Application of the theory of regulatory fit to promote adherence to evidence-based breast cancer screening recommendations: experimental versus longitudinal evidence

Objectives To reduce overtreatment caused by overuse of screening, it is advisable to reduce the demand for mammography screening outside the recommended guidelines among women who are not yet eligible for inclusion in systematic screening programmes. According to principles of regulatory fit theory, people make decisions motivated by either orientation to achieving and maximising gains or avoiding losses. A study developed in two phases investigated whether video messages, explaining the risks and benefits of mammography screening for those not yet eligible, are perceived as persuasive Design Phase 1 was an experimental study in which women’s motivation orientation was experimentally induced and then they were exposed to a matching video message about mammography screening. A control group received a neutral stimulus. Phase 2 introduced a longitudinal component to study 1, adding a condition in which the messages did not match with the group’s motivation orientation. Participants’ natural motivation orientation was measured through a validated questionnaire Participants 360 women participated in phase 1 and another 292 in phase 2. Participants’ age ranged from 30 to 45 years, and had no history of breast cancer or known BReast CAncer gene (BRCA) 1/2 mutation. Results In phase 1, a match between participants’ motivation orientation and message content decreased the intention to seek mammography screening outside the recommended guidelines. Phase 2, however, did not show such an effect. Fear of breast cancer and risk perception were significantly related to intention to seek mammography screening Conclusions Public health researchers should consider reducing the impact of negative emotions (ie, fear of breast cancer) and risk perception when promoting adherence to evidence-based breast cancer screening recommendations.

I, the Submitting Author has the right to grant and does grant on behalf of all authors of the Work (as defined in the below author licence), an exclusive licence and/or a non-exclusive licence for contributions from authors who are: i) UK Crown employees; ii) where BMJ has agreed a CC-BY licence shall apply, and/or iii) in accordance with the terms applicable for US Federal Government officers or employees acting as part of their official duties; on a worldwide, perpetual, irrevocable, royalty-free basis to BMJ Publishing Group Ltd ("BMJ") its licensees and where the relevant Journal is co-owned by BMJ to the co-owners of the Journal, to publish the Work in this journal and any other BMJ products and to exploit all rights, as set out in our licence.
The Submitting Author accepts and understands that any supply made under these terms is made by BMJ to the Submitting Author unless you are acting as an employee on behalf of your employer or a postgraduate student of an affiliated institution which is paying any applicable article publishing charge ("APC") for Open Access articles. Where the Submitting Author wishes to make the Work available on an Open Access basis (and intends to pay the relevant APC), the terms of reuse of such Open Access shall be governed by a Creative Commons licence -details of these licences and which Creative Commons licence will apply to this Work are set out in our licence referred to above.
Other than as permitted in any relevant BMJ Author's Self Archiving Policies, I confirm this Work has not been accepted for publication elsewhere, is not being considered for publication elsewhere and does not duplicate material already published. I confirm all authors consent to publication of this Work and authorise the granting of this licence.  screening can save lives. Assuming that a substantial part of breast cancer screening below the age of 50 is not due to medical indications, to promote the adherence to evidence-based recommendations on breast cancer screening among young women seems to be a vital research mandate. A way to overcome the impact of an individual's involvement and negative emotions as motivational factors for high breast cancer screening demand could arise by the activation of an alternative motivation system, such as the regulatory orientation .

Theory of Regulatory Focus and Regulatory Fit
The Theory of Regulatory Focus  states that people's regulatory orientation is a motivational principle, which influences behavioural choices, and it is characterized either by a promotion or by a prevention orientation. While individuals with a promotion focus are described as eagerly pursuing their goals and striving towards the realization of desired outcomes, those with a prevention focus are defined as being safety-driven and vigilant to prevent errors and undesired results Keller, 2006). The regulatory focus orientation can be primed (Cesario, Higgins, & Scholer, 2008; or either estimated through questionnaires (Higgins et al., 2001).
A phenomenon called 'regulatory fit' occurs    The purpose of the present research was to test whether health messages framed to correspond with a woman's regulatory focus orientation are effective in reducing the intention to ask for medically not indicated breast cancer screening under the age of 50, challenging lay people's common sense. Achievement from the present research would be twofold. Theoretically, significant results will improve knowledge on the impact that regulatory orientation, as a motivational system, has when applied to a counterintuitive topic for laypeople. The practical implication will include the possibility to reduce the demand for regular mammography without medical indications and, then, to moderate the possible harms associated.
Two studies were developed. Study 1 tested the hypothesis that the fit between the message frame and the women's regulatory orientation would reduce their intention to ask for medically not indicated breast cancer screening. Study 2 longitudinally tested the same association comparing two fit conditions vs. two non-fit conditions vs. a control condition. It was expected that the fit conditions would lead to a reduction of the intention to ask for medically not indicated breast cancer screening compared to the non-fit conditions and the control group.

Participants
An a priori power analysis applying G*Power 3.  pre-test (N = 140) and who filled in the entire survey (N = 360). Participants were randomly assigned to prevention fit, promotion fit, and control condition (see Table 1).
No differences were found between the intervention groups and the control group on socio-demographic variables.

Procedure
A pre-post-test design with two experimental conditions and a control group was applied (see Table 2 for full details).
[insert Table 2 here] After the pre-test questionnaires, participants were randomized into promotion fit, prevention fit, or control condition. In the fit conditions, the two regulatory foci were primed  and then followed by video-messages fitting with the primed focus. Immediately after priming, participants in the promotion fit condition watched a video message emphasizing promotion concerns (i.e., they should adhere to evidence-based recommendations on mammography screening for safety and health protection reasons). Participants in the prevention fit condition watched a video emphasizing prevention concerns (i.e., they should not abstain from following the evidence-based recommendations on mammography screening to avoid negative/side effects). Participants in the control group did not receive any priming and read a general health leaflet. Table 3 shows the content of the video messages and leaflet. In a pilot study, 30 women assessed the survey as clear and understandable.
Regulatory Focus Induction. Regulatory fit manipulation was induced by completing the regulatory fit questionnaire . Prevention induced participants were asked to list two of their current obligations and then write down five actions they could take to avoid failure in fulfilling them. Promotion induced participants were asked to list two aspirations and write down five actions they could take to ensure their accomplishment.
Intention to ask for breast cancer screening. Three items measured the intention to have a mammography screening for breast cancer in the next 2-3 years were applied (the measure shows acceptable internal consistency, α = .97, rs > .94).

Analytic Strategy
Data were normalized through reverse scoring and logarithmic transformations.
There were not missing data.

Results
The ANCOVA [F(2, 319) = 49.57, p < .0001, η 2 p = .24] revealed that the promotion fit condition, t(319) = -8.80, p < .0001, r = .44, as well as the prevention fit condition, t(319) = -8.80, p < .0001, r = .44, were both associated with lower intentions to ask for breast cancer screening compared to the control condition. There was no significant difference between the two intervention groups (p > .05). See Table  F Table 4 here] A significant association between promotion fit/control condition and past diagnoses of breast cancer among first grade relatives was found, χ 2

Discussion
Limitations of Study 1 include that women received the intervention one time.
Study 2 was designed to overcome this limitation.

Participants
An a priori power analysis was calculated as for Study 1, and a sample size of 312 was estimated. Nine hundred seventy-three women from 30 to 45 filled in the pretest questionnaires (i.e., pre-test sample). Completed questionnaires (i.e., analytical sample) were returned from 292 women with an attrition rate of 70%. Comparisons between the pre-test sample and the analytical sample did not yield significant differences. The 292 participants were randomly assigned to five conditions: promotion fit, promotion non-fit, prevention fit, prevention non-fit, and control condition (see Table 1). Women aged 30 to 45 living in Ticino and Italy participated.
Italian and Ticinese-Swiss participants share the same culture and language and follow the same rules for their breast cancer screening programs. No differences were found among the five groups regarding socio-demographic variables or other pre-test variables.

Procedure
A pre-post-test longitudinal design was applied with four experimental conditions, two fit conditions (promotion and prevention), two non-fit conditions (promotion and prevention), and a control group (see Table 2). In the pre-test (T0), participants replied to a set of questions comprising socio-demographic variables, covariates, and a questionnaire measuring women's regulatory focus. The latest was applied because working with the trait regulatory focus would be more stable than a primed focus in a longitudinal design. Participants were randomly assigned to the fit or non-fit condition or the control group. In the control group, half of them had a promotion orientation and half prevention orientation. Participants in the fit conditions watched two videos (T1 and T2) emphasizing the fit concerns (see Table 5). In the non-fit conditions, participants watched two videos (T1 and T2) emphasizing the nonfit concerns. In the control group, participants watched two videos (T1 and T2) treating the topic of breast cancer prevention, but without any regulatory prompt. A post-test questionnaire evaluates women's' intention to ask for opportunistic screening (T3). Ten days elapsed between each experimental phase.
[Insert Table 5 here] RCSMedia Group, an Italian-based publishing group that uses participant panels.
Inclusion/exclusion criteria were as for study 1

Patient and Public Involvement
As for Study 1.

Measures
Pre-test covariates were measured as for study 1; the intention was measured both in the pre-test and post-test.
Trait Regulatory Orientation. The Regulatory Focus Questionnaire (Higgins et al., 2001) was applied in the pre-test phase. The questions asked how frequently several specific events occur in the participant's life. Six questions capture the promotion focus, and the other five the prevention focus. Participants replied on a 5-point scale from 1 (never) to 5 (very often). The scores for promotion and prevention scales were calculated averaging the answers on given items: data show good internal consistency for both promotion, α = .66, rs > .33, and prevention, α = .74, rs > .47. The individual's chronic orientation was calculated following the original procedure (see Higgins et al., 2001).

Analytic Strategy
Data were normalized through reverse scoring and logarithmic transformations.
There were not missing data. Repeated measures ANCOVA was applied.

Discussion
The application of regulatory fit in the area of health communication is . Therefore, we could exclude that the two methods have had a differential impact on post-test intention.

Promotion Fit video-message Prevention Fit video-message Control Leaflet
The mammography screening is a method for the early detection of breast cancer. Using x-rays, mammograms can identify very small tumours generally longer before they are palpable. A mammogram is a method that is used early, often even without symptoms. In a screening program, experts recommend mammography from the age of 50. Here in Ticino, women aged 50 and over are invited to voluntarily undergo mammography every two years at one of the accredited Radiology Centres. For most women between the ages of 50 and 69, the benefits of screening are higher than the risks. However, nevertheless, it is essential to be adequately informed to make the best decision about mammography.
To protect their health, women under the age of 50, without a medical indication or family history of cancer, are excluded from the program.
To avoid adverse effects on their health, women under the age of 50 without a medical indication or family history of cancer are excluded from the program.
Now I would like to explain the scientific reasoning behind the recommendation to not undergo a mammogram without a medical reason. Anyway, in case of doubt or symptoms, I suggest to contact your doctor. So, you are asking why women under the age of 50 are excluded from mammography screening. Scientific research shows that for women between the ages of 50 and 69 mammography screening is the most effective method for the early detection of breast cancer and for reducing the mortality rate associated with it. In contrast, for young women between the ages of 30 and 49, the disadvantages and risks to health are greater than the benefits. This is mainly due to the fact that women before menopause have a denser breast tissue.
Given the reasons I have just presented, one should avoid undergoing a mammogram early to prevent negative consequences.
Given the reasons I have just presented, to early undergo a mammogram can lead to negative consequences.
For example, mammography could show anomalies that, after additional diagnostic tests, could be proved to be benign. This type of error is called a false positive. If for women aged between 50 and 69 this risk is minimal, for young women is higher due to the denser breast tissue. In addition, breast cancer could not be seen by mammograms because it is too small and therefore the exam may appear normal although cancer is present. This is a false-negative result. These risks always exist, but they are higher for young women. As all medical testing, waiting for the outcome of the mammogram can generate a state of anxiety and the procedure sometimes can be perceived as painful. Radiation exposure also have health consequences. Although the exposure is minimal, for women under the age of 50 the risk is higher than the benefits of mammography. Furthermore, screening could lead to over-treatment for tumours that are benign. Over diagnosis represents approximately 1-10% of diagnosed cancers. This would expose young women to the negative effects of anti-cancer therapies, without a real need. In the absence of scientific evidence of the effectiveness of mammographic screening for young women, the inclusion of young women in the program would entail additional costs for society. These financial resources could be used to prevent other diseases.
Healthy eating associated with an active lifestyle is a useful way for disease prevention. An adequate and balanced diet plan guarantees an optimal supply of nutrients to meet the needs of your body. A balanced diet also allows receiving substances that play a protective and/or preventive role against diseases.
This booklet -thought for people of all ages without any particular diseases -explains the scientific reasoning behind the recommendation to follow a healthy diet even in the absence of particular weight or health disorders. Anyway, in case of doubts or problems, we suggest contacting your doctor. For these reasons, it is recommended that young women do not ignore the indications related to breast cancer screening.

Eating healthy is easy
Food provides the body with both the necessary energy and nutrients that allow it to function correctly.
The diet must, therefore, provide a correct caloric intake and a sufficient amount of nutrients. No food is so complete that it contains everything the body needs. The basic rule is, therefore, to eat everything and in a varied way.

How do I interpret it?
A balanced diet requires the foods at the base of the pyramid to be consumed in higher quantities.
Climbing up to the vertex, the quantities of food consumed should be limited. Nothing is forbidden, every food finds its place in a balanced diet, but the recommended quantities will depend on its location in the pyramid.
Give food the importance that deserves, eating healthier. Your health and well-being will be better! Note: in bold are shown the specific parts of the promotion focus and prevention focus versions. The Videos created for Study 1 can be retrieved from https://youtu.be/mperSG5_9yQ and https://youtu.be/KnhRUnDoSV0. Both videos last 3:28 minutes. The videos created for Study 2 can be retrieved from https://youtu.be/btM3HrvYDlQ, https://youtu.be/BZPjFPUQuvw, https://youtu.be/-lXzGpcmzD4, https://youtu.be/jRi8Y-sZvSc. A translation of the Italian voice has been provided in Table B1 and B2

STUDY 2
For the early detection of breast cancer, experts recommend mammography to women aged 50 and over. Mammography is the most effective medical examination for the early detection of breast cancer. It consists of an X-ray exposure that allows you to identify even very small tumours before they are palpable or recognizable. Women over the age of 50 are invited to undergo a mammogram every 2 years at an accredited radiology centre. For women between 50 and 69, the benefits of the exam outweigh the risks. And before the age of 50? It is a medical recommendation: before the age of 50, the risks of the examination are higher than the benefits. Mammography is a breast test that allows you to detect even many small tumours. Over 50 years, it is done every two years. Women under the age of 50 are excluded from the breast cancer screening program, except in case of genetic predisposition or family history of breast cancer. What are the reasons for this decision? The risk of false positives in young women is higher because the breast tissue is denser.
It is advised not to make mammograms before the age of 50 in order not to expose themselves to anti-cancer treatments not recommended as they are often directed to benign anomalies. (PROMOTION FOCUS) It is advisable not to make mammograms before the age of 50 to avoid exposure to non-recommended anti-cancer treatments as they are often directed to benign anomalies.

(PREVENTION FOCUS)
The breast cancer screening could lead to an exposition of nonrecommended anti-cancer treatments, as they are often directed to benign anomalies. (CONTROL GROUP) Breast cancer is much rarer in women under the age of 50.
Excluding younger women from screening allows them to be protected from unnecessary radiation exposure. Furthermore, this choice promotes psychological wellbeing against stress and anxiety. For these reasons, mammographic screening involves only women over 50 years. If you are under 50 and want to take care of your health, we recommend that you respect the age threshold.

(PROMOTION FOCUS)
Excluding younger women from screening allows you to avoid unnecessary radiation exposure. Furthermore, this choice avoids psychological discomforts, such as stress and anxiety. For these reasons, mammographic screening involves only women over 50 years. If you are under 50 years old and want to avoid negative consequences for your health, we recommend that you respect the age threshold.

(PREVENTION FOCUS)
Radiation exposure poses health risks. In the case of mammography, the exposure is minimal, but for women, under the age of 50, the risk is higher than the benefits. Furthermore, as any other medical procedure, waiting for the outcome can generate anxiety and stress. For these reasons, mammographic screening involves only women over 50 years. If you are under 50 years old, we recommend that you respect the age threshold. (CONTROL GROUP)

VIDEO 2
Conscious prevention it is worth it! Note: in bold are shown the specific parts of the promotion focus, prevention focus, and control versions.   • An individual's goal-pursuit orientation was induced in Study 1 through a priming technique, and measured through a validated questionnaire in Study 2.

STROBE 2007 (v4) Statement-Checklist of items that should be included in reports of cross-sectional studies
• Messages were tailored to create a match (or not) between message content and the individual's goal-pursuit orientation.
• Limitations of the studies included dropout rates (Study 2) and selection bias (due to cancer fear). Breast cancer is one of the most common forms of cancer in women worldwide and the principal cause of cancer-related death in the female population, [1]. To promote early diagnosis, many EU countries have introduced systematic breast cancer screening programs, [2]. However, the age threshold to start inviting women to screening is in dispute, [3][4][5]. The balance between the benefits (i.e., reducing breast cancer mortality) and the harm associated with mammography (i.e. x-ray exposure, over diagnosis and false positive results; see, [4][5][6][7][8] is uncertain. Technologies for breast cancer screening have been constantly evolving, affecting evidence quality and suggested recommendations, [9]. For these reasons recommended age for starting screening have varied from 40,[10], to 45, [11,12], to 50 years, [13,14].
There has been a vast amount of research investigating the intentions of women to adhere to screening guidelines and encouraging women with characteristics that match with the national guidelines to attend systematic screening, [15][16][17]. However, many women below the established age threshold seek and receive mammography screenings without medical reasons in the U.S., [18,19], Switzerland, [20,21], Germany, [22], and The Netherlands, [22]. Studies show that women tend to overestimate the mortality reduction determined by breast cancer screening, [23,24] and that they have unrealistic expectations regarding screening as reducing the risk of breast cancer, [25]. Moreover, social pressure in favour of breast cancer screening may stimulate a sense of moral obligation to participate, [26,27], even among young women.
Given the above-mentioned considerations, women under the age threshold for systematic breast cancer screening may consider the recommendation to avoid screening as counterintuitive, although scientifically supported, because of social pressure and the belief that cancer screening can save lives. The present research aimed to promote adherence to evidence-based recommendations on breast cancer screening among young women by activating a motivation system, such as regulatory orientation, [28].

Theory of Regulatory Fit
According to a popular psychological theory proposed by [29], people show one of two regulatory orientations, which determines how they pursue their goals. They either show a promotion-focused orientation, meaning they eagerly strive towards the realization of desired outcomes, or they show a prevention-focused orientation, emphasizing the prevention of errors and losses and making them safety-driven, [29,30]. While every individual has a natural tendency to lean more towards one orientation than the other, thus making it a measurable trait, [31], the regulatory orientation can also be experimentally induced, [28,29,32].
If individuals adopt a behaviour or processes a message highlighting goalpursuit strategies that match their regulatory orientation, they experience a phenomenon called "regulatory fit", [28]. For example, if a person with a promotion orientation reads a message highlighting strategies to achieve gains, a fit condition occurs. The same applies to someone with a prevention orientation processing a message emphasizing strategies to avoid losses. Such a fit or match causes an "it just feels right" perception, increasing the perceived value of the behaviour [33].
Regulatory fit has been consistently found to influence outcomes such as evaluation, behaviour and behavioural intention, [34]. Some authors [33] showed that this "it-just-feels-right" experience is also transferred to the context of persuasion, with positive effect of regulatory fit on the perceived persuasiveness of a message. A study by [35] in the context of tobacco use prevention among adolescents is in line with this finding. The effects of regulatory fit have also been extensively studied in the context of disease prevention and health promotion, [35,36]. In particular, some authors [37] applied the principles of regulatory fit to inform people about the benefits of regular HP1: a fit between the message frame and the regulatory orientation would lead to an immediate reduction of the intention to ask for breast cancer screening, in non-at risk women under the age threshold indicated by the local guidelines.
This hypothesis was tested in experimental Study 1.
HP2: a fit between the message frame and the regulatory orientation would lead to a reduction of the intention to ask for breast cancer screening, stable over time. This hypothesis was tested in the longitudinal experimental Study 2.

Participants
An a priori power analysis applying G*Power 3. Participants were randomly assigned to prevention fit, promotion fit, and control condition (see Table 1). No differences were found between the intervention groups and the control group on socio-demographic variables.

Process, Measures and Data Collection
A pre-post-test design with two experimental conditions and a control group was applied (see Figure 1 for full details).

Patient and Public Involvement
Results from previous studies involving participants from Switzerland informed the present research (see, [21]). Participants were not directly involved in the design, conduct, recruitment, reporting or dissemination of the study results. An expert panel, composed of two health communication professionals with expertise on regulatory fit theory, evaluated the message contents and the graphical aspects of the videos. Data were normalized through reverse scoring and logarithmic transformations.

Analytic Strategy
There were no missing data. An ANCOVA tested the main hypothesis (HP1) of the study. The fit vs. control conditions variable was inserted as independent variable. All the variables measured at the pre-test were inserted as covariates. Chi-square tests were conducted to evaluate whether the covariates might interact with the three experimental conditions in determining the intention to ask for breast cancer screening.

Results
The ANCOVA analysis revealed that women in the two experimental conditions showed less intention to ask for breast cancer screening compared to the women in the control condition. Thus, when there is a fit between individual orientation (i.e., a tendency to promote positive expected outcomes or to prevent negative outcomes for one's health) and the given message, then a persuasive effect is induced. There was no meaningful difference between the two manipulation conditions. Older women and women with higher levels of fear of breast cancer showed a greater intention to ask for breast cancer screening than younger ones and those with lower levels of fear. This evidence supports the assumption that regulatory orientation represents a motivational system able to overcome the impact of negative emotions and strengthen an individual's involvement in decision-making orientation.
Descriptive data and results from the ANCOVA are displayed in Table 2.  Within subject comparison between pre-and post-intention: F b (1, 267.91) = 5.10, p = .025, partial η 2 = .02 Between subject comparisons among groups: F b (4, 284) = .43, p > .05 Significant covariates: Fear of breast cancer: t(284) = 2.76, p = .006, B = .24, partial η 2 = .03 (95% Low CI = .07, 95% High CI = .42) Age, t(284) = 6.26, p < .0001, B = .11, partial η 2 = .12 (95% Low CI = .08, 95% High CI = .15), Risk perception, t(284) = 2.26, p = .024, B = .37, partial η 2 = .02 (95% Low CI = .05, 95% High CI = .70), Further analyses were conducted to evaluate whether the covariates might interact with the three experimental conditions in determining the intention to ask for breast cancer screening. Analyses revealed only one association among the three groups of women and the past diagnoses of breast cancer among first degree-relatives, χ 2 (2) = 12.98, p = .002. Women in the promotion fit condition had a lower number of breast cancer diagnoses among first-degree relatives than was expected (z = -1.96), while women in the control condition had a higher number than expected (z = 2.8). The subsequent ANCOVA did not find any significant interaction between past diagnosis of breast cancer among first-degree relatives and the experimental manipulations, therefore demonstrating that the regulatory fit genuinely influences the intention.

Participants
A priori power analysis estimated a sample size of 312. Recruitment took place from June to October 2017. The research was advertised through the Facebook page of the University and by RCSMedia Group, an Italy-based publishing group that uses participant panels. Inclusion/exclusion criteria were as for Study 1, with the addition that participants included in Study 1 could not participate in Study 2. Participants completed a written informed consent as for Study 1, and at the end received a 10 CHF/EU supermarket voucher. 973 women aged from 30 to 45 filled in the pre-test questionnaires (i.e., pre-test sample). Completed questionnaires (i.e., analytical sample) were returned by 292 women with an attrition rate of 70%. Comparisons between the pre-test sample and the analytical sample did not yield significant differences. 292 participants were randomly assigned to five conditions: promotion fit, promotion non-fit, prevention fit, prevention non-fit, and control condition (see Table 1). This time, women aged 30 to 45 living in  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59

Process, Measures and Data Collection
A pre-post-test longitudinal design was applied with four experimental conditions, two fit conditions (promotion and prevention), two non-fit conditions (promotion and prevention) and a control group (see Figure 2).
[Insert Figure 2 approx. here] In the pre-test (T0), participants replied to the same questions as for Study 1 (see online supplemental material). In Study 2, the regulatory focus orientation was measured with a questionnaire (online supplemental material), rather than induced as in Study 1, because working with the trait regulatory focus would be more stable than a primed focus in a longitudinal design. Subsequently, participants were randomly assigned to the fit or non-fit condition or control group. Participants in the fit conditions watched two videos (at T1 and T2) emphasizing the fit concerns (online supplemental material). In the non-fit conditions, participants watched two videos (at T1 and T2) emphasizing the non-fit concerns (online supplemental material). In the control group, participants watched two videos (at T1 and T2) treating the topic of breast cancer prevention, but without any regulatory prompt (online supplemental material).
A post-test questionnaire evaluated the women's intention to ask for opportunistic screening (T3). Ten days elapsed between each experimental phase.

Patient and Public Involvement
As for Study 1.

Analytic Strategy
Data were normalized through reverse scoring and logarithmic transformations. There were no missing data. A repeated measure ANCOVA tested the main hypothesis (HP2) of the study. The fit vs. unfit vs. control conditions variable was inserted as independent variable. All the variables measured at the pre-test were inserted as covariates.

Results
There was a general significant decrease of the intention from pre-to post-evaluation across groups, but no significant differences among them, indicating that the scores of the post-test intention among the five groups were in general the same. Among the covariates older women, greater fear of breast cancer and greater risk perception were associated with greater post-test intention compared to the opposite. Table 2 shows descriptive statistics and results from the analysis.
The intervention effect was not significant either when the two fit conditions and the two non-fit conditions were collapsed into two categories (i.e., comparison among fit condition vs. un-fit condition vs. control) as done in Study 1, even though a general decrease in the post-intention across groups was found as before. Risk perception was tested as a moderator, but the analysis was not significant.

Discussion
The application of regulatory fit in the area of health communication is beneficial across various health contexts and outcomes,[38] (Ludolph & Schulz, 2015).
The scientific community recognizes an undoubtable value in studies, [15][16][17] 1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y investigating the intentions of women to adhere to breast cancer screening guidelines, with the aim to encourage them to attend screening, rather than to avoid it. Anyway, there is a widespread debate on what the age threshold to start inviting women to screening should be, [3][4][5]. Possible beneficial effects of screening and the harm associated with it have to be balanced for informed decision-making. The most recent European Guidelines, [11] suggest that, in absence of risk conditions, women under the age of 45 should not receive breast cancer screenings regularly.
No previous studies have tested messages designed according to the assumptions of regulatory fit to influence the intention to not engage in disease detection screening.
This would challenge the intuitive perception that breast cancer screening leads to a mortality reduction determined by breast cancer, [23,24] and the unrealistic expectations regarding screening as reducing the risk of breast cancer, [25].
The present research shows inconsistent results. Study 1 confirmed the hypothesized effect of the intervention on the intention to seek mammography before the age of 45, with a reduction of the intention when a fit between the message frame and the individual's regulatory focus occurred. Longitudinal results from Study 2 demonstrated that this effect was not significant over one month, although a general decrease of the intention across groups was observed. Even though further evidence is needed to confirm our results, it still seems that the 'just-feels-right' experience appears to be insufficient to convince non-at risk women under the age threshold to avoid systematic breast cancer screening in the long run.
Our results could genuinely reflect the fact that the regulatory fit is not sufficient to induce a long-term decrease in women's intentions or could be an artefact of the research itself. Study 1 and Study 2 applied two different ways to evoke a regulatory orientation. Study 1 primed the individuals' regulatory orientation, whereas Study 2 measured it with a questionnaire to overcome a limitation of Study 1 and explore a  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n l y different aspect of the theory. One could argue that the different ways to induce vs. measure the regulatory orientation could have influenced the persuasiveness of the message and so its effectiveness. However, researchers of regulatory orientation suggest that there is no difference between the two procedures, [32]. Therefore, we could exclude that the two methods have had a differential impact on post-test intention. Possible differences in the cultural milieu of Italian-speaking Swiss and Italian participants might make the population primed to receive or primed to ignore the intervention. However, to the best of our knowledge there is no study comparing different cultural environments in the propensity to be primed or not.
The relatively small sample size and the recruitment strategies could have influenced the power of the analyses, the sample composition and, ultimately, the significance of the results. However, there is no such concern in Study 2 since the effect due to the intervention was not significant either when the two fit conditions and the two non-fit conditions were collapsed into two categories.   1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  The present research has several limitations. We experienced high dropout rates, especially in Study 2. The high dropout rates may be related to the topic of breast cancer itself or the fear associated with it. One could assume that women with a low level of fear of breast cancer may have decided not to take part in our research, and this may have created a selection bias that could affect the generalizability of the results. A second limitation concerns the fact that we measured the intention to ask for breast cancer screening, not the actual behaviour. Although according to many theories in the field of health promotion (e.g. Health Belief Model), the intention is a valid predictor of the actual behaviour, it would be beneficial if future research followed women until the moment they actually have a mammography.
In conclusion, it seems that by framing health messages that conform to a promotion or prevention focus, a decrease in the intention to ask for merely preventive opportunistic mammography screening is observed; but this takes place only immediately after message exposure. The influence decreases over time, and the messages lose their predictive effects after one month. This may be because breast cancer fear/opinions are very deeply ingrained in women and one/two messages cannot  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59

Competing interests statement
None declared.

Acknowledge
We are very grateful to Prof. John Hodgson for his thoughtful revisions.

Pre-test measures
Health Status and Healthy Lifestyle. Questions measured overall health status as perceived by the participants on a 5-point Likert scale and healthy lifestyle behaviours (i.e., diet, physical activity, smoking habits, alcohol consumption; see, [1].

Breast Cancer/Mammography Experience and Knowledge of the Ticino screening program.
Participants replied to a set of questions on: past diagnosis of breast cancer among first-grade relatives, [2], if they had a mammography in the past, if doctor recommended the mammography, if they had a breast biopsy, if they know the breast cancer screening program in Ticino, and its age thresholds. Ego-involvement. The Personal Involvement Inventory, [5] were administered measuring participants' involvement in breast cancer screening through affective and cognitive adjectives because previous research has found that . The scale was administered as a 7-digit semantic differential (e.g., important/unimportant, relevant/irrelevant or worthless/valuable). The original item 'of concern to me/of not concern to me' was deleted based on results of a previous study, [4].
Perceived benefits of mammography screening. The perceived benefit of mammography screening was measured by four items, [6]: 'Having a mammogram will help me find breast lumps ; 'Having a mammogram is the best way for me to find a very small breast lump'; 'Having a mammogram will decrease my chance of dying from breast cancer'. Participants replied on a 5point Likert-scale from 1 (strongly disagree) to 5 (strongly agree). Data from the present sample show that internal consistency was modest, Cronbach's α = .75, rs > .49 and the factor structure was good, χ 2 (1) = .51, p = .47, CFI = 1, RMSEA = .00.

Experimental manipulation
Regulatory Focus Priming Procedure. Prevention induced participants were asked to list two of their current obligations and then write down five actions they could take to avoid failure in fulfilling them, [7]. Promotion induced participants were asked to list two aspirations and write down five actions they could take to ensure their accomplishment, [7].
Video Messages. Participants in the promotion fit condition watched a video message emphasizing promotion concerns (i.e., they should adhere to evidence-based recommendations on mammography screening for safety and health protection reasons). Participants in the prevention fit condition watched a video emphasizing prevention concerns (i.e., they should not abstain from following the evidence-based recommendations on mammography screening to avoid negative/side effects). Participants in the control group did not receive any priming and read a general health leaflet. See Supplemental Table 1 for details of the voice-text of the two video messages and the control leaflet.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46 Note: grey rounded rectangles show the common parts of promotion and prevention video-messages; orange rounded rectangles show the promotion video-message specific parts (text in bold); blue rounded rectangles show the prevention video-message specific parts (text in bold); the green rounded rectangle shows the content of the control leaflet. The Videos created for Study 1 can be retrieved from https://youtu.be/mperSG5_9yQ and https://youtu.be/KnhRUnDoSV0. Both videos last 3:28 minutes. The videos created for Study 2 can be retrieved from https://youtu.be/btM3HrvYDlQ, https://youtu.be/BZPjFPUQuvw, https://youtu.be/-lXzGpcmzD4, https://youtu.be/jRi8Y-sZvSc. A translation of the Italian voice-over has been provided in this

Post-test Measures
Intention to ask for breast cancer screening. Intention was measured by the question "I am evaluating the idea to have a mammography screening for breast cancer in the next 2-3 years", [8].
Two further questions were added: "I have the intention to have a mammography screening for breast cancer in the next 2-3 years" and "I will take an appointment for a mammography screening for breast cancer in the next 2-3 years". Participants replied on a 5-point scale from 1 (definitely yes) to 5 (definitely not). Data from the present sample show that internal consistency was good, Cronbach's α = .97, rs > .94.

Pre-test Measures
Pre-test covariates were measured as for study 1. Intention to ask for breast cancer screening was asked during the pre-test with the three items applied in Study 1.

Experimental manipulation
Trait Regulatory Orientation. The Regulatory Focus Questionnaire, [9] was applied in the pre-test phase. The questions asked how frequently several specific events occur in the participant's life. Six questions capture the promotion focus, and the other five the prevention focus. Participants replied on a 5-point scale from 1 (never) to 5 (very often). The scores for promotion and prevention scales were calculated averaging the answers on given items: data show good internal consistency for both promotion, α = .66, rs > .33, and prevention, α = .74, rs > .47. The individual's chronic orientation was calculated following the original procedure, [9].
Video Messages. Six video-messages were developed for the present study:  Two video-messages emphasising prevention concerns;  Two video-messages emphasising promotion concerns;  Two video-messages without any prompt to regulatory orientation.
Supplemental Table 2 shows the content of the voice-text of the six video-messages.

Strengths and Limitations of this Study
• An experimental study (Phase 1) and an experimental study with a longitudinal component (Phase 2) were implemented applying principles from the theory of regulatory fit.
• An individual's goal-pursuit orientation was induced in Phase 1 through a priming technique, and measured through a validated questionnaire in Phase 2.
• Messages were tailored to create a match (or not) between message content and the individual's goal-pursuit orientation.
In the last years, there has been a vast amount of research on screen intention, including barriers, enablers, and how to get women with characteristics matching with the recommended guidelines to adhere to the screening programs,[see [15][16][17]. There was also a progressive shift from persuading women to undergo screening to increasing their informed decision making [18]. Targeted information programs and invitation materials encouraging women to learn about the screening procedures increased levels of knowledge and supported decision-making about their participation, [19,20]. Web-based dynamic decision aids, including pros, cons, controversies, and overdiagnosis-overtreatment issues, have been found to improve the quality of information without reducing the screening participation rate, [21].
Other research tested communication programs intending to inform women approaching 70 years of age about the benefits and harms of continuing screening, [22,23]. Similarly, non high-risk women below the recommended age threshold seek and receive mammography screenings outside the suggested guidelines in the U.S., [24,25], Switzerland, [26,27], Germany, [28], and The Netherlands, [29].
Given the above-mentioned considerations, women under the age threshold for systematic breast cancer screening may consider the recommendation to avoid screening as counterintuitive, although scientifically supported, because of social pressure and the belief that cancer screening can save lives. The present research aimed to promote adherence to evidence-based recommendations on breast cancer screening among young women by activating a motivation system, such as regulatory orientation, [35].

Theory of Regulatory Fit
According to a popular psychological theory proposed by [36], people show one of two regulatory orientations, which determines how they pursue their goals. They either show a promotion-focused orientation, meaning they eagerly strive towards the realization of desired outcomes, or they show a prevention-focused orientation, emphasizing the prevention of errors and losses and making them safety-driven, [36,37]. While every individual has a natural tendency to lean more towards one orientation than the other, thus making it a measurable trait,[38], the regulatory orientation can also be experimentally induced, [35,36,39].
This would challenge the intuitive perception that breast cancer screening leads to a mortality reduction determined by breast cancer, [30,31] and the unrealistic expectations regarding screening as reducing the risk of breast cancer, [32]. The purpose of the present research was to test whether health messages framed to correspond with a woman's regulatory orientation are effective in reducing the intention to ask for breast cancer screening in non high-risk women under the age of  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   8 45, according to the local mammography screening guidelines. The following hypotheses have been tested: HP1: a fit between the message frame and the regulatory orientation would lead to an immediate reduction of the intention to ask for breast cancer screening, in non-high-risk women under the age threshold indicated by the local guidelines.
HP2: a fit between the message frame and the regulatory orientation would lead to a reduction of the intention to ask for breast cancer screening, stable over time.
To this end, a study has been developed organized in two distinct phases: Phase 1 was an experimental study testing HP1, while Phase 2 added a longitudinal component and tested HP2.

Phase 1
A pre-post-test design with two experimental conditions and a control group was applied (see Figure 1 for full details).
[Insert Figure 1 here] At pre-test, the survey included measures of health status and health behaviours, a set of questions on past diagnosis of breast cancer, mammography,

Phase 2
A pre-post-test longitudinal design was applied with four experimental conditions, two fit conditions (promotion and prevention), two non-fit conditions (promotion and prevention) and a control group (see Figure 2).
[Insert Figure 2 approx. here] In the pre-test (T0), participants replied to the same questions as for Phase 1 (see online supplemental material). In Phase 2, the regulatory focus orientation was measured with a questionnaire (online supplemental material), rather than induced as in Phase 1, because working with the trait regulatory focus would be more stable than a primed focus in a longitudinal design. Women were then identified according to their goal-pursuit main orientation (prevention orientation vs. promotion orientation).
Subsequently, participants were randomly assigned to the fit or non-fit condition or control group. In other words, randomisation was performed separately for preventionoriented women and promotion-oriented women to ensure a balanced representation of

Phase 1 and Phase 2
Results from previous studies involving participants from Switzerland informed the present research (see, [27]). Participants were not directly involved in the design, conduct, recruitment, reporting or dissemination of the study results. An expert panel, composed of two health communication professionals with expertise on regulatory fit theory, evaluated the message contents and the graphical aspects of the videos.

Phase 1 and Phase 2
In both Phase 1 and Phase 2, data were normalized through reverse scoring and logarithmic transformations and there were no missing data. In Phase 2, a repeated measure ANCOVA tested the main hypothesis (HP2) of the study. The fit vs. non-fit vs. control conditions variable was inserted as independent variable. All the variables measured at the pre-test were inserted as covariates.

Phase 1
The ANCOVA analysis revealed that women in the two experimental conditions showed less intention to ask for breast cancer screening compared to the women in the control condition. Thus, when there is a fit between individual orientation (i.e., a tendency to promote positive expected outcomes or to prevent negative outcomes for one's health) and the given message, then a persuasive effect is induced. There was no meaningful difference between the two manipulation conditions. Older women and women with higher levels of fear of breast cancer showed a greater intention to ask for breast cancer screening than younger ones and those with lower levels of fear. This evidence supports the assumption that regulatory orientation represents a motivational system able to overcome the impact of negative emotions and strengthen an individual's involvement in decision-making orientation.
Descriptive data and results from the ANCOVA are displayed in Table 2.  Within subject comparison between pre-and post-intention: F b (1, 267.91) = 5.10, p = .025, partial η 2 = .02 Between subject comparisons among groups: F b (4, 284) = .43, p > .05 Significant covariates: Fear of breast cancer: t(284) = 2.76, p = .006, B = .24, partial η 2 = .03 (95% Low CI = .07, 95% High CI = .42) Age, t(284) = 6.26, p < .0001, B = .11, partial η 2 = .12 (95% Low CI = .08, 95% High CI = .15), Risk perception, t(284) = 2.26, p = .024, B = .37, partial η 2 = .02 (95% Low CI = .05, 95% High CI = .70), Further analyses were conducted to evaluate whether the covariates might interact with the three experimental conditions in determining the intention to ask for breast cancer screening. Analyses revealed only one association among the three groups of women and the past diagnoses of breast cancer among first degree-relatives, χ 2 (2) = 12.98, p = .002. Women in the promotion fit condition had a lower number of breast cancer diagnoses among first-degree relatives than was expected (z = -1.96), while women in the control condition had a higher number than expected (z = 2.8). The subsequent ANCOVA did not find any significant interaction between past diagnosis of breast cancer among first-degree relatives and the experimental manipulations, therefore demonstrating that the regulatory fit genuinely influences the intention.

Phase 2
There was a general significant decrease of the intention from pre-to post-evaluation across groups, but no significant differences among them, indicating that the scores of the post-test intention among the five groups were in general the same. Among the covariates older women, greater fear of breast cancer and greater risk perception were associated with greater post-test intention compared to the opposite. Table 2 shows descriptive statistics and results from the analysis.
The intervention effect was not significant either when the two fit conditions and the two non-fit conditions were collapsed into two categories (i.e., comparison among fit condition vs. un-fit condition vs. control) as done in Phase 1, even though a general decrease in the post-intention across groups was found as before. Risk perception was tested as a moderator, but the analysis was not significant. The present research shows inconsistent results. Phase 1 confirmed the hypothesized effect of the intervention on the intention to seek mammography screening before the age of 45, with a reduction of the intention when a fit between the message frame and the individual's regulatory focus occurred. Longitudinal results from Phase 2 demonstrated that this effect was not significant over one month, although a general decrease of the intention across groups was observed. Even though further evidence is needed to confirm our results, it still seems that the 'just-feels-right' experience appears to be insufficient to convince non high-risk women under the age threshold to avoid systematic breast cancer screening in the long run.

General discussion
Our results could genuinely reflect the fact that the regulatory fit is not sufficient to induce a long-term decrease in women's intentions or could be an artefact of the research itself. Phase 1 and Phase 2 applied two different ways to evoke a regulatory orientation. Phase 1 primed the individuals' regulatory orientation, whereas Phase 2 measured it with a questionnaire to overcome a limitation of Phase 1 and explore a different aspect of the theory. One could argue that the different ways to induce vs. measure the regulatory orientation could have influenced the persuasiveness of the message and so its effectiveness. However, researchers of regulatory orientation suggest that there is no difference between the two procedures, [39]. Therefore, we could exclude that the two methods have had a differential impact on post-test intention. Possible differences in the cultural milieu of Italian-speaking Swiss and Italian participants might make the population primed to receive or primed to ignore the intervention. However, to the best of our knowledge there is no study comparing different cultural environments in the propensity to be primed or not.
The relatively small sample size and the recruitment strategies could have influenced the power of the analyses, the sample composition and, ultimately, the significance of the results. However, there is no such concern in Phase 2 since the  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59 [27,[49][50][51]. Moreover, the benefits of mammography screening often seem to be overestimated, [30,31]. Therefore, it is challenging to develop effective health messages promoting the adherence to breast cancer screening guidelines for young women based on factual information. As messages based on the principles of regulatory fit take the motivational orientations of recipients into account, they go beyond the effectiveness of purely providing information. Here, messages building on the theory of regulatory fit did not seem to offer a new way to overcome the 'emotional barrier' generated by the fear of breast cancer. However, Phase 2 demonstrated a general 'pedagogical effect' deriving from talking about the topic of breast cancer screening without evoking a boomerang effect (i.e. an increase of intention instead of a decrease).
The present research has several limitations. We experienced high dropout rates, especially in Phase 2. The high dropout rates may be related to the topic of breast  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y cancer itself or the fear associated with it. One could assume that women with a low level of fear of breast cancer may have decided not to take part in our research, and this may have created a selection bias that could affect the generalizability of the results. A second limitation concerns the fact that we measured the intention to ask for breast cancer screening, not the actual behaviour. Although according to many theories in the field of health promotion (e.g. Health Belief Model), the intention is a valid predictor of the actual behaviour, it would be beneficial if future research followed women until the moment they actually have a mammography.

Breast Cancer/Mammography Experience and Knowledge of the Ticino screening program.
Participants replied to a set of questions on: past diagnosis of breast cancer among first-grade relatives, [2], if they had a mammography in the past, if doctor recommended the mammography, if they had a breast biopsy, if they know the breast cancer screening program in Ticino, and its age thresholds. Ego-involvement. The Personal Involvement Inventory, [5] were administered measuring participants' involvement in breast cancer screening through affective and cognitive adjectives because previous research, [4]. The scale was administered as a 7-digit semantic differential (e.g., important/unimportant, relevant/irrelevant or worthless/valuable). The original item 'of concern to me/of not concern to me' was deleted based on results of a previous study, [4]. Data from the present sample show that internal consistency was good, Cronbach's α = .91, rs > .71, as well as and the factor structure, χ 2 (5) = 11.34, p = .04, CFI = .99, RMSEA = .06.

Experimental manipulation
Regulatory Focus Priming Procedure. Prevention induced participants were asked to list two of their current obligations and then write down five actions they could take to avoid failure in fulfilling them, [7]. Promotion induced participants were asked to list two aspirations and write down five actions they could take to ensure their accomplishment, [7].
Video Messages. Participants in the promotion fit condition watched a video message emphasizing promotion concerns (i.e., they should adhere to evidence-based recommendations on mammography screening for safety and health protection reasons). Participants in the prevention fit condition watched a video emphasizing prevention concerns (i.e., they should not abstain from following the evidence-based recommendations on mammography screening to avoid negative/side effects). Participants in the control group did not receive any priming and read a general health leaflet. See Supplemental Table 1 for details of the voice-text of the two video messages and the control leaflet.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46 F o r p e e r r e v i e w o n l y Note: grey rounded rectangles show the common parts of promotion and prevention video-messages; orange rounded rectangles show the promotion video-message specific parts (text in bold); blue rounded rectangles show the prevention video-message specific parts (text in bold); the green rounded rectangle shows the content of the control leaflet. The Videos created for Study 1 can be retrieved from https://youtu.be/mperSG5_9yQ and https://youtu.be/KnhRUnDoSV0. Both videos last 3:28 minutes. The videos created for Study 2 can be retrieved from https://youtu.be/btM3HrvYDlQ, https://youtu.be/BZPjFPUQuvw, https://youtu.be/-lXzGpcmzD4, https://youtu.be/jRi8Y-sZvSc. A translation of the Italian voice-over has been provided in this

Post-test Measures
Intention to ask for breast cancer screening. Intention was measured by the question "I am evaluating the idea to have a mammography screening for breast cancer in the next 2-3 years", [8].
Two further questions were added: "I have the intention to have a mammography screening for breast cancer in the next 2-3 years" and "I will take an appointment for a mammography screening for breast cancer in the next 2-3 years". Participants replied on a 5-point scale from 1 (definitely yes) to 5 (definitely not); participants' scores ranged 1-5, M = 2.61, and S.D. = 1.14, with higher scores indicating greater intention. Data from the present sample show that internal consistency was good, Cronbach's α = .97, rs > .94.

Pre-test Measures
Pre-test covariates were measured as for study 1. Intention to ask for breast cancer screening was asked during the pre-test with the three items applied in Study 1.
Trait Regulatory Orientation. The Regulatory Focus Questionnaire, [9] was applied in the pre-test phase. The questions asked how frequently several specific events occur in the participant's life. Six questions capture the promotion focus, and the other five the prevention focus. Participants replied on a 5-point scale from 1 (never) to 5 (very often). The scores for promotion and prevention scales were calculated averaging the answers on given items after reverse score: data show good internal consistency for both promotion, α = .66, rs > .33, and prevention, α = .74, rs > .47. The individual's chronic orientation was calculated by subtracting promotion score to prevention score, [9].

Experimental manipulation
Video Messages. Six video-messages were developed for the present study:  Two video-messages emphasising prevention concerns;  Two video-messages emphasising promotion concerns;  Two video-messages without any prompt to regulatory orientation.
Supplemental Table 2 shows the content of the voice-text of the six video-messages.