Digital Education of Health Professionals on the Management of Domestic Violence: Systematic Review and Meta-Analysis by the Digital Health Education Collaboration

Introduction

Domestic violence (also referred to as family violence) is a complex public health problem [1] that places a notable burden on the health care system [2]. The World Health Organization (WHO) defined domestic violence, an aggressive and oppressive form of interpersonal violence, as a situation where an individual uses control tactics to emotionally, physically, sexually, or economically abuse a family member or past/current romantic partner [3]. Forms of control behavior can include but are not limited to psychological, physical, sexual, financial, and emotional abuse [4].

The WHO commissioned a multinational study on domestic violence with data collected from 10 countries, which showed that 13%-61% of women between the ages of 15-49 years had experienced physical abuse from their intimate partners at least once in their lifetime [3]. However, due to various reasons including shame, embarrassment, social stigma, and fear of and dependency on the abuser, survivors are often unwilling to reveal their difficulties to others [5,6]. Domestic violence can have both short- and long-term effects on the mental and physical well-being of the survivors. Injuries and physical ailments resulting from prolonged exposure to domestic violence include chronic neurological disorders, cardiovascular diseases, respiratory, intestinal and digestive conditions, reproductive disorders, physical injuries, and even death [7-10]. The less visible, but equally detrimental, impact of domestic violence includes psychological and emotional sufferings, anxiety, fear, depression, and posttraumatic stress [11]. Survivors often require treatment and care from a spectrum of health professionals, ranging from family physicians to physical therapists or clinical psychologists [5].

By facilitating early detection and treatment through a well-structured system of education and support, health professionals could play an important role in promoting greater awareness on domestic violence, identifying survivors of violence, and enabling survivor protection [12]. As health professionals are usually the first line of contact for survivors, they would need to undergo special training to identify, support, and treat domestic violence patients. This would be of particular importance to conservative societies where domestic violence is known to be severely underreported [13,14]. Moreover, the immense shortage of health professionals worldwide, especially in developing countries [15], compounded by the lack of training of health professionals, poses immense challenges in tackling the global domestic violence crisis [13,16].

Occasional training programs on domestic violence consisting of seminars and workshops often claim poor retention rates, as they are mostly time-consuming [17], require health professionals to travel to training locations, and are taught by academics who themselves may have had little exposure to people who have experienced domestic violence [18,19]. With the increasing use of information communication technologies in health professions’ education, leveraging on digital education to provide domestic violence management education could help address the various challenges of training and manpower shortage while improving the cost-effectiveness of educational programs [20-22].

The term digital education refers to a range of teaching and learning strategies that utilize digital media and devices for training and as interaction tools [23]. Digital education can be further subclassified into various types according to delivery methods (online or offline), content, learning objectives, pedagogical approaches, and delivery settings [24]. The use of the internet to deliver content is referred to as the online mode of digital education, while the use of software or PowerPoint without the need for the internet to deliver content is referred to as the offline mode of digital education. With its scalability, flexibility, cost-effectiveness, and ability to overcome geographical and temporal constraints, digital education has the potential to provide more independent, customized, and accessible domestic violence training. Studies comparing digital education to traditional methods in various specialties including medical education and engineering have found digital education to be more efficient and effective in building knowledge [25,26].

Although a previous study reviewed domestic violence education among health professionals [19], digital education has been gaining popularity in health professions’ curricula and hence its use in domestic violence education should be studied further. To the best of our knowledge, there are no systematic reviews evaluating the effectiveness of digital health interventions specifically for domestic violence training among health professionals. Hence, the objective of this study was to evaluate the effectiveness of health professions’ digital education on domestic violence compared to that of traditional ways or no intervention.

Methods

Search Strategy

We followed the Cochrane Handbook guidelines for this review. A more detailed description of the methodology is provided in the paper by Car et al [27]. This review is part of a global evidence-synthesis initiative for digital health professions’ education [28-39]. The search for the relevant trials was conducted across 7 databases: MEDLINE (Ovid), EMBASE (Elsevier), the Cochrane Central Register of Controlled Trials (Wiley), PsychINFO (Ovid), Educational Resource Information Centre (Ovid), Cumulative Index to Nursing and Allied Health Literature (EBSCO), and Web of Science Core Collection (Thomson Reuters). The detailed search strategy for MEDLINE is presented in the Multimedia Appendix 1. A manual search was conducted to identify any relevant articles from the reference lists of all included articles. A search was also conducted in the International Clinical Trials Registry Platform Search Portal and the metaRegister of Controlled Trials to identify unpublished trials, meeting abstracts, and doctoral theses from Jan 1990 to August 2017.

Eligibility Criteria

The inclusion criteria are presented in Textbox 1. We adopted a broad definition of domestic violence, encompassing all subcategories of domestic violence, to capture a wide range of studies on the topic.

We included randomized controlled trials (RCTs) and quasi-randomized trials reporting the use of digital education interventions (including blended learning, which is a combination of conventional learning and digital education) to educate health professionals on domestic violence management. RCTs with and without control groups that received traditional interventions delivered by either health professionals or university personnel were included. Studies targeting both practicing health professionals and students were included in this review. No language restrictions were imposed. All digital education interventions were included. Cross-over studies were excluded due to the high likelihood of carry-over effects [40]. Non-RCTs and studies not focusing on computer-based interventions and interventions delivered to individuals other than health professionals were also excluded.

Study Selection

The search results from all the databases were combined in a single Endnote X8 library (Clarivate Analytics, Philadelphia, PA), and all duplicate records were removed. Search filters were used to remove articles not related to digital education for health professionals. Two reviewers (UD and NN) then independently screened the titles, abstracts, and full-text articles to identify studies potentially meeting the inclusion criteria. Disagreements were resolved through discussion between the reviewers. Primary outcomes included knowledge, skills, attitudes, self-efficacy, and satisfaction with the education measured using any validated and nonvalidated instruments. Secondary outcome measures included patient outcomes (eg, feedback from domestic violence survivors seeking treatment), change in health professionals’ behavior (ie, health professionals’ confidence in and ease of identifying and treating domestic violence survivors), and economic impact of the intervention.

Data Extraction

All the relevant data including study characteristics, type of digital education intervention, participant demographics, data for outcome measures, and other publication details were extracted independently by UD and NN using a structured data extraction form. We contacted one study author (Short LM) [43] for missing information.

Risk of Bias Assessment and the Overall Quality of Evidence

UD and NN independently assessed the risk of bias using the Cochrane Collaboration’s risk of bias tool [44]. When it was unclear if a trial was of low or high risk, the field was coded as unclear risk of bias. The following domains were evaluated: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessors, incomplete outcome data, selective outcome reporting, and other bias. The following GRADE (Grading of Recommendations Assessment, Development and Evaluation) criteria for evaluating the overall quality of evidence were used: limitations of studies (risk of bias), inconsistency (heterogeneity), indirectness, imprecision, and publication bias [45].

Data Synthesis

Postintervention mean and SDs were used. The baseline mean value was used to calculate the final posttest mean and SD in studies that presented change scores rather than the final mean. When the studies compared more than two groups, the results from the comparison of the least active control group and the most active intervention group were presented.

Statistical Analysis

We pooled the data using the random-effect model and calculated standardized mean differences (SMDs) with 95% CIs. Statistical heterogeneity across studies was assessed using the Cochran Q test and I² statistics (negligible: 0%-40%, moderate: 30%-60%, or substantial: 50%-90% heterogeneity) [44]. All statistical analyses were conducted using RevMan software (version 5.3; The Nordic Cochrane Centre, Copenhagen, Denmark).

Results

The searches generated a total of 30,073 references. Following abstract and title screening, 144 articles were found to be relevant to domestic violence and selected for full-text screening. Of those, six met our eligibility criteria (Figure 1).

‎View this figure

Study Characteristics

All the 6 included RCTs were published between 2000 and 2014 in high-income countries: 5 were from the United States [43,46-49] and 1 was from the Netherlands [50]. All the studies included were parallel RCTs. Three of the studies were conducted in a university setting, and the remaining three were conducted in community and hospital settings. Two of the studies [49,50] focused on child abuse; one, on intimate partner violence [43], and three, on domestic violence, in general [46-48]. In addition, three studies targeted dental professionals [46,48,49], two targeted physicians [43,47], and one targeted nurses [50].

A total of 631 participants were included in the six studies, of which 420 participants (66%) were dentists and dental students. Three studies used offline and three studies used online modes of delivering digital education intervention. The following primary outcomes were reported: knowledge [43,46-49], attitudes [43,46-48], self-efficacy [43,47,50], and skills [50]. Surveys, questionnaires, and checklist were used to measure these outcomes, of which only two instruments [43,47] were validated. The duration of the intervention varied between 15 minutes and 3 weeks. Tables 1 and 2 summarize the main characteristics of the included studies. The study by Shapiro [49], including second-year dental students, on recognizing child abuse was the only study comparing a digital education intervention with traditional lecture-based learning, whereas all the other studies compared digital education to no intervention.

Effects of Interventions

Meta-analysis of five studies [43,46-49] considered to be sufficiently homogeneous found that digital education (offline and online) may increase knowledge of domestic violence in dentists, physicians, and allied health professionals (510 participants; SMD 0.67, 95% CI 0.38-0.95; I²=59%; low certainty evidence) compared with no intervention and traditional learning postintervention. There was evidence of moderate heterogeneity among the studies (ζ²=0.06; χ²₄=9.7; P=.05; I²=59%; Figure 2).

Meta-analysis of three studies [43,46,48] found that compared to no intervention, digital education (offline and online) may increase postintervention attitude toward domestic violence management in dentists and physicians (339 participants; SMD 0.67, 95% CI 0.25-1.09; I²=68%). There was a substantial level of heterogeneity among the studies (ζ²=0.09; χ²₂=6.3; P=.04; I²=68%).

Meta-analysis of three studies [43,47,50] found that compared to no intervention, digital education (offline and online) may increase postintervention self-efficacy toward domestic violence management in physicians and nurses (174 participants; SMD 0.47, 95% CI 0.16-0.77; I²=0%). The was no evidence of heterogeneity (ζ²=0.00; χ²₂=0.7; P=.71; I²=0%).

One study [50] comparing change of score in skills found that digital education (offline program) may improve domestic violence skills in nurses (25 participants; SMD 0.94, 95% CI 0.11-1.77) compared to no intervention.

‎View this figure

Summary Risk of Bias

Of the six studies, four [43,46,48,50] were found to have an overall low risk of bias and the remaining two [47,49] had a high or an unclear risk of bias.

The random sequence generation method was reported in four [43,46,48,50] of the six studies. Blinding and protection against selective reporting was achieved through the nature of the intervention and the reporting of all the results in all the studies. Attrition and other biases were of low risk for five [43,46,48-50] of the six studies. One study had a high risk of attrition bias resulting from a high drop-out rate (42%). However, details of allocation concealment were not reported in any of the studies, and blinding of outcome assessment was attempted in only one study [50]. Similarly, the method for random sequence generation was not clearly stated in two studies [47,49]. At the individual-study level, of 56 domains, 14 (25%) were reported as unclear and one (2%) was reported as high risk (Figure 3). The summary of findings table shows the evidence to be of low to moderate quality as analyzed per the GRADE criteria (Table 3).

‎View this figure

Discussion

In this paper, we systematically reviewed and pooled data on the use of digital education for domestic violence management. We draw attention to the gap in digital education on domestic violence and the potential benefits of this educational strategy. Our findings provide preliminary evidence to show that using digital education to address socially sensitive issues such as domestic violence may improve certain educational outcomes in health professionals receiving the training. Although competencies, trustable professional activities, knowledge, skills, and attitudes do not automatically translate into change of practice, they are indispensable for improving patient outcomes.

All the studies included in this review were published after the year 2000, which is the period when digital education started becoming more prominent in health care professions’ education [51] and more laws were implemented to tackle domestic violence [5,52,53].

Interestingly, we found that in all the studies, the intervention groups had improved knowledge, skills, attitudes, and self-efficacy, even though the studies employed different methodologies, sample sizes, sampling periods, settings, and types of domestic violence education. Additionally, the changes in primary outcomes were observed within short time periods of up to 2 weeks after the intervention, with only one study [43] measuring retention at the 12-month follow-up. Although the variability suggests that digital education has the versatility to reach a wide range of health professionals in different populations and settings, it underscores the potential of homogeneous short-term digital interventions in improving the quality of care that these professionals provide.

Risk of bias was mostly unclear for blinding of outcome assessment and allocation concealment, but it was mainly low for sequence generation. While the nature of the interventions does not allow blinding of participants, we believe it would not have had any effect on bias risk. We minimized biases by having two reviewers independently assess the articles for inclusion, complete data extraction, risk of bias, and use of the GRADE criteria. The overall quality of the evidence was low or moderate due to the risk of bias and inconsistency across the studies (Table 3).

This review has some important strengths including a strict adherence to the gold-standard Cochrane methods and use of validated, comprehensive, and reproducible searches across seven databases. Our review adds to previous research on domestic violence education for health professionals, as it focuses on the use of digital education, which is a growing area of research. Some weaknesses have to be kept in mind when interpreting the results of this systematic review. For instance, although our searches were comprehensive, we cannot be certain that all relevant trials were included.

However, the evidence evaluated has some limitations. First, only a few studies were published in this area, and they were all from high-income countries, making generalizability challenging. Although digital education may potentially serve as an effective and impactful solution to educating health professionals in domestic violence management, applicability, scalability, and implementation in low- and middle-income countries have to be studied further [54]. Second, we acknowledge that in certain countries such as the United Kingdom, social workers are the first “line of response” to domestic violence. Third, only two studies [47,49] measured and reported learners’ satisfaction as one of the primary outcomes. This further highlights the need for uniform and validated outcomes and methods of measuring them to make conclusive judgements. Moreover, only two studies [43,47] used validated measurement instruments to measure outcomes, thereby making it challenging to compare the use of digital education between settings. Subsequently, the lack of data on retention rate, costs, or patient outcomes prevents policy makers from making informed decisions or assessing the transferability of digital education to other settings. Finally, none of the RCTs reported secondary outcomes such as patient outcomes, health professionals’ behavior change, and economic impact. Hence, we are unable to assess how these outcomes changed with digital education.

Future studies should be designed to evaluate the effect of digital education on these outcomes. We further recommend that future studies consider including other professionals such as social carers, psychologists, counsellors, or teachers. Findings of this review suggest that digital education could contribute to developing the competencies that health professionals need to respond to complex psychosocial problems such as domestic violence. Therefore, future studies should focus on recording more practical outcomes of the trainings such as change in detection and referral rates [55]. This will help ensure a better understanding of the actual value of integrating digital education modules into pre- and postregistration as well as the continuing professional development curricula. In addition, although domestic violence is more accepted and prevalent in low- and middle-income countries, the education gap is wide and digital education is still at the developing stage in these countries [14,56]. Data should be collected beyond geographical regions with inclusion of the cost analysis to obtain a better understanding of the impact and feasibility of integrating electronic learning modules on domestic violence management into the medical curriculum in low- and middle-income countries. We further recommend that future studies be designed with larger, appropriately powered RCTs, in both developed and low- and middle-income countries alike in order to ensure better representation. Researchers could use methods such as the Consolidated Standards of Reporting Trials for Social and Psychological Interventions 2018 (CONSORT-SPI 2018) checklist as guidance to design and report future studies on digital education for domestic violence in order to ensure that the data collected are of high quality and representative [57].

We believe that while digital education could help increase identification of and support to patients experiencing domestic violence, research with study designs incorporating blended learning might hold the highest potential. Such designs would combine the best of digital education, such as smartphones, apps, emails, text messages, and virtual patients, with the best of traditional classroom practices such as personalized contact or feedback, meetings, and discussions.

In conclusion, we found some promising, predominantly low-quality evidence for the effectiveness of digital education on domestic violence. We also highlighted the need for further research evaluating and validating culturally tailored digital education interventions geared toward more holistic management of domestic violence.