Introduction

As radiology continuously advances in subspecialty knowledge, training programs around the world are stretched to adequately cover radiology fundamentals while also providing in-depth subspecialty training. Radiology programs in developing countries face a wide variety of barriers, but a lack of resources such as diagnostic imaging equipment and a scarcity of fellowship programs and subspecialty mentors appear to be a common theme. Within Eastern Africa specifically, a 2017 study surveying 13 countries found that only 2 offered radiology subspecialty fellowship training [1]. A possible reason for the lack of subspecialized physicians in the region is physician attrition. Between 2004 and 2014, Sub-Saharan Africa had one of the highest intensities of physician migration [2]. The lack of subspecialized training in radiology constrains the development of trainees and remains a long-term barrier to the advancement of healthcare. As progress towards permanent subspecialty training programs is made, alternative tools that provide general radiologists and radiology residents access to subspecialty teaching and mentorship are necessary to fill the gap.

Medical educational programs delivered remotely to resource-limited countries have been shown to improve educational outcomes [3,4,5]. Strategies for implementing remote education are diverse and should be tailored to the unique constraints of the region and field of study [6]. Access to the internet in Eastern Africa can occasionally be unreliable and costly and so standard online resources that radiology residents might use, such as online question banks or educational videos, may not be readily accessible. However, the widespread use of smartphones and in particular, the social networking platform WhatsApp, provide a unique opportunity to expand access to educational resources and radiology subspecialty experts. In recent years, WhatsApp has emerged as a popular messaging service due to its cost-free ability to send messages and videos across countries, its compatibility with multiple smartphone types and the reliability of the application (app) in the face of intermittent internet connections [7, 8]. It has been shown that WhatsApp can serve as a platform to connect students with educators and facilitate reliable access to educational materials in certain educational settings [7]. Delivering pediatric radiology case studies through WhatsApp may help radiology residents overcome technological limitations in their education by providing an interactive and asynchronous means of learning directly from subspecialized radiologists. This unconventional educational format may allow the delivery of tailored learning material and facilitate conversations in which residents can ask questions and refine their knowledge. The purpose of this pilot program is to assess WhatsApp as a platform for delivering educational material to residents in Kenya, Rwanda, Tanzania and Uganda and to solicit the feedback of residents regarding the usefulness of WhatsApp as an educational tool.

Methods

Participants were recruited via contacts in Kampala, Uganda (through the Imaging the World Consortium for Health), Rwanda (through Human Resources for Health Initiative Clinton Foundation Grant) and Tanzania (through the international consortium of physicians, nurses, technologists and trainees, Road2IR). These not-for-profit organizations are physician training programs designed to improve access to radiology education. Inclusion criteria required the participant to be a radiology resident at one of the following hospitals; Aga Khan University or Moi University in Kenya, Makerere University at Mulago Hospital in Uganda, Central Hospital University of Kigali in Rwanda or Muhimbili National Hospital in Tanzania. Physicians at the universities recruited participants through word of mouth.

The study was designed by K.D., the president and co-founder of Imaging the World, an American Board of Radiology certified radiologist with 28 years of experience. The questionnaires were designed by J.M., a physician with a background in sociology and 4 years of experience in general radiology. The content was provided by V.R., an American Board of Radiology certified radiologist with Certificate of Added Qualification in Pediatric Radiology and 23 years of experience in pediatric radiology. The module was administered by L.R. and E.R., a fourth year radiology resident and a fourth year undergraduate student,  respectively. Questions raised by residents were answered by the pediatric radiologist (V.R.).

Institutional review board approval was not required and informed consent was obtained from all participants before the module began. Residents were organized into four WhatsApp group chats based on country of origin. The module administrator, radiologist and organizing resident physician from the respective country were also included in the group chats. All module information, questionnaires and case studies were delivered through these chats. An introductory message that described the program and a pre-module questionnaire were distributed to each group chat. The questionnaire contained Likert scale questions that gauged familiarity with WhatsApp, current scholarship engagement and access to resources (Table 1). Participants copied the questionnaire message, filled in their answers and responded in a direct message back to their module administrator. Then participants received a weekly multiple choice case study question in a portable document format (PDF) (Supplementary material 1). Each case was presented as an unknown case with a brief clinical history, multiple conventional radiographs or ultrasound images and a multiple choice question with four choices. Participants viewed the cases on their personal smartphones and were encouraged to use any resources available to answer the question, including conferring with their peers. They then sent their answer along with their level of confidence in a direct message to their module facilitator (Fig. 1). If the resident responded within the week, they individually received a PDF from the module facilitator explaining the correct and incorrect answers supplied by the module administrator. Each answer contained supplemental imaging such that they learned about four distinct pathologies from one case. A total of 12 case studies were administered over 12 weeks, exposing residents to 48 distinct pediatric radiology cases. After each question and a 1-week period that allowed everyone time to answer, residents were able to ask questions or solicit feedback from the pediatric radiologist either privately, through the module facilitator, or directly in their respective group chat (Fig. 1). At the end of the study, participants completed a final questionnaire about their experience that contained both Likert (Table 2) and free response questions (Table 3). Over the course of the module, the facilitator recorded participants’ answers, confidence levels and questionnaire responses in an excel spreadsheet. As an incentive for full participation, if a resident responded to all case questions and completed both questionnaires, they received a Certificate of Completion awarding 12 Continuous Professional Development points.

Table 1 Pre-module questionnaire
Full size table
Fig. 1
figure 1

a Example of a response provided by a participant. b Example of a clarifying question asked by a participant

Full size image
Table 2 Post-module questionnaire, Likert answers
Full size table
Table 3 Post-module questionnaire, free response answers
Full size table

Likert scale question responses were reported as averages. Answers to free response questions were categorized based on response content and reported numerically [9]. A McNamar’s chi-square test was used to compare comfort level with reading pediatric radiology cases before and after the module. A Spearman rank correlation was used to determine the correlation between number of correct responses and reader confidence. Statistical analysis was conducted using R version 4.2.2. (Bell Laboratories, Murray Hill, NJ). Statistical significance was set at P≤ 0.05

Results

Of the 72 residents initially recruited, 40 (56%) completed the pre- and post-module questionnaires and answered all 12 pediatric radiology study questions within the allotted answer period. In total, 8 Tanzanian, 9 Rwandan, 13 Ugandan and 10 Kenyan residents received the Certificate of Completion for their full participation.

Tables 1, 2 and 3 show the average of responses to the pre- and post-module questionnaires and free response answers, respectively.

Participants reported their comfort with reading pediatric radiology studies in the pre- and post-module questionnaires; 22 of 40 participants (55%) reported little or no comfort with reading pediatric radiology studies prior to the module, and then reported being very comfortable after the module (P=0.027) (Table 4).

Table 4 The McNamar’s chi-square test demonstrated a statistically significant difference in participant’s comfortability with reading pediatric radiology studies pre- and post-intervention
Full size table

For each multiple choice question, participants rated their confidence on a 4-point Likert scale with 1 representing not confident and 4 representing very confident. The number of participants who answered incorrectly was plotted against the average confidence of all participants. There was a significant negative correlation between the number of incorrect answers and the average confidence of participants (R =  − 0.74, P = 0.0057). As the number of incorrect answers to a question increased, the average confidence decreased (Fig. 2).

Fig. 2
figure 2

Number of correct answers compared to average confidence in answer

Full size image

The number of questions a participant answered correctly was plotted against year in residency training (Fig. 3). There was no significant correlation between a participant’s year of residency training and the number of questions they answered correctly (R = 0.16, P = 0.32).

Fig. 3
figure 3

Correlation between the year of residency training and number of correct answers (each circle represents an individual resident)

Full size image

Discussion

WhatsApp can be used as an interactive asynchronous pediatric radiology learning platform in Eastern Africa given the widespread use of this technology. This method of delivery may be used as an adjunct to current teaching methods. As an educational tool for learning pediatric radiology, participants perceived the modules positively.

Over half of the residents who participated in this module remained engaged throughout the 12 weeks. There was a high, yet gradual attrition rate throughout module administration which might be attributed to the requirement of submitting answers within a week of receiving the question. If a participant did not complete the answer in time, they were no longer included in the study results and did not receive a certificate, although they were allowed to continue participating. We expect that future renditions of this module will have much higher participation rates as participants will not be required to answer every question each week to receive the certificate.

We sought to evaluate residents’ current exposure level to pediatric imaging to establish the necessity of supplemental pediatric radiology education programs. In the pre-module questionnaire, participants reported that they see pediatric radiology cases nearly every day (average 4.33 on a 1–5 scale) averaging around 100–500 cases per month (2.32 on a 1–5 scale). They infrequently engage in activities such as lecturing in radiology and mentoring radiology students (1.4 on a 1–5 scale and 2.35 on a 1–5 scale respectively) (Table 1). The high volume of pediatric cases seen reflects the overall younger demographic in Eastern Africa [10] and highlights the need for greater emphasis on pediatric radiology subspecialty teaching. In response to the question, “What obstacles do you face in learning pediatric radiology?” 19 participants reported a lack of specialty expertise (Table 2). These findings suggest that residents do not get the necessary exposure and training needed to interpret pediatric studies in their general radiology residency, underscoring the need for expanding pediatric radiology education.

To determine the effectiveness of this module, we sought to evaluate if there was a perceived impact on residents’ comfort with pediatric radiology and to determine if WhatsApp is a suitable platform for providing educational content. There was a statistically significant increase in comfort level with reading pediatric radiology studies after participation in the module, with 55% of participants moving from the little to no comfort category to the comfortable category (Table 3). Pediatric radiology is typically a 1-year fellowship training program. As such, this educational module provided only very limited exposure (totaling 48 cases) and should not have had such a significant impact on a resident’s ability to read pediatric imaging studies. The significant increase in perceived comfort with reading studies could likely be attributed to response bias. The novelty of module content and exposure to pediatric radiologists could have led residents to inflate the reported effect on their comfort level.

It is noteworthy that 7.5% of participants felt less comfortable after participating in the module; this may be explained by an improved self-awareness of their skill level in pediatric radiology in this relatively small number of participants. As expected, if a question was more challenging, more participants got it wrong and reported less confidence in their answer on average (Fig. 2). This indicates that most participants were aware of their skill level to a certain extent. It is interesting to note that there was no correlation between the year of residency training and number of correct answers (Fig. 3). The lack of correlation could be due to the open access to resources which leveled the playing field. It also indicates that all participants benefited from the module since everyone answered questions incorrectly and subsequently learned from their mistakes.

In free response, nearly half of the participants, 18/40, reported that they would like to be shown more cases, reflecting their awareness that increased exposure would improve their comfort with pediatric radiology imaging. Additionally, participants found the content to be useful for their learning (3.78 on a 1–4 scale). They also found that it was moderately challenging (2.23 on a 1–4 scale) (Table 2). These findings support that this educational strategy was perceived positively by the participants.

The post-module questionnaire also contained questions that evaluated the convenience of WhatsApp as a platform for receiving educational content. It was hypothesized that WhatsApp might be particularly suitable in the face of intermittent computer internet access, yet residents reported their internet access to be reliable (3.95 on a 1–5 scale). This discrepancy is likely explained by the residents interpreting this question to mean their smartphone internet access is reliable as opposed to computer internet access which is known to be unreliable in the region.

The app proved convenient, as participants reported that they use WhatsApp nearly every day (4.98 on a 1–5 scale). In the post module questionnaire, they confirmed that WhatsApp is a convenient tool for learning pediatric radiology (3.38 on a 1–4 scale). They also thought the images were clear enough to provide an accurate diagnosis (3.73 on a 1–4 scale) and considered the supplemental explanations helpful (3.8 on a 1–4 scale) (Table 2). These findings demonstrate that WhatsApp is a user-friendly format for learning.

Lastly, we were interested in receiving feedback on how to improve the module for future iterations. In total, 18 participants independently reported they would like to see an increase in the number of case-based questions, 5 participants suggested using a systems-based approach, 4 indicated they would like the feedback to be timelier, 4 requested the open discussion of the case questions and 4 others suggested including additional resources such as reading lists or videos in the explanation for further study.

Limitations to this study have been identified and the following alterations could be made for future iterations of this module. Questions should be delivered every other day to increase the frequency and allow participants to evaluate the imaging and compose their response. The duration of the module should also be increased. Module content should be organized by organ system to ensure logical flow and exposure to multiple subspecialty areas. We considered adding videos to supplement the explanations, but sending videos over WhatsApp is not feasible as data transmission is limited to under 16 MB per transmission. Due to the time-consuming nature of individually sending explanations, these PDFs should be sent once in the groupchat after adequate time has been provided for participants to answer. Participants would then be able to have the option to openly discuss the case with their peers. Board-certified pediatric radiologists would provide additional insight during the discussion.

Future iterations of the module should include as many residents as possible to increase regional exposure to subspecialty content. The expansion of the program may be impeded by the availability of board-certified mentors, yet considering that a group chat can have up to 50 participants and one mentor participate in multiple chats, this is unlikely to be a serious limitation. The program will remain case-based, asynchronous and interactive and could be used as a supplement to existing educational resources such as in-person education and textbooks.

One of the greatest strengths of this delivery format is the opportunity for interactive education. Participants were able to seek written feedback for clarification. Additionally, participants use the app daily for social and professional communication. The frequency of use of a messaging app like WhatsApp makes it surpass other forms of communication such as email. Cellphones are readily accessible and WhatsApp allows residents to easily and seamlessly integrate a case question into their day. While this delivery system is in its early stages of development, for its improvement, the focus should be on increasing its scope and organization. Delivering a total of 12 image-based questions does not begin to address the scope of the case content for subspecialty training and therefore this method does not supplant the need to have in-person education. This point is especially true for pediatric radiology, since modalities such as fluoroscopy and ultrasound are learned through hands-on experience.

There is limited availability of subspecialist radiology training programs in African countries primarily due to funding, equipment, expertise, emigration and politics [11]. While absolutely necessary, establishing subspecialty programs is a very complex task. A pediatric radiology curriculum that can be shared across all of Eastern Africa could provide an immediate and substantial benefit to regions that do not have established fellowship programs. An interactive program that shares curated cases designed to develop diagnostic reasoning skills and expose residents to subspecialty mentorship surpasses traditional didactic materials such as textbooks and websites. The accessibility of a WhatsApp-based educational system also skirts barriers such as computer availability and internet costs.

Establishing these group chats also facilitates conversations, relationships and collaborations between students and experts, which may lead to further education within Eastern African countries if experts were to travel to the region. The ultimate goal is to train the next generation of pediatric radiologists in order to establish accredited pediatric radiology fellowship programs, as has been done in Ethiopia [12]. Many subspecialty fellowship programs have been successfully established in countries all across Africa. With barriers such as funding and the coronavirus disease 2019 pandemic, onsite visits by faculty mentors have been limited. Keys to program success include local university involvement, host country training sites and strong collaboration with international institutions to help coordinate volunteer faculty and staff [13]. A continual drive for higher education and dedicated professionals is needed to advance training in subspecialty areas of radiology. This pilot program is a convenient and useful tool for distributing pediatric radiology educational content throughout radiology residency programs in Eastern Africa.