Describing changes in local short-term outcomes of youth-focused comprehensive tobacco control and their effect on smoking outcomes among cohorts of Minnesota youth, 2000-2005

Introduction. The halting decline in national youth smoking prevalence indicates the need for comprehensive tobacco control (CTC). Social ecological theory postulates that CTC reduces smoking by altering social environments that influence long-term youth tobacco use. At the local level, CTC's short-term outcomes result from (1) policy action, such as ordinances restricting public smoking and youth access to tobacco (2) school-based prevention, such as trained teachers offering proven curricula and (3) community mobilization, such as voluntary home smoking restrictions and youth engagement. Few CTC studies document the effect of a cutoff in CTC funding on these outcomes most studies can only weakly assert that this multilevel intervention causes youth smoking reductions. The Minnesota Adolescent Community Cohort (MACC) study evaluates the state's CTC 2000-2003 program, the Minnesota Youth Tobacco Prevention Initiative (MYTPI), and the effect of its sudden, major funding cut.The current study tested for (1) increases in short-term outcomes during the MYTPI and their sustainability after its shutdown (2) decreases in youth smoking during the MYTPI and increases in youth smoking after the shutdown and (3) a link between baseline levels of and changes in short-term outcomes to reduced youth smoking. Methods. MACC is a multilevel, population-based, observational cohort study of Minnesota youth (n=3,636) nested in geo-political units (GPUs N=60), which signified local communities. A comparison group of youth (n=605) came from six Midwest states that had not dedicated major tobacco control funding. Annual ordinance, youth access enforcement, and school administrator surveys and a twice-annual youth survey capture short-term outcomes in Minnesota. The youth survey also measures smoking outcomes. Data were collected from the MYTPI launch to two years post-shutdown (2000-2005).Analysis employs latent curve modeling, growth curves using structural equation modeling to parameterize the intercept and slope as latent variables. Piecewise latent curves modeled the effect of a sudden drop in tobacco control funding on short-term outcomes and youth smoking. Parallel process latent curves tested for contemporaneous change between short-term outcomes and youth smoking. Random coefficients models accounted for repeated measures and where appropriate, the multilevel nature of the data. Results. Implementation of short-term outcomes from school-based prevention and community-mobilization increased during the MYTPI however, this growth eventually stopped or slowed after the shutdown. Some policy action outcomes increased minimally, and one outcome of policy action declined post-shutdown. Increases in smoking stage were the same between Minnesota and comparison groups, showing no period effects for the MYTPI or shutdown. However, younger cohorts, with early-teen MYTPI exposure, smoked less than older cohorts by the same age. Only youth access ordinances scores and living in homes banning smoking were negatively associated with smoking stage. Discussion. MYTPI implemented less controversial educational and community-mobilization strategies, but was not as successful with permanent policy change before its shutdown. This lack of policy action and youth focus could explain why Minnesota youth did not differ from other states with concurrent national tobacco control. The study offers some evidence supporting MYTPI interventions, but study flaws may make the influence of several of them undetectable. Study limitations include lack of/poor comparison group, secular trends, selection bias, measurement error, and analysis issues.