Beyond Reach and Effectiveness

Abstract

Despite most teenage smokers wanting to quit, their likelihood of success resembles that of flipping a coin. Evidence-based cessation programs, like the American Lung Association’s Not-On-Tobacco (N-O-T) program, are effective. Evaluation of program dissemination is critical. This study uses the RE-AIM framework to evaluate the N-O-T program in West Virginia from 2000 to 2005. RE-AIM components consisted of four measures. Regional dissemination was measured using comparative differences between Regional Educational Service Agency regions (RESAs). Significant associations were found between RESAs for numerous characteristics. Among the RE-AIM components, two measures of Implementation were significantly different between RESAs. Variability between RESAs provided valuable descriptive evidence of N-O-T program dissemination in West Virginia. Therefore, geographical tailoring grounded in community-based participatory research could increase the N-O-T program’s overall dissemination.

Keywords

program evaluation RE-AIM adolescent smoking cessation geographical tailoring regional analysis

Despite 65% of daily teen smokers wanting to quit (Lamkin, Davis, & Kamen, 1998), most are unable to quit on their own. A 4-year study found smokers between the ages of 12 and 19 years had a successful quit rate of 15.6%, or approximately 4% per year (Zhu, Sun, Billings, Choi, & Malarcher, 1999). Comparatively, a current estimate of failed cessation rates among adolescent smokers is 43% (CDC, 2006), with approximately 58% of adolescent smokers attempting at least one quit attempt in the previous year (CDC, 2004). With 4,000 youths initiating cigarette smoking daily, it is critical to identify and disseminate evidence-based youth smoking cessation programs (Horn et al., 2008).

Several reviews report that there is now sufficient evidence to recommend behaviorally-based programs for helping youths quit smoking (Curry, Mermelstein, & Sporer, 2009; Grimshaw & Stanton, 2006; Sussman & Sun, 2009; Sussman, Sun, & Dent, 2006). Sussman et al. (2006) found 48 experimental and quasi-experimental studies published between 1970 and 2003 for a meta-analysis. A 2.9% absolute advantage in quitting over control conditions was found, and an increased likelihood of quitting of 46%. Only two federally recognized, evidence-based behavioral programs—Project EX (Sussman, Dent, & Lichtman, 2001) and the American Lung Association’s (ALA) Not-On-Tobacco (N-O-T) Program (Dino, Horn, Goldcamp, Fernandes, et al., 2001; Horn, Dino, Kalsekar, & Mody, 2005)—were reported. The Cochrane Collaboration’s (Grimshaw & Stanton, 2006) meta-analysis included youth cessation intervention studies used worldwide. Stringent inclusion criteria resulted in 15 studies for review. The Cochrane meta-analysis reported that tobacco cessation treatment significantly increased the likelihood of cessation over control conditions. This review identified N-O-T as the only program of promise based on the number and outcome of evaluation studies.

The efficacy and effectiveness of a teen smoking cessation intervention are necessary but not sufficient conditions determining a program’s suitability for widespread dissemination (Glasgow, Lichtenstein, & Marcus, 2003). Other factors include feasibility, relevance, acceptability, adaptability, and economic cost. The RE-AIM framework (e.g., Glasgow et al., 2003) directs intervention development with a high degree of translatability, and evaluating effectiveness. RE-AIM includes: Intervention Reach, Efficacy/Effectiveness, Adoption, Implementation, and Maintenance (Glasgow, Vogt, & Boles, 1999; RE-AIM, 2010).

The N-O-T program, an evidence-based, theory-driven teen smoking cessation program, was developed and evaluated consistent with the RE-AIM framework. ALA adopted N-O-T and facilitates national program training, marketing, dissemination, and implementation. A decade of research indicates N-O-T is cost-effective, adoptable, and suitable (Dino, Horn, Abdulkadri, Kalsekar, & Branstetter, 2008; Horn et al., 2005; Joffe et al., 2009). N-O-T studies between 1998 and 2003 showed end-of-program intent-to-treat quit rates between 15% and 19% (Horn et al., 2005), among the highest reported in the literature (Sussman et al., 2001). Furthermore, N-O-T is the most widely disseminated (Curry, Sporer, Pugach, Campbell, & Emery, 2007) teen smoking cessation program in the nation.

N-O-T has been evaluated consistent with RE-AIM. Reach was assessed by tracking recruitment strategy effectiveness (Massey et al., 2003) and monitoring participant characteristics across sites and states (Horn et al., 2005). This information informed national and local ALA training. Effectiveness evaluation included assessing multiple outcome measures (i.e. reduction of and/or quitting smoking, stage of change, and nicotine dependence) across states and sites (Dino, Kamal, Horn, Kalsekar, & Fernandes, 2004; Horn et al., 2005; Horn, Fernandes, Dino, Massey, & Kalsekar, 2003). Results related to program effectiveness were used to inform a 2008 program revision and guide national research agendas. Adoption included assessment of facilitators and program adoption barriers. Findings were used in the national training and program revision (Horn & Dino, 2009). We assessed Implementation using cost-effectiveness data (Dino et al., 2008) and an ongoing facilitator process evaluation. Finally, Maintenance was explored by monitoring program adoption over time by state and site and examination of the longer term (6 months or more) effects of N-O-T on smoking cessation and reduction outcomes (Dino, Horn, Goldcamp, Fernandes, et al., 2001; Horn & Dino, 2009; Horn, Dino, Kalsekar, & Fernandes, 2004).

The current study uses the RE-AIM framework to evaluate N-O-T at the state level. First, we assessed the public health impact of N-O-T in West Virginia from 2000 to 2005. Second, we examined the impact of place by assessing the comparative public health impacts of N-O-T among Regional Educational Service Agency regions (RESAs), institutionalized service regions in West Virginia (West Virginia Board of Education, 2010). These RESAs provided an understanding of the geographical contexts that effect the dissemination of the N-O-T program in West Virginia. The use of the RE-AIM framework provides a more comprehensive program assessment than the more common efficacy evaluation.

Method

Participants

This evaluation is a function of nonresearch, that is, real-world N-O-T effectiveness evaluation from 2000 to 2005 (T2R2, 2008). In West Virginia, from 2000 to 2005, approximately 1,500 middle school and high school students participated in N-O-T programs. Data were obtained from the American Lung Association of West Virginia (ALA-WV) as routinely collected from N-O-T facilitators using a standardized N-O-T evaluation protocol (Prevention Research Center of West Virginia University, 2002).

Once entered into a database, participants were excluded if they (a) participated in N-O-T for educational rather than intervention purposes (e.g., health classes), (b) were nonsmokers, and/or (c) were smokeless tobacco users only. Participants were also excluded if they were missing county and school identifiers. These criteria resulted in a final sample of 959 participants. This study was approved as exempt by the Institutional Review Board at West Virginia University.

Instruments

Data collection instruments were consistent with the three standard evaluation instruments found in the N-O-T curriculum. These included the Individual Information Form (e.g., demographic information), About Me 1Form (e.g., baseline information about smoking habits), and About Me 2 Form (e.g., follow-up information about smoking habits). Occasionally, schools using N-O-T make modifications to the tools to better meet their needs. To ensure consistency, we created a set of common variables for similar items, but modified some items, from these instruments for inclusion into the final database. See Table 1 for a complete listing.

Table 1

Not-On-Tobacco (N-O-T) Study Variables

Variables	Response options
Basic demographics
Age	12-19 years
Gender	Male or female
Grade	6th-12th
School (based on grade)	Middle, high, alternative
Race	White, African American, American Indian/Alaskan Native, Asian American/Asian, Hispanic, Native Hawaiian/Pacific Islander, Other, Mixed
Year	2000-2005
Smoking status and history
Number of cigarettes smoked per weekday	Cigarettes per day (min 0 to max 140)
Number of cigarettes smoked per weekend day	Cigarettes per day (min 0 to max 280)
Age at which first tried smoking	2-17 years
Ever tried to quit	Yes or no
If yes, how many times	1-2, 3-4, 5-7, 7 or more
Longest period smokefree	Less than a week, ≤1 month, 1-6 months, >6 months
Last quit attempt	Days ago, weeks ago, months ago, >1 year
How did you try to quit?
On my own	Yes or No
In a group
NRT
Health provider
Other
Ever participated in N-O-T
Smoking status of parents	Yes or No (current use)
Smoking status of siblings
Smoking status of friends
Intervention readiness
Motivation to quit (Scaled 1-5)	None, low, medium, high, very high
Confidence to quit (Scaled 1-5)	None, low, medium, high, very high
Stage of change (Scaled 1-5)	Do not plan to quit in the next 6 months (precontemplation), plan to quit in the next 6 months (contemplation), plan to quit in next 30 days (preparation), made a serious quit attempt in the past 6 months (action), quit less than 6 months ago (maintenance)

Note: NRT = nicotine replacement therapy.

N-O-T program data (e.g., facilitator name, program dates, number of participants, school name, and location of offering) were extracted from the ALA-WV Facilitator Registration Form database (e.g., facilitator name, gender, street address, county residence, and date trained) and the N-O-T Facilitator Process Form (e.g., facilitator name, school, and suggested changes/comments).

Procedures

Table 2 lists specific measurements used in the RE-AIM component measures. Reach was assessed by determining the percentage and characteristics of West Virginia middle/high school smokers enrolled in the N-O-T program during 2000 to 2005. Characteristics included demographic and psychological data from the Individual Information Form and About Me 1 Form and About Me 2 Form. The Reach dimension included participants’ age, race, gender, smoking status and history, quit attempts, and family/friend smoking status, motivation to quit, confidence to quit, and stage of change (DiClemente, Prochaska, Fairhurst, et al., 1991).

Table 2

Evaluation Methods and Measures Using RE-AIM Framework

RE-AIM	Indicators	Measures
Reach: N-O-T program participation	No. of participants % first-time participants % female participants % minority participants	Individual-level data representing program participation using data collected by N-O-T practitioners (e.g., attendance at sessions, sociodemographic characteristics)
Effectiveness: N-O-T program outcomes	Rates of tobacco cessation and reduction	Individual-level data representing program participants’ behavioral outcomes as a result of the N-O-T program (e.g., number of N-O-T program participants who quit smoking)
Adoption: Program offerings	Average number of offerings annually Average no. of participants per N-O-T offering Average no. of facilitators trained annually % female facilitators	Program-level data representing proportion and representativeness of N-O-T program offerings (e.g., characteristics of N-O-T program offerings)
Implementation
Individual level	% annual retention rate % active facilitators who offered the N-O-T program	Individual-level data representing N-O-T program adherence (e.g., following the N-O-T program guidelines)
Program level	No. of offerings per facilitator % counties offering N-O-T program	Program-level data representing the extent at which N-O-T program facilitators adhered to the N-O-T program (e.g., following the N-O-T program guidelines)
Maintenance: Changes during the last 2 years of study	Proportion of participants who have never previously participated in the N-O-T program Quit rate (ITT) Number of N-O-T program offerings Number of facilitators	Individual-level and program-level data representing the extent at which the N-O-T program has sustained over time (e.g., has the number of offering increased over the past 2 years or has the number of ITT increased during the past 2 years, etc.)

Note: N-O-T = Not On Tobacco; ITT = intent to treat.

The Reach component measures evaluating the N-O-T program across RESAs included total number of participants, percentage of first-time participants, percentage of female participants, and percentage of minority participants.

Assessment of Effectiveness was based on the participants’ smoking outcome, defined as favorable change or quit rate at the end of the program (3 months postbaseline). The first two Effectiveness measures, reduction rates and favorable change, were calculated for participants who did not quit. Favorable change measured the percentage of individuals who quit or reduced smoking any amount below the baseline rate of smoking. Reduction rates were calculated only among continuing smokers.

Quit rates were defined by two methods: compliant subsample (CSS) and intent-to-treat (ITT) methods (Dino, Horn, Goldcamp, Fernandes, et al., 2001; Dino, Horn, Goldcamp, Kemp-Rye, et al., 2001; Dino, Horn, Goldcamp, Maniar, et al., 2001; Sussman, 2002; Sussman et al., 2001). ITT analysis is the more conservative measure examining the total number of participants at baseline and assuming participants who were not available at follow-up failed to quit or reduce smoking (Dino et al., 2008; Horn et al., 2005). In contrast, CSS analysis assumes that with appropriate attrition analysis, participants’ available postprogram data are representative of the total sample (Little, 1993). Both analyses have been used in past N-O-T research although with no systematic or consistent differences by age, gender, daily smoking, and other key variables between completers and noncompleters (Dino, Horn, Goldcamp, Maniar, et al., 2001).

Participants were used as the unit of analysis in calculating quit rates for RESAs. Quit was defined by a self-reported no response to the question “Are you still smoking?” in the About Me 2 Form, which assumes at least 24-hr abstinence (Horn & Dino, 2009). Continuous days of abstinence were also reported. The CSS quit rate uses a population denominator composed from participants completing baseline (About Me 1 Form) and postprogram (About Me 2 Form) assessments, whereas the population denominator of ITT was constructed from the total number of participants at baseline.

Measures of Adoption address N-O-T program offerings and ALA-WV-trained N-O-T facilitators. Adoption measures included average number of N-O-T offerings annually; average number of participants per N-O-T program offering; average number of ALA-WV-trained N-O-T program facilitators annually; and percentage of female ALA-WV-trained N-O-T program facilitators. The N-O-T program Offering and Facilitator Registration Form databases provided these data.

Implementation included individual- and program-level measures. Annual retention rate (individual level) refers to the percentage of individual adherence to the program requirements (e.g., completion and follow-through of the program). Participant adherence was measured by participants’ completion of baseline (About Me 1 Form) and postprogram (About Me 2 Form) evaluations. The final individual-level measure was percentage of ALA-WV-trained N-O-T program facilitators who actually offered the program. There were two program-level measures: ratio of N-O-T program offerings per active ALA-WV-trained N-O-T program facilitators and the percentage of counties housing the N-O-T program.

We assessed Maintenance using four measurements derived from the difference between the last 2 years of the study period. The first measurement analyzed changes in the percentage of all participants who had never previously participated in the N-O-T program. The second measure analyzed changes in quit rates (ITT). The third measure examined changes in the number of N-O-T offerings. The final measure referred to changes in the number of ALA-WV-trained N-O-T program facilitators.

Data Analysis

This study used N-O-T program data collected from 2000 to 2005. Data were analyzed using SPSS, version 16. A random 40% error check was conducted to ensure data entry accuracy. Analytic techniques were used to account for missing data, and the study set entry criteria for participants with missing data. Some participant forms had missing or incomplete information, especially data related to grade; missing data is not uncommon with evaluation data collected under typical field conditions. Specifically, correlation analyses were used to investigate missing values for grade. Pearson’s correlation between age and grade was found to be .851 (p < .01). A strong correlation between these two variables supports the use of age to calculate missing grade and of grade to compute missing age.

Participant data were then assigned to one of eight RESAs based on county and school. We assessed regional performance by conducting descriptive analyses for demographics, smoking history, and psychosocial factors and by evaluating RE-AIM component measures across RESAs.

To examine variability in the regional RE-AIM measures, we calculated the average, lowest, and highest regional values for baseline characteristics and RE-AIM component measures. Pearson’s chi-square and analysis of variance (ANOVA) tests were used to compare RESAs. We used Bonferroni correction to adjust for multiple comparisons.

Each RE-AIM composite score was constructed from four measures. As suggested by Glasgow et al. (1999), measures were converted to normalized z scores for easier comparison. Then, based on the z score for each measurement, the study obtained an average z score for each RE-AIM component. RESA performance was then assessed using a composite z score, constructed by summing across each region’s RE-AIM component z scores. The highest and lowest RESA composite z scores were then plotted for comparative analysis.

Results

Baseline Characteristics

The average age across the RESAs was 15.8 (SD = 1.4). Approximately 53% of the sample was female and more than 90% were white. Typically, participants smoked 14.4 (SD = 13.2) cigarettes per weekday and 20.3 (SD = 20.1) cigarettes per weekend day. The mean age participants first tried smoking was 10.7 (SD = 2.8), and approximately 79% of participants tried to quit. Participants observed smoking more often among friends (98.4%) than parents (78.5%) and siblings (60%). Participants’ intervention readiness measures of motivation to quit was 3.1 (SD = 1.1), confidence to quit was 3.0 (SD = 1.1), and stage of change was 2.7 (SD = 1.3) across RESAs. There were significant differences (p < .05) among RESAs including age, number of cigarettes smoked per weekday and weekend day, age first tried smoking, parents smoking, motivation to quit, and stage of change. Post hoc analysis of baseline characteristics after Bonferroni adjustment found significant difference(s) between pair(s) of RESAs for all variables except for number of cigarettes smoked per weekend day.

RE-AIM Component Measures and Performance Values

RE-AIM component measures and RESA values (average, lowest, and highest), p values, and post hoc tests results are provided in Table 3. No significant differences were found between RESAs for the Reach, Effectiveness, and Adoption measures. Implementation measures of retention rate and percentage of active facilitators were significantly different among RESAs. Conversely, the ratio of N-O-T offerings by active facilitators and percentage of counties in the RESA that offered N-O-T were not. Maintenance measures were not tested because of missing data. Despite the majority of the RE-AIM measures not reaching statistical significance, the variability of these measures provided great insight into regional N-O-T dissemination characteristics.

Table 3

RE-AIM Measures and RESA Region Values 2000-2005

	RESA region values
RE-AIM Components/Measures	Average	Lowest	Highest	p value^a
Reach
Total number of participants	119	51	297	na
First-time participants, %	82.0	72.5	93.0	.393
Female participants, %	53.6	42.6	59.3	.125
Minority participants, %	9.8	7.1	14.3	.962
Effectiveness
Quit rate
CSS, %	24.2	17.5	38.2	.336
ITT, %	18.0	12.8	21.3	.713
Favorable change—reduced in smoking cigarette, %	76.1	60.0	89.5	.176
Reduction rate among continuing smokers, %	67.7	48.4	85.2	.170
Adoption
The average number of N-O-T offerings annually	4	2	8	.637
The average number of participant per N-O-T offerings	5	4	6	.785
The average number of ALA-WV N-O-T trained facilitators annually	14	7	20	na
Female ALA-WV N-O-T trained facilitators, %	72.7	59.3	85.1	.052
Implementation
Retention rate, %	74.7	55.7	78.4	.034*^b
Active facilitators (who offered N-O-T programs), %	14.3	6.5	32.4	.000**^b
The ratio of N-O-T offerings by active facilitators (who offered N-O-T program)	1.7	1.1	2.2	.331
Counties in the RESA that offered N-O-T, %	63.6	25	100	.175
Maintenance
Changes in proportion of participants who have never previously participated in the N-O-T program in the last 2 years of study, %	−2.7	−18.8	13.1	na
Changes in quit rate (ITT) in the last 2 years of study	7.4	−7.3	21.0	na
Changes in numbers of N-O-T offerings in the last 2 years of study	1	−10	7	na
Changes in numbers of trained facilitator in the last 2 years of study	5	0	16	na

Note: a. Pearson c² for percentage distribution and ANOVA test for mean value. CSS = compliant subsample; ITT = intention to treat; ALA-WV = American Lung Association of West Virginia; RESA = Regional Educational Service Agency region; na = not available.

The post hoc test showed only 1 of 28 RESA pairs as significantly different.

Significant at α = .05. **Significant at α = .001

Highest and Lowest Plotted Analysis (Figure 1)

There was little difference between the highest and lowest RESA z scores for program Reach. Effectiveness for the highest RESA z score was approximately 1.5 standard deviations from the lowest RESA z score. Adoption was significantly different (SD >2.0) between the highest and lowest RESA z scores. In terms of Implementation, the lowest and highest RESA z scores were separated by 1 standard deviation. In Maintenance, the lowest RESA z score was slightly higher than the highest RESA z score.

Figure 1

RE-AIM Component Performance for Lowest RESA z Score and Highest RESA z Score

Discussion

We applied the RE-AIM framework to a research-tested, nationally available teen smoking cessation program (N-O-T) as implemented across West Virginia between 2000 and 2005. RE-AIM is a valuable tool for guiding program evaluation and dissemination (Glasgow et al., 2003; Klesges, Estabrooks, Dzewaltowski, Bull, & Glasgow, 2005). Each RE-AIM component illustrated unique characteristics for the various regions of West Virginia, each affecting program adoption and implementation. To our knowledge, this is the first published examination of RE-AIM in this context.

Reach findings revealed that N-O-T reached the intended audience of daily adolescent smokers within the 14- to 19-year age range across West Virginia. Important differences were found in teens’ social and familial contexts (i.e., parental smoking was found to be significantly different between RESAs).

Regional enrollments fluctuated over time yet statewide enrollment increased each year. Calculating absolute Reach as recommended by the RE-AIM reach calculator (total teen smokers enrolled per RESA/total possible teen smokers per RESA) is difficult because the total number of possible smokers per RESA is extremely limited (e.g., we can only estimate total possible teen smokers). Available youth smoking data for 2000, 2002, and 2005 reveal an average West Virginia youth smoking prevalence rate of 33.3% (West Virginia Bureau of Public Health, 2007). Therefore, we estimate that N-O-T reaches <1% of West Virginia’s teen smokers. This is clearly unacceptable program Reach. Identifying unique characteristics of N-O-T participants in each RESA may serve as a guide to future recruitment or program content adaptations.

Overall, about 20% of West Virginia N-O-T participants reported quitting smoking at 3 months postbaseline; this is consistent with national program effectiveness data (Horn et al., 2005). Effectiveness data revealed variations in quit and reduction rates by year and region. Nonetheless, N-O-T quit rates showed a positive upward trend for five of eight regions. This finding supports future examination of regional factors that may influence program Effectiveness. Importantly, this finding begs the question “What conditions lead to higher program quit rates in West Virginia?” Identification of factors in regions with higher quit rates may guide program improvement across the state.

Adoption indicators support more N-O-T facilitator training and continued program use. From 2000 to 2005, a total of 664 N-O-T facilitators were trained in West Virginia, with an average of 14 facilitators trained annually per RESA. Future research is needed to determine the optimal number of new facilitators per year. The number of N-O-T offerings increased over the years, suggesting sustained interest in N-O-T. However, the Reach data strongly supports increased dissemination efforts statewide.

The RE-AIM assessment framework helps to identify areas in the state most needing improvement. Our data minimally informed Implementation issues, particularly related to facilitator adherence to the N-O-T curriculum. Of note, sometimes facilitators modified evaluation forms that constrained use of a common set of variables. Major discrepancies were found between the number of total trained facilitators and those actively offering the N-O-T program. Although 664 total facilitators were trained from 2000 to 2005, only 96 facilitators actively offered the N-O-T program during this period, with facilitator distribution across RESAs being variable. Nonetheless, inconsistencies between number of facilitators trained and numbers using the program, additional efforts are needed to increase facilitator Implementation. A significant difference was found in retention rates for only 1 of the 28 comparative pairs (e.g., lowest-value RESA vs. highest-value RESA). Therefore, regional implementation issues should be addressed in depth to more closely examine program fidelity across regions.

Finally, program Maintenance examined progress from 2004 thorough 2005. Selected measures represent how well each RESA region carried out the N-O-T program. Although more participants quit smoking, less youth smokers participated in the N-O-T program for the first time. In addition, RESAs trained more N-O-T facilitators but not all new facilitators offered N-O-T during the last 2 years of the study.

Plotting the lowest and highest RESA z score values for each component composite score can depict strengths and weaknesses of each region (Farris, Will, Khavjou, & Finkelstein, 2007). The highest RESA had higher composite scores for Reach, Effectiveness, Adoption, and Implementation of N-O-T dissemination; however, Maintenance of the N-O-T program during the last 2 years of the study was higher in the lowest RESA.

A few limitations warrant discussion in this study. First, response bias is of concern. N-O-T participation is voluntary; therefore, the sample may reflect teen smokers who are moderately motivated to quit. Second, self-reported data may overestimate quit rates; however, previous studies have shown that self-report and biochemically valid quit rates are independently consistent, with 81% to 96% sensitivity and 77% to 95% specificity (Kentala, Utrainen, Pahkala, & Mattila, 2004).

To enhance data quality, N-O-T uses standardized evaluation procedures, trained facilitators. Furthermore, participant surveys are manually reviewed to eliminate errors, inaccuracies, and unreliable data (Horn et al., 2005).

Conclusion

In keeping with Glasgow et al. (RE-AIM, 2010), the RE-AIM framework informed the “robustness, translatability, and public health impact” of the 2000-2005 West Virginia N-O-T program. N-O-T demonstrated respectable effectiveness and low to moderate levels of maintenance. Despite reaching the intended audience, the absolute reach was alarmingly low. This may be partially due to lack of Implementation among trained facilitators. Without increased program offerings, Reach is limited. Overall program dissemination was fruitful. However, limitations effecting regional dissemination were identified and examined providing information that would have been otherwise overlooked.

Footnotes

The Prevention Research Center at West Virginia University is a member of the Prevention Research Centers Program, supported by the Centers for Disease Control and Prevention cooperative agreement number 1-U48-DP-001921. This study was supported through grant funding from the West Virginia Bureau for Public Health/Division of Tobacco Prevention.

References

Board of Education, W. V. (2009). West Virginia RESAs. Retrieved from http://resa2.k12.wv.us/wvresa.asp

Centers for Disease Control and Prevention. (2004). 2004 National Youth Tobacco Survey. Retrieved from http://www.cdc.gove/tobacco/NYTS/nyts2004.htm

Centers for Disease Control and Prevention. (2006). Tobacco use among adults—United States, 2005. Morbidity and Mortality Weekly Report, 55, 1145-1148.

Curry

S. J.

Mermelstein

R. J.

Sporer

A. K.

(2009). Therapy for specific problems: Youth tobacco cessation. Annual Review of Psychology, 60, 229-255.

Curry

S. J.

Sporer

A. K.

Pugach

Campbell

R. T.

Emery

(2007). Use of tobacco cessation treatments among young adult smokers: 2005 National Health Interview Survey. American Journal of Public Health, 97, 1464-1469.

DiClemente

C. C.

Prochaska

J. O.

Fairhurst

S. K.

Velicer

W. F.

Velasquez

M. M.

Rossi

J. S.

(1991). The Process of Smoking Cessation: An Analysis of Precontemplation, Contemplation, and Preparation Stages of Change. Journal of Consulting and Clinical Psychology, 59, 295-304.

Dino

Horn

Abdulkadri

Kalsekar

Branstetter

(2008). Cost-effectiveness analysis of the Not On Tobacco Program for adolescent smoking cessation. Prevention Science, 9, 38-46.

Dino

Horn

Goldcamp

Fernandes

Kalsekar

Massey

(2001). A 2-year efficacy study of Not On Tobacco in Florida: An overview of program successes in changing teen smoking behavior. Preventive Medicine, 33, 600-605.

Dino

Horn

Goldcamp

Kemp-Rye

Westrate

Monaco

(2001). Teen smoking cessation: Making it work through school and community partnerships. Journal of Public Health Management and Practice, 7, 71-80.

10.

Dino

Horn

Goldcamp

Maniar

Fernandes

Massey

(2001). Statewide demonstration of Not On Tobacco: A gender-sensitive teen smoking cessation program. Journal of School Nursing, 17, 90-97.

11.

Dino

Kamal

Horn

Kalsekar

Fernandes

(2004). Stage of change and smoking cessation outcomes among adolescents. Addictive Behaviors, 29, 935-940.

12.

Farris

R. P.

Will

J. C.

Khavjou

Finkelstein

E. A.

(2007). Beyond effectiveness: Evaluating the public health impact of the WISEWOMAN Program. American Journal of Public Health, 97, 641-647.

13.

Glasgow

R. E.

Lichtenstein

Marcus

A. C.

(2003). Why don’t we see more translation of health promotion research to practice? Rethinking the efficacy-to-effectiveness transition. American Journal of Public Health, 93, 1261-1267.

14.

Glasgow

R. E.

Vogt

T. M.

Boles

S. M.

(1999). Evaluating the public health impact of health promotion interventions: The RE-AIM framework. American Journal of Public Health, 89, 1322-1327.

15.

Grimshaw

Stanton

(2006). Tobacco cessation interventions for young people. Cochrane Database of Systematic Reviews (4). doi:10.1002/14651858.CD003289.pub410.1002/14651858.CD003289.pub4

16.

Horn

Dino

(2009). A comparative review of the effectiveness trial of N-O-T in Alabama: Guideposts for future research. American Journal of Health Education, 40, 47-55.

17.

Horn

Dino

Branstetter

Zhang

Kelley

Noerachmanto

Tworek

(2008). A profile of teen smokers who volunteered to participate in school-based smoking intervention. Tobacco Induced Diseases, 4, 6.

18.

Horn

Dino

Kalsekar

Fernandes

(2004). Appalachian teen smokers: Not On Tobacco 15 months later. American Journal of Public Health, 94, 181-184.

19.

Horn

Dino

Kalsekar

Mody

(2005). The impact of Not On Tobacco on teen smoking cessation: End-of-program evaluation results, 1998 to 2003. Journal of Adolescent Research, 20, 640-661.

20.

Horn

Fernandes

Dino

Massey

C. J.

Kalsekar

(2003). Adolescent nicotine dependence and smoking cessation outcomes. Addictive Behaviors, 28, 769-776.

21.

Joffe

McNeely

Colantuoni

Wang

Scharfstein

(2009). Evaluation of school-based smoking-cessation interventions for self-described adolescent smokers. Pediatrics, 124, e187-e194.

22.

Kentala

Utrainen

Pahkala

Mattila

(2004). Verification of adolescent self-reported smoking. Addictive Behaviors, 29, 405-411.

23.

Klesges

Estabrooks

Dzewaltowski

Bull

Glasgow

R. E.

(2005). Beginning with the application in mind: Designing and planning health behavior change interventions to enhance dissemination. Annals of Behavioral Medicine, 29(2), 66-75.

24.

Lamkin

Davis

Kamen

(1998). Rationale for tobacco cessation interventions for youth. Preventive Medicine, 27(5), A3-A8.

25.

Little

R. J. A.

(1993). Pattern-mixture models for multivariate incomplete data. Journal of the American Statistical Association, 88(421), 125-134.

26.

Massey

C. J.

Dino

G. A.

Horn

K. A.

Lacey-McCracken

Goldcamp

Kalsekar

(2003). Recruitment barriers and successes of the American Lung Association’s Not-On-Tobacco program. Journal of School Health, 73(2), 58-63.

27.

Prevention Research Center of West Virginia University. (2002). Not On Tobacco. New York, NY: American Lung Association.

28.

RE-AIM (2010). Applying the RE-AIM Framework to Health Behavior Interventions. Retrieved from http://www.re-aim.org/2003/researchers/framework_res.html

29.

Sussman

(2002). Effects of sixty six adolescent tobacco use cessation trials and seventeen prospective studies of self-initiated quitting. Tobacco Induced Diseases, 1, 35-81.

30.

Sussman

Dent

C. W.

Lichtman

K. L.

(2001). Project EX: Outcomes of a teen smoking cessation program. Addictive Behaviors, 26, 425-438.

31.

Sussman

Sun

(2009). Youth tobacco use cessation: 2008 update. Tobacco Induced Diseases, 5, 3.

32.

Sussman

Sun

Dent

C. W.

(2006). A meta-analysis of teen cigarette smoking cessation. Health Psychology, 25, 549-557.

33.

Translational Tobacco Reduction Research Program (T2R2) (2008). An evaluation report of the Not On Tobacco Program in West Virginia, 2000-2005. Morgantown: West Virginia University.

34.

West Virginia Board of Education. (2010). Regional Education Service Agency (RESA). Retrieved from http://wvde.state.wv.us/ed_directory/#resa

35.

West Virginia Bureau of Public Health. (2007). 2005 West Virginia Youth Tobacco Survey Report. Charleston: West Virginia Department of Health and Human Resources.

36.

Zhu

S. H.

Sun

Billings

S. C.

Choi

W. S.

Malarcher

(1999). Predictors of smoking cessation in U.S. adolescents. American Journal of Preventive Medicine, 16, 202-207.