Thank you for the opportunity to express our views on the draft research plan for tobacco and nicotine use prevention in children and adolescents through primary care interventions. The National Center for Health Research is a nonprofit think tank that conducts, analyzes, and scrutinizes research, policies, and programs on a range of issues related to health and safety. We do not accept funding from companies that make products that are the subject of our work, so we have no conflicts of interest.
Tobacco use remains the number one cause of preventable deaths in the United States, and almost 90% of those who die from smoking-related illness began using tobacco products at or prior to age 18.1 In the United States, smoking-related illness costs approximately $300 billion and kills more than 480,000 Americans every year.2 Preventing and stopping nicotine/tobacco use in children and adolescents would benefit our society greatly by decreasing the disease burden of tobacco use.
We support the efforts of the U.S. Preventive Services Task Force (USPSTF) to carefully draft a research plan to guide the systematic review of available evidence for primary care interventions to reduce tobacco and nicotine use in children and adolescents.3 USPSTF last reviewed the literature in 2013 and recommended a “B” grade for primary care interventions.4 Changing medical practice and the evolving use of smoking products by children and adolescents (e.g. e-cigarettes, smokeless tobacco, and hookah) could alter the benefit/risk ratio of these interventions.
However, we are concerned that the proposed key questions (KQ) seem to focus on evaluating the effects of primary care interventions in general, rather than determining which, if any, interventions are effective for which children.
We have four methodological suggestions that would greatly strengthen the research plan:
1) The research plan should focus on determining the effectiveness of specific types of interventions in primary care settings. A 2017 analysis found that many national smoking cessation guidelines differed in fundamental ways, including the content of interventions and how they were delivered.5 If the research plan evaluated the specific types of interventions, and if possible, the most effective components of these interventions, that information would be more useful than a more general overview of intervention effectiveness. In addition, researchers should analyze the data to evaluate effectiveness for particular age groups, demographic groups, or settings.
One way to assess the impact of specific components is to conduct mediator and moderator analyses. These analyses would help healthcare practitioners, researchers, and policymakers draw more specific conclusions about what kinds of specific interventions work for specific kids under specific circumstances. For example, the type of primary care intervention (i.e. family counseling sessions, anti-tobacco messages, tele-counseling sessions, pharmacotherapy) may greatly influence outcomes. In addition, the effectiveness of interventions may depend on the characteristics of the children/adolescents being targeted, such as age, gender, geographical location, SES, and age at first use of tobacco/nicotine products. Moderator analyses will help to determine the key components of primary care interventions that lead to success, which in turn will help providers implement effective techniques.
2) For key questions KQ1 and KQ2, we suggest broadening inclusion criteria to include brief/short-term therapeutic interventions with less than 6 months of follow-up. Given a strong evidence base as well as limited consultation time, brief targeted interventions are standard practice in the primary care setting. These interventions are solution-focused and active, with clearly and collaboratively defined goals related to specific behavior change.6 Even if effects are short-lasting, this systematic review would contribute to our understanding of positive outcomes we can work to maintain. For example, this might mean adding on a management component to a brief intervention that was highly effective at 2 months of follow-up.
3) The research plan should appropriately manage dependent effect sizes. Dependent effect sizes arise when multiple effect sizes are estimated from the same sample. Not accounting for dependent effect sizes overestimates confidence in the findings and may bias the interpretations made by policymakers and practitioners. Fortunately, meta-analytic methods have made it possible to handle dependent effect sizes easily and efficiently.7 The research plan should include language about how the meta-analysts plan to deal with this statistical phenomenon.
4) Unpublished research should be included if it is well-designed and has been reviewed by peers. It is certainly likely that there are evaluations of interventions within the primary care setting that have not yet been published and/or may never be published, especially if the interventions were not effective. Research indicates that findings from unpublished studies tend to have smaller effects than published findings. Excluding unpublished findings has the potential to bias the results.
In conclusion, we support many of the methodological proposals included in USPSTF’s draft research plan to study the impact of interventions to prevent and reduce tobacco and nicotine use in children and adolescents. However, the draft approach would not evaluate the effectiveness of specific types of interventions on subgroups of children and adolescents. We strongly recommend making the above changes to strengthen the utility of the research findings that USPSTF will rely on for its recommendations.
Thank you for the opportunity to share our perspective.
For questions or more information, please contact Megan Polanin, PhD at firstname.lastname@example.org or at (202) 223-4000.
1) U.S. Department of Health and Human Services, Office of Adolescent Health. (2016, July 26). Tobacco Use in Adolescence: Preventing and Reducing Teen Tobacco Use. Retrieved from https://www.hhs.gov/ash/oah/adolescent-development/substance-use/drugs/tobacco/index.html#_ftn4
2) Centers for Disease Control and Prevention. (20 June, 2017). Burden of Tobacco Use in the U.S. Retrieved from https://www.cdc.gov/tobacco/campaign/tips/resources/data/cigarette-smoking-in-united-states.html
3) U.S. Preventive Services Task Force (2017, June). Draft Research Plan for Tobacco and Nicotine Use Prevention in Children and Adolescents: Primary Care Interventions. Retrieved from https://www.uspreventiveservicestaskforce.org/Page/Document/draft-research-plan/tobacco-and-nicotine-use-prevention-in-children-and-adolescents-primary-care-interventions
4) U.S. Preventive Services Task Force (2013, August). Final Recommendation Statement. Tobacco Use in Children and Adolescents: Primary Care Interventions. Retrieved from https://www.uspreventiveservicestaskforce.org/Page/Document/UpdateSummaryFinal/tobacco-use-in-children-and-adolescents-primary-care-interventions
5) Verbiest, M., Brakema, E., van der Kleij, R., Sheals, K., Allistone, G., Williams, S., … & Chavannes, N. (2017). National guidelines for smoking cessation in primary care: a literature review and evidence analysis. NPJ Primary Care Respiratory Medicine, 2. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5434788/
6) U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration. Behavioral Health in Primary Care: Integrating Behavioral Health into Primary Care. Retrieved from https://www.integration.samhsa.gov/integrated-care-models/behavioral-health-in-primary-care
7) Hedges, L. V., Tipton, E., & Johnson, M. C. (2010). Robust variance estimation in meta‐regression with dependent effect size estimates. Research Synthesis Methods, 1(1), 39-65. Retrieved from http://onlinelibrary.wiley.com/doi/10.1002/jrsm.5/full