In studies aimed at encouraging healthier lifestyles, participants may be given free access to both a set of website tools as well as an online community of similarly minded people. Automated tracking of passive (e.g., reading the posts, viewing profiles) and active (e.g., writing posts, sending messages) community engagement can then be used to classify study participants as non-users, passive users, or active community users. Differences in observed outcome rates by community exposure status crossed with levels of website use lack a causal interpretation, as they fail to account for between-group imbalances on demographic and psychosocial characteristics. Inverse probability of treatment weighting (IPTW) can partly correct for such imbalances. George Papandonatos and his students have sought to incorporate such approaches in the analysis of data from online smoking interventions. They are also actively researching ways in which they can transport findings among study participants willing to enroll in online interventions to the full population subjects were drawn from, when random sampling assumptions cannot be assumed to hold.
Behavioral Medicine
-
George Papandonatos
Professor of Biostatistics (R)
Researching Effective Strategies to Prevent Opiod Death (RESPOND)
Researching Effective Strategies to Prevent Opioid Death (RESPOND) engages a team of investigators building a compartmental simulation model of opioid use disorder and its treatment. The model will simulate the population of people with opioid use disorder in Massachusetts and it will be calibrated to state-specific estimates of disease prevalence, admissions to treatment facilities, and overdose deaths.
-
Stavroula Chrysanthopoulou
Assistant Professor of Biostatistics, Director of the Master's Graduate Program in Biostatistics
Optimizing Text Messaging to Improve Adherence to Web-Based Cessation Treatment
This study is comprised of 2 phases to study the impact of a text message system on overall adherence to a web-based smoking cessation program. Phase I uses a full factorial design to identify the most optimal text message intervention to maximize adherence to the BecomeAnEX.org smoking cessation program. Phase II is a randomized controlled trial that compares regular BecomeAnEX users to those who will receive the optimized text message intervention (from Phase I) in conjunction with BecomeAnEX for impact on long-term abstinence.
-
George Papandonatos
Professor of Biostatistics (R)
Implementing a Virtual Tobacco Treatment in Community Oncology Practices
Implementing a Virtual Tobacco Treatment in Community Oncology Practices. In Phase II trial studies how well smoking cessation treatment plans work in tobacco-dependent cancer patients when delivered virtually as part of their cancer care in community oncology practices. Virtual information and counseling sessions may help cancer patients quit smoking.
-
Ilana F. Gareen
Associate Professor of Epidemiology (Research)
Behavioral Activation and Varenicline for Smoking Cessation in Depressed Smokers
Persons who struggle with depression smoke at high rates and experience low quit rates in treatment. The best way to improve cessation treatment for this underserved population remains unknown. The proposed trial tests whether the combination of varenicline and behavioral mood management treatment enhances long-term abstinence for depressed smokers and, if so, whether this treatment achieves its effects through addressing the unique psychological factors that appear to maintain tobacco dependence for these smokers.
-
George Papandonatos
Professor of Biostatistics (R)