BMC Public Health

Dating SOS: a systematic and theory-based development of a web-based tailored

intervention to prevent dating violence among Brazilian youth.
--Manuscript Draft--

Manuscript Number: PUBH-D-19-05397R2

Full Title: Dating SOS: a systematic and theory-based development of a web-based tailored
intervention to prevent dating violence among Brazilian youth.

Article Type: Study protocol

Section/Category: Mental health

Funding Information: Coordenação de Aperfeiçoamento de Dr. Sheila Giardini Murta

Pessoal de Nível Superior
(88887.310840/2018-00)

Abstract: Background

Dating violence has an alarming prevalence among Brazilian adolescents. School-

based preventive programs have been implemented, but remain isolated initiatives with
low reach. Health communication strategies based on innovative technologies with a
high potential of diffusion are urgent. This study aimed to develop a computer-tailored
intervention to prevent victimization and perpetration of dating violence among
Brazilian youth.

Methods

The intervention, called Dating SOS (SOS Namoro), is based on the I-Change Model
and Attachment Theory and is a comprehensive preventive program targeted to young
people with a current partner. The intervention design included the stages of needs
assessment; definition of objectives of change; development of the library of
messages; elaboration of a questionnaire for tailoring feedbacks according to the
relevant variables; integration of the content in the software Tailor Builder; pre-testing;
and usability and efficacy evaluation planning.

Results

Dating SOS is composed of four online sessions. The first session gives a tailored
orientation on attachment style and risk perception of violence. The second session
addresses knowledge on conflict management, positive and negative social models of
intimate relationships and an action plan to improve everyday interactions. The third
session covers social norms, self-efficacy and an action plan to cope with conflicts. The
fourth session discusses attitudes, social support and an action plan to protect from
violence. Improvements on the interface and tailoring refinement was done after pre-
testing to improve attractiveness and decrease risk of iatrogenic effects.

Discussion

The principal merit of the present study resides in the development of an innovative
strategy based on the qualified use of the internet for education surrounding romantic
relationships and the prevention of dating violence among adolescent and young
Brazilians, a hitherto unaddressed need in the field. The intervention usability and
efficacy should be investigated in further studies.

Corresponding Author: Sheila Giardini Murta

Universidade de Brasilia Instituto de Psicologia
BRAZIL

Corresponding Author E-Mail: giardini@unb.br

Corresponding Author Secondary

Information:

Corresponding Author's Institution: Universidade de Brasilia Instituto de Psicologia

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
Corresponding Author's Secondary
Institution:

First Author: Sheila Giardini Murta

First Author Secondary Information:

Order of Authors: Sheila Giardini Murta

Priscila Oliveira Parada

Sara da Silva Menezes

João Victor Venâncio Medeiros

Amanda Balbino

Marina Caricatti Rodrigues

Marco Akira Miura

Thiago André Araújo dos Santos

Hein de Vries

Order of Authors Secondary Information:

Response to Reviewers: 1. Abstract structure

Please ensure that the abstract contains the following headings only: Background,
Methods, Discussion and Trial Registration.

Authors: it was done (please see page 2)

2. Trial registration
Thank you for registering the trial to the Brazilian Clinical Trials Registry (ReBec).
Alongside the URL, please include the following information:
• Date of registration
• Retrospectively or prospectively registered (if the trial was registered after the first
participant was enrolled, please state that the trial was retrospectively registered.
• Full name of clinical trials registry
• Full accession code

Authors: it was done (please see page 3)

3. Conclusion
As per the submission guidelines for Study Protocols (please see the following link for
more information: https://bmcpublichealth.biomedcentral.com/submission-
guidelines/preparing-your-manuscript/study-protocol), please remove the Conclusion
heading from the main body of text and incorporate this section into the Discussion.

Authors: it was done (please see last paragraph of Discussion, page 26).

4. Funding
In the Funding statement of the Declarations, please describe the role of the funding
bodies in the design of the study, the collection, analysis, and interpretation of data,
and in writing the manuscript.

Authors: it was done (please see last paragraph of page 26).

5. Figure 1 permissions
Please clarify whether Figure 1 was adapted from another source. If so, please
acknowledge the source, and if it is under copyright also state the written permission
given to use and adapt it in the figure legend.

Authors: the source was acknowledged (please see page 35, bellow the Figure 1). The
author is Hein de Vries, the last author of this article. He stated that the Figure is not
under copyright.

Additional Information:

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
Question Response

Does your study have ethical approval? Yes, and I have included the relevant documentation as an additional file

Has your study received funding? Yes, the funding is external and not industry funded, and I have included the relevant
documentation as an additional file

Has your study undergone full external No

peer review as part of your funding
application?
 as follow-up to "Has your study
received funding?"

What stage of participant recruitment is Participant recruitment has not started

your study at?

Has this manuscript been submitted No

before to this journal or another journal in
the <a
href="https://www.biomedcentral.com/p/th
e-bmc-series-journals#journallist"
target="_blank" >BMC series</ a>?

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
Manuscript Click here to access/download;Manuscript;Dating SOS-
BMCPH-27.2.2020-final.docx
Click here to view linked References

TITLE PAGE
1
2
3 Dating SOS: a systematic and theory-based development of a web-based
4 tailored intervention to prevent dating violence among Brazilian youth.
5
6
7
8 Sheila Giardini Murta1
9 Priscila Oliveira Parada2
10 Sara da Silva Meneses3
11
João Victor Venâncio Medeiros4
12
13 Amanda Balbino5
14 Marina Caricatti Rodrigues6
15 Departament of Clinical Psychology
16 Institute of Psychology
17
18
University of Brasília
19
20 Marco Akira Miura7
21 Department of Design
22 Institute of Arts
23
24 University of Brasília
25
26 Thiago André Araújo dos Santos8
27 Department of Computer Sciences
28
29
Institute of Exact Sciences
30 University of Brasília
31
32 Hein de Vries9
33 Department of Health Promotion
34
35 CAPHRI School for Public Health and Primary Care
36 Maastricht University
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54 Corresponding author:
55
56 Sheila Giardini Murta, Departamento de Psicologia Clínica, Instituto de Psicologia,
57 Universidade de Brasília, Campus Darcy Ribeiro, Brasília, DF, Brazil, CEP 70910-900. E-
58
59
mail: giardini@unb.br. Phone: 55 (61) 99633 9631.
60
61
62 1
63
64
65
 Abstract
1
2
3 Background: Dating violence has an alarming prevalence among Brazilian adolescents.
4
5
6 School-based preventive programs have been implemented, but remain isolated initiatives with
7
8 low reach. Health communication strategies based on innovative technologies with a high
9
10
11 potential of diffusion are urgent. This study aimed to develop a computer-tailored intervention
12
13 to prevent victimization and perpetration of dating violence among Brazilian youth.
14
15
16
17
Methods: The intervention, called Dating SOS (SOS Namoro), is based on the I-Change Model
18
19 and Attachment Theory and is a comprehensive preventive program targeted to young people
20
21 with a current partner. The intervention design included the stages of needs assessment;
22
23
24 definition of objectives of change; development of the library of messages; elaboration of a
25
26 questionnaire for tailoring feedbacks according to the relevant variables; integration of the
27
28
29 content in the software Tailor Builder; pre-testing; and usability and efficacy evaluation
30
31 planning. Dating SOS is composed of four online sessions. The first session gives a tailored
32
33
34
orientation on attachment style and risk perception of violence. The second session addresses
35
36 knowledge on conflict management, positive and negative social models of intimate
37
38 relationships and an action plan to improve everyday interactions. The third session covers
39
40
41 social norms, self-efficacy and an action plan to cope with conflicts. The fourth session
42
43 discusses attitudes, social support and an action plan to protect from violence. Improvements
44
45
46 on the interface and tailoring refinement was done after pre-testing to improve attractiveness
47
48 and decrease risk of iatrogenic effects.
49
50
51
52
Discussion: The principal merit of the present study resides in the development of an innovative
53
54 strategy based on the qualified use of the internet for education surrounding romantic
55
56 relationships and the prevention of dating violence among adolescent and young Brazilians, a
57
58
59
60
61
62 2
63
64
65
 hitherto unaddressed need in the field. The intervention usability and efficacy should be
1
2 investigated in further studies.
3
4
5
6 Trial registration: Brazilian Registry of Clinical Trials. RBR-9frj8q. Prospectively
7
8 registered on July 25, 2019. http://www.ensaiosclinicos.gov.br/rg/RBR-9frj8q/
9
10
11
12
13 Keywords: prevention, domestic violence, dating violence, health education, adolescent
14
15 health
16
17
18
19
20 Background
21
22
23
Violence received from and practiced against a romantic partner is widespread among
24
25 Brazilian adolescents. Evidence shows that eight of every ten adolescents (age 15 to 19) have
26
27 already been victims of or have practiced some type of aggression against an intimate partner
28
29
30 at some point in their lives, particularly of a sexual or emotional nature [1]. The predominant
31
32 constructs of femininity and masculinity contribute to the normalization of dating violence and
33
34
35 make its recognition difficult among Brazilian adolescents, as violent manifestations are taken
36
37 as expressions of love or playfulness inherent to intimacy [2, 3]. This violence invisibility
38
39
40
impedes help-seeking and sustains victimization and perpetration of violence by intimate
41
42 partners in adolescence. Findings of a national study show that adolescents rarely seek help
43
44 when they are victims of their intimate partner, but when they do so, they prioritize friends and
45
46
47 family as the sources while professional help is sought less often [4].
48
49 Effective prevention and early intervention services aimed at this type of violence in the
50
51
52 adolescent public [5] as well as studies with these purposes are lacking. Results of an evaluation
53
54 of the implementation quality a law aimed at the eradication, prevention, and punishment of
55
56
57 domestic violence against women (the Maria da Penha Law), reveal that educational actions
58
59 focused on prevention of violence in adolescents, boys or girls, are rare [6]. The paucity of
60
61
62 3
63
64
65
 preventive services identified in public policies is also reflected in research. As opposed to other
1
2 countries in the Americas, which produce frequent research on the prevention of dating violence
3
4
5 among adolescents, there is a scarcity of evidence-based prevention interventions in Brazil [7].
6
7 Recent Brazilian interventions against partner violence are offered face-to-face in the
8
9
10 school environment and conducted in a workshop format by previously trained facilitators [8-
11
12 12]. Evidences has pointed reduction in sexist and homophobic beliefs [8] and dating violence
13
14
15
perpetration [10], increasing in emotion regulation skills [9], and intervention’s acceptability
16
17 [11, 12]. Even with their promising results, such interventions require institutional support,
18
19 space in the school’s curriculum schedule, and qualified personnel for their execution in new
20
21
22 contexts, all of which hinder their diffusion at a national level resulting in limited public health
23
24 impact.
25
26
27 An alternative mode of intervention that that may be easier to disseminate and adopted
28
29 can be computer-tailored intervention (CTI). A CTI consists of the adaptation of health
30
31
32
education material to one specific person after an individualized evaluation by an algorithmic
33
34 process to generate customized feedback and guidance for the user [13]. As in traditional
35
36 clinical counseling, individual evaluations are followed by feedback and guidance tailored to
37
38
39 the participants’ beliefs, behaviors, and needs. However, in contrast to the traditional clinical
40
41 approach, such feedback and tailored guidance is offered through printed letters, email, mobile
42
43
44 phones, and other mobile devices instead of the face-to-face mediation of a professional.
45
46 CTI differs from general internet intervention modes (eHealth) that do not use tailored
47
48
49 guidance derived from individual evaluations (commonly called computer-based interventions
50
51 or internet-based interventions) [14]. Several studies of communication in health care reveal
52
53
54
consistent evidence that tailored messages, when compared to general messages, tend to be read
55
56 more often and are remembered and considered more relevant and appealing [15, 16].
57
58 Therefore, tailored interventions are more effective and efficient as a health education strategy
59
60
61
62 4
63
64
65
 than other means of general mass communication. Additionally, CTI seems to be highly
1
2 appropriate for adolescents who are more accustomed to technology than older generations.
3
4
5 While the application of tailored internet-mediated interventions to promote positive
6
7 health habits is broad [17, 18], their application in the prevention of dating violence is still
8
9
10 limited and recent. The first study focusing on this in the literature was conducted by Levesque,
11
12 Johnson and Prochaska [19], who developed an intervention called Teen Choices, based on the
13
14
15
transtheoretical model, which sought to prevent dating violence and foster healthy relationships
16
17 in adolescence. The intervention, composed of a single session, contains different guidance for
18
19 participants not dating, dating with a low risk of violence, and dating with a high risk of
20
21
22 violence. The feedback is tailored according to readiness to engage in healthy relationships and
23
24 in the seeking and offering of help. The results showed that the Teen Choices participants,
25
26
27 compared to the control group, presented a significant reduction in perpetration and
28
29 victimization of physical and emotional violence at 6 and 12 months [20]. The results were
30
31
32
more expressive in those who presented a prior history of violence in intimate relationships.
33
34 In Brazil, the use of information and communication technologies for mental health
35
36 promotion is rare [21] and studies using CTI are still embryonic [22] and generally address the
37
38
39 reduction of addictive behaviors, such as smoking [23]. Given this situation, the current study
40
41 innovates by developing an intervention to prevent dating violence among Brazilian
42
43
44 adolescents. The intervention, called SOS Namoro (hereafter Dating SOS), was designed to
45
46 reduce victimization and perpetration of dating violence and increase relationship quality,
47
48
49 conflict management skills, and help seeking. It is a universal prevention strategy, targeting
50
51 youths between 15 and 29, who have an affective-sexual partner at the beginning of the
52
53
54
intervention, regardless of being in violent relationship or being exposed to dating violence risk
55
56 factors. It is based on two theoretical models: the integrated change model, also known as the
57
58 I-Change Model, and attachment theory.
59
60
61
62 5
63
64
65
 Theoretical basis
1
2
3 The integrated change model [24, 25] is a composite of several models and socio-
4
5 cognitive theories, including Azjen’s Theory of Planned Behavior [26], Bandura’s Social
6
7
8 Cognitive Theory [27], Prochaska’s Transtheoretical Model [28], Janz and Becker’s Health
9
10 Beliefs Model [29], and Goal Setting Theories [30]. According to the model (see Figure 1),
11
12
13 behavior and behavior change is a process of several phases and each phase is determined by
14
15 different types of factors.
16
17
Insert Figure 1 here
18
19
20 First of all, a person needs to become aware of the problem, which is determined by
21
22 awareness of his/her own behavior (i.e. is the behavior that I am doing appropriate, am I
23
24
25 behaving in a violent way), knowledge (i.e. what are considered to be sexually violent
26
27 behaviors?); risk perceptions (i.e. am I at risk of violating someone and what are the
28
29
30 consequences); and perceived cues (i.e. what are the situations that may prompt me to become
31
32 sexually violent). Second, when a person becomes aware of a problem and its own role in the
33
34
35
problem, he/she may become open to become motivated to change unhealthy behavior. Factors
36
37 that are important here are: attitudes towards the behavior and the perceived pros and cons of it
38
39 (e.g. when I am sexually violent I can show that I am in charge of things), the perceived social
40
41
42 support (e.g. my mates think it is okay to be sexually violent), and self-efficacy (e.g. when I
43
44 feel stressed I find it difficult to control myself and not become sexually violent) . The
45
46
47 combination of positive attitudes, supportive norms and high self-efficacy will lead to a positive
48
49 intention to perform a healthy behavior. Third, a positive intention does not automatically
50
51
52 translate into the desired goal behavior but requires action planning (e.g. what will I do when I
53
54 feel stressed), actual realization of these plans, the skills, and sometimes the removal of certain
55
56
barriers.
57
58
59
60
61
62 6
63
64
65
 Lastly, the integrated change model postulates that these factors are influenced by a wide
1
2 array of predisposing factors, such as psychological, behavioral, socio-environmental factors.
3
4
5 Additionnaly, the model postulates that awareness, motivation and action can be influenced by
6
7 developing communicative actions, of which the effects are determined by personal factors (e.g.
8
9
10 level of education), message factors (e.g. clearly written with clear instructions, tailored to a
11
12 specific person), source factors (e.g. using peers that provide information), and channel factors
13
14
15
(e.g. the choice for interpersonal, personal or eHealth channels).
16
17 Dating SOS will operate through three mechanisms outlined by the I-Change Model. The
18
19 first mechanism consists in raising awareness of dating violence as a problem by promoting
20
21
22 knowledge of the phenomenon and its impact on health. The second comprises strengthening
23
24 motivation to protect oneself against violence from an intimate partner by increasing attitudes
25
26
27 favorable to the non-violent management of conflicts and unfavorable to the acceptance of
28
29 violence, enhancing critical thinking about pro-violence social norms, aiding in the
30
31
32
identification of healthy models of loving relationships and sources of social support when
33
34 facing violence, and developing self-efficacy to deepen emotional intimacy in the relationship
35
36 and reduce the use and acceptance of violent conflict resolution strategies. The third covers
37
38
39 fostering self-protective intentions by means of developing clear action plans for facilitating
40
41 day-to-day communication with the partner, managing conflicts without violence, and
42
43
44 protecting oneself should violence arise (for example, seeking help). Applications of the I-
45
46 Change Model to study and intervene on violence, showed the importance of these determinants
47
48
49 [31], as well as the efficacy of intervening on these determinants to reduce violence successfully
50
51 [32, 33].
52
53
54
In addition to the I-Change Model, attachment theory was adopted as a secondary
55
56 theoretical basis for Dating SOS [34]. Although this theory has been extensively used to explain
57
58 how attachment may explain the process leading to intimate partner violence [35] and
59
60
61
62 7
63
64
65
 prevention to dating violence [36], its use in CTI to this end has not yet been recorded. The
1
2 choice of attachment theory to base Dating SOS upon resides in the evidence that adolescents
3
4
5 with insecure affective bonding styles tend to use more destructive conflict management
6
7 strategies and are more frequently victims of dating violence than those securely attached [37,
8
9
10 38]. Evidence from a study of a Brazilian sample with couples in various stages of romantic
11
12 relationships corroborates these findings [39].
13
14
15
Insecure attachment styles stem from rejective, inconsistent, intrusive, or abusive
16
17 parenting practices that foment the development of internal working models characterized by
18
19 distrust and insecurity in the establishment of loving bonds as well as an image of themselves
20
21
22 as undeserving of love [40, 41]. In turn, secure attachment style results from parental care
23
24 responsive to the needs of the child who, as a consequence, develops lifelong internal working
25
26
27 models predictive of a positive image of oneself and one’s partner, facilitating the building of
28
29 intimate relationships in adulthood and increasing the likelihood that they be longer-lived,
30
31
32
healthier, and more secure. In this way, romantic love in adolescence and adulthood can be
33
34 understood as an attachment process stemming from the first attachment experiences in
35
36 childhood.
37
38
39 Low levels of anxiety and avoidance in affective relationships characterize secure
40
41 attachment style, whereas insecure attachment styles present elevated levels of anxiety and/or
42
43
44 avoidance [42]. Avoidance is related to the degree of comfort individuals experience in the face
45
46 of emotional closeness and intimacy. Applied to romantic relationships, highly-avoidant
47
48
49 individuals are reluctant to establish intimate relationships and seek emotional comfort based
50
51 on the concept that intimacy is threatening and the seeking of emotional support undesirable
52
53
54
and its offer impossible. Anxiety denotes the degree to which individuals worry about being
55
56 rejected or abandoned by their romantic partners. High levels of anxiety manifest themselves
57
58 in vigilance for signs of partner distancing as well as elevated investment in maintaining
59
60
61
62 8
63
64
65
 closeness by compulsively searching for support, which can overwhelm partners [40, 43] and
1
2 translate into violent outbursts in addition to indifference, threats, insults, and possessiveness.
3
4
5 Secure attachment is a protective factor for the quality of romantic relationships [44], while
6
7 insecure attachment is one of the risk factors for dating violence, contributing to deficient
8
9
10 emotional regulation [41] and conflict resolution abilities [38, 39].
11
12 In line with attachment theory, it is hoped that Dating SOS can broaden the knowledge
13
14
15
of its participants regarding their own attachment styles and how their anxiety levels and
16
17 avoidance affect their romantic relationships. In due course, this should boost action plans for
18
19 emotion-regulating, including understanding aspects of the relationship that evoke negative
20
21
22 emotions (such as exposure to violence, its perpetration, or both) as well as awareness of one’s
23
24 own emotions and managing how they are expressed. It is expected that, upon receiving
25
26
27 guidance tailored to their own attachment style and to the levels of dating violence experienced,
28
29 the participants will recognize opportunities for implementing action plans more responsive to
30
31
32
their needs and those of their partners. This in turn should reduce the use of negative conflict
33
34 resolution strategies, such as distancing and excessive control, and augment the use of positive
35
36 ones, such as offering to help and emotional closeness – depending on the dynamics of the
37
38
39 relationship (violent or not) – thereby improving the quality of the romantic relationship and
40
41 reducing victimization and perpetration of violence against one’s partner. Finally, one
42
43
44 anticipates that the intervention can boost action plans and skills for seeking help or ending a
45
46 relationship for those participants exposed to violence as opposed to tolerating violence,
47
48
49 deflecting them from strategies intended to cope with insecure attachment and instead leading
50
51 them to strategies related to a secure attachment style.
52
53
54
This article is intended describe the development process of Dating SOS, a computer
55
56 tailored intervention grounded in theory for preventing dating violence and improving the
57
58 romantic relationship quality of Brazilian youths. It is anticipated that the study can advance
59
60
61
62 9
63
64
65
 the production of knowledge in the field in three ways. First, it adopts dating violence as the
1
2 focus, something little done thus far in computer tailored interventions. Second, it bases its
3
4
5 intervention design on a collection of empirical studies of dating violence among young
6
7 Brazilians, a population largely ignored in preventive actions and unstudied in this area. Third
8
9
10 it selects, in an original way, theoretical models yet untested in CTI for dating violence
11
12 prevention.
13
14
15
16
17 Method
18
19
20
21
Intervention
22
23
SOS Dating is an online intervention in which the participants receive tailored guidance
24
25
26 based on their attachment style, their experience of having been involved in a romantic
27
28 relationship with or without violence, their knowledge of dating violence, their attitudes
29
30
31 towards dating violence, their positive and negative relationship models derived from those
32
33 they spend their time and live with, the social norms regarding dating violence in their social
34
35
36 circles, their sources of social support available for coping with violence, their self-efficacy for
37
38 coping with conflicts nonviolently, and their action plans for improving the day-to-day quality
39
40
41
of their romantic relationship and for managing conflict and protecting against violence. The
42
43 tailoring also take in account the participants’ gender, sexual orientation and status of violence.
44
45 Status of violence in the romantic relationship ranges from an absence of violence through
46
47
48 episodes of violence to regular violence. In cases of regular violence or episodes of violence
49
50 the direction (victimization, perpetration, bidirectional) is also considered.
51
52
53 The intervention comprises four individual sessions, available at
54
55 http://sosnamoro.geppsvida.com.br. The content, objectives, and variables used in the
56
57
58 customization are described on Table 1. Each session lasts approximately fifteen minutes and
59
60 occurs one week after the previous one. The feedback and participant guidance customizations
61
62 10
63
64
65
 were made using the TailorBuilder software, used in CTIs already described in literature [45,
1
2 46].
3
4
5 Insert Table 1 here
6
7 In Session 1, called “My Relationship,” data from the attachment style and dating
8
9
10 violence scales are used to generate tailored feedback. Based on the answers given to those
11
12 scales, the participants get immediate feedback about their attachment style (secure or
13
14
15
insecure/evasive, insecure/anxious or insecure/preoccupied), the degree of violence in the
16
17 relationship (no violence, episodes of violence, or high levels of violence), and the direction of
18
19 violence (victimization, perpetration, or mutual).
20
21
22 In Sessions 2 to 4, the participants initially answer a questionnaire focused on the dating
23
24 violence determinants from the I-Change Model, including knowledge, social modeling for
25
26
27 violent or positive intimate relations, action plans to improve the day-to-day relationship quality
28
29 (Session 2 – “Improving Coexistence”), social norms, self-efficacy, action plans for conflict
30
31
32
management (Session 3 – “Dealing with conflicts”), attitudes, social support, and action plans
33
34 for self-protection in the face of violence (Session 4 – “Protecting Yourself Against Violence”).
35
36 After this, they receive tailored feedback, as in Session 1.
37
38
39 In each session, opening and concluding narratives were used to help the participants
40
41 identify with the intervention. The feedback is offered via text passages, available as narratives
42
43
44 or lists of items, and self-monitoring exercises. They are posted to a website in a password-
45
46 protected area and sent via e-mail as a PDF attachment as well. Email reminders are sent to
47
48
49 participants one week after every session, if they have not completed the previous session or
50
51 started the next one. All interactions between the participant and SOS Dating happen online,
52
53
54
without the mediation of a human facilitator.
55
56
57
58
59
60
61
62 11
63
64
65
 Procedures
1
2
3 The development of SOS dating comprised eight steps: needs evaluation, definition of
4
5 the intervention’s objective, construction of the customization questionnaire, elaboration of the
6
7
8 message library, entering the content into the customization software, intervention pre-test, and
9
10 evaluation planning [47].
11
12
13 Step 1: needs evaluation
14
15 The needs evaluation consisted of three empirical studies with youths, both victims and
16
17
not of dating violence, to identify relevant objectives for the intervention and develop its content.
18
19
20 The first study had the purpose to examine the psychosocial determinants of dating
21
22 violence and was conducted with seven focus groups of adolescents (n = 12) coming from
23
24
25 public and private schools and young students (n = 16) from private and public universities.
26
27 The focus groups explored the psychosocial determinants of dating violence, as recommended
28
29
30 by the I-Change Model, and were conducted separately according to gender and
31
32 school/university. The data, examined by deductive thematic analysis, was categorized as
33
34
35
knowledge, social norms, social support, models, attitudes, self-efficacy, action plans, and
36
37 suggestions for preventive initiatives against dating violence. The results pointed to gaps in the
38
39 knowledge of the psychological and material manifestations of dating violence, with physical
40
41
42 and sexual being the most acknowledged forms. Similarly, ignorance of non-violent conflict
43
44 resolution among students was apparent, as well as what positive and healthy aspects of
45
46
47 romantic relationships would entail. Dating violence was reported as acceptable in several
48
49 situations, such as in the case of betrayal, when attacked first, and if one or both partners are
50
51
52 intoxicated. Pro-violence social norms attached to sexist stereotypes were commonly reported.
53
54 The participants reported financial dependence and having kids as conditions that can promote
55
56
submission to violence and inhibit social support for leaving the abusive relationship. Finally,
57
58
59 they reported action plans with potentially negative impacts, such as resigning themselves to
60
61
62 12
63
64
65
 violence, acting indifferently to the partner, and responding to violence with violence, but there
1
2 were some positive plans as well, such as terminating the relationship or looking for
3
4
5 professional help. The participants suggested that preventive interventions might develop
6
7 critical thinking about gender stereotypes and foster respect for the partner’s individuality.
8
9
10 The second study investigated the characteristics, triggers and consequences of violent
11
12 conflicts in dating relationships. It was conducted via individual interviews with ten participants,
13
14
15
six men and four women, aged 17 to 27, who had experienced dating violence, either as victims
16
17 or as perpetrator-victims (i.e. bidirectional violence). The results were analyzed using deductive
18
19 thematic analysis. The participants described their relationships as a mix of positive and
20
21
22 negative experiences, with strong affective involvement but with the use of ineffective
23
24 strategies of conflict management. Among those, resignation to the partner’s impositions was
25
26
27 reported; threats, possessiveness, and casting blame to control the other’s behavior; poor
28
29 emotional control (cursing and bouts of rage) accompanied by psychological, physical, and
30
31
32
mental aggressions; attempts at reconciliation and repairing; and the withdrawal into silence
33
34 and indifference toward the partner. It was reported that all these strategies resulted in new
35
36 conflicts and subsequent future violence, causing great suffering to the parties involved.
37
38
39 Furthermore, data indicated the hyperactivation of the system of attachment in the face of
40
41 signals interpreted as abandonment by the partner, for example silence and distancing
42
43
44 accompanied by anxious responses, such as emotional outbursts to make the other listen.
45
46 The third study, aimed to examine emotional intimacy and safety in the romantic
47
48
49 relationship, keeping attachment theory as the theoretical lens. It used semi-structured
50
51 individual interviews with youths between sixteen and twenty-three years of age who had dating
52
53
54
experience. The results revealed variability among the interviewees regarding the intimacy
55
56 experience, with higher levels of intimacy being connected to the exchange of social support
57
58 between partners and sensitivity to each other’s needs, and threatening levels of intimacy
59
60
61
62 13
63
64
65
 connected to evasiveness when seeking support from the partner and negative conflict
1
2 resolution strategies. The data aligned with the recommendations of attachment theory, which
3
4
5 are that partner’s availability and accessibility are the central path for promoting emotional
6
7 closeness in romantic relationships.
8
9
10 In short, the needs evaluation demonstrated the urgency of the intervention, taking into
11
12 account its main findings: violence is seen as natural in romantic relationships; gender
13
14
15
stereotypes justifies attitudes of acceptance of violence; negative social modeling for romantic
16
17 relationships are largely present in closer relationships; psychological and material violence are
18
19 barely recognized as violence; betrayal is the most difficult situation to not using violence
20
21
22 against the partner; there are negative strategies to cope with conflicts, including submission,
23
24 emotional dysregulation and withdrawal; and violent conflicts are related to non responsive and
25
26
27 hostile partners’ answers.
28
29
30
31
32 Step 2: defining the target group and intervention objectives
33
34 The data collected from all three needs evaluation studies underpinned the elaboration of
35
36 the intervention’s objectives (Table 1) and target group. The target group selected for the
37
38
39 intervention was youths (15 - 29 years old) who had a romantic partner. The results of needs
40
41 assessment suggested that the intervention should (a) prescribe objectives to address
42
43
44 psychosocial determinants of dating violence which were also in line with socio-cognitive
45
46 ecological models such as the I-Change Model, (b) address not only violence but relationship
47
48
49 quality too, focusing on the partners’ mechanisms of availability and accessibility to strengthen
50
51 emotional closeness, as informed by attachment theory, (c) discuss gender stereotypes and how
52
53
54
they are connected to attitudes and social norms that sustain violence, (d) consider attachment
55
56 styles and their impacts on emotional regulation as a barrier to self-efficacy for coping with
57
58 conflict and breaking the connection with the partner, and (e) consider contextual diversity in
59
60
61
62 14
63
64
65
 action plans, ranging from daily interactions, to conflict situations as well as currently being in
1
2 a violent situation.
3
4
5 Step 3: construction of questionnaire for customization
6
7 In order to address the various objectives to be addressed in a computer tailored
8
9
10 intervention, questionnaire based on the I-Change Model was developed, titled the Dating
11
12 Violence Determinant Questionnaire. The instrument considers several determinants predicted
13
14
15
in the integrated change model: awareness (eight items, for instance, “Conflict can occur in a
16
17 relationship without there being violence”); attitudes (six items, for example, “When you
18
19 depend on the partner’s money to live, it’s better to endure the violence”); social norms (nine
20
21
22 items, for instance, “Someone who cheats deserves to be attacked”); models (two items, for
23
24 example, “Of the people you know, who has a romantic relationship (dating or marriage) that
25
26
27 is positive, with companionship, respect, and affection? You can choose more than one option:
28
29 parents, aunts and uncles, grandparents, friends, neighbors, nobody”); social support (one item,
30
31
32
“Who would you ask for help to get out of a violent relationship? You can choose more than
33
34 one option: father, mother, siblings, friends, church group, police, nobody”); self-efficacy (eight
35
36 items, for example, “How hard would it be for you to ‘keep cool’ and not be violent towards
37
38
39 your partner if he or she had lied to you”); and action plans, subdivided into action plans to help
40
41 in daily life with the partner (six items, for instance, “If I was down, I would tell him/her my
42
43
44 problems and ask for help”), dealing with conflicts (14 items, for instance, “I’d get my revenge,
45
46 I’d get back at him/her”), and protecting against violence (seven items, for example, “I would
47
48
49 end the relationship”).
50
51 We chose a closed question format with multiple choice answers for the following
52
53
54
determinants: awareness, social support and action plans. A Likert scale was used to evaluate
55
56 the determinants: attitude (4 points, from disagree completely to agree completely), social
57
58 norms (5 points, from almost everyone to almost no one), and self-efficacy (4 points, from very
59
60
61
62 15
63
64
65
 easy to very difficult). The instrument was submitted to semantic validation; confusing terms
1
2 or those with double meanings were corrected.
3
4
5 Step 4: elaboration of the message library.
6
7 Once the intervention objectives had been established and the Dating Violence
8
9
10 Determinant Questionnaire had been created, the elaboration of a repository of advice messages
11
12 was started, resulting in a message library. The message library was developed according to
13
14
15
recommendations of Kreuter et al. [47]. First, tables were built for each determinant predicted
16
17 in the integrated change model (awareness, attitudes, social support, social norms, models, self-
18
19 efficacy, and positive and negative action plans) with the message title (for identification
20
21
22 purposes), objective (changes targeted by that message), origin (which items in the
23
24 questionnaire supplied the data that is the basis of that message), format (didactic resource for
25
26
27 message expression, such as narration, vignette, comparative table, video, etc.), theory and base
28
29 evidence (studies that support the message’s content), main problem (principal presumed life
30
31
32
context of user), tone (emotional aspect of the message, i.e., empathetic or informative),
33
34 concepts (message content), and message (the guidance itself).
35
36 After this systematic planning, the messages were elaborated by psychology students,
37
38
39 who were told to write freely, as if they were giving advice to a friend. This strategy was
40
41 adopted to maximize the accessibility and friendly tone of the messages. Next, the messages
42
43
44 were reviewed by researchers with expertise in the research subject. At this stage, it was verified
45
46 that the elaborated guidance covered the objectives and concepts expected of it.
47
48
49 This process resulted in 67 messages. In session 1, four variations were created based on
50
51 attachment style and eight based on the status of violence in the current relationship. In session
52
53
54
2, eleven messages were elaborated regarding awareness of dating violence, six about violent
55
56 relationship models, six about non-violent relationship models, and lastly four action plans were
57
58 created to cover daily life with the partner. In session 3, six message variations were developed
59
60
61
62 16
63
64
65
 about social norms, three on self-efficacy, and four of action plans for conflict management.
1
2 Finally, in session 4, six messages were composed concerning attitudes, six about social support,
3
4
5 and three for action plans to deal with dating violence. Table 1 shows the objectives of each
6
7 session and the customization variables by session.
8
9
10 The messages consist of a combination of text segments, with the presence and position
11
12 of each in the presented message body determined by the participant’s answers. The answers to
13
14
15
the items are used as input, generating a final classification of attachment style. This
16
17 classification becomes the variable responsible for determining which segments are used in the
18
19 message subsequently presented to the participant. Moreover, it is possible to use this
20
21
22 previously calculated variable as input for the succeeding questionnaires.
23
24 Care was taken that the combination of various messages would result in coherent texts
25
26
27 (see Figure 2 for an example of one segment of text). The text segments were constant only
28
29 within a specific set of classifications, serving as introductory, connecting, or concluding
30
31
32
passages, whereas the remaining part was tailored based on the response of the participant to
33
34 the question.
35
36 The Portuguese language requires special attention to grammatical gender. To write a text
37
38
39 that seems reader-directed, it is necessary to use the correct gender for the participant and
40
41 partner. In the example “Try to be more understanding with your partner”, the construction
42
43
44 assumes it is a message directed to a man who has a relationship with a woman. Since gender
45
46 use occurs according to well-defined rules, it was possible to adapt the text for other possibilities.
47
48
49 Step 5: entering the content into the software for customization
50
51 In parallel with the previous step, the intervention’s webpage and its visual identity were
52
53
54
developed. After finishing the message library, all instruments were entered into TailorBuilder,
55
56 which then customized the guidance to the participants’ profiles. This stage involved the
57
58
59
60
61
62 17
63
64
65
 employment of a programmer and a UX (user experience) expert to conceive and implement
1
2 the interaction with the SOS Dating participants.
3
4
5 Step 6: intervention pre-testing
6
7 Next, the intervention’s pre-testing was performed. Three pre-testing sessions were held,
8
9
10 the first with three young people potential users, the second with three professionals working
11
12 at centers for victims of domestic violence, and the third with twenty university students. The
13
14
15
main findings of pre-testing were related to the format and content:
16
17  The young people potential users thought the messages were too long and that SOS
18
19 Dating’s interface was not clear;
20
21
22  The professionals noted a risk for side-effects in the guidance regarding the bidirectional
23
24 violence and conflict management for participants whose relationships were already
25
26
27 violent.
28
29  All participants reported perceiving the intervention as relevant.
30
31
32
Responding to these findings, an extensive review was performed of the intervention’s
33
34 visual aspects and content.
35
36  A new visual identity, aiming for greater interface elements clarity, was proposed and
37
38
39 content readability was improved.
40
41  The messages were shortened to reduce the cognitive demand on the participants;
42
43
44  Customization adjustments were made in conflict management and actions plans
45
46 regarding the status of violence in the relationship, with the purpose of preventing
47
48
49 iatrogenic effects and boosting participant identification with the intervention.
50
51 After these adjustments, new wave of pre-testing was done with two potential users and
52
53
54
two experts in domestic violence, which approved the aforementioned changes.
55
56 Step 7: evaluation planning.
57
58 Two evaluation modes were planned, efficacy and usability.
59
60
61
62 18
63
64
65
 Efficacy Evaluation
1
2
3 Objectives
4
5
6 The intervention was designed to reduce victimization and perpetration of physical,
7
8
9 sexual, and psychological violence in dating (primary outcomes) and to boost the quality of the
10
11 romantic relationship as well as conflict management skills and help seeking (secondary
12
13
14
outcomes).
15
16
17 Calculation of the sample size
18
19
20 The calculation of the sample size for this study uses a conservative procedure to identify
21
22 a 10% reduction in dating violence prevalence on the intervention group compared to the
23
24
25 control group. Adopting a significance level of 5%, a testing power of 80%, and a 50% dropout
26
27 rate over the course of the study, it was calculated that each condition should have 610
28
29
30
individuals.
31
32
33 Participants and design
34
35
36 An experimental design with a pre-test evaluation and three-month follow-up will be used.
37
38 Youths between 15 and 29 years of age and their current partners will be invited to participate.
39
40
41 A sample of 1220 youths, which will be randomly and evenly divided between the experimental
42
43 and control groups, is expected. Participants younger than 15 and older than 29 or not currently
44
45
46
in a dating relationship will be excluded from the intervention.
47
48 The intervention group will receive the SOS Dating intervention, while the control group
49
50 will receive general information about dating violence, life skills, and sexual and reproductive
51
52
53 rights, available in the work “Telling a night out from a trap: a Guide for the Empowerment of
54
55 Teens in Intimate Relationships” [48], available at
56
57
58 http://www.geppsvida.com.br/publicacoes/cartilhas-e-guias/.
59
60
61
62 19
63
64
65
 Randomization
1
2
3 The randomization for the experimental and control group will be performed after the
4
5 eligibility evaluation, the express agreement to participate in the intervention through the
6
7
8 consent form, and the baseline evaluation. The placement into one group or the other will be
9
10 performed automatically by the TailorBuilder software, as has been done before in other similar
11
12
13 interventions [45, 46].
14
15
16 Recruitment
17
18
19 Several strategies will be used to advertise SOS Dating and recruiting participants: (a)
20
21 posting on the social networks of social movements related to fighting violence against women,
22
23
24 LGBT rights, and youth engagement; (b) advertising on e-mail lists of research associations
25
26 and research groups interested in related topics; (c) informing the Federal District Ministry of
27
28
29 Education, which is connected to public schools frequented by the target audience; and (d)
30
31 communicating it to student health and guidance services of Brazilian public universities.
32
33
34
35
36 Instruments and procedures
37
38 Eligibility Evaluation
39
40
After clarifications regarding the research and declaration of voluntary participation via
41
42
43 the consent form, the participants will answer a social-demographic survey composed of closed
44
45 questions about age, gender, level of education, socio-economic level, religion, sexual
46
47
48 orientation, parents’ professions, parents’ educations, previous dating experience, romantic
49
50 status (whether single or dating), and how long have they been dating. The participants between
51
52
53 15 and 29 years of age who have a current romantic partner will be included in the intervention,
54
55 while those who do not meet both criteria will be excluded (Figure 2).
56
57
58
59
60 Insert Figure 2 here
61
62 20
63
64
65
 Baseline
1
2 The included participants will next respond to the baseline evaluation, which will assess
3
4
5 attachment style; conflict resolution skills; victimization and perpetration of psychological,
6
7 physical, and/or sexual violence in their partners; and relationship quality.
8
9
10
11
12 Attachment
13
14
15
Attachment will be evaluated by the Experience in Close Relationships Scale, short version
16
17 (ECR-Short). The original scale comprises twelve items and was developed by Wei et al. [49]
18
19 from the full Experience in Close Relationships Scale, which has 36 items, by Brennan, Clark
20
21
22 and Shaver [50]. In Brazil, the adaptation was done by Natividade and Shiramizu [51], and
23
24 contains 10 items that evaluate two factors: anxiety related to attachment (preoccupation with
25
26
27 maintaining the romantic relationship and the partner’s responsiveness and the need for
28
29 physical and emotional closeness, for example: “I need a lot of assurance that I am loved by my
30
31
32
partner”) and avoidance related to attachment (discomfort with emotional proximity, with
33
34 dependence on a romantic partner, and a preference for emotional detachment, for example: “In
35
36 general, I try to avoid strong affective closeness with my partner”). Low avoidance and low
37
38
39 anxiety indicate secure attachment, while high rates of avoidance and/or anxiety reveal insecure
40
41 attachment styles. The items are answered on a seven-point Likert scale. Higher scores indicate
42
43
44 higher levels of anxiety and avoidance and thus an insecure attachment style.
45
46
47 Conflict resolution skills
48
49
50 Conflict resolution skills are evaluated through the Conflict Resolution Behavior Questionnaire
51
52 (CRBQ), developed by Rubenstein and Feldman [52] and adapted to Brazil by Delatorre and
53
54
55 Wagner [53]. The instrument is composed of 22 items that evaluate the frequency with which
56
57 strategies of conflict resolution by avoidance (for example: “I stay cool and distant and don’t
58
59
60 pay attention to the other person”), attack (for instance: “I am sarcastic”), and agreement (for
61
62 21
63
64
65
 instance: “I apologize to the other person”) are employed. A five-point Likert scale is used,
1
2 ranging from 1 (never) to 5 (always). In the study by Delatorre and Wagner [53], internal
3
4
5 consistency levels of 0.68 to 0.79 were found.
6
7
8
9
10 Victimization and perpetration of dating violence
11
12 The Conflict in Adolescent Dating Relationship Inventory (CADRI) [54], Brazilian version by
13
14
15
Oliveira et al. [1] will be used to evaluate victimization and perpetration of dating violence. In
16
17 this study, a short version of CADRI will be used. This version is composed of 24 items that
18
19 evaluate perpetration and victimization various types of violence: psychological (16 items, for
20
21
22 instance: “He/she talked to me in a hostile or mean way”); physical (four items, for example:
23
24 “He/she pushed me or he shook me”); and sexual (4 items, for instance: “I touched him/her
25
26
27 sexually when he/she did not want to be”). The participants are requested to indicate the
28
29 frequency of violent events in the current relationship on a four-point scale (always = it
30
31
32
happened six or more times in my current relationship, sometimes = it happened three to five
33
34 times in my current relationship, rarely = it happened once or twice in my current relationship,
35
36 never = it has never happened in my current relationship).
37
38
39
40
41 Relationship quality
42
43
44 A short version of the instrument Romantic Relationship Quality Scale, developed by Murta et
45
46 al. [55], will be used, with 21 items on a five-point Likert scale (1 = nothing to do with us;
47
48
49 5 = everything to do with us). The scale is composed of two factors. The first is sexual-affective
50
51 closeness (alpha = 0.87), comprising 12 items (for instance “I’m affectionate with him/her”),
52
53
54
which evaluates mutual help and support, appreciation of each other’s company, self-revelation,
55
56 empathetic listening, respecting the other’s world (including his/her family), and enjoying
57
58 physical and sexual intimacy with the other person. The second factor is being in harmony
59
60
61
62 22
63
64
65
 (alpha = 0.88), comprising 9 items (for example “I trust him/her”), which evaluate trust in the
1
2 other person, admiration for the other person, the perception of what the other person likes or
3
4
5 feels, commitment to the relationship, and acceptance of the other’s individuality
6
7
8 Follow-up
9
10
11 At three months after the end of the intervention, the experimental and control group
12
13
14
participants will again respond to measurements of conflict resolution skills, victimization and
15
16 perpetration of violence, and relationship quality. Additionally, they will also fill out the
17
18 Seeking Help Questionnaire. This instrument, developed by the authors, is composed of three
19
20
21 closed questions that inquire whether the respondents sought any sort of help to deal with their
22
23 affective relationship since they first entered the research website, whom they asked for help,
24
25
26 and what the perceived efficacy of the help received was.
27
28 Additionally, the Dating SOS participants will be asked to evaluate the impact of the
29
30
31
intervention by responding to the Dating SOS Impact Questionnaire, composed of one closed
32
33 question (How would you evaluate the effects of Dating SOS in your life?) with a multiple
34
35 choice answer (e.g.: It helped me to end the violent relationship) and three open questions (If
36
37
38 Dating SOS made no difference in your life, what do you attribute this to?; If Dating SOS had
39
40 some effect in your life, what do you attribute this to?; Comment freely about what you thought
41
42
43 of Dating SOS).
44
45
46
47
48
49
Usability Evaluation
50
51 With the aim of evaluating the attractiveness of each session and identifying aspects for
52
53
54 possible improvement, a usability evaluation will be done at the conclusion of each session of
55
56 the intervention.
57
58
59
60
61
62 23
63
64
65
 A questionnaire with both open and closed questions, which evaluate the satisfaction of
1
2 the user with the guidance received and interaction with the tool will be used. Six closed
3
4
5 questions investigate satisfaction with the theme of the session, questions, guidance, navigation,
6
7 quantity of information, and errors. Three open questions examine satisfaction with the layout,
8
9
10 guidance content, and suggestions for improvement. Finally, the user is asked which aspect of
11
12 the Dating SOS was most attractive and which was the least attractive.
13
14
15 Statistical analysis
16
17
18 Descriptive statistics will be used to describe the characteristics of the participants in the
19
20
21 baseline and usability data. The principal analyses will consist of intra- and intergroup
22
23 comparisons between the baseline and the three-month follow-up on the outcomes of
24
25
26 victimization and perpetration of physical, psychological, and sexual violence; quality of the
27
28 romantic relationship; and conflict management skills. Multivariate analysis of covariance will
29
30
31
be used to examine the efficacy of the intervention on these outcomes. The attachment style
32
33 and the relationship’s violence status will be used as covariates. In addition to this, intergroup
34
35 comparisons will be conducted using the t-test on the three-month follow-up outcome of
36
37
38 seeking help. Moderation analysis will be employed to evaluate the differential effects of the
39
40 intervention based on gender, age, and sexual orientation. A significance level of 0.05 will be
41
42
43 adopted.
44
45 Omitted or incomplete data will be handled via multiple imputation using the Missing
46
47
48 Value Analysis procedure of SPSS Statistics. Intention-to-treat analyses will be carried out, and
49
50 all randomized participants will be posteriorly included regardless of having completed the
51
52
intervention or not. The effects of Dating SOS will be estimated utilizing Cohen’s d (where
53
54
55 values around 0.2 = small effect size; around 0.5 = medium effect size; and around 0.8 = large
56
57 effect size).
58
59
60
61
62 24
63
64
65
 Ethical aspects
1
2
3 This study was approved by the Committee for Ethics in the Human and Social Sciences
4
5
6 of the University of Brasília (approval CAAE 71749317.9.0000.5540).
7
8
9 Discussion
10
11
12
13 We believe the principal merit of the present study resides in the development of an
14
15 innovative strategy based on the qualified use of the internet for education surrounding romantic
16
17
18 relationships and the prevention of dating violence among adolescent and young Brazilians, a
19
20 hitherto unaddressed need in the field. It is hoped that studies based on Dating SOS and the
21
22
23 intervention itself will serve, to one extent or another, to strengthen professional practices in
24
25 prevention and psychotherapy as applied to dating violence, supported by a more nuanced
26
27
28
understanding of the phenomenon.
29
30 The primary limitation of this study lays in data collection in the pre-testing stage from
31
32 youths from Brasília, who could be distant in culture and sociodemographic characteristics from
33
34
35 their counterparts in other regions of Brazil. Given the vastness of the country and its cultural
36
37 diversity, it is possible this would result in a loss of cultural sensitivity for the intervention when
38
39
40 implemented with youths in small cities, rural areas, and traditional communities, such as
41
42 indigenous ones. It is also possible that the omission of an evaluation of partner mutability, that
43
44
45
is, how influenceable the partner is or is not by the efforts of the intervention user constitutes
46
47 an important limitation. This could be a target of future studies.
48
49 Should there be positive evidence for the intervention’s usability and efficacy, it will
50
51
52 become a useful tool for bringing scientific knowledge to adolescents and will contribute to the
53
54 debate around the broadening of educational services and early interventions that integrate the
55
56
57 scope of public policies directed at dealing with violence against women and the promotion of
58
59 mental health in adolescents of both genders in Brazil. Should the evidence be insufficient or
60
61
62 25
63
64
65
 negative regarding the intervention’s efficacy, new studies should correct problems, remedy
1
2 gaps, and undertake additional testing.
3
4
5 In conclusion, dating violence has an elevated prevalence among Brazilian adolescents.
6
7 School-based preventive programs have been implemented, but remain isolated initiatives with
8
9
10 insufficient reach. Health communication strategies based on scalable technologies are
11
12 compelling. Taking this into account, we developed Dating SOS, an innovative computer
13
14
15
tailored intervention designed to prevent dating violence and improve the quality of affective-
16
17 sexual relationship among Brazilian adolescents. Its production was based on clear theoretical
18
19 basis and empirical evidences derived from Brazilian context. The romantic partner
20
21
22 changeability and a broader geographic diversity sample should be considered in the design of
23
24 similar technologies. The usability and efficacy of Dating SOS should be examined in future
25
26
27 studies.
28
29
30
31 Abbreviations
32 CTI - computer-tailored intervention
33 I-Change Model – integrated change model
34 UX – user experience
35
36 ECR-short - Experience in Close Relationships Scale, short version
37 CRBQ - Conflict Resolution Behavior Questionnaire
38 CADRI - Conflict in Adolescent Dating Relationship Inventory
39
40
41 Declarations
42
43 Acknowledgements
44 Not applicable.
45
46
47 Authors’ contributions
48 SGM conceived the study, analyzed the data and wrote the paper. POP, SSM, JVVM, AB, &
49 MCR developed the library of messages. MAM and TAAS collaborated in the intervention
50 design. HV collaborated in the research design and writing review. All authors read and
51
52 approved the final manuscript.
53
54 Funding
55 Brazilian Coordination for the Improvement of Higher Education Personnel (CAPES) funded
56 this study. Funding bodies were not involved in study design, data collection, analysis, the
57
58 interpretation of data, or in writing the manuscript.
59
60 Availability of data and material
61
62 26
63
64
65
 Data are available from the corresponding author on reasonable request.
1
2
3 Ethics approval and consent to participate
4 The study was approved by the Committee for Ethics in Human and Social Sciences of the
5
6 University of Brasília (CAAE 71749317.9.0000.5540). All participants of the study provided
7 written informed consent after they were informed of the general aims of the study and their
8 confidentiality and privacy ensured. Written informed consent for participation in the needs
9 assessment phase was obtained with participants adolescents under 16 years old from their
10
11
parent or guardian.
12
13 Consent for publication
14 Not applicable.
15
16
17 Competing interests
18 The authors declare that they have no competing interests.
19
20 Author details
21 1, 2, 3, 4, 5, 6
Departament of Clinical Psychology, Institute of Psychology, University of Brasília,
22
23 Brasília, Brazil. 7 Department of Design, Institute of Arts, University of Brasília, Brasília,
24 Brazil. 8 Department of Computer Sciences, Institute of Exact Sciences, University of
25 Brasília, Brasília, Brasil. 9 Department of Health Promotion, CAPHRI, School for Public
26 Health and Primary Care, Maastricht University, Maastricht, The Netherlands.
27
28
29
30
31 References
32
33 1. Oliveira QB, Assis SG, Njaine K, Oliveira RV. Violência nas relações afetivo-sexuais. In:
34
35
36 Minayo MM, Assis SG, Njaine K, editors. Amor e violência: um paradoxo das relações de amor
37
38 e do ficar. Rio de Janeiro: Fiocruz; 2011. p. 87-140.
39
40
41 2. Brancaglioni BC, Fonseca RM. Violência por parceiro íntimo na adolescência: uma análise
42
43 de gênero e geração. Rev Bras Enferm [Internet]. 2016; doi:10.1590/0034-7167-2016-0408.
44
45 3. Nascimento FS, Cordeiro RL. Violência no namoro para jovens moradores de Recife. Psicol
46
47
48 Soc. 2011; doi:10.1590/S0102-71822011000300009.
49
50 4. Soares JS, Lopes MJ, Njaine K. Violência nos relacionamentos afetivo-sexuais entre
51
52
53 adolescentes de Porto Alegre, Rio Grande do Sul, Brasil: busca de ajuda e rede de apoio. Cad.
54
55 Saúde Pública [Internet]. 2013; doi:10.1590/S0102-311X2013000600009.
56
57
58 5. Silva LM, Oliveira CM, Santana MG, Veríssimo AV, Santos TM, Cardoso MD. Violência
59
60 entre namorados adolescentes em Pernambuco, Brasil. Adolesc Saude. 2017;14(3):63-70.
61
62 27
63
64
65
 6. Murta SG, Ramos CE, Cangussú ED, Tavares TN, Costa MS. Desenvolvimento de um
1
2 website para prevenção à violência no namoro, abandono de relações íntimas abusivas e apoio
3
4
5 aos pares. Contextos Clínic [Internet]. 2014; doi:10.4013/ctc.2014.72.01.
6
7 7. Murta SG, Santos BR, Martins CP, Oliveira B. Prevenção primária à violência no namoro:
8
9
10 uma revisão de literatura. Contextos Clínic [Internet]. 2013; doi:10.4013/ctc.2013.62.05.
11
12 8. Murta SG, Santos BR, Nobre LA, Araújo IF, Miranda AA, Rodrigues ÍO, et al. Prevenção à
13
14
15
violência no namoro e promoção de habilidades de vida em adolescentes. Psicol USP [Internet].
16
17 2013; doi:10.1590/S0103-65642013000200005.
18
19 9. Murta SG, Moore RA, Miranda AA, Cangussú ED, Santos KB, Bezerra KL, et al. Efeitos
20
21
22 de um programa de prevenção à violência no namoro. Psico-USF [Internet]. 2016;
23
24 doi:10.1590/1413-82712016210214.
25
26
27 10. Priolo Filho, SR. Avaliação de uma intervenção para prevenção da violência no namoro
28
29 [dissertation on the Internet]. São Carlos, Brazil: Federal University of São Carlos; 2017 [cited
30
31
32
2019 Nov 20]. Available from: https://repositorio.ufscar.br/handle/ufscar/9024.
33
34 11. Santos KB, Murta SG. Peers as agents of dating violence prevention: feasibility analysis of
35
36 an intervention. Trends Psychol [Internet]. 2019; doi:10.9788/TP2019.3-03.
37
38
39 12. Santos KB, Murta SG, Vinha LG, Deus JS. Efficacy of a bystander intervention for
40
41 preventing dating violence in Brazilian adolescents: short-term evaluation. Psicol Refl Crít.
42
43
44 2019; doi:10.1186/s41155-019-0133-4.
45
46 13. Hawkins RP, Kreuter M, Resnicow K, Fishbein M, Dijkstra A. Understanding tailoring in
47
48
49 communicating about health. Health Educ Res. 2008; doi:10.1093/her/cyn004.
50
51 14. Richards D, Richardson T. Computer-based psychological treatments for depression: a
52
53
54
systematic review and meta-analysis. Clin Psychol Rev. 2012; doi:10.1016/j.cpr.2012.02.004.
55
56
57
58
59
60
61
62 28
63
64
65
 15. Lustria ML, Cortese J, Noar SM, Glueckauf RL. Computer-tailored health interventions
1
2 delivered over the Web: review and analysis of key components. Patient Educ Couns. 2009;
3
4
5 doi:10.1016/j.pec.2008.08.023.
6
7 16. Noar SM, Benac CN, Harris MS. Does tailoring matter? Meta-analytic review of tailoring
8
9
10 print health behavior change interventions. Psychol Bull. 2007; doi:10.1037/0033-
11
12 2909.133.4.673.
13
14
15
17. Krebs P, Prochaska JO, Rossi JS. A meta-analysis of computer-tailored interventions for
16
17 health behavior change. Prev Med. 2010; doi:10.1016/j.ypmed.2010.06.004.
18
19 18. Neville LM, O’Hara B, Milat AJ. Computer-tailored dietary behaviour change
20
21
22 interventions: a systematic review. Health Educ Res. 2009; doi:10.1093/her/cyp006.
23
24 19. Levesque DA, Johnson JL, Prochaska JM. Teen Choices, an online stage-based program
25
26
27 for healthy, non-violent relationships: development and feasibility trial. J Sch Violence. 2016;
28
29 doi:10.1080/15388220.2016.1147964.
30
31
32
20. Levesque DA, Johnson JL, Welch CA, Prochaska JM, Paiva AL. Teen dating violence
33
34 prevention: cluster-randomized trial of Teen Choices, an online, stage-based program for
35
36 healthy, nonviolent relationships. Psychol Violence. 2016; doi:10.1037/vio0000049.
37
38
39 21. Hallberg SC, Lisboa CS, Souza DB, Mester A, Braga AZ, Strey AM, et al. Systematic
40
41 review of research investigating psychotherapy and information and communication
42
43
44 technologies. Trends Psychiatry Psychother [Internet]. 2015; doi:10.1590/2237-6089-2014-
45
46 0055.
47
48
49 22. Gomide HP, Martins LF, Ronzani TM. É hora de investirmos em intervenções
50
51 comportamentais computadorizadas no Brasil?. Psicol Estud [Internet]. 2013;
52
53
54
doi:10.1590/S1413-73722013000200011.
55
56 23. Gomide HP. Desenvolvimento e avaliação de uma intervenção para tabagismo mediada por
57
58 internet [master's thesis on the Internet]. Juiz de Fora, Brazil; Federal University of Juiz de
59
60
61
62 29
63
64
65
 Fora; 2014 [cited 2019 Nov 20]. Available from:
1
2 https://repositorio.ufjf.br/jspui/handle/ufjf/1710.
3
4
5 24. De Vries H, Mesters I, van de Steeg H, Honing C. The general public’s information needs
6
7 and perceptions regarding hereditary cancer: an application of the Integrated Change Model.
8
9
10 Patient Educ Couns. 2005; doi:10.1016/j.pec.2004.01.002.
11
12 25. De Vries H. An integrated approach for understanding health behavior; the I-Change Model
13
14
15
as an example. Psychol Behav Sci Int J. 2017; doi:10.19080/PBSIJ.2017.02.555585.
16
17 26. Ajzen I. The theory of planned behavior. Organ Behav Hum Decis Process. 1991;
18
19 doi:10.1016/0749-5978(91)90020-T.
20
21
22 27. Bandura A. Social foundations of thought and action: a social cognitive perspective.
23
24 Englewood Cliffs, NJ: Prentice-Hall; 1986.
25
26
27 28. Prochaska JO, DiClemente CC. Stages and processes of self-change of smoking: toward an
28
29 integrative model of change. J Consult Clin Psychol. 1983; doi: 10.1037//0022-006x.51.3.390.
30
31
32 29. Janz NK, Becker MH. The health belief model: a decade later. Health Educ Behav. 1984;
33
34 doi:10.1177/109019818401100101.
35
36
30. Gollwitzer PM. Implementation intentions. Strong effects of simple plans. Am Psychol.
37
38
39 1999; doi:10.1037/0003-066X.54.7.493.
40
41 31. De Vries H, Eggers SM, Jinabhai C, Meyer-Weitz A, Sathiparsad R, Taylor M.
42
43
44 Adolescents’ beliefs about forced sex in KwaZulu-Natal, South Africa. Arch Sex Behav. 2014;
45
46 doi:10.1007/s10508-014-0280-8.
47
48
49 32. Mathews C, Eggers SM, Townsend L, Aarø LE, De Vries PJ, Mason-Jones AJ, et al. Effects
50
51 of PREPARE, a multi-component, school-based HIV and Intimate Partner Violence (IPV)
52
53
54
prevention programme on adolescent sexual risk behaviour and IPV: cluster randomised
55
56 controlled trial. AIDS Behav. 2016; doi:10.1007/s10461-016-1410-1.
57
58
59
60
61
62 30
63
64
65
 33. McClinton AT, Eggers SM, de Vries PJ, de Vries H, Lund C, Mathews C. The impact of
1
2 participation in research about abuse and intimate partner violence: an investigation of harms,
3
4
5 benefits, and regrets in young adolescents in the Western Cape of South Africa. J Interpers
6
7 Violence. 2017; doi:10.1177/0886260517691522.
8
9
10 34. Bowlby J. Developmental psychiatry comes of age. Am J Psychiatry. 1988;
11
12 doi:10.1176/ajp.145.1.1.
13
14
15
35. Velotti P, Zobel SB, Rogier G, Tambelli R. Exploring relationships: a systematic review on
16
17 intimate partner violence and attachment. Front Psychol. 2018; doi: 10.3389/fpsyg.2018.01166.
18
19 36. Murta SG, Pires MR, Tavares AS, Cordeiro MA, Teixeira EG, Adorno N. Intimacy and
20
21
22 attachment in dating relationships: implications from case studies for violence prevention.
23
24 Contextos Clínic [Internet]. 2019; doi:10.4013/ctc.2019.121.09.
25
26
27 37. Bonache H, Gonzalez-Mendez R, Krahé B. Romantic attachment, conflict resolution styles,
28
29 and teen dating violence victimization. J Youth Adolesc. 2017; doi:10.1007/s10964-017-0635-
30
31
32
2.
33
34 38. Bonache H, Gonzalez-Mendez R, Krahé B. Adult attachment styles, destructive conflict
35
36 resolution, and the experience of intimate partner violence. J Interpers Violence. 2019;
37
38
39 doi:10.1177/0886260516640776.
40
41 39. Scheeren P, Vieira RV, Goulart VR, Wagner A. Marital quality and attachment: the
42
43
44 mediator role of conflict resolution styles. Paidéia [Internet]. 2014; doi: 10.1590/1982-
45
46 43272458201405.
47
48
49 40. Hazan C, Shaver P. Romantic love conceptualized as an attachment process. J Pers Soc
50
51 Psychol. 1987; doi:10.1037/0022-3514.52.3.511.
52
53
54
41. Riggs SA, Cusimano AM, Benson KM. Childhood emotional abuse and attachment
55
56 processes in the dyadic adjustment of dating couples. J Couns Psychol. 2011; doi:
57
58 10.1037/a0021319.
59
60
61
62 31
63
64
65
 42. Bartholomew K, Horowitz LM. Attachment styles among young adults: a test of a four-
1
2 category model. J Pers Soc Psychol. 1991;61(2):226-244.
3
4
5 43. Simpson JA, Rholes WS. Adult attachment, stress, and romantic relationships. Curr Opin
6
7 Psychol. 2017; doi:10.1016/j.copsyc.2016.04.006.
8
9
10 44. Tan JS, Hessel ET, Loeb EL, Schad MM, Allen JP, Chango JM. Long-term predictions from early
11
12 adolescent attachment state of mind to romantic relationship behaviors. J Res Adolesc. 2016;
13
14
15
doi:10.1111/jora.12256.
16
17 45. Cremers HP, Mercken L, Oenema A, de Vries H. A web-based computer-tailored smoking
18
19 prevention programme for primary school children: intervention design and study protocol.
20
21
22 BMC Public Health. 2012; doi:10.1186/1471-2458-12-277.
23
24 46. Voncken-Brewster V, Moser A, van der Weijden T, Nagykaldi Z, de Vries H, Tange H.
25
26
27 Usability evaluation of an online, tailored self-management intervention for chronic obstructive
28
29 pulmonary disease patients incorporating behavior change techniques. JMIR Res Protoc. 2013;
30
31
32
doi:10.2196/resprot.2246.
33
34 47. Kreuter M, Farrell D, Olevitch L, Brennan L. Tailoring health messages: customizing
35
36 communication with computer technology. New York: Routledge; 2012.
37
38
39 48. Murta SG, Santos BR, Nobre LA, Oliveira SA, Diniz GR, Rodrigues ÍO, et al.
40
41 Diferenciando baladas de ciladas: um guia para o empoderamento de adolescentes em
42
43
44 relacionamentos íntimos. Brasília: Letras Livres; 2011.
45
46 49. Wei M, Russell DW, Mallinckrodt B, Vogel DL. The Experiences in Close Relationship
47
48
49 Scale (ECR)-short form: reliability, validity, and factor structure. J Pers Assess. 2007;
50
51 doi:10.1080/00223890701268041.
52
53
54
50. Brennan KA, Clark CL, Shaver PR. Selfreport measurement of adult attachment: an
55
56 integrative overview. In: Simpson JA, Rholes WS, editors. Attachment theory and close
57
58 relationships. New York: Guilford; 1998. p. 46-76.
59
60
61
62 32
63
64
65
 51. Natividade JC, Shiramizu VK. Uma medida de apego: versão brasileira da Experiences in
1
2 Close Relationship Scale – Reduzida (ECR-R-Brasil). Psicol USP [Internet]. 2015;
3
4
5 doi:10.1590/0103-656420140086.
6
7 52. Rubenstein JL, Feldman SS. Conflict-resolution behavior in adolescent boys: antecedents
8
9
10 and adaptational correlates. J Res Adolesc. 1993; doi:10.1207/s15327795jra0301_3.
11
12 53. Delatorre MZ, Wagner A. Estratégias de resolução de conflitos conjugais: evidências de
13
14
15
validade do CRBQ. Aval Psicol [Internet]. 2015;14(2):233-242.
16
17 54. Wolfe DA, Scott K, Reitzel-Jaffe D, Wekerle C, Grasley C, Straatman A. Development and
18
19 validation of the Conflict in Adolescent Dating Relationships Inventory. Psychol Assess. 2001;
20
21
22 doi:10.1037//1040-3590.13.2.277.
23
24 55. Murta SG, Ramos CE, Tavares TN, Cangussú ED, Costa MS. Libertando-se de namoros
25
26
27 violentos. Novo Hamburgo: Sinopsys; 2014.
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
61
62 33
63
64
65
 Figure Legends
1
2 Fig. 1 Integrated Model of Change
3
4 Fig. 2 Evaluation flowchart
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
61
62 34
63
64
65
 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 Fig. 1 Integrated Model of Change (Source: De Vries H. An integrated approach for
32 understanding health behavior; the I-Change Model as an example. Psychol and Behav Sci
33 Intern Journal, 2017;2(2). DOI:10.19080/PBSIJ.2017.02.555585)
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
61
62 35
63
64
65
 Table 1 Content and objectives of the SOS Dating
1
2 Content Objectives
3
4 Increase knowledge of attachment styles* and their impact on romantic
Knowledge of attachment style
5 relationships.
Session 1

6
7
Boost risk perception regarding violence experienced in a romantic
8 Violence risk perception
relationship.
9
10
Identify types of dating violence: physical, psychological, sexual, and
11 material.
12 Knowledge of violence, Recognize the characteristics that classify relationships as safe and
13 relationship quality, and conflict gratifying as opposed to insecure and violent.
14 management
15 Differentiate negative dating conflict management strategies from
16 positive ones and its relation with attachment style.
17 Identify models of violent loving relationships: friends, aunts and
Session 2

18 Positive and negative models of uncles, cousins, parents, grandparents, stepparents, and neighbors
19 loving relationships Identify models of positive loving relationships: friends, aunts and
20 uncles, cousins, parents, grandparents, stepparents, and neighbors
21 Identify damage potential in risky action plans for daily interaction: lack
22 of communication, ignoring the partner's needs, shutting down and
23 Action plans for increasing ignoring one's own needs, and not respecting the partner's individuality.
24 relationship quality in day-to-day
coexistence Construct positive action plans for daily interactions: noting the partner's
25
needs and providing help sensitively, sharing personal needs with the
26 partner, asking for help, and respecting the partner's individuality.
27
28 Social norms regarding dating Weaken the acceptance of socially disused gender stereotypes that
29 violence justify violence.
30 Recognize the most challenging dating conflict situations to deal with in
Self-efficacy in handling dating
31 a non-violent way: cheating, lying, provocation, an attack by one's
conflicts
partner, a bad temper, being drunk, and repeated violence.
Session 3

32
33 Identify the potential for damage in risky action plans for handling
34 conflict: retaliation, resignation, use of alcohol or drugs, withdrawal,
35 Action plans for non-violent empathy (for victimization).
36 dating conflicts
37 Construct positive action plans for handing conflict: regulation of
38 emotions, assertiveness, recognizing errors and apologizing, empathy
39 (for perpetration), and ending the relationship.
40 Strengthen the understanding that violence can be detrimental by
41 harming the relationship, the victim, the perpetrator, and future
42 generations.
Attitudes toward dating violence
43 Weaken the perception that violence can be advantageous by granting
44 power, self-defense, affectional gains, and financial gains.
45
46 Recognize sources of social support for protection in the face of
Session 4

47 relationship violence.
48 Actively seek social support and give clear indications that support is
49 Social support for protecting needed, as people may be reluctant to be invasive and help.
50 oneself from dating violence
51 Think critically and refuse social support for accepting violent
52 relationships on the basis that they are a source of economic support,
wealth, power, or that one should honor the relationship or protect the
53
children.
54
55 Action plans for freeing oneself Construct positive action plans for protecting oneself against violence:
56 from dating violence separating from the partner, seeking help, and ending the relationship.
57
58
59
60
61
62 36
63
64
65
 1 Eligibility evaluation:
2
3 age and dating status
4
5
6
7
8 Excluded:
Those not currently in a relationship or Included:
9
Those in a relationship and aged 15 to 29
10 outside of the 15 to 29 age range
11
12
13
14
15 Control group Experimental group
16
17
18
19
20 Baseline evaluation: attachment style, dating relationship quality, dating violence
21
22 victimization or perpetration, conflict management skills
23
24
25
26 Session 1: Evaluation and feedback
27 tailored according to attachment style and
28 Week 1
29 General information about romantic having experienced relationship violence.
30 relationship quality and dating violence Usability evaluation.
31 via the book Telling a Night Out from a
32
33
Trap: a guide for empowering teens in
intimate relationships Session 2: Evaluation and feedback
34
35 tailored according to knowledge, models, Week 2
36 and action plans for improving
37 coexistence. Usability evaluation.
38
39
40
41 Session 3: Evaluation and feedback
42 tailored according to social norms, self- Week 3
43 efficacy, and action plans for handling
44
conflicts. Usability evaluation.
45
46
47
48 Session 4: Evaluation and feedback
49
50 tailored according to attitudes, social Week 4
51 support, and action plans for dealing with
52 violence. Usability evaluation.
53
54
55
56 Follow-up: romantic relationship quality, dating violence victimization and perpetration,
conflict management skills, and help seeking.
3 months
57
58
59
60 Fig. 2 Evaluation flowchart
61
62 37
63
64
65
Ethics Approval

Click here to access/download

Supplementary Material
Ethics Approval.pdf
Funding

Click here to access/download

Supplementary Material
CAPES Funding.pdf
Funding_English

Click here to access/download

Supplementary Material
CAPES Funding-English.docx.pdf
Ethics_Approval-English

Click here to access/download

Supplementary Material
Ethics Approval-English.docx.pdf