Evaluation of the International Diffusion of the European Code Against Cancer Through Mobile Telephony

A type-2 hybrid effectiveness-implementation study was designed to primarily test the impact of a mHealth implementation strategy on implementation outcomes (e.g., adoption, fidelity of intervention delivery) while simultaneously gathering information on the effectiveness of the intervention. The feasibility single-blind randomised controlled trial has been carried out on a sample of the general population in the Autonomous Community of Andalusia, the most populated region of Spain, with more than 8.5 million residents (17.9%). The Andalusian Health Service (SAS, acronym in Spanish) is the main healthcare provider of the Andalusian Public Health System (SSPA, acronym in Spanish), part of the decentralised Spanish National Health System, which provides free universal health insurance. The SAS oversees "Salud Responde" (SR, acronym in Spanish), a multi-channel platform that provides a broad online portfolio of services to respond to health-related issues and administrative management needs of citizens and health professionals, complementing the care provided by the health care units. It allows 24 hours a day access to the services and benefits of the SSPA, from any point in Andalusia, through telephone, e-mail, short message system (SMS) and a mobile application (app) to manage appointments and reminders, and to disseminate personalized health information. The app has a high coverage with more than 5.7 million users (67% of the Andalusian's population) registered in its system in 2023; yet the SR platform maintains the information system service via SMS for specific campaigns to those users who do not have or want the app installed on their mobile phone (around 9700 registered users).

The inclusion criteria were being 18 years or over, being entitled to health care benefits in SSPA, having a mobile phone, and being able to read messages on them. A sample of 1991 users from the User Database (BDU, in Spanish) of the SSPA, that in 2019 included more than 8.5 million users (REF), was used to randomly assign subjects to the different groups. The first randomization was performed according to the recruitment procedure (users registered or not in the SR platform) and the technology used to disseminate the messages (app or SMS). Group 1 (G1), subjects registered in the SR app service platform, and Group 2 (G2) subjects registered in the SR SMS service platform, were recruited through phone call. The rest of subjects needed to reach the target sample size were recruited from those included in the BDU database through a phone call. A maximum of eight attempts were done before considering no response. To be able to calculate participation rate (positive response rate + acceptance to participate), no replacement was performed.

For those not registered in the SR platform who accepted to participate in the study, a further distribution was done based on a discriminatory question: those users with internet connection in their mobile phone both through 3G/4G and Wi-Fi were allocated in Group 3 (G3) and encouraged to download the app to receive messages, and those without internet were allocated in Group 4 (G4) to receive messages by SMS. Informed consent to participate was requested for tehe investigators and the voluntary and anonymous nature of the participation was ensured. The simple randomisation to the different study groups was performed using the EPIDAT 4.0 programme. The second randomization was performed based on the messages' delivery frequency: frequency of 2 SMS or 2 app notifications ("push") per week, or frequency of 7 SMS or 7 app notifications per week.

Intervention After the second randomisation, an intervention was conducted for 4 weeks where participants were sent app push notifications or SMS messages via the SR platform. Each push notification or SMS message included a cancer prevention recommendation from the ECAC 4th edition (in Spanish) from the following list: "Don't smoke. Do not use any type of tobacco"; "Maintain a healthy weight"; "Exercise daily. Limit time spent sitting"; "Eat lots of whole grains, legumes, fruits and vegetables"; "Limit high-calorie foods (foods rich in sugar or fat) and avoid sugary drinks"; "Avoid processed meat; limit consumption of red meat and salt-rich foods"; "Limit alcohol consumption, although it is best for cancer prevention to avoid alcoholic beverages"; and the URL link to visit the Spanish version of the ECAC's website to consult additional information, through the mobile phone or other digital device. All messages were endorsed and signed by the Head of the Andalusian Regional Ministry of Health.

Data collection was conducted for the investigators at two time points: at baseline during the recruitment call, a pre-intervention survey was performed to gather basic socio-demographic and patterns of use of new technologies (e.g., type of mobile device, internet access from phone or computer, etc.); at the end of the intervention, from one week to one month after receiving the last message, a post-intervention survey was carried out by telephone to all participants.

Variables and categorization The sample was described for the investigators according to the following socio-demographic variables of the participants: sex, age, employment status, level of education, living alone or not, and province of residence. The mHealth-related variables included were the type of mobile phone, access to the Internet through the mobile phone or other digital device, permanent access to Internet data or only through Wi-Fi, ability to browse on the Internet and from which type of device, and willingness to receive messages in the future (acceptability of a future intervention). The survey questions were structured for the investigators to estimate intervention-related variables: the adoption of the innovation (number of messages read) for the participants, the fidelity to the mHealth implementation strategy (receiving and reading the number of messages expected), the preferred frequency and time to receive messages, willingness to seek further information (number of visits to the ECAC website), and perceived utility and understandability of the messages.

Statistical analysis The sample size was calculated for the investigators considering a confidence level of 95% (5% alpha error), a precision of 5% (width of the confidence interval), an expected proportion of 50% (maximum indeterminacy) in the main variables from the Conceptual Framework for Implementation Outcomes (adoption, fidelity, appropriateness, and acceptability), and a non-response rate of 5%.

Participation rate was considered for the investigators an outcome and therefore calculated as the rate of those who accepted participate in the study among all those who responded to the recruitment call. A descriptive analysis was performed for the investigators to assess homogeneity across groups by sociodemographic data by means of relative frequencies, using Chi-square test for categorical covariates and t-tests for continuous covariates, in the univariate analysis. The sample was also described at follow-up, according to socio-demographic data and patterns of mHealth use by messaging channel. The implementation research outcomes of adoption, fidelity, appropriateness, and acceptability, and willingness to visit the ECAC's website (coded as a form of adoption) for the participants, were calculated with relative frequencies by dissemination channel and intervention group. Finally, a multivariate logistic regression was conducted to characterize the participant socio-demographic characteristics associated with having read at least one message weekly. Crude and adjusted Odds Ratios (cOR and aOR) and its 95% confidence intervals (95% CI) were calculated. Significant association was defined with p≤0.05. STATA 17.0 and SPSS V.21.0. were used in the analysis.