Uptake of Tailored Text Message Smoking Cessation Support in Pregnancy When Advertised on the Internet (MiQuit): Observational Study.

BACKGROUND: Smoking in pregnancy is a major public health concern. Pregnant smokers are particularly difficult to reach, with low uptake of support options and few effective interventions. Text message-based self-help is a promising, low-cost intervention for this population, but its real-world uptake is largely unknown. OBJECTIVE: The objective of this study was to explore the uptake and cost-effectiveness of a tailored, theory-guided, text message intervention for pregnant smokers ("MiQuit") when advertised on the internet. METHODS: Links to a website providing MiQuit initiation information (texting a short code) were advertised on a cost-per-click basis on 2 websites (Google Search and Facebook; £1000 budget each) and free of charge within smoking-in-pregnancy webpages on 2 noncommercial websites (National Childbirth Trust and NHS Choices). Daily budgets were capped to allow the Google and Facebook adverts to run for 1 and 3 months, respectively. We recorded the number of times adverts were shown and clicked on, the number of MiQuit initiations, the characteristics of those initiating MiQuit, and whether support was discontinued prematurely. For the commercial adverts, we calculated the cost per initiation and, using quit rates obtained from an earlier clinical trial, estimated the cost per additional quitter. RESULTS: With equal capped budgets, there were 812 and 1889 advert clicks to the MiQuit website from Google (search-based) and Facebook (banner) adverts, respectively. MiQuit was initiated by 5.2% (42/812) of those clicking via Google (95% CI 3.9%-6.9%) and 2.22% (42/1889) of those clicking via Facebook (95% CI 1.65%-2.99%). Adverts on noncommercial webpages generated 53 clicks over 6 months, with 9 initiations (9/53, 17%; 95% CI 9%-30%). For the commercial websites combined, mean cost per initiation was £24.73; estimated cost per additional quitter, including text delivery costs, was £735.86 (95% CI £227.66-£5223.93). Those initiating MiQuit via Google were typically very early in pregnancy (median gestation 5 weeks, interquartile range 10 weeks); those initiating via Facebook were distributed more evenly across pregnancy (median gestation 16 weeks, interquartile range 14 weeks). CONCLUSIONS: Commercial online adverts are a feasible, likely cost-effective method for engaging pregnant smokers in digital cessation support and may generate uptake at a faster rate than noncommercial websites. As a strategy for implementing MiQuit, online advertising has large reach potential and can offer support to a hard-to-reach population of smokers.

Cognitive and Behavioral Predictors of Quit Attempts and Biochemically-Validated Abstinence During Pregnancy.

INTRODUCTION: Initiating a quit attempt and achieving abstinence are distinct behaviors that have distinct correlates in general smokers. Studies predicting prenatal smoking have not addressed this. METHODS: Pregnant smokers (N = 207), recruited to a cessation intervention trial, were used as an observational cohort. Women completed measures at baseline and 12-week follow-up (mid-late pregnancy). Outcomes were having made at least one quit attempt since baseline, and cotinine-validated 7-day abstinence at follow-up in attempters. Baseline predictors included demographics (age, deprivation, partner's smoking), smoking behaviors (nicotine dependence, quit attempt history, previous prenatal smoking), and smoking beliefs (self-efficacy, determination, intention to quit, nonsmoker identity, social support, pregnancy-outcome beliefs). For each outcome, variables reaching p < .1 in logistic regression analyses were entered into a multivariate model controlling for trial arm. A complete case analysis was undertaken, with missing data assumptions tested in sensitivity analyses. RESULTS: One hundred seventy-five women (85%) completed follow-up. Intention and determination to quit (p < .001), self-efficacy, nonsmoker identity, and not having previously smoked in pregnancy (p < .05) were univariate predictors of making a quit attempt, with stronger intention to quit the only independent predictor (multivariate odds ratio [OR] = 2.41, 95% confidence interval [CI] 1.19-4.87). Only nicotine dependence predicted validated abstinence among those who made a quit attempt (multivariate OR = 0.25, 95% CI 0.10-0.60). CONCLUSIONS: Initiating a quit attempt and achieving abstinence during pregnancy were found to have different correlates. For women yet to make a quit attempt in their pregnancy, smoking beliefs may be important intervention targets, but once they are engaged in quitting, nicotine dependence appears of prime importance. IMPLICATIONS: This study suggests that cognitive, particularly motivational, variables predict whether pregnant smokers will make a quit attempt, but they do not predict successful abstinence in those who attempt to quit, where nicotine dependence dominates. Interventions should facilitate quit attempts by targeting motivational variables among pregnant women who continue to smoke, but should focus on managing withdrawal once a woman initiates a quit attempt.