Gingiss 2006.

Study characteristics
Methods	Trial name: Texas Tobacco Prevention Initiative Study design: non‐randomised control (comparison group) Intervention duration: 2 years Length of follow‐up from baseline: 2 years (baseline in 2000 and 2 years later) Differences in baseline characteristics: no significant differences between schools Unit of allocation: schools Unit of analysis: schools
Participants	School type: middle and high schools Region: East Texas, USA Demographic/socioeconomic characteristics: not reported Inclusion/exclusion criteria: not reported Number of schools allocated: 134 Numbers by trial group n (controls baseline) = not reported n (controls follow‐up) = not reported n (interventions baseline) = not reported n (interventions follow‐up) = not reported Recruitment School:the baseline sample of schools was randomly selected by Texas Department of State Health Services (TDSHS) in school year (SY) SY2000 from schools serving students in grades 6–12 in the East Texas study area. A representative sample of schools was selected. Participating schools were located in 69 districts in a 7‐county study area. The original SY2000 sample of 171 schools was drawn with probability proportional to study‐area size and school condition (intervention or comparison). Among these schools, 134 participated in the baseline study. This sample was retained and used in this SY2002 follow‐up. Student:not reported Recruitment rate: 134/171 (78.4%)
Interventions	Number of experimental conditions: 2 (1 intervention, 1 control) Policies, practices or programmes targeted by the intervention ‐ Conduct tobacco‐use prevention education with the designated curriculums of Project Towards No Tobacco (TNT) at the middle school level and Not On Tobacco (NOT) at the high school level. ‐ Conduct ≥ 1 tobacco prevention event. ‐ Provide education and training for parents and staff regarding local policies and ordinances as well as state tobacco laws. ‐ Establish STARS, PALS, or Teens Against Tobacco Use (TATU) groups or other peer mentor programmes at the high school level. Implementation strategies EPOC: local consensus processes Contractual agreements were issued to intervention schools to: ‐ Participate in training organised by their Education Service Centres (ESC) on tobacco issues and curricula. ‐ Conduct tobacco‐use prevention education with the designated curriculums of Project TNT at the middle school level and NOT at the high school level. EPOC: educational outreach visits ‐ The Texas Tobacco Prevention Initiative placed a Tobacco Specialist at each of 4 regional ESC serving the East Texas study area. Their responsibilities included co‐ordination, distribution and management of funding, training and TA for intervention schools in respective service areas. EPOC: educational meetings ‐ Training to encourage planning and implementing the above activities based on the guidelines for school health programmes to prevent tobacco use and addiction. EPOC: external funding ‐ Each school received an allocation of approximately USD2000 per year to be used for materials, supplies and small equipment, in‐service release time, training and travel to tobacco‐related meetings. Theoretical underpinning: not reported Description of control: not reported, but assume usual practice
Outcomes	Outcome relating to the implementation of school policies, practices or programmes ‐ Enforcement of school policy on tobacco use ‐ Instruction on tobacco prevention education ‐ Assessment of prevention programmes ‐ Student cessation support ‐ Teacher training for tobacco prevention education ‐ Establish or change school policy on tobacco use ‐ Faculty and staff cessation support ‐ Family involvement in student tobacco programmes ‐ Parental involvement in policy ‐ > 10 lessons Data collection method: self‐report surveys were mailed to the schools for both principals and health co‐ordinators. Validity of measures used: not reported/self‐report methods Data collection method: self‐report surveys were mailed to the schools for both principals and health co‐ordinators. Validity of measures used: not reported Outcome relating to cost: not reported Outcome relating to adverse consequences: not reported Outcome relating to child diet, PA or weight status: not reported
Notes	Note: the Texas Tobacco Initiative was also a non‐randomised trial that did not report baseline data for the implementation outcomes, however, was included as the authors state that there were no differences between groups at baseline on these measures. Research funding: supported by funding from Texas Department of State Health Services. Conflicts of interest: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Non‐randomised control (comparison group). It appeared that schools were not randomly assigned to the intervention or the comparison group and, therefore, high risk of selection bias.
Allocation concealment (selection bias)	High risk	Non‐randomised trial and there is no indication that allocation was concealed and, therefore, were at high risk of selection bias.
Blinding of participants and personnel (performance bias) Implementation outcome	High risk	Outcome group: there was no mention that participants or personnel were blinded to experimental group and, therefore, were at high risk of performance bias.
Blinding of outcome assessment (detection bias) Implementation outcome	High risk	Outcome group: there was no blinding of principals or health co‐ordinators described and the outcomes were likely to be influenced by the use of self‐reported questionnaires.
Incomplete outcome data (attrition bias) Implementation outcome	High risk	Outcome group: high attrition as 25 (19%) schools were lost for the principal survey and 50 (37%) schools for the health co‐ordinator survey.
Selective reporting (reporting bias)	Unclear risk	There was no study protocol, therefore, it was unclear if there was selective outcome reporting.
Other bias	Unclear risk	Study description unclear. The initiation of mandate to implement and evaluate tobacco prevention in East Texas may have contaminated comparison schools.
Potential confounding	High risk	There was no mention of measurement of potential confounders, or any attempt to adjust for confounders.
Overall risk of bias assessment	High risk	Most domains were at high risk of bias.