Peer- and Mentor-Enhanced Web-Based Training on Substance Use Disorders: A Promising Approach in Low-Resource Settings

To build capacity in managing substance use disorders in primary care, we developed two online-based training courses with peer activities and local mentorship for two audiences: the Substance Use Screening Course for Lay Healthcare Workers (“screening course”) and the Substance Use Disorders Management Course for Primary Care Practitioners (“management course”). These courses follow evidence-based recommendations for online learning by health care professionals. Key competencies outlined in the mhGAP and screening and brief intervention manuals linked with WHO’s Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) (2, 5, 9) were addressed by curating existing, high-quality, online materials based on these competencies.

The screening course is tailored to community health care workers in public facilities and registration officers or receptionists in private facilities. It consists of four substance use disorder–related modules: mental health and substance use services and screening for substance use disorders; substance use disorder–related stigma and determinants and how they can be addressed; introduction to interventions for substance use disorders in urgent situations or with common comorbidities; and communication strategies with people seeking care, their families, and caregivers (including the stigma-related issues of confidentiality and discussing sensitive topics). The main distinction between this training and the management course is the depth of training in clinical management offered; the management course adds modules on best practices for substance use disorder management in primary care. This includes training on the full brief intervention from the ASSIST package and on other related psychiatric and general medical conditions (e.g., depression, suicidality, psychosis, HIV, and liver and lung diseases) as per the mhGAP and similar programs from WHO (and other institutions).

Beyond these didactic components, both training courses include peer and mentored activities, discussion forums, and role-playing exercises. Each trainee is assigned a local mentor (a trained psychologist or physician) to support his or her learning. Communications between mentors and course registrants can be by phone, e-mail, or video links; through the online learning platform; or in person. Mentors answer trainees’ questions on course materials and clinical practice, demonstrate clinical and communication skills, and evaluate trainees’ competencies—e.g., by discussing students’ reflections and grading a standardized patient-simulated encounter. The achievement of competencies is also assessed through formative quizzes for each module and a summative test (final exam) at the end of each course.

Change in stigma was assessed by applying t tests to the change in pre- and postcourse scores; scores were obtained from a modified version of a validated scale for health workers’ stigma toward those with mental illnesses (10). Trainees also completed a demographic survey, an online self-evaluation, and a course evaluation.

In the 14 health care facilities engaged in eDATA K in 2014, a total of 97 individuals registered to take either the screening (N=58) or the management (N=39) course. The overall completion rate for both courses was 50%. Sociodemographic characteristics and economic assets varied little between course completers and noncompleters, except that the noncompleters were more likely than the completers to personally own computers and modems. Participants in both the screening and management courses averaged high scores on the final exam (90% and 88%, respectively, with only one failure in the screening course). These high scores were achieved even though some participants in the final examination appeared to have completed only some of the modules. However, trainees reported sometimes completing modules together and logging into only one participant’s account, which rendered exact tracking impossible.

During the pilot and stakeholder engagement phase, we found through field observation, meeting notes, and focus groups that among Kenyan study participants and staff, many doubted the feasibility of the proposed program. Similar skepticism has been noted in most locales around the world when we begin this model of uniquely accredited and globally free training. Concerns were raised regarding the difficulties of imparting clinical skills through Web-based courses, despite the availability of local mentors. Others questioned the cultural appropriateness of the available online materials or the likelihood that untrained community health workers or administrative staff would be able to acquire the knowledge, skills, and change in attitude to effectively screen for substance use disorders. Others expected that Internet and computer literacy would be too low in this group of trainees. Therefore, all these factors were the object of specific qualitative research questions regarding acceptability and feasibility of the training. Through incorporation of stakeholder feedback during the pilot phase, the training courses were improved and shortened. At the implementation stage, qualitative data revealed that the previously expressed concerns did not emerge, or if the suspected challenges arose, they did not significantly affect the success of the trainees. More specifically, trainees found the material both clinically and culturally relevant, even though most materials were sourced from high-income countries. Trainees also indicated that the assembled material had a manageable level of complexity for new knowledge and skills acquisition.

Although all groups of participants expressed high levels of motivation to learn from the substance use disorder courses, primary care workers needed less support from study staff because of their previous experience with computers and patient care—and, for a minority, their past online-education experience. However, primary care workers, especially those in larger urban centers, had many other competing training and advancement opportunities, and a smaller proportion of them completed the NextGenU.org courses, compared with those in more rural settings.

Overall, in course evaluation, 92% of the respondents said that they would prefer to take a NextGenU.org-based course over a traditional classroom-based course, and 100% said that they would recommend NextGenU.org screening or management courses to their peers. The learners’ self-evaluations supported the positive findings of the course evaluation: 100% felt very confident that they had gained substantial knowledge of the subject matter, 97% in the screening course and 100% in the management course reported spending adequate time reviewing the online learning resources, and 100% of respondents felt comfortable learning independently online and reported that they learned new and useful skills.

Notably, many clinical settings, especially those in rural areas and urban regions outside the largest cities, had acquired computer equipment quite recently, and most had not been connected to the Internet; 40% of course completers had never used a computer before or had used one for less than 1 year.

Among the trainees who completed the stigma survey (N=99), pre- to posttraining changes in stigma scores were not statistically significant for those who did not complete the course; however, among course completers, scores measuring stigma toward people with tobacco use disorder (p=.049) and other substance use disorders (p=.048) decreased significantly, but scores measuring stigma toward those with alcohol use disorders did not.

Our findings contribute to the literature demonstrating that Web-based training is an efficient, cost-effective, and fast-emerging means of delivering education and can be as effective in decreasing stigma as the best-practice, in-person forms of training, with the advantage of being more efficient and accessible (6). The training was acceptable and feasible and was shown in our study to be effective in improving not only knowledge but also skills and attitudes across a wide range of ages, education levels, socioeconomic and professional backgrounds, facility types (from primary care centers to outpatient hospital departments), and locations (large urban centers, small urban areas, and rural settings). These results are critical because they contribute to reassuring stakeholders about the potential generalizability of this educational model across all these circumstances.

Furthermore, the demonstrated ability of computer-naïve trainees, whose education level was less than high school, to complete the courses and attain competency is crucial, because these workers constitute a large pool of previously untapped care providers. The training course had a transformational impact on many of these workers, who noted that they were now more valued by their community for performing previously unavailable and effective interventions. This finding of a change in how the course completers were perceived by others was supported by field observations as well as by testimonies collected in focus groups and interviews. Testimonies even supported a change in attitude in the broader community about the value of treating people who have substance use disorders. Therefore, our findings support an expansion of the role of community health workers to the area of substance use disorders and highlight the priority of addressing substance use disorders at the primary care level.

This model can be key in alleviating the large shortage of qualified health workers in most low- and middle-income countries and in low-resource settings in high-income countries. Importantly, our course completion rate of approximately 50% is much higher than the average completion rate of <8% for many Massive Open Online Courses (MOOCs) (7, 11), supporting the acceptability and feasibility of this model. Developed in 2001, which predates the development of MOOCs, this model is a DOOHICHE (a Democratically-Open Outstanding Hybrid of Internet-aided, Computer-aided, and Human-aided Education)—pronounced “doohickey,” and also known as a “gadget” (7). The NextGenU.org/DOOHICHE model is now being used by more than 7,000 registered users in 191 (of 193) United Nations member states and countries, offers the first globally free, accredited degree—a master’s degree in public health—and launched in August 2019. Current courses span the academic spectrum—from college-level pre–health sciences and community health worker training courses through graduate training in public health. A MedSchoolInABox curriculum that includes graduate medical education was codeveloped with Stanford University, University of Toronto, and University of Central Florida. The NextGenU.org courses (including these Kenya-tested courses) are competency based and recognized by a free certificate from partner universities, which is given to participants after they complete the course and are evaluated through online didactic components, participation in a Web-based global peer community of practice, and skills-based mentorship. Our accredited partners, North American universities that are outstanding in each particular course topic, give learners credit for this training (or institutions can adopt the courses and use them with their students)—all courses are accessible and advertisement free and carbon free through NextGenU.org at no cost. We have tested this free model with North American medical, public health, and undergraduate students and with community health workers and primary care physicians in India and published reports on its use. Participants report as much knowledge gain as with traditional courses and greater satisfaction, and by participating, they join a global community of practice.

We must increase health care workers’ substance use disorder–related knowledge and skills and decrease their substance use disorder–related stigmatizing attitudes to successfully implement evidence-based interventions for the vast populations without access to clinical care for substance use disorders. Therefore, it is critical to know that a Web-based, peer- and mentor-enhanced training program can address these aspects all at once, with an educational model that is cost-effective (free) and easily scalable and which has global application in all other didactic fields. Although more research can further inform the model’s impact in different contexts, this study supports the conclusion that freely available Web-based training in addiction treatment is effective, even in low-resourced settings, and leads to increases in knowledge and skills and decreases in stigma that are similar to those seen in current best-practice, contact-based education models, which are costly or impossible to scale (6, 7).