 21.7 20.6 16.9 14.9 12.9 11.2 9.7 All policies with $3.00 tax 21.7 20.6 16.5 14.5 12.4 10.8 9.3 % Change in Prevalence from Status Quob Status quo               $0.50 tax increase     −1.8 −2.4 −2.9 −3.3 −3.7 $1.00 tax increase     −3.4 −4.4 −5.3 −6.2 −6.8 $2.00 tax increase     −6.1 −7.7 −9.3 −10.7 −11.8 $3.00 tax increase     −8.2 −10.3 −12.4 −14.2 −15.6 Clean air laws     −3.1 −3.4 −3.7 −3.9 −4.2 Media campaign     −5.5 −6.0 −6.5 −6.9 −7.4 Cessation treatment     −3.1 −5.8 −7.8 −9.4 −11.2 Education programs     0.0 −0.7 −2.9 −4.8 −5.9 Youth access policies     0.0 −1.1 −2.8 −4.3 −5.1 All policies with $1.00 tax     0.0 −19.7 −25.7 −30.5 −34.0 All policies with $2.00 tax     −14.3 −22.5 −28.9 −33.8 −37.6 All policies with $3.00 tax     −16.7 −24.7 −31.3 −36.4 −40.3 SOURCE: SimSmoke model. a Policies are implemented and maintained from year 2005 forward. b Percent changes calculated relative to the status quo rate at (Policy Rate − Status Quo Rate)/Status Quo rate.

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation REFERENCES CDC (Centers for Disease Control and Prevention). 1996. Cigarette smoking before and after an excise tax increase and an antismoking campaign--Massachusetts, 1990–1996. Morbidity and Mortality Weekly Report 45(44):966-970. CDC (2004). State medicaid coverage for tobacco-dependence treatments—United States, 1994–2002. Morbidity and Mortality Weekly Report 53(3):54-57. DHHS (Department of Health and Human Services). 1989. Reducing the Health Consequences of Smoking: 25 Years of Progress: A Report of the Surgeon General. Rockville, MD: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. DHHS. 1990. The Health Benefits of Smoking Cessation: A Report of the Surgeon General. Atlanta, Georgia: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, Office on Smoking and Health. DHHS. 1994. Preventing Tobacco Use Among Young People (A Report of the Surgeon General). Rockville, MD: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, Center for Health Promotion and Education, Office on Smoking and Health. DHHS. 2000. Reducing Tobacco Use: A Report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. DHHS. 2004. The Health Consequences of Smoking: A Report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. Farrelly MC, Davis KC, Haviland ML, Messeri P, Healton CG. 2005. Evidence of a Dose—Response Relationship Between “truth” Antismoking Ads and Youth Smoking Prevalence. American Journal of Public Health 95(3):425-431. Farrelly MC, Pechacek TF, Chaloupka FJ. 2003. The impact of tobacco control program expenditures on aggregate cigarette sales: 1981–2000. Journal of Health Economics 22(5):843-859. Farrelly Matthew C, Pechacek Terry F, Chaloupka Frank J. 2003. The impact of tobacco control expenditures on aggregate cigarette sales: 1981–2000. Journal of Health Economics 22:843-859. Fiore MC, Bailey WC, Cohen SJ. 2000. Treating Tobacco Use and Dependence (Clinical Practice Guideline). Rockville, MD: U.S. Department of Health Human Services. Public Health Service. Friend K, Levy DT. 2002. Reductions in smoking prevalence and cigarette consumption associated with mass-media campaigns. [Review]. Health Education Research 17(1):85-98. Gilpin EA, Emery SL, Farkas AJ, Zhu SF, et al. (2003). Tobacco Control Successes in California: A Focus on the Young People, Results from the California Tobacco Control Surveys, 1990–2002. La Jolla, CA: University of California, San Diego. Gilpin EA, Pierce JP, Farkas AJ. 1997. Duration of smoking abstinence and success in quitting. Journal of the Nationall Cancer Institute 89(8):572-576. Hughes JR. 2000. Reduced smoking: an introduction and review of the evidence. Addiction 95 (Suppl 1):S3-S7. Hymowitz N, Sexton M, Ockene J, Grandits G. 1991. Baseline factors associated with smoking cessation and relapse. MRFIT Research Group. Preventive Medicine 20(5):590-601. Hymowitz N, Cummings KM, Hyland A, Lynn WR, Pechacek TF, Hartwell TD. 1997. Predictors of smoking cessation in a cohort of adult smokers followed for five years. Tobacco Control 6 (Suppl 2):S57-S62. Jha P, Chaloupka FJ. 2000. The economics of global tobacco control. British Medical Journal 321:358-361. Levy DT, Bauer J, Ross H, Powell L, Nikolayev N. 2004a. The Role of Public Policies in Reducing Smoking Prevalence and Deaths Caused by Smoking in Arizona: Results from the Arizona Tobacco Policy Simulation Model. Calverton, MD. Levy DT, Chaloupka F, Gitchell J, Mendez D, Warner KE. 2002. The use of simulation models for the surveillance, justification and understanding of tobacco control policies. Health Care Management Science 5(2):113-120. Levy DT, Cummings KM, Hyland A. 2000a. A simulation of the effects of youth initiation policies on overall cigarette use. American Journal of Public Health 90(8):1311-1314.

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation Levy DT, Cummings KM, Hyland A. 2000b. Increasing taxes as a strategy to reduce cigarette use and deaths: results of a simulation model. Preventive Medicine 31(3):279-286. Levy DT, Friend KB. 2000. A simulation model of tobacco youth access policies. Journal of Health Politics, Policy and Law 25(6):1023-1050. Levy DT, Friend K. 2001. A computer simulation model of mass media interventions directed at tobacco use. Preventive Medicine 32(3):284-294. Levy DT, Friend K. 2002a. A simulation model of policies directed at treating tobacco use and dependence. Medical Decision Making 22(1):6-17. Levy DT, Friend K. 2002b. Examining the effects of tobacco treatment policies on smoking rates and smoking related deaths using the SimSmoke computer simulation model. Tobacco Control 11(1):47-54. Levy DT, Friend K, Holder H, Carmona M. 2001a. Effect of policies directed at youth access to smoking: results from the SimSmoke computer simulation model. Tobacco Control 10(2):108-116. Levy DT, Friend K, Polishchuk E. 2001b. Effect of clean indoor air laws on smokers: the clean air module of the SimSmoke computer simulation model. Tobacco Control 10(4):345-351. Levy DT, Mumford E, Pesin B. 2003. Tobacco control policies and reductions in smoking rates and smoking-related deaths: Results from the SimSmoke Model. Expert Review of Pharmacoeconomics and Outcomes Research 3(4): 457-468. Levy DT, Nikolayev N, Mumford EA Compton C. 2004b. The Healthy People 2010 smoking prevalence and tobacco control objectives: results from the SimSmoke Tobacco Control Policy Simulation Model. Cancer Causes and Control 16(4):359-371. Levy DT, Nikolayev N, Mumford EA. 2004c. The Role of Public Policies in Reducing Smoking Prevalence and Deaths Caused by Smoking in California:Results from the California Tobacco Policy Simulation Model. Calverton, Maryland: Pacific Institute. Levy DT, Nikolayev N, Mumford EA. 2004e. A validation of the SimSmoke simulation model: The role of tobacco control policies in recent years. PIRE Working Paper. Calverton, MD. Levy DT, Nikolayev L, Mumford E. 2005a. Recent trends in smoking and the role of public policies: results from the SimSmoke tobacco control policy simulation model. Addiction 100(10):1526-1536. Levy DT, Nikolayev L, Mumford E, Compton C. 2005b. The Healthy People 2010 smoking prevalence and tobacco control objectives: results from the SimSmoke tobacco control policy simulation model (United States). Cancer Causes Control 16(4):359-371. McPhillips-Tangum C, Cahill A, Bocchino C, Cutler CM. 2002. Addressing tobacco in managed care: Results from the 2000 survey. Preventive Medicine in Managed Care 3(3):85-95. McWhorter WP, Boyd GM, Mattson ME. 1990. Predictors of quitting smoking: the NHANES I followup experience. Journal of Clinical Epidemiology 43(12):1399-1405. Metzger KB, Mostashari F., Kerker BD. 2005. Use of pharmacy data to evaluate smoking regulations' impact on sales of nicotine replacement therapies in New York City. American Journal of Public Health 95(6):1050-1055. Miller N, Frieden TR, Liu SY, Matte T, Mostashari F, Deitcher D, Cummings K, Chang C, Bauer U, Bassett M. 2005. Effectiveness of a large-scale distribution programme of free nicotine patches: a prospective evaluation. Lancet 365(9474):1849-1854. NCI (National Cancer Institute). 2000. State and Local Legislative Action to Reduce Tobacco Use. Bethesda MD: U.S. DHHS, National Institutes of Health, National Cancer Institute. Orzechowski W, Walker RC. 2003. The Tax Burden on Tobacco: Historical Compilation, Vol. 37, 2002. Arlington, VA: The Tobacco Institute Snyder LB, Hamilton MA, Mitchell EW, Kiwanuka-Tondo J, Fleming-Milici F, Proctor D. 2004. A meta-analysis of the effect of mediated health communication campaigns on behavior change in the United States. Journal of Health Communication 9 (Suppl 1):71-96. Tauras JA, Chaloupka FJ, Farrelly MC, Giovino GA, Wakefield M, Johnston LD, O'Malley PM, Kloska DD, Pechacek TF. 2005. State tobacco control spending and youth smoking. American Journal of Public Health 95(2):338-344. Thun MJ, Henley SJ, Miracle-McMahill HL. 2001. Health consequences of tobacco use: total mortality. In Women and Smoking: A Report of the Surgeon General. Atlanta, GA: DHHS. Wakefield M, Chaloupka F. 2000. Effectiveness of comprehensive tobacco control programmes in reducing teenage smoking in the USA. Tobacco Control 9(2):177-186. West R, DiMarino ME, Gitchell J, McNeill A. 2005. Impact of UK policy initiatives on use of medicines to aid smoking cessation. Tobacco Control 14(3): 166-171.

OCR for page 578
Ending the Tobacco Problem: A Blueprint for the Nation Zhu SH, Anderson CM, Tedeschi GJ, Rosbrook, B, Johnson CE, Byrd M, Gutierrez-Terrell E. 2002. Evidence of real-world effectiveness of a telephone quitline for smokers. New England Journal of Medicine 347(14):1087-1093.