UBC Faculty Research and Publications

A tailored online safety and health intervention for women experiencing intimate partner violence: the… Ford-Gilboe, Marilyn; Varcoe, Colleen; Scott-Storey, Kelly; Wuest, Judith; Case, James; Currie, Leanne M; Glass, Nancy; Hodgins, Marilyn; MacMillan, Harriet; Perrin, Nancy; Wathen, C. N Mar 21, 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12889_2017_Article_4143.pdf [ 477.43kB ]
JSON: 52383-1.0343285.json
JSON-LD: 52383-1.0343285-ld.json
RDF/XML (Pretty): 52383-1.0343285-rdf.xml
RDF/JSON: 52383-1.0343285-rdf.json
Turtle: 52383-1.0343285-turtle.txt
N-Triples: 52383-1.0343285-rdf-ntriples.txt
Original Record: 52383-1.0343285-source.json
Full Text

Full Text

STUDY PROTOCOL Open AccessA tailored online safety and healthintervention for women experiencingintimate partner violence: the iCAN Plan 4Safety randomized controlled trial protocolMarilyn Ford-Gilboe1* , Colleen Varcoe2, Kelly Scott-Storey3, Judith Wuest3, James Case4, Leanne M. Currie2,Nancy Glass4, Marilyn Hodgins3, Harriet MacMillan5, Nancy Perrin4 and C. Nadine Wathen6AbstractBackground: Intimate partner violence (IPV) threatens the safety and health of women worldwide. Safetyplanning is a widely recommended, evidence-based intervention for women experiencing IPV, yet fewerthan 1 in 5 Canadian women access safety planning through domestic violence services. Rural, Indigenous,racialized, and immigrant women, those who prioritize their privacy, and/or women who have partners otherthan men, face unique safety risks and access barriers. Online IPV interventions tailored to the unique featuresof women’s lives, and to maximize choice and control, have potential to reduce access barriers, and improvefit and inclusiveness, maximizing effectiveness of these interventions for diverse groups.Methods/Design: In this double blind randomized controlled trial, 450 Canadian women who haveexperienced IPV in the previous 6 months will be randomized to either a tailored, interactive online safetyand health intervention (iCAN Plan 4 Safety) or general online safety information (usual care). iCAN engageswomen in activities designed to increase their awareness of safety risks, reflect on their plans for theirrelationships and priorities, and create a personalize action plan of strategies and resources for addressingtheir safety and health concerns. Self-reported outcome measures will be collected at baseline and 3, 6,and 12 months post-baseline. Primary outcomes are depressive symptoms (Center for EpidemiologicalStudies Depression Scale, Revised) and PTSD Symptoms (PTSD Checklist, Civilian Version). Secondaryoutcomes include helpful safety actions, safety planning self-efficacy, mastery, and decisional conflict.In-depth qualitative interviews with approximately 60 women who have completed the trial and websiteutilization data will be used to explore women’s engagement with the intervention and processes ofchange.Discussion: This trial will contribute timely evidence about the effectiveness of online safety and healthinterventions appropriate for diverse life contexts. If effective, iCAN could be readily adopted by health andsocial services and/or accessed by women to work through options independently. This study will producecontextualized knowledge about how women engage with the intervention; its strengths and weaknesses;whether specific groups benefit more than others; and the processes explaining any positive outcomes.Such information is critical for effective scale up of any complex intervention.(Continued on next page)* Correspondence: mfordg@uwo.ca1Arthur Labatt Family School of Nursing, University of Western Ontario, FNB2302, 1151 Richmond St., London, ON NBA 5C1, CanadaFull list of author information is available at the end of the article© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Ford-Gilboe et al. BMC Public Health  (2017) 17:273 DOI 10.1186/s12889-017-4143-9(Continued from previous page)Trial registration: Clinicaltrials.gov ID NCT02258841 (Registered on Oct 2, 2014).Keywords: Intimate partner violence against women, Randomized controlled trial, Online interventions,e-health, Safety planning, Mental health, Technology, Mastery, Self-efficacy, Public health informatics,Computerized decision supportBackgroundIntimate partner violence (IPV), a pattern of physical,sexual and/or emotional abuse in the context of coercivecontrol [1], affects 1 in 3 Canadian women in their life-times [2] and often persists even after separation [3].Over time, chronic stress and fear associated with ex-periencing IPV contributes to depression, anxiety, andPost-Traumatic Stress Disorder(PTSD) [4, 5]. In a meta-analysis of studies of female IPV victims, the meanprevalence of depression was estimated at 47.6%, and ofPTSD at 63.8% (three- to five-fold and five-fold increasesover general female population rates, respectively) [4].Recognizing that a relationship is abusive and sortingout what to do about it is often a long, complex processduring which women learn by trial and error about therisks to their physical and emotional safety and health,the resources and supports available to them, and theconsequences of various strategies [6, 7]. Women’s cap-acity to address their safety is intertwined with theirhealth. For example, women with poorer mental healthare more likely to return to an abusive partner after ini-tially separating [8]. To improve the mental health andquality of life of women experiencing IPV, gender- andcontext-specific, culturally safe interventions areurgently needed.However, few interventions have been shown to im-prove the safety, health or quality of life of womenexperiencing IPV. Safety planning, including educationto increase women’s awareness about IPV, and personal-ized support to identify options for reducing risks andsupport for accessing resources, is one of the mostwidely recommended IPV interventions [9]. There is evi-dence from systematic reviews that providing informa-tion and support to engage women in safety planningand accessing services that meet their needs reducesboth physical and psychological violence [9, 10], but theeffects on women’s mental health and well-being areunknown. Although violence-specific services are a keysource of help with safety planning, fewer than 1 in 5Canadian women who experience IPV access such ser-vices [11]. Rural, Aboriginal, racialized and immigrantwomen, women with partners other than men, and thosewho prioritize privacy (including more affluent women)face unique safety risks and barriers to accessing infor-mation and support for safety planning [12–15]. Effect-ive, accessible, personalized interventions are needed tosupport women’s capacity to engage in actions to pro-mote their safety and well-being, and potentially, toimprove their sense of confidence, control and mentalhealth.Effective online health interventions have been devel-oped in varied domains, including smoking cessation,depression, anxiety and sexual health [16–18]. Evidenceshows that interaction, feedback, and tailoring areimportant features of these online interventions [17].For women experiencing IPV, personalized, online inter-ventions could improve access to information andsupport for safety planning and well-being, particularlyfor women who are reluctant or unable to access formalresources. In a recent pilot test of the first interactive,online safety decision aid for women experiencing IPV,called IRIS, Glass, Eden, Bloom and Perrin [19] foundthat women’s safety planning increased and decisionalconflict decreased immediately after one use of thisresource. The effectiveness of IRIS in improving the men-tal health of women exposed to IPV was recently tested ina randomized controlled trial in the United States [20],the results of which are pending. Two other variations ofthis online intervention (isafe and I-DECIDE) are cur-rently being tested in randomized controlled trials in NewZealand [21] and Australia [22].Drawing on these online interventions and our long-standing history of conducting research with womenwho have experienced IPV, we developed an evidence-based online safety and health intervention for diverseCanadian women experiencing IPV, called ‘iCAN Plan 4Safety’ (hereafter referred to as iCAN). In the develop-ment phase, we drew on insights from our completedresearch examining women’s lives and health after separ-ation from abusive partner [23–27], including researchto develop and test a complex health promotion inter-vention (called iHEAL) designed to support health andquality of life of women transitioning out of an abusiverelationship [28, 29], and current research testing anadaptation of iHEAL for Indigenous women with histor-ies of IPV [30]. In developing the intervention, we drewon these studies, our theoretical grounding and currentliterature to: a) broaden the concept of women’s safety,shifting beyond a focus on immediate physical risks toemphasize safety as an ongoing issue for women (includ-ing after separation), especially emotional safety, and avariety of issues that are closely linked to safetyFord-Gilboe et al. BMC Public Health  (2017) 17:273 Page 2 of 12concerns, including women’s health and well-being overtime; b) focus on developing a personalized (tailored)intervention that would be inclusive of and ‘fit’ with theuniqueness and complexity of Canadian women’s lives,particularly those facing the most substantial barriers tosupport; c) adopt an approach that recognizes and fur-ther builds women’s strengths and capacities, includingtheir awareness of their own priorities and risks; know-ledge of a wide range of options for addressing safety,health and life concerns; and confidence in their abilityto try out these actions, while keeping control andchoice in women’s hands. During the developmentphase, the online intervention was refined by iterativelyintegrating feedback about both content and usabilityuntil it was deemed acceptable by potential end-users. Inthis paper, we describe the trial protocol for testing theeffectiveness of our online safety and health intervention(iCAN) among Canadian women.Objectives of the trialThe primary aim of this study is to test the effectivenessof iCAN, a tailored, online safety and health interven-tion, in improving the health and safety actions of adultwomen experiencing IPV. A group of women complet-ing iCAN will be compared to a control group whoreceive general online safety information as a proxy forusual care. We hypothesize that, compared to the con-trol group, women in the intervention group will: 1)have reduced symptoms of depression and PTSD(primary outcomes), and increase self-efficacy for safetyplanning, use of helpful safety actions, and mastery(secondary outcomes) at 6 and 12 months post-baseline;and 2) reduced decisional conflict immediately post-baseline (after completing the intervention one time)and at 6 and 12 months post-baseline. Additional aimsare to: a) describe how women engage with the interven-tion, and their perceptions of its strengths and weak-nesses; b) examine whether repeat engagement or doseare related to intervention effects; c) explore whetherspecific groups of women benefit in more or differentways from the intervention than others; and, d) examinethe processes of change that explain how any positiveoutcomes are achieved.MethodsTrial designA two arm, double-blind randomized controlled trial isbeing conducted to test the effectiveness of iCAN in adiverse sample of Canadian women who are experien-cing IPV. The protocol was developed using CONSORTe-health guidelines [31]. Outcomes will be measuredpre-intervention (baseline) and 3, 6 and 12 months post-intervention; decisional conflict will also be measuredimmediately post-intervention (after a single completionof the intervention). Website utilization data are beingcollected to capture access to and use of the interventionand control conditions. Women’s experiences of en-gaging with the intervention and control condition willbe explored after completion of the 12-month study visitusing their ratings of potential benefits and harms andcomments about their experiences (collected online) andvia in-depth qualitative interviews conducted with a sub-sample of approximately 60 women. Ethics approval forthis study was obtained in July 2014 from the Institu-tional Research Ethics Boards at the University ofWestern Ontario, University of British Columbia, andUniversity of New Brunswick.Participant inclusion criteriaThe target population for this trial is English speakingwomen aged 19 years or older, living in the provinces ofOntario, New Brunswick, or British Columbia who self-identify as experiencing IPV (physical, sexual or emotionalabuse within the context of coercive control) by a currentor former partner (boyfriend, girlfriend, husband, wife, orintimate partner) in the previous 6 months. Exposure toIPV will be determined by a positive response to at leastone question on a modified 4-item version of the AbuseAssessment Screen, a standardized IPV screening tool thathas been used in both research and clinical settings [25, 32,33]. Specifically, women will be asked whether, in the pre-vious 6 months, a partner or ex-partner ever: a) hit, kickedor otherwise physically hurt her; b) forced her to have sex-ual activities against her will; c) did things to make her feelafraid of them; or d) did things to try and intimidate her orto control her thoughts, feelings or actions. Eligible partici-pants must have access to a safe computer (to access thestudy website), safe email address (to which study-relatedinformation will be sent), and a safe mailing address (e.g.,that of a friend or family member) to which the participa-tion fee can be sent. Women not meeting all of these cri-teria will be excluded from participation.RecruitmentWomen will be recruited to the trial using advertise-ments posted in a variety of online spaces (e.g., classifiedads, health or violence websites, online discussion for-ums), supplemented by more traditional approaches toadvertising in community settings (e.g., libraries), orthrough organizations, associations or service agenciesaccessed by women. Women seeking more informa-tion about the study will be directed to the studywebsite (www.icanplan4safety.ca) or asked to contacta Research Assistant (RA) by email or a toll-free tele-phone number. This approach is similar to that uti-lized by the IRIS trial, and mirrors the varied waysthat women might discover and access the onlineintervention outside of a research context.Ford-Gilboe et al. BMC Public Health  (2017) 17:273 Page 3 of 12Number of participants requiredA sample of 450 women (225 per group) will be re-cruited for this trial based on the statistical powerneeded to both test the impact of the tailored onlineintervention on primary outcomes using a general esti-mating equations (GEE) approach, and, subsequently, toexamine mechanisms by which the intervention exertsits effects on these outcomes using Structural EquationModeling (SEM). We based the power analysis on twokey outcomes: decreased symptoms of depression, andof PTSD. Baseline means and standard deviations fromthe IRIS study [34], a similar trial in the USA, were usedin the power analysis. In estimating the sample size, wevaried the percent change over time in the interventiongroup relative to the control group from 15% to 20%which corresponds to moderate effects sizes, 0.29 and0.52, found in previous studies of IPV interventions. Thepower analysis was based on repeated measures ANOVAsince specific estimates of variance needed to calculatesample size for GEE are not available. However, in theGEE we will be able to retain all cases in the analysis, in-cluding those with missing data. With an alpha of .05,15% greater change in the intervention group, relative tothe control group, a sample size of 400 participants pro-vides .73 power for depression and .92 power for PTSD.With a 20% greater change in the intervention, power is.94 for depression. We assume that attrition will beapproximately 10%, based on the retention rate in therecently completed US trial (<10% at 12 month followup) [35] and our recent success in retaining 81% of acohort of women who had experienced IPV over a 4-year period. Attrition of 10% at 12 months would resultin a final sample of 405 women, adequate for allanalyses.Trial processWomen who are interested in taking part in this studywill contact a RA by telephone. The RA will review thestudy purpose with each woman, assess her eligibilityusing a Recruitment and Eligibility Form, and invite alleligible women to participate. Initially, the RA will ob-tain informed verbal consent by reading the letter of in-formation to the woman and answering questions shemay have; as noted below, women will provide writtenconsent to participate when they login to the website.Each woman who consents will have her name, safeemail address and preferred contact information, prov-ince of residence, whether she has children or not, andhow she learned about the study recorded on theEligibility and Recruitment Form by the RA. The RA willdiscuss safety concerns with the woman and develop aplan for safe contact with the study. The RA will alsoenter this information into an electronic tracking data-base, housed on a secure server at The University ofWestern Ontario. The woman will then be asked to cre-ate a security question and answer for the purpose ofidentification in the case of a lost password.Entry of the woman’s safe email address into the track-ing database will initiate her enrollment into the study.At this point, the system will generate a unique studyID, randomly assign the woman to either the interven-tion or control group, and send out an automated ‘wel-come’ email to the woman’s safe email addresscontaining a link to the letter of information and con-sent, and a corresponding URL, username and password.The women will be instructed to use these credentials tologin to the website. The first time the woman logs intothe website, she will be asked to re-affirm her consent toparticipate by selecting 3 radio buttons indicating that:a) she has read the letter of information, b) has had allquestions about the study answered, and, c) agrees toparticipate. These questions must be answered affirma-tively in order to proceed.At baseline, the woman will be encouraged to loginand complete the baseline measures and interventionactivities on the same day she is randomized and re-ceives her study website password. The woman will beallowed to complete the baseline study visit in morethan one sitting; completed responses will be automatic-ally saved so that the woman can continue where she leftoff when she logs in at other times. However, she willhave 6 weeks from the time of enrollment in which tocomplete the baseline measures and accrue to the trial.Automated and manual emails will be sent at regular in-tervals over the 6-week period until the woman eithercompletes baseline measures or the time expires. Shouldthe time expire, an email message will be sent to in-form the woman that she is no longer able tocomplete the tool. Based on the IRIS trial and ourpilot study, baseline study measures are expected totake about 30 min to complete. The additional timerequired to complete intervention activities is 30-60min for women in the intervention arm and 20-30min for those in the control arm.For 3, 6 and 12-month follow-up, automated reminderemails will be sent to the woman’s safe email address no-tifying her that she is due to complete another session.The woman’s username and password, along with thestudy URL will be included in the email notification.When the woman returns to the website, the study mea-sures will be slightly different for the 3, 6 and 12 monthassessments. At these follow-up visits, women in theintervention group will be invited to complete the inter-vention activities again if they wish, and encouraged toconsider this option if their situations have changed.Once enrolled, women will be able to visit their assignedwebsite between scheduled study visits to review theirresponses. Women in the intervention arm will be ableFord-Gilboe et al. BMC Public Health  (2017) 17:273 Page 4 of 12to update some activities, such that the online tool canbe used as an ongoing resource that changes and evolvesto fit with their changing life contexts (see Interventionarm for details).Website utilization data for both study arms will becollected automatically to: 1) track the number of timesthe site is accessed within scheduled follow-up windowsand between windows; 2) the time spent on a sessionand within specific sections; 3) item completion rates;and, 4) materials accessed and skipped. This informationwill be used to describe the how women engage with theintervention and control websites as well as inform uson dosing.Randomization and blindingOn enrollment, participants will be automatically com-puter randomized using a stratified blocked randomizedscheme based on whether she has children or not andprovince of residence using information gathered duringrecruitment that has been entered into the electronictracking data base by a RA. Once participants are ran-domly allocated to either the intervention or controlgroup, they will be automatically sent a link to theappropriate website. Women will not be informed oftheir group assignment, although they may guess whichonline intervention they are using based on the extent towhich they receive personalized (versus general) infor-mation and suggestions. Members of the research team,with the exception of the programmer (JC) and statisti-cian (NP), will also be blind to the group allocation ofparticipants until 12-month data have been collected sothat research staff, who may be contacted by women fortechnology or other support, are not in a position toprovide preferential treatment to women based on groupassignment.Online InterventionsThe online interventions that will be delivered to womenin both arms are organized into the same sections andfocus on: 1) questions about women’s background char-acteristics; 2) women’s plans for their relationship withtheir abusive partner, their priorities and current safetyrisks; 3) an action plan containing safety planning infor-mation and resources, and 4) a structured debriefing.The content of each section by study arm is summarizedin Table 1, with details of each provided in the text thatfollows.Intervention arm (tailored, online safety and healthintervention)Women will initially respond to questions about theirpersonal context (such as whether they have children,are employed, were born in Canada). In the next section,“My Decision, My Priorities, My Risks”, women willcomplete a series of questions and interactive activities.First, women will be asked to reflect on their relation-ship with their abusive partner and to report on theircurrent plan from 5 options (i.e., stay with partner,separate, remain separated, return to partner, unsure).Second, women will be invited to complete an inter-active priority setting exercise in which they will rate therelative importance of 5 factors (my feelings for my part-ner, my health and well-being, having resources, my con-cern for safety, and my children’s well-being) in makingdecisions about their unsafe relationship. Women will beshown pairs of priorities positioned at opposite ends of ahorizontal line and then slide a toggle bar toward thepriority that is more important to them. Weights foreach priority will be computed mathematically and realtime feedback provided in the form of a graph, alongwith examples of safety strategies in the action plan thatfit with the top priority identified. Third, women willcomplete the Danger Assessment (DA), or DA-Revised(for women with partners other than men), a validatedrisk assessment [36, 37]. Using the DA, women will beasked to mark the type and timing of episodes of phys-ical and/or sexual violence from their partner in the past12 months on a calendar, and then respond to 19Table 1 Comparison of content of intervention and control armsSection Study arm Study arms differTailored intervention Usual care controlBackground questions Questions about women’s demographiccharacteristics and living situationQuestions about women’s demographiccharacteristics and living situationNoMy decisions, priorities,risks• Plan for Relationship• Priorities Exercise with personalized feedback• Danger Assessment with personalized feedback• Plan for Relationship• General Information about Priorities• General Information about risk of IPVNoYesYesMy action plan Strategies organized in 8 categories with associatedresources• Recommended based on her responses to questionsand activities; woman can further customizeEmergency safety planning and child safetystrategies and resources• General information not personalized to thewoman’s situation; no opportunity to modifyYesDebriefing Standardized information about symptoms of astress reaction and strategies for managingStandardized information about symptoms ofa stress reaction and strategies for managingNoFord-Gilboe et al. BMC Public Health  (2017) 17:273 Page 5 of 12questions assessing risk factors for serious or lethalviolence. A standardized weighted score (range 0-38),calculated using these responses, will then be providedto the women on a graph with information about thecorresponding level of risk, ranging from variable dangerto extreme danger, and a brief personal message abouttheir level of danger and suggested actions.Next, in “My Action Plan”, women will create atailored Action Plan for themselves by selecting from 46strategies organized in 8 different groups (EmergencyPlan; Domestic Violence Information and Services,Security Measures, Basic Resources, Health, Supportfrom Family, Friends and Community, Child Safety andLegal Strategies). Each strategy conveys detailed infor-mation about actions women could take to address theirsafety and health concerns. To engage diverse groups ofwomen, these strategies are intentionally written in aconversational tone, use a coaching approach thatincludes ‘tips’ for accessing services (e.g., how specificservices work and how to access them more quickly) orusing a strategy (e.g., a guided video to demonstrategrounding exercises), and comments about conditionsthat might alter the strategies (e.g., If you live in a ruralcommunity…..”). Linked resources include contact andother information for services, and relevant websites andvideos. When women begin this exercise, some strategieswill be marked as recommended for each woman basedon her unique responses to previously completed ques-tions and activities (e.g., whether parenting children,plan for the relationship, DA score). Women will beinstructed to select each strategy to open and read it,and to further personalize their action plan by selectingadditional strategies important to them, and/or remov-ing (de-selecting) those which they do not want to keepin their plan. Encouraging customization by the womanis an approach that reinforces her control in creating theplan. Women will be encouraged to save, print or accessthe Action Plan online at any time if safe to do so, andto visit the website to update their Action Plans shouldtheir circumstances change.As the final step in the intervention, women will beprovided with information about the signs of a stressreaction, advised that this is a normal reaction that mayoccur following a study visit, and provided with sugges-tions for managing this type of reaction, if it occurs. Thisis part of a standardized debriefing protocol we haveused successfully in other studies [38] and an importantstep in promoting women’s emotional safety whenengaging with the study.Control arm (General Online Risk and Safety PlanningInformation)Women in the control arm will complete the same back-ground questions and question about their plans for therelationship as those in the intervention arm. However,in Section 2, which focusses on decisions, priorities andrisks, brief general information will be provided aboutthe importance of considering priorities when makingimportant decisions along with risk factors for IPV, butwomen will not complete the interactive Priorities Exer-cise or the DA, or be provided with individualized feed-back. Next, women will be provided with an Action Plancontaining emergency safety strategies (1 of 8 categoriesof strategies provided to the intervention group), includ-ing safety of children, and some general informationabout online and community resources that is not tai-lored to the woman’s situation. The type of informationincluded will be consistent with what women would findonline if they did their own search. In this sense, it con-stitutes a proxy for usual care. As with the interventiongroup, women will be able to save or print their plan forlater use, or revisit the website at any time during the12-month period of the study. After the final study visithas been completed, women in the control group will beoffered an opportunity to complete the tailored, onlineintervention and provided with a clickable link to accessit if they wish.RetentionWe will use a retention protocol based on proven tech-niques for tracking women in longitudinal IPV studies[25]. Retention strategies will include: a) asking womenfor names and contact information for up to 6 people(family, friends, employers, service providers) who wouldknow of their whereabouts and could be used as alter-nate safe contacts, along with their consent to contactpeople to release this information to us, b) contactingparticipants mid-way between the 4 study visits (at 1.5,4.5, and 9 months post baseline) to confirm contactinformation and any anticipated changes to this; c)providing a small incentive for completing each of the 4study visits, with the value increasing over time (i.e.,$20, $30, $40 and $50); and, d) after each email or tele-phone contact, providing the woman with the name ofthe local RA, the study phone number and email, andthe date of the next study visit. Efforts will be made bythe RAs to use friendly, warm communication strategiesto establish rapport with participants. It is hoped that byusing telephone-based enrollment, women will feel morecomfortable with the research team, and be more likelyto ask for support if needed, improving retention.Safety and securityThe physical and emotional safety of women who par-ticipate in the iCAN Plan 4 Safety trial is a primaryconcern. We will adopt many safety procedures used inthe IRIS trial [19] with success (i.e., that trial reportedno adverse events). All research staff will be trained inFord-Gilboe et al. BMC Public Health  (2017) 17:273 Page 6 of 12the use of these safety procedures. We expect the mostsignificant safety risks faced by women in the trial willbe from an abusive partner who becomes angry if theydiscover that she is participating in this research. Anumber of strategies will be used to prevent risk ofdetection by abusive partners and to ensure that partici-pation is private and confidential. Information will beprovided to all participants regarding the safe (private)use of computers and the Internet. As necessary, the RAwill help the woman brainstorm a safe location to assessa computer or tablet (e.g., family, friend, public library,community agency, work, etc.). Women will be encour-aged to not use a computer or tablet where the partnercould potentially witness her participating in the studyor access her Internet history. The RA will also offertraining to the women about how to safely use the Inter-net. For example, the RA will provide information aboutdeleting browser history and send this information towomen via email if requested; this information will alsobe posted on the home page of the study website. Whenwomen load the iCAN website, a pop up dialogue boxappears with information and instructions for openingthe site in private mode (Incognito mode). The websiteis also equipped with a ‘quick escape’ bar that whenclicked, leads to two actions: 1) provides immediate exitfrom the website and brings the user to the Googlehomepage; 2) pushes the tab running iCAN to the wea-ther network webpage. All manual and automated emailssent to participants are addressed with the subjectheader “Women’s Health Study” to further reduce risk ofdetection.Given the potential for questions about sensitivetopics to trigger women with experiences of violenceand trauma, and their experiences of stigma and judg-ment by others, women’s emotional safety was alsoconsidered in developing both the online surveys andthe interventions to limit any potential emotionalharms of participating. The language and content ofthe surveys and intervention was carefully drafted andreviewed to ensure inclusiveness and applicability towomen from various backgrounds so that womencould see themselves in the study, and to avoidphrasing that could be interpreted as blaming or stig-matizing. To increase women’s sense of comfort, weattempted to use an informal, conversational tone, in-cluded messages that acknowledged when questionsor activities might create distress for women, and en-couraged women to take a break if needed or other-wise care for themselves. As previously noted, weadded a structured debriefing at the end of eachstudy visit to remind the woman that she might ex-perience a stress reaction after the session, that thisis a normal response and to provide options for man-aging it.Risk of suicidality is heightened in women with histor-ies of IPV [39]. Although we do not anticipate that studyparticipation will increase this risk, we adapted a proto-col developed for the IRIS study for use in both studyarms. Safety programming has been integrated into theonline interventions to: a) identify women who are atrisk of suicide using their responses to standardizedquestions for depressed mood and suicide attempts, b)advise women directly if they are at-risk of suicide andprovide options for managing the situation, including afollow up contact with an RA; and; c) alert RAs by emailwhen a woman at-risk wishes to be contacted. Follow-upcontact with women will be guided by a standard proto-col, based on current best practices in suicide manage-ment. All research assistants will complete the AppliedSuicide Intervention Skills Training (ASIST). The princi-pal investigators at each site, who are Registered Nurses,will be available to consult with and support RAs, ifneeded.An independent Data Safety Monitoring Committee(DSMC) will meet every 6 months to review reportsof potential harms (e.g., exposure to abuse, risk ofsuicide) and adherence to safety study protocols pre-pared by our statistician. The committee will recom-mend investigations and/or follow up actions aboutany safety concerns which they identify to the Co-PIs(MFG, CV, KSS, JW).OutcomesThe primary outcomes are:(a)Depressive Symptoms measured by the Center forEpidemiologic Studies Depression Scale, Revised(CESD-R) [40]. The CESD-R is a 20-item self-reportmeasure of symptoms reflective of the DSM-IVcriteria for depression. Women rate how often theyhave experienced each symptom in the past weekusing 5 options that range from ‘not at all or lessthan 1 day’ to ‘nearly every day for 2 weeks’. Totaland subscale scores are computed by summingresponses to applicable items. The revised CESDcorrelates highly with the original scale, anddemonstrates good to excellent face and constructvalidity, as well as excellent internal consistency(α = .90-.96) [40, 41].(b)PTSD Symptomology, measured on the PTSDchecklist, Civilian Version (PCL-C) [42, 43]. ThePCL-C is a 17 item self-report measure designed foruse in community samples to assess the probabilityof meeting DSM-IV diagnostic criteria for PTSD.The PCL-C asks about symptoms in relation togeneric stressful experiences; women rate howbothered they are by each symptom during thepast month using a 5 point Likert-type scale,Ford-Gilboe et al. BMC Public Health  (2017) 17:273 Page 7 of 12with a range of 1 (not at all) to 5 (extremely).Total summation score range from 17-85, withhigher scores indicative of greater symptomatology.The PCL-C has demonstrated validity and excellentinternal consistency reliability (.94 for the total scaleand .82 to .94 for subscales) [44].The secondary outcomes are:(a)Decisional Conflict, measured on an adapted 13 itemversion of the low literacy Decisional Conflict Scale(DCS) [45]. The DCS assesses the extent to whichwomen understand the advantages and disadvantagesof safety planning options and their values related tothese decisions [46]. The DCS discriminates betweenpeople who make decisions and those who delaymaking decisions [47]. Women self-report on a 3point Likert-type scale as to which option they prefer,‘yes’ (0), ‘unsure’ (2), or ‘no’ (4). Scoring is summative,with higher scores representing higher levels ofdecision conflict and anxiety [45]. The lowliteracy version of the DCS has shown goodtest-retest reliability (.75) and internal consistency(α = .72) [47].(b)Use of Safety Actions, measured on a 22 itemself-report scale, adapted from several sources,including the Safety Behavior Checklist [48],Intimate Partner Violence Strategies Index [49]and our previous longitudinal research [50].Women are asked to indicate whether they haveused a variety of safety behavior strategies/actions(yes/no) within the past 12 months and, if used,how helpful this strategy was it in dealing withthe violence (on a 5-point scale ranging from‘not at all helpful’ to ‘very helpful’).(c)Mastery, measured on Pearlin’s 7-item MasteryScale. The Mastery Scale is a self-report measurethat taps into perceptions of personal control overone’s life. Women respond to how much theyagree (1 = strongly agree) or disagree (7 = stronglydisagree) with each item. Scores are computed bysumming responses to all items, where higher scoresindicating greater perceived mastery. The MasteryScale has demonstrated good internal consistency(α = .75 -.78) and has been widely used in a varietyof populations providing evidence of strong facevalidity [51–53].(d)Self-efficacy for Safety Planning, measured on 2visual analogue scales (VAS), developed for thisstudy. Women are asked to rate their confidence inmaking a safety plan for themselves on a 100 mmhorizontal line, with anchors of ‘not at all confident’and ‘completely confident.” Women with childrenwill be asked to complete a second VAS to rate theirconfidence in making a safety plan for their children.VAS scores are recorded by the website as thedistance in mm from the left anchor (0) to thelocation of the mark on the line. Scores rangefrom 0 to 100, with greater scores representgreater self-efficacy for safety planning.Other outcomes:(a)Level of Coercive Control, measured on theWomen’s Experiences with Battering (WEB) Scale[54]. The WEB is a 10-item scale designed to measurethe intensity of experiences of psychologicalvulnerability from IPV and the impact of coercivecontrol [54–56]. For each item, woman are asked torespond on a 6-point Likert scale, ranging from 1(Strongly agree) to 6 (Strongly disagree). Totalsummation scores range from 10 to 60; lower scoresrepresent less abusive behavior and loss of power andcontrol. The WEB has demonstrated high internalconsistency (α = .99) and construct validity [54].(b)Severity of Intimate Partner Violence, measured onthe 30-item Composite Abuse Scale (CAS), askswomen to rate the frequency of experiencing specificabusive acts in the previous 12 months on a 6-pointscale ranging from ‘never’ (0) to ‘daily’ (5). Usingcut-off scores, women’s responses are categorized as‘positive’ or ‘negative’ for exposure to 4 types of IPV:physical abuse, emotional abuse, harassment andsevere combined abuse. A total summative scorecan be derived, with higher scores indicative ofgreater severity of abuse [57]. The CAS hasdemonstrated good internal consistency [58] andevidence of content, construct, criterion, andfactorial validity [59]. In this study, the 3 sexualabuse items were modified to make them moreconsistent with current theory and measurementapproaches in the field [60].(c)Social Support, measured using a 5-item modifiedscale from the Medical Outcomes Study SocialSupport Survey (MOS-SSS). The shortenedMOS-SSS assesses the perceived availability ofemotional, informational, and instrumental support.For each item, the availability of support is ratedon a 5-point Likert-type scale, ranging from 1(none of the time) to 5 (all of the time). Totalsummative scores are computed, with higher scoressuggestive of greater perceived support. The 5-itemversion has demonstrated good internal consistency(α = .87) and validity [61].Data collectionCollection of outcome data will be integrated into boththe intervention and control group websites and will beFord-Gilboe et al. BMC Public Health  (2017) 17:273 Page 8 of 12collected at baseline (prior to completion of the inter-vention) and 3-, 6-, and 12-months post-intervention. Ateach study visit, participants in both groups willcomplete self-report measures of mental health, deci-sional conflict, use of safety actions, self-efficacy forsafety planning, social support, mastery, IPV severity andcoercive control. Demographic information will be col-lected at baseline and updated at subsequent waves, ifneeded. Website utilization data for both study arms willbe collected automatically to track the number of timesthe site is accessed, time spent, and materials accessedand skipped. This information will be used to describehow women engage with the intervention and controlwebsites (including completion rates of specific sec-tions). Feedback on the helpfulness of the online inter-vention will also be collected at the end of each studysession. Women will be asked to complete a brief exitsurvey immediately following the 12-months study visitto gather their ratings of perceived comfort, harms andbenefits of the research using questions adapted fromother studies, and any additional comments they wish toprovide.Statistical analysisExamination of outcomes by study armThe effectiveness of the intervention will be assessed bycomparing the intervention and control groups onchanges in outcome measures, between the baseline andthe 3, 6, and 12 month post-baseline assessments. Wewill test the hypothesis that, at 3, 6 and 12 months post-baseline, the intervention group will have improvedmental health (our primary outcomes), in comparison tothe control group, using intent-to-treat principles withgeneralized estimating equations (GEE). The sameapproach will be used for testing secondary outcomeswhich hypothesize decreased decisional conflict, in-creased safety seeking behavior, mastery and self-efficacy.Separate analyses will be conducted for each outcome.The parameter of interest is the group (intervention vs.control) by time interaction, which if significant meansthat change over time differs for intervention and controlgroups. Significant findings will be graphed to determinethe nature of the effect. Potential confounding variablesfor which intervention and control groups are significantlydifferent at baseline and variables that are related to miss-ingness will be entered in the model. These analyses willestablish overall effect sizes, for specific outcomes, for theintervention.Analysis of differential intervention effectsTo test for differential effects of the intervention for spe-cific subgroups of women, we will either enter a group-ing variable (e.g., rural/urban) and its three-wayinteraction with intervention group and time, or, ifsubgroups are reasonably large, we will analyze differ-ences in intervention effects using multi-group analysistechniques. These analyses are not fully powered; there-fore, we will interpret differences in effect sizes acrossthe specific subgroups rather than rely on statistical sig-nificance. Subgroups to be examined include geographiclocation (rural/urban), partner status (separated frompartner or not), continuing exposure to abuse (yes/no),mothering (yes/no), Indigenous identification (yes/no).Testing the mechanisms that explain the interventionseffectsTo better understand how the intervention achieves itseffects, we will employ structural equation modeling(SEM) techniques for longitudinal data to test a theoret-ical model specifying the relationships between changein the intervention, a set of mediating variables (deci-sional conflict, safety strategies, support, mastery, abuseseverity) and women’s mental health. Latent variableswill be developed for each construct from existing mani-fest variables. Goodness of fit between the model andthe data will be assessed. The repeated measures charac-ter of the data will be integrated in the model by specify-ing latent growth curve trajectories for the outcome andsome of the mediating variables.Process evaluationIn-depth qualitative interviews will be conducted with asub-sample of approximately 60 women after they havecompleted the trial to explore how they engaged withthe intervention; their perceptions of interventionstrengths and weaknesses, risks and benefits, and help-fulness; and processes of change that may explain inter-vention effects. At the end of the 12-month study visit,all participants (both groups) will be asked to indicatetheir interest in participating in a qualitative interviewabout their experiences taking part in the study, and toprovide permission for a RA to contact them using theirsafe email address. Maximum variation sampling will beused to identify and recruit a sample of women with var-ied sociodemographic backgrounds (e.g., age, whetherthey are mothers, geographic location) and histories withtheir partners (e.g., abuse history, partner gender, livingwith partner versus separated) from those who consentto be contacted. A RA will contact these women to pro-vide information about the interview process, seek sep-arate informed consent and to arrange a time for theinterview. Interviews will be conducted by an investiga-tor or trained RA by telephone or Skype using existingstudy protocols for safe contact. A structured interviewguide will be used to explore how women engaged withthe online interventions and how information or re-sources were used; positive and negative aspects and im-pacts; and how the context of women’s lives shaped theirFord-Gilboe et al. BMC Public Health  (2017) 17:273 Page 9 of 12intervention use and impacts. Interviews will be audio-recorded and transcribed verbatim and analyzed usingcontent analysis techniques.The process evaluation will also draw on data collectedin the 12-month exit survey offered to all women,administrative (use) data collected by the online system(both groups), and report of helpfulness of the onlineinterventions in supporting women’s decision making(collected at baseline and 3, 6, and 12-month study visitsusing the Preparation for Decision-Making Scale previ-ously described) in order to fully develop, explain, andcontextualize the findings.DiscussionThis trial builds on and extends research on onlinesafety interventions for women experiencing IPV inthree key ways. First, the intervention tested in thisstudy integrates increased attention to both health andemotional safety – both during completion of the onlineintervention and in the strategies offered to women aspart of the intervention. Second, building on lessonslearned from IRIS, we have streamlined data that will becollected from the control group women (e.g., no com-pletion of the Danger Assessment), in an effort to reduceany measurement effects. Finally, we will be able to testfor effects in a range of specific contexts, including var-ied geographies, women who have and have not sepa-rated from their abusive partners and who are/are notexperiencing ongoing violence. Therefore, study findingswill contribute timely evidence about the effectiveness ofonline safety and health interventions appropriate fordiverse life contexts. Moreover, this study will producecontextualized knowledge about how women engagewith the intervention; its strengths and weaknesses;whether specific groups benefit more than others; andthe processes that explain any positive outcomes. Ifeffective, this information is critical for successful scaleup of complex interventions, such as iCAN, into healthand social services and/or for public access for womenwho prefer to work through the interventionindependently.Trial statusActive follow up.AbbreviationsIPV: Intimate partner violence; iCAN: iCAN Plan 4 Safety; RA: Researchassistant; PTSD: Post-traumatic stress disorder; DA: Danger AssessmentAcknowledgementsWe would like to thank the iCAN trial staff, Joanne Hammerton (ResearchManager), Meghan Fluit, Koushambhi Khan, Phoebe Long, Jeannie Malcolm,and Sarena McLean (Research Coordinators/Assistants) for their contributionsto development of website content and study protocols.FundingThis study was funded by the Canadian Institutes of Health Research (CIHR),Grant #123323, Dr. M. Ford-Gilboe (PI). Marilyn Ford-Gilboe is supported bythe Women’s Health Research Chair in Rural Health at Western University.Harriet MacMillan is supported by the Chedoke Health Chair in ChildPsychiatry.Availability of data and materialsNot applicable.Authors’ contributionsMFG, KSS, CV, JW had major responsibility for the design and conductof the study and, with NG, conceived the study. MFG, KSS, CV and JWconceptualized the intervention, developed the website content andworked with the programmer (JC) to create and pilot test the websites.All authors contributed to the trial protocol. NP developed the plan forstatistical analysis. MFG prepared the first draft of this manuscript andcoordinated revisions. KSS and CV provided substantial early input andrevisions to the manuscript. All other authors reviewed and edited themanuscript. All authors read and approved the final manuscript.Competing interestsThe authors declare that they have no competing interests.Consent for publicationNot applicable.Ethics approval and consent to participateThis protocol was reviewed and approved by the following Research EthicsBoards: University of Western Ontario (105436); University of British Columbia(H14-01817), University of New Brunswick (2013-103). Informed Consent toparticipate in this study was obtained from all participants as a condition ofenrollment, as described in this manuscript.Author details1Arthur Labatt Family School of Nursing, University of Western Ontario, FNB2302, 1151 Richmond St., London, ON NBA 5C1, Canada. 2School of Nursing,University of British Columbia, Vancouver, BC, Canada. 3Faculty of Nursing,University of New Brunswick, Fredericton, NB, Canada. 4School of Nursing,Johns Hopkins University, Baltimore, MD, USA. 5Departments of Psychiatryand Behavioural Neuroscience, and Pediatrics, Offord Center for ChildStudies, McMaster University, Hamilton, ON, Canada. 6Faculty of Informationand Media Studies, University of Western Ontario, London, ON, Canada.Received: 12 January 2017 Accepted: 23 February 2017References1. Tjaden P, Thoennes N. Extent, Nature and Consequences of Intimate PartnerViolence: Findings from the National Violence Against Women Survey.National Institute of Justice and the Center for Disease Control andPrevention. Washington, DC: U.S. Department of Justice, Office of JusticePrograms, National Institute of Justice; 2000.2. Cohen MM, MacLean H. Violence against Canadian women. In: DesmeulesM, Stewart D, Kazanjian A, Maclean H, Payne J, Vissandjée B, editors.Women’s Health Surveillance Report: A Multidimensional Look at the Healthof Canadian Women. Ottawa: Canadian Institute for Health Information;2004.3. Spiwak R, Brownridge DA. Separated Women’s Risk for Violence: An Analysisof the Canadian Situation. J Divorce Remarriage. 2005;43(3/4):105–17.4. Golding JM. Intimate partner violence as a risk factor for mental disorders:A meta-analysis. J Fam Violence. 1999;6:81–95.5. Lagdon S, Armour C, Stringer M. Adult experience of mental healthoutcomes as a result of intimate partner violence victimisation: A systematicreview. Eur J Psychotraumatol. 2014;5. http://doi.org/10.3402/ejpt.v5.24794.6. Anderson DK, Saunders DG. Leaving an abusive partner. Trauma ViolenceAbuse. 2003;4:163–91.7. Wuest J, Merritt-Gray M. Not going back: Sustaining the separation in theprocess of leaving abusive relationships. Violence Against Women. 1999;5:110–33.Ford-Gilboe et al. BMC Public Health  (2017) 17:273 Page 10 of 128. Abdulmohsen Alhalal E, Ford-Gilboe M, Kerr M, Davies L. Identifying factorsthat predict women’s inability to maintain separation from an abusivepartner. Issues Ment Health Nurs. 2012;33:838–50.9. Tirado-Muñoz J, Gilchrist G, Farré M, Hegarty K, Torrens M. The efficacy ofcognitive behavioural therapy and advocacy interventions for womenwho have experienced intimate partner violence: A systematic review andmeta-analysis. Ann Med. 2014;46:567–86.10. Ramsay J, Carter Y, Davidson L, Dunne D, Eldridge S, Feder G, Hegarty K,Rivas C, Taft A, Warburton A. Advocacy interventions to reduce or eliminateviolence and promote the physical and psychosocial well-being of womenwho experience intimate partner abuse (Review). Cochrane Database SystRev. 2009;8(3):CD005043. doi:10.1002/14651858.CD005043.pub2.11. Barrett BJ, Pierre MS. Variations in Women’s Help Seeking in Response toIntimate Partner Violence: Findings From a Canadian Population-BasedStudy. Violence Against Women. 2011;17:47.12. Riddell T, Ford-Gilboe M, Leipert B. Strategies used by rural women to stop, avoid,or escape from intimate partner violence. Heal Care Women Int. 2009;30:134–59.13. Weisz AN. Reaching African American battered women: Increasing theeffectiveness of advocacy. J Fam Violence. 2005;20:91–9.14. Yoshihama M. Battered women’s coping strategies and psychologicaldistress: differences by immigration status. Am J Community Psychol. 2002;30:429–52.15. Messinger AM. Invisible victims: same-sex IPV in the National ViolenceAgainst Women Survey. J Interpers Violence. 2011;26:2228–43.16. Bailey JV, Murray E, Rait G, Mercer CH, Morris RW, Peacock R, Cassell J,Nazareth I. Interactive computer-based interventions for sexual healthpromotion. Cochrane Database Syst Rev. 2010;9:CD006483. doi:10.1002/14651858.CD006483.pub2.17. Civljak M, Sheikh A, Stead LF, Car J. Internet-based interventions for smokingcessation. Cochrane Database Syst Rev 2010;9:CD007078. doi:10.1002/14651858.CD007078.pub3.18. Griffiths KM, Farrer L, Christensen H. The efficacy of internet interventions fordepression and anxiety disorders: a review of randomised controlled trials.Med J Aust. 2010;192:S4–S11.19. Glass N, Eden KB, Bloom T, Perrin N. Computerized Aid Improves SafetyDecision Process for Survivors of Intimate Partner Violence. J InterpersViolence. 2010;25:1947–64.20. Eden KB, Perrin NA, Hanson GC, Messing JT, Bloom TL, Campbell JC, GielenAC, Clough AS, Barnes-Hoyt JS, Glass NE. Use of Online Safety Decision Aidby Abused Women. Am J Prev Med. 2015;48:372–83.21. Koziol-McLain J, Vandal AC, Nada-Raja S, Wilson D, Glass NE, Eden KB,McLean C, Dobbs T, Case J. A web-based intervention for abused women:the New Zealand isafe randomised controlled trial protocol. BMC PublicHealth. 2015;15:56.22. Hegarty K, Tarzia L, Murray E, Valpied J, Humphreys C, Taft A, Gold L, Glass N.Protocol for a randomised controlled trial of a web-based healthy relationshiptool and safety decision aid for women experiencing domestic violence (I-DECIDE). BMC Public Health. 2015;15:736.23. Ford-Gilboe M, Wuest J, Merritt-Gray M. Strengthening capacity to limitintrusion: Theorizing family health promotion in the aftermath of womanabuse. Qual Heal Res. 2005;15:477–501.24. Davies L, Ford-Gilboe M, Willson A, Varcoe C, Wuest J, Campbell J, Scott-Storey K. Patterns of Cumulative Abuse Among Female Survivors of IntimatePartner Violence Links to Women’s Health and Socioeconomic Status.Violence Against Women. 2015;21:30–48.25. Ford-Gilboe M, Wuest J, Varcoe C, Davies L, Merritt-Gray M, Hammerton J,Wilk P, Campbell J. Modelling the effects of intimate partner violence andaccess to resources on women’s health in the early years after leaving anabusive partner. Soc Sci Med. 2009;68:1021–9.26. Wuest J, Ford-Gilboe M, Merritt-Gray M, Wilk P, Campbell JC, Lent B, VarcoeC, Smye V. Pathways of chronic pain in survivors of intimate partnerviolence: Considering abuse-related injury, symptoms of posttraumaticstress disorder, depressive symptoms, and child abuse. J Women’s Heal.2010;19:1665–74.27. Wuest J, Ford-Gilboe M, Merritt-Gray M, Berman H. Intrusion:The central problem for family health promotion among childrenand single mothers after leaving an abusive partner. Qual Health Res.2003;13:597–622.28. Ford-Gilboe M, Merritt-Gray M, Varcoe C, Wuest J. A theory-based primaryhealth care intervention for women who have left abusive partners. ANSAdv Nurs Sci. 2011;34:198–214.29. Wuest J, Merritt‐Gray M, Dubé N, Hodgins MJ, Malcolm J, Majerovich JA,Scott‐Storey K, Ford‐Gilboe M, Varcoe C. The process, outcomes, andchallenges of feasibility studies conducted in partnership with stakeholders:a health intervention for women survivors of intimate partner violence. ResNurs Health. 2015;38:82–96.30. Varcoe C, Brown AJ, Ford-Gilboe M, Stout MD, McKenzie HA, Price R,Bungay V, Skye V, Inyallie J, Day L, Khan KB, Heino A, Merritt-Gray M.Reclaiming Our Spirits: Development and pilot testing of a healthpromotion intervention for Indigenous women who have experiencedintimate partner violence. RINAH (In Press).31. Eysenbach G, CONSORT-EHEALTH Group: CONSORT-EHEALTH: improvingand standardizing evaluation reports of Web-based and mobile healthinterventions. J Med Internet Res. 2011;13:e126. doi:10.2196/jmir.1923.32. Laughon K, Renker P, Glass N, Parker B. Revision of the abuse assessmentscreen to address nonlethal strangulation. JOGNN - J Obstet GynecolNeonatal Nurs. 2008;37:502–7.33. McFarlane J, Parker B, Soeken K, Bullock L. Assessing for abuse duringpregnancy: Severity and frequency of injuries and associated entry intoprenatal care. JAMA. 1992;267:3176–8.34. Glass NE, Perrin NA, Hanson GC, Bloom TL, Messing JT, Campbell JC, Case J,Eden K. The longitudinal impact of an internet safety decision aid for abusedwomen. Am J Prev Med. 2017. http://dx.doi.org/10.1016/j.amepre.2016.12.014.35. Eden KB, Perrin NA, Hanson GC, et al. Use of online safety decision aid byabused women: Effect on decisional conflict in randomized controlled trial.Am J Prev Med; 2015:48(4):372–83. doi:10.1016/j.amepre.2014.09.027.36. Campbell JC, Webster DW, Glass N. The Danger Assessment: Validationof a Lethality Risk Assessment Instrument for Intimate Partner Femicide.J Interpers Violence. 2008;24:653–74.37. Glass N, Perrin N, Hanson G, Bloom T, Gardner E, Campbell JC. Risk for reassaultin abusive female same-sex relationships. Am J Public Health. 2008;98:1021.38. Ford-Gilboe M, Varcoe C, Noh M, Wuest J, Hammerton J, Alhalal E, BurnettC. Patterns and predictors of service use among women who haveseparated from an abusive partner. J Fam Violence. 2015;30(4):419–31.doi:10.1007/s10896-015-9688-8.39. Devries KM, Mak JY, Bacchus LJ, Child JC, Falder G, Petzold M, Astbury J,Watts CH. Intimate Partner Violence and Incident Depressive Symptoms andSuicide Attempts: A Systematic Review of Longitudinal Studies. PLoSMedicine. 2013;10:e1001439. doi:10.1371/journal.pmed.1001439.40. Eaton WW. Center for Epidemiologic Studies Depression Scale: Review andrevision (CESD and CESD-R). In The use of psychological testing fortreatment planning and outcomes assessment. 3rd edition. Edited byMaurish ME. Mahwah, NJ: Lawrence Erlbaum Associates, Inc; 2004.41. Van Dam NT, Earleywine M. Validation of the Center for EpidemiologicStudies Depression Scale - Revised (CESD-R): Pragmatic depressionassessment in the general population. Psychiatry Res. 2011;186:128–32.42. Weathers FW, Huska JA, Keane TM: The PTSD Checklist—Civilian Version(PCL-C) Available from FW Weathers National Center for PTSD, BostonVeterans Affairs Medical Center 150 S. West, M, Livesley, WJ, Reiffer, L,Sheldon, A(1986) place Attach life events Model Stress illness Can JPsychiatry 1991, 31:202207.43. Wilkins KC, Lang AJ, Norman SB. Synthesis of the psychometric propertiesof the PTSD checklist (PCL) military, civilian, and specific versions. DepressAnxiety. 2011;28:596–606.44. Blanchard EB, Jones-Alexander J, Buckley TC, Forneris CA. Psychometricproperties of the PTSD Checklist (PCL). Behav Res Ther. 1996;34:669–73.45. O’Connor A. User Manual – Decisional Conflict Scale. Ottawa, Canada:Ottawa Hospital Research Institute; 1993.46. O’Connor AM. Validation of a decisional conflict scale. Med Decis Mak. 1995;15:25–30.47. O’Connor AM. Decisional Conflict Scale. Ottawa: Loeb Health ResearchInstitute; 1999.48. McFarlane J, Parker B: Preventing abuse during pregnancy: an assessment andintervention protocol. MCN Am J Matern Child Nurs 1994, 19:321–324 4p.49. Goodman L, Dutton MA, Weinfurt K, Cook S. The Intimate Partner ViolenceStrategies Index. Violence Against Women. 2003;9:163.50. Varcoe C, Hankivsky O, Ford-Gilboe M, Wuest J, Wilk P, Hammerton J,Campbell J. Attributing Selected Costs to Intimate Partner Violence in aSample of Women Who Have Left Abusive Partners: A Social Determinantsof Health Approach. Can Public Policy. 2011;37:359–80.51. Nolen-Hoeksema S, Larson J, Grayson C. Explaining the gender difference indepressive symptoms. J Pers Soc Psychol. 1999;77:1061–72.Ford-Gilboe et al. BMC Public Health  (2017) 17:273 Page 11 of 1252. Turner HA, Pearlin LI, Mullan JT. Sources and Determinants of Social Supportfor Caregivers of Persons with AIDS. J Health Soc Behav. 1998;39:137–51.53. Brady T. Measures of self-efficacy, helplessness, mastery, and control: TheArthritis Helplessness Index (AHI)/Rheumatology Attitudes Index (RAI),Arthritis Self-Efficacy Scale (ASES), Children’s Arthritis Self-Efficacy Scale(CASE), Generalized Self-Efficacy Scal. Arthritis Care Res (Hoboken). 2003;49:S147–64.54. Smith PH, Earp JA, DeVellis R. Measuring battering: development of theWomen’s Experience with Battering (WEB) Scale. Women’s Heal (Hillsdale,NJ). 1995;1:273.55. Smith PH, Tessaro I, Earp JL. Women’s experiences with battering:A conceptualization from qualitative research. Women’s Heal Issues. 1995;5:173–82.56. Smith PH, Smith JB, Earp JAL. Beyond the Measurement Trap AReconstructed Conceptualization and Measurement of WomanBattering. Psychol Women Q. 1999;23:177–93.57. Hegarty K. Composite abuse scale manual. Melbourne Department ofGeneral Practice, University of Melbourne, Australia. 2014.58. Hegarty K, Bush R, Sheehan M. The composite abuse scale: furtherdevelopment and assessment of reliability and validity of amultidimensional partner abuse measure in clinical settings. ViolenceVict. 2005;20:529–47.59. Hegarty K, Sheehan M, Schonfeld C. A multidimensional definition ofpartner abuse: development and preliminary validation of the CompositeAbuse Scale. J Fam Violence. 1999;14:399–415.60. Ford-Gilboe M, Wathen CN, Varcoe C, MacMillan HL, Scott-Storey K, MantlerT, Hegarty K, Perrin N. Development of a brief measure of intimate partnerviolence experiences: the Composite Abuse Scale (Revised)-Short Form(CASR-SF). BMJ Open. 2016;6:e012824.61. Mccarrier K, Bushnell D, Martin M, Paczkowski R, Nelson DR, Buesching D:Validation and Psychometric Evaluation of a 5-item Measure of PerceivedSocial Support. In ISPOR 16th Annual International Meeting. Baltimore,Maryland; 2011.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Ford-Gilboe et al. BMC Public Health  (2017) 17:273 Page 12 of 12


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items