Evaluations

• Initial evaluation of a real-world Internet smoking cessation system.

  Cobb NK, Graham AL, Bock BC, Papandonatos G, Abrams DB. 2005. Nicotine and Tobacco Research. 7(2): 207-216. doi: 10.1080/14622200500055319

  Summary: Newly registered users on QuitNet were invited to complete a survey 90 days after registration. Of the 1,501 users sent surveys, 385 completed the survey. Surveys assessed seven-day point prevalence abstinence, actual quit date, number of quit attempts of at least 24 hours in the past three months, past 3-month use of cessation treatment, and smoking rate. Participants who had not quit were asked stage of change, desire to, and confidence in, quitting, and time to first cigarette after waking. Additional data was extracted from registration and site usage information on QuitNet. At registration, 30% of users indicated that they had already quit smoking. Among participants who were smoking at registration, rate of seven-day point prevalence abstinence was 7%, but among all participants (including those who had already quit at registration), rate of seven-day point prevalence abstinence was 11.5%. Of participants reporting seven-day point prevalence abstinence, 90% reported having been abstinent for at least 30 days. Participants who had not quit at follow-up smoked significantly fewer cigarettes relative to baseline. Participants who quit had logged into the website more than participants who were still smoking at follow-up.

  Take Away: QuitNet may promote abstinence from smoking, particularly among users who had already quit when they joined QuitNet.

• Utilization of smoking cessation international, interactive, and online community resources as predictors of abstinence: Cohort study.

  An LC, Schillo BA, Saul JE, et al. 2008. Journal of Medical Internet Research. 10(5): e55. doi: 10.2196/jmir.1018

  Summary: Researchers evaluated relationships between use of different features of a version of QuitNet personalized for users in Minnesota and cessation outcomes. For this version of QuitNet, content was unchanged, but branding was specific to Minnesota. Researchers recruited 607 new users who reported currently using tobacco at registration. Participants completed an online survey of reported abstinence in the past 30 days and use of cessation medication six months after registration. If participants did not respond to the online survey, researchers called participants to interview them. Researchers extracted information about readiness to quit, cigarettes smoked per day, and time to first cigarette from registration information and participant usage data. Past 30-day abstinence was related to use of informational resources, quit planning tools, counselor resources, and actively interacting with the user community. Use of the quit planning tools and privately messaging other users predicted abstinence in the past 30 days. Usage data indicated that informational resources, particularly the quit planning tools, were more popular than interactive features. Participants who interacted with the community were more likely to do so passively (e.g. reading posts) than actively (e.g. writing their own post).

  Take Away: Quit planning tools are a popular feature of a Minnesota-branded version of QuitNet and predict abstinence, but another predictor of abstinence, private messaging is used less frequently than informational features and passive interaction with the community.

• Social network structure of a large online community for smoking cessation.

  Cobb NK, Graham AL, Abrams DB. 2010. American Journal of Public Health. 100(7): 1282-1289. doi: 10.2105/AJPH.2009. 165449

  Summary: Researchers extracted registration and site utilization data about users who had used one of the QuitNet social features (private messaging, forums, buddy lists) during a six-month study period. From the large initial lists of users, researchers created sets of users who could easily be connected with each other through their connections with other users, were included on many buddy lists and had many communications with others (strongly connected), were mutually buddies with many users and had at least five communications with at least one buddy (densely connected), were new registrants, were connected to new users, and were highly connected to the entire community. The initial sample consisted of 7569 participants, and 45.4% of these participants were abstinent during the study period. Users who were strongly or densely connected were more likely to be abstinent during the study period. Smoking was less likely among users who were connected to more users, but the number of smokers in a user’s network was positively related to smoking.

  Take Away: Abstinent users may be important members of social networks on QuitNet and connections with other users may promote abstinence.

• Sentiment analysis to determine the impact of online messages on smokers’ choices to use varenicline.

  Cobb NK, Mays D, Graham AL. (2013). Journal of the National Cancer Institute Monographs. 47: 224-230. doi: 10.1093/jncimonographs/lgt020

  Summary: Researchers compared use of varenicline by new QuitNet users and sentiment of posts mentioning varenicline posted in threads that these users participated in. Posts mentioning varenicline (n=127,959) written during a period spanning before and after varenicline’s approval as a cessation aid were extracted for analyses. Researchers used sentiment analysis software to code post sentiment (i.e. positive or negative). New users (n=2,132) had to have registered after the approval of varenicline. Researchers extracted information about use of pharmaceutical cessation aids from registration data and updates about their use of medication made during thirty days after registration. About one quarter of posts (25.9%) mentioning varenicline were positive and 31.5% were negative. At thirty days post-registration, 68.6% of users reported not using varenicline and 31% reported using varenicline at registration, but terminated their use during the thirty-day observation period. 17.4% of users initiated varenicline and continued use at 30 days was reported by 13.9% of users. Users who initiated or maintained varenicline during the observation period were more active in discussions about varenicline, exposed to significantly more positive and negative sentiments about varenicline, and to higher rates of positive sentiments relative to negative sentiments. The probability of varenicline initiation and maintenance during the 30-day observation period increased as users’ rate of positive sentiment relative to negative sentiment exposure increased. Replication of analyses with bupropion showed similar, but weaker relationships.

  Take Away: QuitNet user sentiments about pharmaceutical cessation aids may influence new users’ decisions to use or maintain use of pharmaceutical cessation aids.

• Related Articles

  Characteristics of nurses who used the internet-based nurses QuitNet for smoking cessation. Bialous SA, Sarna L, Wells M, Elashoff D, Wewers ME, Froelicher ES. 2009. Public Health Nursing. 26(4): 329-338. doi: 10.1111/j.1525-1446.2009.00787.x

  Content-driven analysis of an online community for smoking cessation: Integration of qualitative techniques, automated text analysis, and affiliation networks. Myneni S, Fujimoto K, Cobb N, Cohen T. 2015. American Journal of Public Health. 105(6):1206-1212. doi: 10.2105/AJPH.2014.302464

  In pursuit of theoretical ground in behavior change support systems: Analysis of peer-to-peer communication in a health-related online community. Myneni S, Cobb N, Cohen T. 2016. Journal of Medical Internet Research. 18(2): e28. doi: 10.2196/jmir.4671