Abstract
Introduction:Despite a substantial increase in use of SMS text messages for collecting smoking-related data, there is limited knowledge on the parameters of response. This study assessed response rates, response speed, impact of reminders and predictors of response to text message assessments among smokers.Methods:Data were from two SMS cessation intervention trials using clinical samples of pregnant (n = 198) and general smokers (n = 293) sent text message assessments during 3-month cessation programs. Response rates were calculated using data from the host web-server. Changes in response over time, impact of reminders and potential demographic (age, gender, ethnicity, parity, and deprivation) and smoking (nicotine dependence, determination to quit, prenatal smoking history, smoking status at follow-up) predictors of response were analyzed.Results:Mean response rates were 61.9% (pregnant) and 67.8% (general) with aggregated median response times of 0.35 (pregnant) and 0.64 (general) hours. Response rate reduced over time (P = .003) for general smokers only. Text message reminders had a significant effect on response (Ps < .001), with observed mean increases of 13.8% (pregnant) and 17.7% (general). Age (odds ratio [OR] = 0.95, 95% confidence interval [CI] 0.90–1.00) and deprivation (OR = 0.98, 95% CI 0.96–1.00) weakly predicted response among pregnant smokers and nonsmoking status at 4 weeks follow-up (OR = 8.63, 95% CI 3.03–24.58) predicted response among general smokers.Conclusions:Text message assessments within trial-based cessation programs yield rapid responses from a sizable proportion of smokers, which can be increased using text reminders. While few sources of nonresponse bias were identified for general smokers, older and more deprived pregnant women were less likely to respond.Implications:This study demonstrates that most pregnant and general smokers enrolled in a cessation trial will respond to a small number of questions about their smoking sent by text message, mostly within 1 hour of being sent the assessment text message. For those who do not initially respond, our findings suggest that 24- and 48-hour text message reminders are likely to increase response a small but meaningful amount. However, older age and higher deprivation among pregnant smokers and relapse among general smokers is likely to reduce the chance of response.
Highlights
Despite a substantial increase in use of SMS text messages for collecting smoking-related data, there is limited knowledge on the parameters of response
The response rate significantly declined from weeks 3 to 7 by 15.5% (GEE Wald test: χ2(1) = 8.56 P = .003)
Impact of Reminders on Response For pregnant smokers, after the reminders were sent, increases in response rate were 16.3%, 12.0%
Summary
Despite a substantial increase in use of SMS text messages for collecting smoking-related data, there is limited knowledge on the parameters of response. SMS text messages are recommended as a method for collecting data from participants of medical research[1] and are increasingly used for doing so, including for smoking behavior.[2] Text messages have high reach potential for data collection due to very high mobile phone ownership in developed nations and rapidly increasing ownership in developing nations.[3] While smoking studies using SMS assessments often report response rates, they seldom explore time to response, and have not examined interventions to increase response or predictors of response Bridging these knowledge gaps is important to help determine the utility of text messaging for measuring smoking characteristics and to inform the design of digital smoking cessation interventions. Estimating the speed of response profile and the impact of reminders would help optimize the use and timing of reminders and help inform interventions contingent on data input during delivery such as tailored interventions.[4,5] A record of lapses or relapse to smoking, changes in determinants of relapse or engagement in a period of abstinence can be used by eHealth and mHealth interventions to dynamically trigger support and tailor the type and intensity of that support.[6,7] Increasing our knowledge of factors associated with response would help identify potential biases when text messages were used exclusively for assessment purposes and identify groups who might become underserved by interventions where effectiveness may be partially dependent on interaction
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
