For Peer Review Only Journal: BMJ Open

BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

Perception and use of massive open online courses among medical students of a developing country: multicenter cross-sectional study

For peer review only Journal: BMJ Open

Manuscript ID: bmjopen-2014-006804

Article Type: Research

Date Submitted by the Author: 01-Oct-2014

Complete List of Authors: Aboshady, Omar; Faculty of Medicine, Menoufia University, 6th Year Medical Student Radwan, Ahmed; Faculty of Medicine, Menoufia University, 6th Year Medical Student Eltaweel, Asmaa; Faculty of Medicine, Alexandria University, 6th Year Medical Student Azzam, Ahmed; Faculty of Medicine, Al-Azhar University in Cairo, 6th Year Medical Student Aboelnaga, Amr; Faculty of Medicine, Tanta University, 5th Year Medical Student Hashem, Heba; Faculty of Medicine, Beni Suef University, 6th Year Medical Student Darwish, Salma; Faculty of Medicine, Suez Canal University, 4th Year http://bmjopen.bmj.com/ Medical Student Salah, Rehab; Faculty of Medicine, Benha University, Intern Kotb, Omar; Faculty of Medicine, Assiut University, 5th Year Medical Student Afifi, Ahmed; Faculty of Medicine, Ain Shams University, 4th Year Medical Student Noaman, Aya; Faculty of Medicine, Cairo University, 5th Year Medical Student Salem, Dalal; Faculty of Medicine, Cairo University, 6th Year Medical Student on September 24, 2021 by guest. Protected copyright. Hassouna, Ahmed; Faculty of Medicine, Ain Shams University, MD, Department of Cardiothoracic Surgery

Primary Subject Medical education and training Heading:

Secondary Subject Heading: Medical education and training

Computer-Assisted Instruction , Medical Education , Distance Education , Keywords: MOOCs, Egypt

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 1 1 2 3 4 Title Page 5 6 7 Title 8 9 10 Perception and use of massive open online courses among medical students of a 11 12 13 developing country: multicenter cross-sectional study 14 15 Authors For peer review only 16 17 18 - Omar Aboshady 19 20 6th year medical student, Faculty of Medicine, Menoufia University, Shebin El-Kom, Menoufia, Egypt. 21 22 [email protected] 23 24 25 - Ahmed E. Radwan 26 27 6th year medical student, Faculty of Medicine, Menoufia University, Shebin El-Kom, Menoufia, Egypt. 28 29 [email protected] 30 31 - Asmaa R. Eltaweel 32 33

th http://bmjopen.bmj.com/ 34 6 year medical student, Faculty of Medicine, Alexandria University, Alexandria, Egypt. 35 36 [email protected] 37 38 - Ahmed Azzam 39 40 6th year medical student, Faculty of Medicine, Al-Azhar University in Cairo, Cairo, Egypt. 41 42 on September 24, 2021 by guest. Protected copyright. 43 [email protected] 44 45 - Amr A. Aboelnaga 46 47 5th year medical student, Faculty of Medicine, Tanta University, Tanta, Egypt. 48 49 [email protected] 50 51 52 - Heba A. Hashem 53 th 54 6 year medical student, Faculty of Medicine, Beni Suef University, Beni Suef, Egypt. 55 56 [email protected] 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 2 1 2 3 - Salma Y. Darwish 4

5 rd 6 3 year medical student, Faculty of Medicine, Suez Canal University, Ismailia, Egypt. 7 8 [email protected] 9 10 - Rehab Salah 11 12 Intern, Faculty of Medicine, Benha University, Benha, Egypt. 13 14 15 [email protected] peer review only 16 17 - Omar N. Kotb 18 19 th 20 5 year medical student, Faculty of Medicine, Assiut University, Assiut, Egypt. 21 22 [email protected] 23 24 - Ahmed M. Afifi 25 26 3rd year medical student, Faculty of Medicine, Ain Shams University, Cairo, Egypt. 27 28 29 [email protected] 30 31 - Aya M. Noaman 32 33 th

5 year medical student, Faculty of Medicine, Cairo University, Cairo, Egypt. http://bmjopen.bmj.com/ 34 35 [email protected] 36 37 38 - Dalal S. Salem 39 40 6th year medical student, Faculty of Medicine, Cairo University, Cairo, Egypt. 41 42 [email protected] on September 24, 2021 by guest. Protected copyright. 43 44 45 - Ahmed Hassouna 46 47 MD, Department of Cardiothoracic Surgery, Faculty of Medicine, Ain Shams University, Cairo, Egypt. 48 49 [email protected] 50 51 52 Corresponding author: 53 54 Omar Ali Aboshady 55 56 th 57 6 year medical student, 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 3 1 2 3 Faculty of Medicine, Menoufia University. 4 5 6 Address: 20 Sadat School St, Shanawan, Shebin El-kom, Menoufia, Egypt. 7 8 Tel: +2-048-2282698 / +2- 01010747627 9 10 11 E-mail: [email protected] 12 13 Fax: +2-048-2326810 14 15 Postal code: 32718For peer review only 16 17 18 19 20 Key Words: 21 22 23 Computer-Assisted Instruction (MeSH terms); Medical Education (MeSH terms); Distance Education 24 25 (MeSH terms); MOOCs; Egypt. 26 27 28 Word Count: 29 30 - Title: 18 words (114 characters) 31 32 - Abstract: 294 words 33 34 http://bmjopen.bmj.com/ - Text: 3335 words 35 36 37 - Number of figures and tables: 5 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 4 1 2 3 4 ABSTRACT 5 6 7 8 9 Objectives: To primarily assess the prevalence of awareness and use of massive open online courses 10 11 12 (MOOCs) among medical undergraduates in Egypt as a developing country, besides identifying the 13 14 limitations and satisfaction of using these courses. 15 For peer review only 16 Design: A multi-center, cross-sectional study using a web-based, pilot-tested and self-administered 17 18 19 questionnaire. 20 21 Settings: Ten randomly selected medical schools in Egypt. 22 23 24 Participants: Randomly selected 2700 undergraduate medical students with an equal allocation of 25 26 participants in each university and each study year. 27 28 Primary and secondary outcomes measures: The primary outcome measures were the percentages of 29 30 31 students who knew about MOOCs, students who enrolled and students who obtained a certificate. 32 33 Secondary outcome measures included the limitations and satisfaction of using MOOCs through 5- 34 http://bmjopen.bmj.com/ 35 point Likert scale questions. 36 37 38 Results: Of 2527 eligible students, 2106 filled the questionnaire (response rate 83.3%). Of these 39 40 students, 456 (21.7%) knew the term MOOCs or websites providing these courses. Out of the latter, 41 on September 24, 2021 by guest. Protected copyright. 42 136 students (29.8%) had enrolled in at least one course, but only 25 (18.4%) of them completed 43 44 45 courses earning certificates. Clinical years’ students showed significantly higher rates of knowledge 46 47 (P= .009) and enrollment (P< .001) than academic year students. The primary reasons for incompletion 48 49 50 of courses included lack of time (105; 77.2%) and slow internet speed (73; 53.7%). Of the 25 students 51 52 who completed courses, 21 (84%) were satisfied with the overall experience. However, there was less 53 54 satisfaction regarding student-instructor (8; 32%) and student-student (5; 20%) interactions. 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 5 1 2 3 Conclusions: Approximately one-fifth of Egyptian medical undergraduates have heard about MOOCs 4 5 6 with only about 6.5% actively enrolled in courses. However, students who actively participated showed 7 8 a positive attitude toward the experience, but better time management skills and faster internet 9 10 11 connection speeds are required. Further studies are needed to address the enrolled students for a better 12 13 understanding of their experience. 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 6 1 2 3 4 5 6 STRENGTHS AND LIMITATIONS OF THIS STUDY 7 8 9 10 11 12 - This study is the first to assess the actual prevalence awareness and use of MOOCs in Egypt and in 13 14 the medical field. 15 For peer review only 16 - This study included a large representing sample of ten Egyptian institutions covering nearly the entire 17 18 19 geographic area of Egypt. 20 21 - Data obtained from students in all six undergraduate years. 22 23 24 - There were relatively low returned number of participants who enrolled and who had certificates, 25 26 which makes analysis of limitations and satisfactions less reliable. 27 28 - The study results can not be generaziable to all developing countries. 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 7 1 2 3 4 INTRODUCTION 5 6 7 8 Massive open online courses (MOOCs) have been recently proposed as a disruptive innovation with 9 10 1 11 high expectations to solve challenges facing higher education. The idea behind MOOCs is to offer 12 13 world-class education to a (massive) number of students around the globe with internet access (online) 14 15 for little or no feesFor (open). Thepeer courses consist review of prerecorded videoonly lectures, computer-graded tests and 16 17 2 18 discussion forums to talk over course materials or to get help. These courses have gained immense 19 20 popularity over a short period, attracting millions of participants and crossing the barriers of location, 21 22 gender, race and social status; making 2012 the year of MOOCs according to NewYork journal.3 23 24 25 Coursera, the largest MOOCs provider, in its lastest infograph in October 2013 showed an 26 27 extraordinary growth reaching more than 100 institutional partners, more than 500 courses and more 28 29 than 5 million students.4 30 31 32 33 In medical education, the number of related MOOCs is steadily increasing. In a recent study in 34 http://bmjopen.bmj.com/ 35 2014, it was found that 98 free courses were offered during 2013 in the fields of health and medicine 36 37 5 38 with an average length of 6.7 weeks. These courses were introduced as a possible solution that may 39 40 help solving great challenges facing medical education nowdays.6 These challenges including the issues 41 on September 24, 2021 by guest. Protected copyright. 42 of quality, costs and the ability to deliver education to enough students who will cover the health care 43 44 7 45 system’s needs. There are uprising discussions to determine which roles MOOCs can play in the 46 47 medical field. Despite that, there is still limited information about how medical students perceive such 48 49 50 courses, especially in the developing countries where high-quality learning is often scarce. 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 8 1 2 3 Although there is a great hope that MOOCs can play a role in solving developing countries’ lack of 4 5 6 high quality education, the current demographic data reveal that most of the MOOCs’ participants are 7 8 from the developed countries with very low participation rates from low-income countries, especially in 9 10 4 11 Africa. These low rates were thought to be due to various complicated conditions, such as lack of 12 8 13 access to digital technologies, linguistic and cultural barriers and low computer skills. In addition, lack 14 15 of awareness of Forthe presence peer of this newly reviewintroduced concept mayonly be considered another problem. 16 17 18 19 To our knowledge, there are no available cross-sectional studies that assessed the actual prevalence 20 21 of awareness and use of MOOCs among medical communities in the developing countries, including 22 23 24 Egypt. Our study primarily aims to assess the prevalence of awareness and use of these courses among 25 26 Egyptian undergraduate medical students, as an example of a developing country. Second, the study 27 28 will assess the limitations that prevent students to enroll and complete courses, besides assessing the 29 30 31 satisfaction level of using MOOCs to better understand of the role these courses in medical education. 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 9 1 2 3 4 METHODOLOGY 5 6 7 8 This is a multi-center, cross-sectional study utilizing a structured, web-based, pilot-tested and self- 9 10 11 administered questionnaire. The study was ethically approved by the institutional review board at 12 13 Faculty of Medicine, Menoufia University, Egypt. 14 15 For peer review only 16 Study Population and Sample 17 18 19 20 Our target population was undergraduate medical students in Egypt enrolled in 19 medical schools 21 22 for the academic year 2013/14. We randomly selected ten medical schools to be our study settings. 23 24 25 These were Ain Shams, Al-Azhar medical school in Cairo, Alexandria, Assiut, Benha, Beni Suef, 26 27 Cairo, Menoufia, Suez Canal, and Tanta medical schools. 28 29 30 31 Students in these schools are enrolled in a six-year MBBCh program, in which the first three years 32 33 are called academic years and the last three years are called clinical years. According to a confidence 34 http://bmjopen.bmj.com/ 35 interval (CI) of 99%, margin of error 3%, and response distribution of 50%; 1784 students were 36 37 38 required to represent the study population. We used a stratified simple random technique to select our 39 40 sample with an equal allocation of participants in each university and each study year. Accordingly, 41 42 on September 24, 2021 by guest. Protected copyright. 43 using the registered students’ names lists, we randomly selected 270 students from each faculty (45 for 44 45 each study year) for a total of 2700 participants. We excluded non-Egyptians students and those who 46 47 changed their enrollment school at the time of data collection. 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 10 1 2 3 Data collection 4 5 6 7 We invited the selected participants via e-mail and social media websites to complete our survey 8 9 using a unique code for each participant during the period of March–April 2014. We used an online 10 11 12 survey program to administer the questionnaire (SurveyGizmo; Boulder, Colorado, US). Students who 13 14 did not have access to the internet at the time of data collection were allowed to record their responses 15 For peer review only 16 using a self-administered paper version of the questionnaire. We sent up to five reminder messages for 17 18 19 participants to complete the survey. The participants were informed about the study aims in the cover 20 21 letter, and they voluntarily consented to participate with no incentives. 22 23 24 25 Questionnaire Development 26 27 28 The study questionnaire was developed by the research team through group discussions after an 29 30 31 extensive literature review. The draft was then reviewed by two experts in the fields of medical 32 33 education and Biostatistics. We used the final draft in a pilot testing on 175 students in all participating 34 http://bmjopen.bmj.com/ 35 medical schools. Detailed feedback about the format, clarity and completion time were collected and 36 37 38 we made minor changes in response to participants’ comments. We did not include the pilot responses 39 40 in our analysis. 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 The questionnaire was in Arabic, the participants’ native language, and it comprised 29 questions in 45 46 four sections using branching logic function (Figure 1). The first section addressed study aims, consent 47 48 and participants’ personal information. This section was followed by a main question asking about their 49 50 51 knowledge about MOOCs. Based on this answer, participants were directed to different sections. 52 53 Students who knew about MOOCs were asked how they heard about it and their state of enrollment. If 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 11 1 2 3 the participant was not enrolled in any course, he/she was asked about the limitations, and then the 4 5 6 questionnaire ends. 7 8 9 Enrolled students were directed to the next section, which assessed their perspectives and experience 10 11 12 with MOOCs. For students who gained certificates, further questions were asked regarding their level 13 14 of satisfaction as well as any obstacles they might have faced. Finally, four questions were addressed to 15 For peer review only 16 assess students’ opinion about integration of MOOCs in the medical field. 17 18 19 20 Most of the questions were single answer multiple-choice questions. However, there were three 21 22 multi-select check-box questions. For assessment of limitations, satisfaction and opinions, a 5-point 23 24 25 Likert scale between 1 (strongly agree/satisfied) and 5 (strongly disagree/unsatisfied) was used. 26 27 28 Statistical analysis: 29 30 31 32 Results were presented as numbers and perecentages with confidence interval at 99%. The 33 http://bmjopen.bmj.com/ 34 significance of the association between qualitative variables of interest was analyzed using Chi-square 35 36 37 test or Fisher’s exact test, as indicated. In order to focus on clear opinions, the 5-point Likert scale of 38 39 limitations, satisfaction and opinions were collapsed into 3 categories (agree/satisfied, neutral, and 40 41 disagree/unsatisfied). Class year was recoded as a dichotomous variable in order to compare results for 42 on September 24, 2021 by guest. Protected copyright. 43 44 students in academic versus clinical education. The acknowledgment of the importance of getting a 45 46 certificate before enrollment was also recoded as a dichotomous variable (important/very important 47 48 versus limited importance/not important) in order to test the significance of association between the 49 50 51 primarily reported importance of acquiring a certificate and the actual possession of the certificate by 52 53 McNemar test. All tests were bilateral and a P value of 0.01 was the limit of statistical significance. 54 55 56 Statistical analysis was performed using the IBM SPSS statistical software package version 22. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 12 1 2 3 4 RESULTS 5 6 7 8 Respondent characteristics 9 10 11 12 Of 2700 total participants, 62 (2.3%) were excluded being non-Egyptians or having changed their 13 14 enrollment school, in addition to 111 (4.1%) students’ whose contact information could not be reached 15 For peer review only 16 with final eligible 2527 students. During the data collection phase, 2357 (93.3%) online questionnaire 17 18 19 invitations and 170 (6.7%) paper versions were sent out. Out of these distributed questionnaires, 2016 20 21 responses were received (response rate 83.3%). Table 1 show participants’ demographics regarding 22 23 24 school, class and gender. 25 26 27 Knowledge about MOOCs 28 29 We found that 456 (21.7% [99% CI, 19.4%–24%]) students had heard about MOOCs or websites 30 31 32 providing such courses. There was no statistically significant difference in knowledge between males 33 http://bmjopen.bmj.com/ 34 and females (43.6% vs 56.4%, 99 CI, P = .8). However, clinical year students had higher rates of 35 36 37 knowledge than students in the academic years (P< .001) (Table 1). Additionally, there was no 38 39 difference between medical schools in students’ knowledge about MOOCs (P=.04). 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 After clarifying the concept of MOOCs to students who did not know about it, 1342 (81.3% [99% 44 45 CI, 78.8%–83.8%]) students showed an interest to participate with a significant difference among 46 47 different medical schools (P< .001). 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 13 1 2 3 Table 1. Participants’ demographics and their state of knowledge, enrollment and certificate 4 5 6 attainment. 7 8 9 Knowledge about MOOCs P Enrollment in courses P Certificate Attainment P 10 value value value 11 Total (%) Yes (%) No (%) Total Yes (%) No (%) Total Yes (%) No (%) 12 (n=2106) (n=456) (n=1650) (n=456) (n=136) (n=320) (n=136) (n=25) (n=111) 13 Faculty Ain Shams 207 38 169 38 13 25 13 3 10 14 15 (9.8%) For(18.4%) (81.6peer %) review(34.2%) (65.8%) only (23.1%) (76.9%) 16 Al-Azhar 216 42 174 42 11 31 11 1 10 17 (10.3%) (19.4%) (80.6%) (26.2%) (73.8%) (9.1%) (90.9%) 18 Alexandria 222 48 174 48 19 29 19 4 15 19 (10.5%) (21.6%) (78.4%) (39.6%) (60.4%) (21.1%) (78.9%) 20 Assuit 180 33 147 33 6 27 6 2 4 21 (8.5%) (18.3%) (81.7%) (18.2%) (81.8%) (33.3%) (66.7%) 22 Benha 205 57 148 57 16 41 16 0 16 23 (9.7%) (27.8%) (72.2%) (28.1%) (71.9%) (0.0%) (100.0%) 24 Beni Suef 220 38 182 P= 38 6 32 P= 6 0 6 P= 25 (10.4%) (17.3%) (82.7%) .04 (15.8%) (84.2%) .13 (0.0%) (100.0%) .02 26 Cairo 188 39 149 39 12 27 12 2 10 27 (8.9%) (20.7%) (79.3%) (30.8%) (69.2%) (16.7%) (83.3%) 28 Menoufia 248 53 195 53 22 31 22 10 12 29 (11.8%) (21.4%) (78.6%) (41.5%) (58.5%) (45.5%) (54.5%) 30 31 Suez 199 59 140 59 20 39 20 2 18 32 Canal (9.4%) (29.6%) (70.4%) (33.9%) (66.1%) (10.0%) (90.0%) 33 Tanta 221 49 172 49 11 38 11 1 10 34 (10.5%) (22.2%) (77.8%) (22.4%) (77.6%) (9.1%) (90.9%) http://bmjopen.bmj.com/ 35Class Academic 1076 176 900 176 40 136 40 4 36 36 (51.2%) (16.4%) (82.6%) P< (22.7%) (77.3%) P= (10.0%) (90.0%) P= 37 Clinical 1024 280 744 .001 280 96 184 .01 96 21 75 .1 38 (48.8%) (27.3%) (72.7%) (34.3%) (65.7%) (21.9%) (78.1%) 39 Gender Male 926 199 730 199 71 128 71 17 54 40 41 (44.1%) (21.4%) (78.6%) P= (35.7%) (64.3%) P= (23.9%) (76.1%) P= 42 Female 1174 257 920 .83 257 65 192 .02 65 8 57 .08 on September 24, 2021 by guest. Protected copyright. 43 (55.9%) (21.8%) (78.2%) (25.3%) (74.7%) (12.3%) (87.7%) 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 14 1 2 3 Enrollment and certificate attainment 4 5 6 7 Of those who knew about MOOCs, 136 (29.8% [99% CI, 24.3%–35.3%]) were enrolled in at least 8 9 one course. Most students (125; 91.9%) registered in 1–5 courses, while only 113 (83.1%) student 10 11 12 reported watching at least one video lecture. Home (109; 99%) was the first place where they watched 13 14 these videos. There was no statistically significant difference in enrollment state between males and 15 For peer review only 16 females (52.2% vs 47.8%, 99% CI, P= .016). However, there was a significant difference between 17 18 19 students’ class and their enrollment (P=.009) (Table 1). Coursera was the most commonly used website 20 21 (99; 72.8%), followed by Edx (14; 10.3%). 22 23 24 25 Only 25 students (18.4% [99% CI, 9.8%–26.9%]) completed courses and attained one certificate or 26 27 more with 81.6% dropout rate. Interestingly, more than half of students who earned certificates (13; 28 29 52% [99% CI, 26.3%–77.7%]) have verified them from the universities that proposed the courses. The 30 31 32 vast majority of enrolled students assumed that getting a certificate is important to them (32 [23.5%] 33 http://bmjopen.bmj.com/ 34 very important, 37 [27.2%] important, 50 [36.8%] important to some extent, and 17 [12.5%] not 35 36 37 important). Out of the 69 students who assumed that getting a certificate is important before enrollment 38 39 (important/very important), 17 were finally certified (24.6%); compared to only 8 certified students out 40 41 of the 67 who were not concerned with having certificates (important to some extent/not important; 42 on September 24, 2021 by guest. Protected copyright. 43 44 11.9%); P< .001. 45 46 47 Ways of knowledge and students’ motivations 48 49 50 To assess how students knew about MOOCs and what were their motivations, two multi-select 51 52 questions were addressed. Social media was the main way through which 206 (45.2%) students knew 53 54 about MOOCs, while knowledge through a friend was the second (184; 40.4%). Using web-search 55 56 57 engines (87; 19.1%) was in the third place, followed by extracurricular activities (46; 10.1%). MOOCs 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 15 1 2 3 providers’ advertisements played a very small role (27; 5.9%) in reaching students as did medical 4 5 6 schools’ official websites (15; 3.3%). Notably, there was no association between the ways through 7 8 which student learned about MOOCs and their enrollment. Nevertheless, students who knew through 9 10 11 extracurricular activities were found to enroll more frequently (P= .005). 12 13 14 Concerning students’ motives, most students reported that their main motivation was “to learn new 15 For peer review only 16 things” followed by “to help me studying medicine” (Figure 2). Interestingly, the students who enrolled 17 18 19 aiming to have a certificate or to help them in obtaining a future job were significantly more likely to 20 21 complete the courses (P= .001) and (P= .008), respectively. 22 23 24 25 MOOCs and Medicine 26 27 By asking the enrolled students (n=136) about their experience and attitude toward medical MOOCs, 28 29 103 (75.7% [99% CI,66.2%–85.2%]) declared participation in at least one medical course. Of them, 24 30 31 32 students (17.6% [99% CI, 7.9%–27.3%]) had completed medical courses and earned certificates. 33 http://bmjopen.bmj.com/ 34 Regarding their medical MOOCs experience, 102 (75%) students agreed that MOOCs helped them in 35 36 37 developing their theoretical background about the topic discussed. However, there was less agreement 38 39 (68; 50%) on the role of MOOCs in developing their practical skills. Most students (89; 86.4%) agreed 40 41 that MOOCs help in studying medicine, while 83 (61%) believed that MOOCs will help them in getting 42 on September 24, 2021 by guest. Protected copyright. 43 44 a better future job opportunity. 45 46 47 Limitations of MOOCs 48 49 50 51 Our study reported two types of limitations: enrollment and completion. Students who knew about 52 53 MOOCs, but did not enroll in any courses (n=320) were asked about their enrollment limitations. The 54 55 56 majority of students (226; 70.4%) agreed that lack of time was the main limitation, while 147 (45.9%) 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 16 1 2 3 agreed that slow internet speed was another cause (Figure 3). Regarding completion limitations, the 4 5 6 enrolled students (n=136) were asked to assess the limitations that made them drop out of courses. 7 8 Similar to the enrollment limitations, it was obvious that lack of time (105; 77.2%) and slow internet 9 10 11 speed (73; 53.7%) were the main obstacles. While lack of technology access, computer literacy, 12 13 language difficulty and culture conflicts had less agreement on their roles as limitations (Figure 14 15 3). Only 16 (11.8%)For students peer agreed that thereview scientific content wasonly difficult. In addition, 93 (68.4%) 16 17 18 students disagreed that “low content than expected” is to be a limitation. 19 20 21 For further assessment of the internet speed, we asked the enrolled students to rate their internet 22 23 24 speed. Sixty students (44.1%) reported that the speed was reasonable while 55 (40.4%) reported slow 25 26 speed and only 21 (15.4%) had a higher connection speed. When we compared the students’ evaluation 27 28 of internet speed and if they watched video lectures or not, we did not find a significant association (P= 29 30 31 .69). 32 33 http://bmjopen.bmj.com/ 34 Students’ satisfaction of MOOCs 35 36 37 38 The 25 students who obtained certificates were asked to report their opinions about each part of the 39 40 MOOCs experience. The results showed that most students (21; 84%) were satisfied with the overall 41 42 on September 24, 2021 by guest. Protected copyright. 43 experience, including video lectures (18; 70%), exams and assignments (16; 64%), quality of the 44 45 presented materials (21; 84%), and the technology used (20; 80%). However, there was less satisfaction 46 47 regarding student–student (5; 20%) and student–instructor (8; 32%) interactions (Figure 4). 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 17 1 2 3 4 DISCUSSION 5 6 7 8 Available information about MOOCs participants is data obtained from course-end demographics, 9 10 11 which usually reports heterogeneous populations of different age groups and educational levels from 12 13 different countries globally. These data show that most MOOCs’ users are well-educated males with 14 15 low participationFor from developing peer countries review and undergraduates. only9-11 To our knowledge, this study is the 16 17 18 first in the medical field and in one of the developing countries to use a cross-sectional study design in 19 20 a homogeneous population for assessment of prevalence and uptake of such courses among 21 22 undergraduate medical students. 23 24 25 26 Knowledge and Enrollment 27 28 29 Our results show a funnel-shaped participation pattern, with 22.7 % of the respondents knowing 30 31 32 about MOOCs and 6.5% actually enrolled. Moreover, only 5.4% watched videos and 1.2% obtained 33 http://bmjopen.bmj.com/ 34 certificates. Although there are no similar cross-sectional studies with which our results can be 35 36 37 compared, the knowledge of about one-fifth of the Egyptian medical students about MOOCs is 38 39 considered promising in a developing country that depends mainly on regular education. Additionally, 40 41 these courses are still new and there was little role of MOOCs providers’ advertisements for reaching 42 on September 24, 2021 by guest. Protected copyright. 43 44 students beside that there is no any medical MOOC which is given by an Egyptian institution till now. 45 46 Social media and personal experience transfer among friends played a vital role in the spreading of the 47 48 MOOCs’ idea, raising students’ awareness to this level. This is in line with the uprising role of social 49 50 51 media websites in medical students’ life with more than 90% of medical students in the US using social 52 53 media.12 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 18 1 2 3 Notably, it was obvious that there is a disproportion between knowledge about MOOCs and 4 5 6 enrollment with only one-third of students having the awareness registered in courses. The students 7 8 reported lack of time and low internet speed as the main limitations. From these enrolled students, 9 10 11 18.4% (23.3% for medical courses) completed courses and earned certificates. These completion rates 12 13 are higher than the reported average completion rates in the course demographics. In 2013, The 14 15 Chronicle of HigherFor Education peer suggested anreview average of 7.5% completion only rate 13, while a recent study in 16 17 14 18 2014 reported a rate of about 6.5%. This may be explained by the reported importance of certificates 19 20 for students to add to their resumes hoping for better future chances. It was interesting to note that 21 22 about half of them paid money to verify their certificates, although there is no academic credit for 23 24 15 25 undergraduates for any MOOCs from any medical school in the US and Egypt till now. 26 27 28 Although there was no association between gender and students’ knowledge or enrollment, class had 29 30 31 a significant association. Clinical year students were found to have higher knowledge and enrollment 32 33 rates. This may be due to the high stress and pressures experienced by first years medical students 34 http://bmjopen.bmj.com/ 35 adapting new systems with little time available for extracurricular activities.16 In contrast, final year 36 37 16-18 38 students were reported to have less stress with more attention to their career plans by searching for 39 40 new learning channels to increase their competitiveness. 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 MOOCs and Medicine 45 46 47 Of the enrolled students, 75.7% participated in at least one medical course with 23.3% completion 48 49 50 rate. They strongly agreed that these courses helped them develop theoretical background about the 51 52 topics discussed with less agreement on their role in developing their clinical skills. This raises 53 54 questions about the effectiveness of MOOCs with the current lecture-based teaching style in covering 55 56 57 the different aspects of medical education, including its clinical part, which needs student–patient 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 19 1 2 3 interaction. However, in the new evolving era of online learning, a question arises: “why to waste 4 5 6 precious class time on a lecture?” Students may watch the instructor’s lecture remotely in their homes 7 8 and utilize class time for learning clinical skills.19 Most of the current opinions expect a complementary 9 10 11 role of MOOCs in undergraduate education with an increasing role in educating those students after 12 15 13 their graduation in continuing medical education. 14 15 MOOCs limitationsFor in Egypt peer review only 16 17 18 19 Lack of time and slow internet speed were the two main limitations reported for causing low 20 21 MOOCs enrollment and course completion rates. MOOCs, being a self-learning educational system, 22 23 24 require a considerable amount of time to choose courses, watch videos, take exams and interact through 25 26 discussions. This imposes burden on students, leading to the need of increased commitments beside 27 28 their busy regular medical education. Time management, either in the design of courses or from 29 30 31 participants, is critically needed to enhance their performance and increase completion rates. 32 33 http://bmjopen.bmj.com/ 34 Low internet speed is a commonly reported problem facing online education in developing 35 36 20 37 countries. This problem prolongs the time needed to watch high-quality videos or to download course 38 39 content, rendering students less adherent and more susceptible to dropout. The main solution to this 40 41 problem is enhancing the internet infrastructure in Egypt. Liyanagunawardena et al. suggested allowing 42 on September 24, 2021 by guest. Protected copyright. 43 44 lower resolution versions of the videos as an alternative solution to help engaging students with limited 45 46 bandwidth.8 Interestingly, we did not find computer literacy, language or culture as barriers, although it 47 48 was expected that they would represent problems in Egypt, being a developing country. 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 20 1 2 3 MOOCs experience satisfaction 4 5 6 7 Encouragingly, most of the participants who completed MOOCs (n= 25) were satisfied with the 8 9 overall experience. However, there was an obvious dissatisfaction regarding student–student and 10 11 12 student–instructor interactions. This problem is in pervasive in online education in general, with a lack 13 14 of face-to-face interaction leading to some feelings of isolation and disconnectedness, which are 15 For peer review only 16 thought to be two main factors in dropout rates.21 Some MOOCs providers such as Coursera support 17 18 19 efforts beside the usual discussion forums for overcoming this point. These include more peer 20 21 assessments, social media groups, Google hangouts and real in-person Meetups. Despite that, more 22 23 24 involvement of participants is still needed to ensure the full psychological presence. 25 26 27 Study strengths and limitations 28 29 The strength of our study is that it included participants from all study years in 10 institutions, 30 31 32 covering nearly the entire geographic area of Egypt with high confidence interval (99%) and high 33 http://bmjopen.bmj.com/ 34 response rate (83.3%). However, our main limitation was the relatively low returned number of 35 36 37 participants who enrolled (n=136) and who had certificates (n=25), which makes analysis of limitations 38 39 and satisfactions less reliable. However, these results are important as a first start to make an evidence 40 41 about the real prevalence of MOOCs in Egypt to help the future studies to bluid upon and take more 42 on September 24, 2021 by guest. Protected copyright. 43 44 representative samples to the students who knew about MOOCs for a better understanding of their 45 46 experience. 47 48 49 50 Conclusions: 51 52 53 Approximately one-fifth of undergraduate medical students in Egypt have heard about MOOCs. 54 55 56 Students who actively participated showed a positive attitude toward the experience, but better time 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 21 1 2 3 management skills and faster internet connection speeds are required. Furthor studies are needed to 4 5 6 address the enrolled students to assess their experience in large representative samples. In addition, 7 8 more efforts are needed to be done to raise the awareness of students of such courses as most of 9 10 11 students who did not hear about MOOCs, showed interest to participate. 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 22 1 2 3 4 STATEMENTS: 5 6 7 8 Acknowledgements: The authors deeply acknowledge Hadeer Alsayed, Islam Shedeed (Menoufia 9 10 11 University), Zyad Abdelaziz, Dina Maklad, Ahmed Gebreil, Mahmoud Medhat (Alexandria university), 12 13 Mohammed Alhendy, Aya Sobhy, Omar Azzam (Al-Azhar University in Cairo), Hassan Aboul Nour, 14 15 Sara Elganzory (TantaFor university), peer Mohamed review Eid, Aya Talaat, Mohamedonly Emad (Beni Suef university), 16 17 18 Mohamed Abdelzaheer, Ahmed Abdelhamed, Ahmed Saleh (Suez Canal university), Ahmed Zain, 19 20 Khaled Ghaleb, Yossri Mohamed (Benha university), Ahmed Alaa, Mohamed Gamal (Assuit 21 22 university), Marina Nashed, Ibrahim Abdelmone'm (Ain Shams university), Bassant Abdelazeim, 23 24 25 Ramadan Zaky (Cairo university) for their assistance in data collection. None of them received 26 27 compensation for their assistance. 28 29 30 31 Contributors: Aboshady, Radwan and Hassouna were responsible for the conception and design of the 32 33 study. Aboshady and Radwan coordinated the study and managed the data collection. Aboshady, 34 http://bmjopen.bmj.com/ 35 Radwan, Eltaweel, Azzam, Aboelnaga, Darwish, Hashem, Salah, Kotb, Afifi, Noaman and Salem 36 37 38 collected the data. Hassouna did the analyses, Aboshady, Radwan, Hassouna, Eltaweel, Kotb and 39 40 Aboelnaga contributed to interpretation of the findings. Aboshady, Eltaweel and Azzam wrote the first 41 on September 24, 2021 by guest. Protected copyright. 42 draft of the manuscript while Radwan, Hassouna, Aboelnaga, Kotb, Hashem, Salah, Darwish, Salem, 43 44 45 Afifi and Noaman made a critical revision of the manuscript for important intellectual content. All 46 47 authors approved the final version of the manuscript. 48 49 50 51 Funding: All funding required was provided by Aboshady and Radwan on their own expenses. 52 53 54 support for this project. 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 23 1 2 3 Competing interests: None. 4 5 6 7 Ethics approval: Institutional Review Board at Menoufia University, Faculty of Medicine, Egypt. 8 9 10 11 Data sharing statement: No additional data are available. 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 24 1 2 3 4 REFERENCES 5 6 7 8 1. Gooding I, Klaas B, Yager JD, Kanchanaraksa S. Massive Open Online Courses in Public Health. 9 10 11 Front Public Health 2013;1 doi: 10.3389/fpubh.2013.00059 12 13 14 2. Hoy MB. MOOCs 101: an introduction to massive open online courses. Med Ref Serv Q 15 For peer review only 16 2014;33(1):85-91 doi: 10.1080/02763869.2014.866490. 17 18 19 20 3. Pappano L. The Year of the MOOC. The New York Times 2013. 21 22 23 24 4. A Triple Milestone: 107 Partners, 532 Courses, 5.2 Million Students and Counting! Coursera Blog: 25 26 Coursera 2013. 27 28 29 30 5. Liyanagunawardena TR, Williams SA. Massive open online courses on health and medicine: review. 31 32 J Med Internet Res 2014;16:e191. doi:10.2196/jmir.3439 33 34 http://bmjopen.bmj.com/ 35 6. Mehta NB, Hull AL, Young JB, Stoller JK. Just imagine: new paradigms for medical education. 36 37 38 Acad Med 2013;88(10):1418-23 doi:0.1097/ACM.0b013e3182a36a07. 39 40 41

7. Cooke M, Irby DM, O'Brien BC. Educating physicians: a call for reform of medical school and on September 24, 2021 by guest. Protected copyright. 42 43 44 residency: John Wiley & Sons, 2010; 25(2): 193–195 45 46 47 8. Liyanagunawardena T, Williams S, Adams A. The impact and reach of MOOCs:a developing 48 49 50 countries’ perspective. eLearning Papers 2013(33) 51 52 53 9. Emanuel EJ. Online education: MOOCs taken by educated few. Nature 2013;503(7476):342-42 doi: 54 55 56 10.1038/503342a. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 25 1 2 3 10. Group ME. MOOCs @ Edinburgh 2013: Report #1: the University of Edinburgh, 2013. 4 5 6 7 11. Huhn C. UW‐Madison Massive Open Online Courses (MOOCs): Preliminary Participant 8 9 Demographics: Academic Planning and Institutional Research, 2013. 10 11 12 13 12. Bosslet GT, Torke AM, Hickman SE, Terry CL, Helft PR. The patient-doctor relationship and 14 15 online social networks:For results peer of a national review survey. J Gen Intern only Med 2011;26(10):1168-74 doi: 16 17 18 10.1007/s11606-011-1761-2|. 19 20 21 13. Kolowich S. The professors who make the MOOCs. The Chronicle of Higher Education 2013;25 22 23 24 25 14. Jordan K. Initial trends in enrolment and completion of massive open online courses. The 26 27 International Review of Research in Open and Distance Learning 2014;15(1) 28 29 30 31 15. Harder B. Are MOOCs the future of medical education? Bmj 2013;346:f2666 doi: 32 33 10.1136/bmj.f2666|. 34 http://bmjopen.bmj.com/ 35 36 37 16. Dahlin M, Joneborg N, Runeson B. Stress and depression among medical students: a cross-sectional 38 39 study. Med Educ 2005;39(6):594-604 doi: 10.1111/j.1365-2929.2005.02176.x. 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 17. Guthrie E, Black D, Bagalkote H, Shaw C, Campbell M, Creed F. Psychological stress and burnout 44 45 in medical students: a five-year prospective longitudinal study. J R Soc Med 1998;91(5):237-43 46 47 48 49 18. Bassols AM, Okabayashi LS, Silva AB, et al. First- and last-year medical students: is there a 50 51 difference in the prevalence and intensity of anxiety and depressive symptoms? Rev Bras Psiquiatr 52 53 (Sao Paulo, Brazil : 1999) 2014;0:0 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 26 1 2 3 19. Frehywot S, Vovides Y, Talib Z, et al. E-learning in medical education in resource constrained low- 4 5 6 and middle-income countries. Hum Resour Health 2013;11(1):4 doi: 10.1186/1478-4491-11-4. 7 8 9 20. Angelino LM, Williams FK, Natvig D. Strategies to Engage Online Students and Reduce Attrition 10 11 12 Rates. Journal of Educators Online 2007;4(2):n2 13 14 15 21. Prober CG, HeathFor C. Lecture peer halls without review lectures--a proposal only for medical education. N Engl J 16 17 18 Med 2012;366(18):1657-9 doi: 10.1056/NEJMp1202451. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 Fig. 1: Questionnaire branching logic questions and the number of responders to each one. 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 Fig. 2: Students’ motives for enrollment in MOOCs reported by 136 students. 4 5 6 7 8 0.9 9 10 0.8 11 12 13 0.7 14 15 For peer review only 16 0.6 17

18 0.5 19 20 21 0.4 22 23 24 0.3 25 26 0.2 27 28 0.1 29 30 31 32 0 33 Lack of Time Low Internet Difficulty of Lack of Computer Beliefs Conflict http://bmjopen.bmj.com/ 34 Speed Language Technology Literacy 35 Access 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 Fig. 3: Enrollment and completion limitations. 4 5 6 7 8 9 10 11 Enrollment Limitations Completion limitations 12 n= 6$ 13 n= # 14 15 For peer review only 16 Lack of Time 70% 14% 11% 77% 13% 10% 17 18 54% 10% 35% 19 Low Internet Speed 46% 17% 34% 20 21 Computer Literacy 13% 16% 64% 10%8% 79% 22 Agree 23 Neutral 24 Difficulty of Language 12% 32% 44% 14% 23% 62% 25 Disagree 26 Lack of Technology 10%11% 73% 12% 11% 74% 27 Access 28 29 Beliefs Conflict 3% 20% 63% 3%8% 82% 30 31 0% 20% 40% 60% 80% 100% 32 0% 20% 40% 60% 80% 100% 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 Fig. 4: Student satisfaction regarding each part of MOOCs experience. 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 32 of 32 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 STROBE Statement—Checklist of items that should be included in reports of cross-sectional studies 3 Item 4 No Recommendation 5 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 6 7 (done) 8 (b) Provide in the abstract an informative and balanced summary of what was done 9 and what was found (done) 10 11 Introduction 12 Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 13 (done) 14 Objectives 3 State specific objectives, including any prespecified hypotheses (done) 15 For peer review only 16 Methods 17 Study design 4 Present key elements of study design early in the paper (done) 18 Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, 19 exposure, follow-up, and data collection (done) 20 21 Participants 6 (a) Give the eligibility criteria, and the sources and methods of selection of 22 participants (done) 23 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect 24 modifiers. Give diagnostic criteria, if applicable (done) 25

26 Data sources/ 8* For each variable of interest, give sources of data and details of methods of 27 measurement assessment (measurement). Describe comparability of assessment methods if there is 28 more than one group (Not applicable) 29 Bias 9 Describe any efforts to address potential sources of bias (done) 30 31 Study size 10 Explain how the study size was arrived at (done) 32 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, 33 describe which groupings were chosen and why (done) http://bmjopen.bmj.com/ 34 Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 35 (done) 36 37 (b) Describe any methods used to examine subgroups and interactions (done) 38 (c) Explain how missing data were addressed (Not applicable) 39 (d) If applicable, describe analytical methods taking account of sampling strategy 40 (Not applicable) 41 42 (e) Describe any sensitivity analyses (Not applicable) on September 24, 2021 by guest. Protected copyright. 43 Results 44 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially 45 46 eligible, examined for eligibility, confirmed eligible, included in the study, 47 completing follow-up, and analysed (done) 48 (b) Give reasons for non-participation at each stage 49 (c) Consider use of a flow diagram (done) 50 51 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and 52 information on exposures and potential confounders 53 (b) Indicate number of participants with missing data for each variable of interest 54 (done) 55 Outcome data 15* Report numbers of outcome events or summary measures(done) 56 57 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and 58 their precision (eg, 95% confidence interval). Make clear which confounders were 59 adjusted for and why they were included (done) 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml1 Page 33 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 (b) Report category boundaries when continuous variables were categorized (Not 3 applicable) 4 (c) If relevant, consider translating estimates of relative risk into absolute risk for a 5 6 meaningful time period (Not applicable) 7 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and 8 sensitivity analyses(Not applicable) 9 10 Discussion 11 Key results 18 Summarise key results with reference to study objectives (done) 12 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or 13 imprecision. Discuss both direction and magnitude of any potential bias (done) 14 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, 15 For peer review only 16 multiplicity of analyses, results from similar studies, and other relevant evidence 17 (done) 18 Generalisability 21 Discuss the generalisability (external validity) of the study results (done) 19 20 Other information 21 Funding 22 Give the source of funding and the role of the funders for the present study and, if 22 applicable, for the original study on which the present article is based (Not 23 applicable) 24

25 26 *Give information separately for exposed and unexposed groups. 27 28 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 29 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 30 31 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 32 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 33 available at www.strobe-statement.org. 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml2 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

Perception and use of massive open online courses among medical students of a developing country: multicenter cross-sectional study

For peer review only Journal: BMJ Open

Manuscript ID: bmjopen-2014-006804.R1

Article Type: Research

Date Submitted by the Author: 20-Nov-2014

Primary Subject Medical education and training Heading:

Secondary Subject Heading: Medical education and training

Computer-Assisted Instruction , Medical Education , Distance Education , Keywords: MOOCs, Egypt

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 1 1 2 3 4 Title Page 5 6 7 Title 8 9 10 Perception and use of massive open online courses among medical students of a 11 12 13 developing country: multicenter cross-sectional study 14 15 Authors For peer review only 16 17 18 - Omar A. Aboshady 19 20 6th year medical student, Faculty of Medicine, Menoufia University, Shebin El-Kom, Menoufia, Egypt. 21 22 [email protected] 23 24 25 - Ahmed E. Radwan 26 27 6th year medical student, Faculty of Medicine, Menoufia University, Shebin El-Kom, Menoufia, Egypt. 28 29 [email protected] 30 31 - Asmaa R. Eltaweel 32 33

th http://bmjopen.bmj.com/ 34 6 year medical student, Faculty of Medicine, Alexandria University,Alexandria, Egypt. 35 36 [email protected] 37 38 - Ahmed Azzam 39 40 6th year medical student, Faculty of Medicine, Al-Azhar University in Cairo, Cairo, Egypt. 41 42 on September 24, 2021 by guest. Protected copyright. 43 [email protected] 44 45 - Amr A. Aboelnaga 46 47 5th year medical student, Faculty of Medicine, Tanta University, Tanta, Egypt. 48 49 [email protected] 50 51 52 - Heba A. Hashem 53 th 54 6 year medical student, Faculty of Medicine, BeniSuef University, BeniSuef, Egypt. 55 56 [email protected] 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 2 1 2 3 - Salma Y. Darwish 4

on September 24, 2021 by guest. Protected copyright. 42 43 44 45 46 METHODOLOGY 47 48 49 50 This is a multi-center, cross-sectional study utilizing a structured, web-based, pilot-tested and self- 51 52 administered questionnaire. The institutional review board at Faculty of Medicine, Menoufia 53 54 University, Egypt, ethically approved the study. 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 9 1 2 3 Study Population and Sample 4 5 6 7 Our target population was undergraduate medical students in Egypt enroled in 19 medical schools for 8 9 the academic year 2013/14. We selected ten out of the 19 medical schools to be our study settings using 10 11 12 simple random sampling technique. The sample included Ain Shams, Al-Azhar medical school in 13 14 Cairo, Alexandria, Assiut, Benha, BeniSuef, Cairo, Menoufia, Suez Canal and Tanta medical schools. 15 For peer review only 16 17 18 Students in these schools are enroled in a six-year MBBCh program, in which the first three years are 19 20 called academic years and the last three years are called clinical years. According to 99% confidence 21 22 interval (CI), 3% margin of error and 50% response distribution; 1784 students were required to 23 24 25 represent the study population. We used a stratified simple random technique to select our sample with 26 27 an equal allocation of participants in each university and each study year. Accordingly, using the 28 29 registered students`names lists, we randomly selected 270 students from each faculty (45 for each study 30 31 32 year) for a total of 2700 participants. We excluded non-Egyptians students and those who changed their 33 http://bmjopen.bmj.com/ 34 enrolment school at the time of data collection. 35 36 37 38 Data collection 39 40 41 We invited the selected participants via e-mail and social media websites to take our survey using a 42 on September 24, 2021 by guest. Protected copyright. 43 44 unique code for each participant during the period of March–April 2014. We used an online survey 45 46 program to administer the questionnaire (Survey Gizmo; Boulder, Colorado, US). Students who did not 47 48 have access to the internet at the time of data collection were allowed to record their responses using a 49 50 51 self-administered paper version of the questionnaire. We sent up to five reminder messages for 52 53 participants to complete the survey. The participants were informed about the study aims in the cover 54 55 56 letter, and they voluntarily consented to participate with no incentives. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 10 1 2 3 Questionnaire Development 4 5 6 7 The study questionnaire was developed by the research team through group discussions after an 8 9 extensive literature review. The draft was then reviewed by two experts in the fields of medical 10 11 12 education and Biostatistics. We used the final draft in a pilot testing on 175 students in all participating 13 14 medical schools. Detailed feedback about the format, clarity and completion time was collected and we 15 For peer review only 16 made minor changes in response toparticipants`comments. We did not include the pilot responses in our 17 18 19 analysis. 20 21 22 The questionnaire was in Arabic, the participants`native language, and it comprised 29 questions in 23 24 25 four sections using branching logic function (Figure 1). The first section addressed study aims, consent 26 27 and participants` personal information. This section was followed by a main question asking about their 28 29 knowledge about MOOCs. Based on this answer, participants were directed to different sections. 30 31 32 Students who knew about MOOCs were asked how they heard about it and their state of enrolment. If 33 http://bmjopen.bmj.com/ 34 the participant was not enrolled in any course, he/she was asked about the limitations, and then the 35 36 37 questionnaire ends. 38 39 40 Enrolled students were directed to the next section, which assessed their perspectives and experience 41 42 on September 24, 2021 by guest. Protected copyright. 43 with MOOCs. For students who gained certificates, further questions were asked regarding their level 44 45 of satisfaction as well as any obstacles they might have faced. Finally, four questions were addressed to 46 47 assess students`opinion about integration of MOOCs in the medical field. 48 49 50 51 Most of the questions were single answer multiple-choice questions. However, there were three 52 53 multi-select check-box questions. For assessment of limitations, satisfaction and opinions, a five-point 54 55 56 Likert scale between one (strongly agree/satisfied) and five (strongly disagree/unsatisfied) was used. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 11 1 2 3 Statistical analysis: 4 5 6 7 Results were presented as numbers and percentages with confidence interval at 99%. The significance 8 9 of the association between qualitative variables of interest was analyzed using Chi-square test or 10 11 12 Fisher`s exact test, as indicated. In order to focus on clear opinions, the five-point Likert scale of 13 14 limitations, satisfaction and opinions were collapsed into three categories (agree/satisfied, neutral and 15 For peer review only 16 disagree/unsatisfied). Class year was recoded as a dichotomous variable in order to compare results for 17 18 19 students in academic versus clinical education. The acknowledgment of the importance of getting a 20 21 certificate before enrolment was also recoded as a dichotomous variable (important/very important 22 23 24 versus limited importance/not important) in order to test the significance of association between the 25 26 primarily reported importance of acquiring a certificate and the actual possession of the certificate by 27 28 McNemar test. All tests were bilateral and a P value of 0.01 was the limit of statistical significance. 29 30 31 Statistical analysis was performed using the IBM SPSS statistical software package version 22. 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 12 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 RESULTS 34 http://bmjopen.bmj.com/ 35 36 37 Respondent characteristics 38 39 40 Of 2700 total participants, 62 (2.3%) were excluded being non-Egyptians or having changed their 41 42 on September 24, 2021 by guest. Protected copyright. 43 enrolment school, in addition to 111 (4.1%) students` whose contact information could not be reached 44 45 with final eligible 2527 students. During the data collection phase, 2357 (93.3%) online questionnaire 46 47 invitations and 170 (6.7%) paper versions were sent out. Out of these distributed questionnaires, 2016 48 49 50 responses were received (response rate 83.3%). Table 1 showes participants`demographics regarding 51 52 school, class and gender. 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 13 1 2 3 Knowledge about MOOCs 4 5 6 We found that 456 (21.7% [99% CI, 19.4%–24%]) students had heard about MOOCs or websites 7 8 providing such courses. There was no statistically significant difference in knowledge between males 9 10 11 and females (43.6% vs. 56.4%, 99 CI, P = .8). However, clinical years` students had higher rates of 12 13 knowledge than students in the academic years (P< .001) (Table 1). Additionally, there was no 14 15 difference betweenFor medical peerschools in students` review knowledge about only MOOCs (P=.04). 16 17 18 19 After informing the students who did not know about MOOCs that this system provides scientific 20 21 courses in different disciplines by specialists from top universities worldwide for no or low fees 22 23 24 through the internet, 1342 (81.3% [99% CI, 78.8%–83.8%]) students showed an interest to participate 25 26 with a significant difference among different medical schools (P< .001). 27 28 29 Table 1. Participants’ demographics and their state of knowledge, enrollment and certificate 30 31 32 attainment. 33 34 http://bmjopen.bmj.com/ 35 Knowledge about MOOCs P Enrollment in courses P Certificate Attainment P 36 value value value 37 Total (%) Yes (%) No (%) Total Yes (%) No (%) Total Yes (%) No (%) 38 (n=2106) (n=456) (n=1650) (n=456) (n=136) (n=320) (n=136) (n=25) (n=111) 39 40Faculty Ain Shams 207 38 169 38 13 25 13 3 10 41 (9.8%) (18.4%) (81.6 %) (34.2%) (65.8%) (23.1%) (76.9%) 42 Al-Azhar 216 42 174 42 11 31 11 1 10 on September 24, 2021 by guest. Protected copyright. 43 (10.3%) (19.4%) (80.6%) (26.2%) (73.8%) (9.1%) (90.9%) 44 Alexandria 222 48 174 48 19 29 19 4 15 45 (10.5%) (21.6%) (78.4%) (39.6%) (60.4%) (21.1%) (78.9%) 46 Assuit 180 33 147 33 6 27 6 2 4 47 (8.5%) (18.3%) (81.7%) (18.2%) (81.8%) (33.3%) (66.7%) 48 Benha 205 57 148 57 16 41 16 0 16 49 (9.7%) (27.8%) (72.2%) (28.1%) (71.9%) (0.0%) (100.0%) 50 Beni Suef 220 38 182 P= 38 6 32 P= 6 0 6 P= 51 (10.4%) (17.3%) (82.7%) .04 (15.8%) (84.2%) .13 (0.0%) (100.0%) .02 52 Cairo 188 39 149 39 12 27 12 2 10 53 (8.9%) (20.7%) (79.3%) (30.8%) (69.2%) (16.7%) (83.3%) 54 Menoufia 248 53 195 53 22 31 22 10 12 55 56 (11.8%) (21.4%) (78.6%) (41.5%) (58.5%) (45.5%) (54.5%) 57 Suez 199 59 140 59 20 39 20 2 18 58 Canal (9.4%) (29.6%) (70.4%) (33.9%) (66.1%) (10.0%) (90.0%) 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 14 1 2 3 Tanta 221 49 172 49 11 38 11 1 10 4 (10.5%) (22.2%) (77.8%) (22.4%) (77.6%) (9.1%) (90.9%) 5 6 Class Academic 1076 176 900 176 40 136 40 4 36 7 (51.2%) (16.4%) (82.6%) P< (22.7%) (77.3%) P= (10.0%) (90.0%) P= 8 Clinical 1024 280 744 .001 280 96 184 .01 96 21 75 .1 9 (48.8%) (27.3%) (72.7%) (34.3%) (65.7%) (21.9%) (78.1%) 10Gender Male 926 199 730 199 71 128 71 17 54 11 (44.1%) (21.4%) (78.6%) P= (35.7%) (64.3%) P= (23.9%) (76.1%) P= 12 Female 1174 257 920 .83 257 65 192 .02 65 8 57 .08 13 (55.9%) (21.8%) (78.2%) (25.3%) (74.7%) (12.3%) (87.7%) 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Enrolment and certificate attainment 31 32 33 34 Of those who knew about MOOCs, 136 (29.8% [99% CI,24.3%–35.3%]) were enroled in at least one http://bmjopen.bmj.com/ 35 36 course. Most students (125; 91.9%) registered in 1–5 courses, while only 113 (83.1%) students reported 37 38 39 watching at least one video lecture. Home (109; 99%) was the first place where they watched these 40 41 videos. There was no statistically significant difference in enrolment state between males and females 42 on September 24, 2021 by guest. Protected copyright. 43 (52.2% vs. 47.8%, 99% CI, P= .016). However, there was a significant difference between students` 44 45 46 class and their enrolment (P=.009) (Table 1). Coursera was the most commonly used website (99; 47 48 72.8%), followed by Edx (14; 10.3%). 49 50 51 52 Only 25 students (18.4% [99% CI, 9.8%–26.9%]) completed courses and attained one certificate or 53 54 more with 81.6% dropout rate. Interestingly, more than half of students who earned certificates (13; 55 56 52% [99% CI,26.3%–77.7%]) have verified them from the universities that proposed the courses. The 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 15 1 2 3 vast majority of enrolled students assumed that getting a certificate is important to them (32 [23.5%] 4 5 6 very important, 37 [27.2%] important, 50 [36.8%] important to some extent and 17 [12.5%] not 7 8 important). Out of the 69 students who assumed that getting a certificate is important before enrolment 9 10 11 (important/very important), 17 were finally certified (24.6%); compared to only 8 certified students out 12 13 of the 67 who were not concerned with having certificates (important to some extent/not important; 14 15 11.9%); P< .001For. peer review only 16 17 18 19 Ways of knowledge and students`motivations 20 21 To assess how students knew about MOOCs and what were their motivations, two multi-select 22 23 24 questions were addressed. Social media was the main way through which 206 (45.2%) students knew 25 26 about MOOCs, while knowledge through a friend was the second (184; 40.4%). Using web-search 27 28 engines (87; 19.1%) got the third place, followed by extracurricular activities (46; 10.1%). MOOCs 29 30 31 providers` advertisements played a very small role (27; 5.9%) in reaching students as did medical 32 33 schools` official websites (15; 3.3%). Notably, there was no association between the ways through 34 http://bmjopen.bmj.com/ 35 which students learned about MOOCs and their enrolment. Nevertheless, students who knew through 36 37 38 extracurricular activities were found to enrol more frequently (P= .005). 39 40 41 Concerning students` motives, most students reported that their main motivation was “to learn new 42 on September 24, 2021 by guest. Protected copyright. 43 44 things” followed by “to help me studying medicine” (Figure 2). Interestingly, the students who enrolled 45 46 aiming to have a certificate or to help them in obtaining a future job were significantly more likely to 47 48 complete the courses (P= .001) and (P= .008), respectively. 49 50 51 52 MOOCs and Medicine 53 54 By asking the enrolled students (n=136) about their experience and attitude toward medical MOOCs, 55 56 57 103 (75.7% [99% CI, 66.2%–85.2%]) declared participation in at least one medical course. Of them, 24 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 16 1 2 3 students (17.6% [99% CI, 7.9%–27.3%]) had completed medical courses and earned certificates. 4 5 6 Regarding their medical MOOCs experience, 102 (75%) students agreed that MOOCs helped them in 7 8 developing their theoretical background about the topic discussed. However, there was less agreement 9 10 11 (68; 50%) on the role of MOOCs in developing their practical skills. Most students (89; 86.4%) agreed 12 13 that MOOCs help in studying medicine, while 83 (61%) believed that MOOCs will help them in getting 14 15 a better future jobFor opportunity. peer review only 16 17 18 19 Limitations of MOOCs 20 21 22 Our study reported two types of limitations: enrolment and completion. Students who knew about 23 24 25 MOOCs, but did not enrol in any courses (n=320) were asked about their enrolment limitations. The 26 27 majority of students (226; 70.4%) agreed that lack of time was the main limitation, while 147 (45.9%) 28 29 agreed that slow internet speed was another cause (Figure 3). Regarding completion limitations, the 30 31 32 enrolled students (n=136) were asked to assess the limitations that made them drop out of courses. 33 http://bmjopen.bmj.com/ 34 Similar to the enrolment limitations, it was obvious that lack of time (105; 77.2%) and slow internet 35 36 37 speed (73; 53.7%) were the main obstacles. While lack of technology access, computer literacy, 38 39 language difficulty and culture conflicts had less agreement on their roles as limitations (Figure 40 41 3). Only 16 (11.8%) students agreed that the scientific content was difficult. In addition, 93 (68.4%) 42 on September 24, 2021 by guest. Protected copyright. 43 44 students disagreed that “low content than expected” is to be a limitation. 45 46 47 For further assessment of the internet speed, we asked the enrolled students to rate their internet 48 49 50 speed. Sixty students (44.1%) reported that the speed was reasonable, while 55 (40.4%) reported slow 51 52 speed and only 21 (15.4%) had a higher connection speed. When we compared the students` evaluation 53 54 of internet speed and if they watched video lectures or not, we did not find a significant association (P= 55 56 57 .69). 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 17 1 2 3 Students` satisfaction of MOOCs 4 5 6 7 The 25 students who obtained certificates were asked to report their opinions about each part of the 8 9 MOOCs experience. The results showed that most students (21; 84%) were satisfied with the overall 10 11 12 experience, including video lectures (18; 70%), exams and assignments (16; 64%), quality of the 13 14 presented materials (21; 84%) and the technology used (20; 80%). However, there was less satisfaction 15 For peer review only 16 regarding student–student (5; 20%) and student–instructor (8; 32%) interactions (Figure 4). 17 18 19 20 21 22 23 24 25 26 27 28 DISCUSSION 29 30 31 32 Available information about MOOCs participants is data obtained from course-end demographics, 33 http://bmjopen.bmj.com/ 34 which usually reports heterogeneous populations of different age groups and educational levels from 35 36 37 different countries globally. These data show that most MOOCs` users are well-educated males with 38 39 low participation from developing countries and undergraduates.9-11 To our knowledge, this study is the 40 41 first in the medical field and in one of the developing countries to use a cross-sectional study design in 42 on September 24, 2021 by guest. Protected copyright. 43 44 a homogeneous population for assessment of prevalence and uptake of such courses among 45 46 undergraduate medical students. 47 48 49 50 Knowledge and Enrolment 51 52 53 Our results show a funnel-shaped participation pattern, with 22.7 % of the respondents knowing 54 55 56 about MOOCs and 6.5% actually enrolled. Moreover, only 5.4% watched videos and 1.2% obtained 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 18 1 2 3 certificates. Although there are no similar cross-sectional studies with which our results can be 4 5 6 compared, the knowledge of about one-fifth of the Egyptian medical students about MOOCs is 7 8 considered promising in a developing country that depends mainly on regular education. Additionally, 9 10 11 these courses are still new and there was little role of MOOCs providers`advertisements for reaching 12 13 students beside that there is no any medical MOOC which is given by an Egyptian institution till now. 14 15 Social media andFor personal experiencepeer transfer review among friends played only a vital role in the spreading of the 16 17 18 MOOCs` idea, raising students` awareness to this level. This is in line with the uprising role of social 19 20 media websites in medical students` life with more than 90% of medical students in the US using social 21 22 media.12 23 24 25 26 Notably, it was obvious that there is a disproportion between knowledge about MOOCs and 27 28 enrolment with only one-third of students having the awareness registered in courses. The students 29 30 31 reported lack of time and low internet speed as the main limitations. Out of these enrolled students, 32 33 18.4% (23.3% for medical courses) completed courses and earned certificates. These completion rates 34 http://bmjopen.bmj.com/ 35 are higher than the reported average completion rates in the course demographics. In 2013, The 36 37 13 38 Chronicle of Higher Education suggested an average of 7.5% completion rate , while a recent study in 39 40 2014 reported a rate of about 6.5%.14 This may be explained by the reported importance of certificates 41 on September 24, 2021 by guest. Protected copyright. 42 for students to add to their resumes hoping for better future chances. It was interesting to note that 43 44 45 about half of them paid to verify their certificates, although there is no academic credit for 46 47 undergraduates for any MOOCs from any medical school in the US 15 and Egypt until now. 48 49 50 51 Although there was no association between gender and students` knowledge or enrolment, class had 52 53 a significant association. Clinical year students were found to have higher knowledge and enrolment 54 55 56 rates. This may be due to the high stress and pressures experienced by first years` medical students 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 19 1 2 3 adapting new systems with little time available for extracurricular activities.16 In contrast, final year 4 5 16-18 6 students were reported to have less stress with more attention to their career plans by searching for 7 8 new learning channels to increase their competitiveness. 9 10 11 12 MOOCs and Medicine 13 14 15 Of the enrolledFor students, 75.7%peer participated review in at least one medical only course with 23.3% completion 16 17 18 rate. They strongly agreed that these courses helped them develop theoretical background about the 19 20 topics discussed with less agreement on their role in developing their clinical skills. This raises 21 22 questions about the effectiveness of MOOCs with the current lecture-based teaching style in covering 23 24 25 the different aspects of medical education, including its clinical part, which needs student–patient 26 27 interaction. However, in the new evolving era of online learning, a question arises: “why to waste 28 29 precious class time on a lecture?” Students may watch the instructor`s lecture remotely in their homes 30 31 19 32 and utilize class time for learning clinical skills. Most of the current opinions expect a complementary 33 http://bmjopen.bmj.com/ 34 role of MOOCs in undergraduate education with an increasing role in educating those students after 35 36 15 37 their graduation in continuing medical education. 38 39 MOOCs limitations in Egypt 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 Lack of time and slow internet speed were the two main limitations reported for causing low 44 45 MOOCs enrolment and course completion rates. MOOCs, being a self-learning educational system, 46 47 require a considerable amount of time to choose courses, watch videos, take exams and interact through 48 49 50 discussions. This imposes burden on students, leading to the need of increased commitments besides 51 52 their busy regular medical education. Time management, either in the design of courses or from 53 54 participants, is critically needed to enhance their performance and increase completion rates. 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 20 1 2 3 Low internet speed is a commonly reported problem facing online education in developing 4 5 20 6 countries. This problem prolongs the time needed to watch high-quality videos or to download course 7 8 content, rendering students less adherent and more susceptible to dropout. The main solution to this 9 10 11 problem is enhancing the internet infrastructure in Egypt. Liyanagunawardena et al. suggested allowing 12 13 lower resolution versions of the videos as an alternative solution to help engaging students with limited 14 15 bandwidth.8 Interestingly,For wepeer did not find computerreview literacy, language only or culture as barriers, although it 16 17 18 was expected that they would represent problems in Egypt, being a developing country.MOOCs 19 20 experience satisfaction 21 22 23 24 Encouragingly, most of the participants who completed MOOCs (n= 25) were satisfied with the 25 26 overall experience. However, there was an obvious dissatisfaction regarding student–student and 27 28 student–instructor interactions. This problem is common in online education in general, with a lack of 29 30 31 face-to-face interaction leading to some feelings of isolation and disconnectedness, which are thought 32 33 to be two main factors in dropout rates.21 Some MOOCs providers such as Coursera support efforts 34 http://bmjopen.bmj.com/ 35 beside the usual discussion forums for overcoming this point. These include more peer assessments, 36 37 38 social media groups, Google hangouts and real in-person Meetups. Despite that, more involvement of 39 40 participants is still needed to ensure the full psychological presence. 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 Study strengths and limitations 45 46 The strength of our study is that it included participants from all study years in 10 institutions, 47 48 covering nearly the entire geographic area of Egypt with high confidence interval (99%) and high 49 50 51 response rate (83.3%). However, our main limitation was the relatively low returned number of 52 53 participants who enrolled (n=136) and who had certificates (n=25), which makes analysis of limitations 54 55 56 and satisfactions less reliable. However, these results are important as a first start to make evidence 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 21 1 2 3 about the real prevalence of MOOCs in Egypt to help the future studies to build upon and take samples 4 5 6 that are representative to the students who knew about MOOCs for a better understanding of their 7 8 experience. 9 10 11 12 Conclusions: 13 14 15 ApproximatelyFor one-fifth ofpeer undergraduate review medical students inonly Egypt have heard about MOOCs. 16 17 18 Students who actively participated showed a positive attitude toward the experience, but better time 19 20 management skills and faster internet connection speeds are required. Further studies are needed to 21 22 address the enrolled students to assess their experience in large representative samples. In addition, 23 24 25 more efforts are needed to be done to raise the awareness of students of such courses as most of 26 27 students who did not hear about MOOCs, showed interest to participate. 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 STATEMENTS: 24 25 26 27 Acknowledgements: The authors deeply acknowledge HadeerAlsayed, Islam Shedeed (Menoufia 28 29 30 University), ZyadAbdelaziz,Dina Maklad, Ahmed Gebreil, Mahmoud Medhat (Alexandria university), 31 32 Mohammed Alhendy, AyaSobhy, Omar Azzam (Al-Azhar University in Cairo), Hassan AboulNour, 33 http://bmjopen.bmj.com/ 34 Sara Elganzory (Tanta university), Mohamed Eid, AyaTalaat, Mohamed Emad (BeniSuef university), 35 36 37 Mohamed Abdelzaheer, Ahmed Abdelhamed, Ahmed Saleh (Suez Canal university), Ahmed Zain, 38 39 KhaledGhaleb, Yossri Mohamed (Benha university), Ahmed Alaa, Mohamed Gamal (Assuit 40 41 university), Marina Nashed, Ibrahim Abdelmone'm (Ain Shams university), BassantAbdelazeim, 42 on September 24, 2021 by guest. Protected copyright. 43 44 Ramadan Zaky (Cairo university) for their assistance in data collection. None of them received 45 46 compensation for their assistance. 47 48 49 50 Contributors: Aboshady, Radwan and Hassouna were responsible for the conception and design of the 51 52 study. Aboshady and Radwan coordinated the study and managed the data collection. Aboshady, 53 54 Radwan, Eltaweel, Azzam, Aboelnaga, Darwish, Hashem, Salah, Kotb, Afifi, Noaman and Salem 55 56 57 collected the data. Hassouna did the analyses, Aboshady, Radwan, Hassouna, Eltaweel, Kotb and 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 23 1 2 3 Aboelnaga contributed to interpretation of the findings. Aboshady, Eltaweel and Azzam wrote the first 4 5 6 draft of the manuscript while Radwan, Hassouna, Aboelnaga, Kotb, Hashem, Salah, Darwish, Salem, 7 8 Afifi and Noaman made a critical revision of the manuscript for important intellectual content. All 9 10 11 authors approved the final version of the manuscript. 12 13 14 Funding: All funding required was provided by Aboshady and Radwan on their own expenses. 15 For peer review only 16 17 18 support for this project. 19 20 21 Competing interests: None. 22 23 24 25 Ethics approval: Institutional Review Board at Menoufia University, Faculty of Medicine, Egypt. 26 27 28 Data sharing statement: No additional data are available. 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 REFERENCES 22 23 24 25 1. Gooding I, Klaas B, Yager JD, Kanchanaraksa S. Massive Open Online Courses in Public Health. 26 27 Front Public Health 2013;1 doi: 10.3389/fpubh.2013.00059 28 29 30 31 2. Hoy MB. MOOCs 101: an introduction to massive open online courses. Med Ref Serv Q 32 33 2014;33(1):85-91 doi: 10.1080/02763869.2014.866490. 34 http://bmjopen.bmj.com/ 35 36 37 3. Pappano L. The Year of the MOOC.The New York Times 2013. 38 39 40 4. A Triple Milestone: 107 Partners, 532 Courses, 5.2 Million Students and Counting! Coursera Blog: 41 42 on September 24, 2021 by guest. Protected copyright. 43 Coursera 2013. 44 45 46 5. Liyanagunawardena TR, Williams SA. Massive open online courses on health and medicine: review. 47 48 49 J Med Internet Res 2014;16:e191. doi:10.2196/jmir.3439 50 51 52 6. Mehta NB, Hull AL, Young JB, Stoller JK. Just imagine: new paradigms for medical education. 53 54 Acad Med 2013;88(10):1418-23 doi:0.1097/ACM.0b013e3182a36a07. 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 25 1 2 3 7. Cooke M, Irby DM, O`Brien BC. Educating physicians: a call for reform of medical school and 4 5 6 residency: John Wiley & Sons, 2010; 25(2): 193–195 7 8 9 8. Liyanagunawardena T, Williams S, Adams A. The impact and reach of MOOCs:a developing 10 11 12 countries` perspective. eLearning Papers 2013(33) 13 14 15 9. Emanuel EJ. OnlineFor education: peer MOOCs reviewtaken by educated few. only Nature 2013;503(7476):342-42 doi: 16 17 18 10.1038/503342a. 19 20 21 10. Group ME. MOOCs @ Edinburgh 2013: Report #1: the University of Edinburgh, 2013. 22 23 24 25 11. Huhn C. UW‐Madison Massive Open Online Courses (MOOCs): Preliminary Participant 26 27 Demographics: Academic Planning and Institutional Research, 2013. 28 29 30 31 12. Bosslet GT, Torke AM, Hickman SE, Terry CL, Helft PR. The patient-doctor relationship and 32 33 online social networks: results of a national survey. J Gen Intern Med 2011;26(10):1168-74 doi: 34 http://bmjopen.bmj.com/ 35 10.1007/s11606-011-1761-2|. 36 37 38 39 13. Kolowich S. The professors who make the MOOCs. The Chronicle of Higher Education 2013;25 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 14. Jordan K. Initial trends in enrolment and completion of massive open online courses. The 44 45 International Review of Research in Open and Distance Learning 2014;15(1) 46 47 48 49 15. Harder B. Are MOOCs the future of medical education? Bmj 2013;346:f2666 doi: 50 51 10.1136/bmj.f2666|. 52 53 54 16. Dahlin M, Joneborg N, Runeson B. Stress and depression among medical students: a cross-sectional 55 56 57 study. Med Educ 2005;39(6):594-604 doi: 10.1111/j.1365-2929.2005.02176.x. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 26 1 2 3 17. Guthrie E, Black D, Bagalkote H, Shaw C, Campbell M, Creed F. Psychological stress and burnout 4 5 6 in medical students: a five-year prospective longitudinal study. J R Soc Med 1998;91(5):237-43 7 8 9 18. Bassols AM, Okabayashi LS, Silva AB, et al. First- and last-year medical students: is there a 10 11 12 difference in the prevalence and intensity of anxiety and depressive symptoms? Rev Bras Psiquiatr 13 14 (Sao Paulo, Brazil : 1999) 2014;0:0 15 For peer review only 16 17 18 19. Frehywot S, Vovides Y, Talib Z, et al. E-learning in medical education in resource constrained low- 19 20 and middle-income countries. Hum Resour Health 2013;11(1):4 doi: 10.1186/1478-4491-11-4. 21 22 23 24 20. Angelino LM, Williams FK, Natvig D. Strategies to Engage Online Students and Reduce Attrition 25 26 Rates. Journal of Educators Online 2007;4(2):n2 27 28 29 30 21. Prober CG, Heath C. Lecture halls without lectures--a proposal for medical education. N Engl J 31 32 Med 2012;366(18):1657-9 doi: 10.1056/NEJMp1202451. 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 1 1 2 3 4 Title Page 5 6 7 Title 8 9 10 Perception and use of massive open online courses among medical students of a 11 12 13 developing country: multicenter cross-sectional study 14 15 Authors For peer review only 16 17 18 - Omar A. Aboshady 19 20 6th year medical student, Faculty of Medicine, Menoufia University, Shebin El-Kom, Menoufia, Egypt. 21 22 [email protected] 23 24 25 - Ahmed E. Radwan 26 27 6th year medical student, Faculty of Medicine, Menoufia University, Shebin El-Kom, Menoufia, Egypt. 28 29 [email protected] 30 31 - Asmaa R. Eltaweel 32 33

th http://bmjopen.bmj.com/ 34 6 year medical student, Faculty of Medicine, Alexandria University,Alexandria, Egypt. 35 36 [email protected] 37 38 - Ahmed Azzam 39 40 6th year medical student, Faculty of Medicine, Al-Azhar University in Cairo, Cairo, Egypt. 41 42 on September 24, 2021 by guest. Protected copyright. 43 [email protected] 44 45 - Amr A. Aboelnaga 46 47 5th year medical student, Faculty of Medicine, Tanta University, Tanta, Egypt. 48 49 [email protected] 50 51 52 - Heba A. Hashem 53 th 54 6 year medical student, Faculty of Medicine, BeniSuef University, BeniSuef, Egypt. 55 56 [email protected] 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 2 1 2 3 - Salma Y. Darwish 4

5 year medical student, Faculty of Medicine, Cairo University, Cairo, Egypt. http://bmjopen.bmj.com/ 34 35 [email protected] 36 37 38 - Dalal S. Salem 39 40 6thyear medical student, Faculty of Medicine, Cairo University, Cairo, Egypt. 41 42 [email protected] on September 24, 2021 by guest. Protected copyright. 43 44 45 - Ahmed Hassouna 46 47 MD, Department of Cardiothoracic Surgery, Faculty of Medicine, Ain Shams University, Cairo, Egypt. 48 49 [email protected] 50 51 52 Corresponding author: 53 54 Omar AliAboshady 55 56 th 57 6 year medical student, 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 3 1 2 3 Faculty of Medicine, Menoufia University. 4 5 6 Address: 20 Sadat School St, Shanawan, Shebin El-kom, Menoufia, Egypt. 7 8 Tel:+2-048-2282698 / +2-01010747627 9 10 11 E-mail:[email protected] 12 13 Fax:+2-048-2326810 14 15 Postal code:32718For peer review only 16 17 18 19 20 Key Words: 21 22 23 Computer-Assisted Instruction (MeSH terms); Medical Education (MeSH terms); Distance Education 24 25 (MeSH terms); MOOCs; Egypt. 26 27 28 Word Count: 29 30 - Title: 18 words (114 characters) 31 32 - Abstract: 297 words 33 34 http://bmjopen.bmj.com/ - Text: 3340 words 35 36 37 - Number of figures and tables: 5 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 4 1 2 3 4 ABSTRACT 5 6 7 8 9 Objectives: To primarily assess the prevalence of awareness and use of massive open online courses 10 11 12 (MOOCs) among medical undergraduates in Egypt as a developing country, besides identifying the 13 14 limitations and satisfaction of using these courses. 15 For peer review only 16 Design:A multi-center, cross-sectionalstudy using a web-based,pilot-tested and self-administered 17 18 19 questionnaire. 20 21 Settings: Ten out of 19 randomly selected medical schools in Egypt by simple random sampling 22 23 24 technique.Ten randomly selected medical schools in Egypt. 25 26 Participants: Randomly selected 2700 undergraduate medical students with an equal allocation of 27 28 participants in each university and each study year. 29 30 31 Primary and secondary outcomes measures: The primary outcome measures were the percentages of 32 33 students who knew about MOOCs, students who enrolled and students who obtained a certificate. 34 http://bmjopen.bmj.com/ 35 Secondary outcome measures included the limitations and satisfaction of using MOOCs through 5- 36 37 38 point Likert scale questions. 39 40 Results: Of 2527 eligible students, 2106 filled the questionnaire (response rate 83.3%). Of these 41 on September 24, 2021 by guest. Protected copyright. 42 students, 456 (21.7%) knew the term MOOCs or websites providing these courses. Out of the latter, 43 44 45 136 students (29.8%) had enrolled in at least one course, but only 25 (18.4%) of them completed 46 47 courses earning certificates. Clinical years’years`students showed significantly higher rates of 48 49 50 knowledge (P= .009) and enrollmentenrolment (P< .001) than academic year students. The primary 51 52 reasons for incompletion of courses included lack of time (105; 77.2%) and slow internet speed (73; 53 54 53.7%). Of the 25 students who completed courses, 21 (84%) were satisfied with the overall 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 32 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 5 1 2 3 experience. However, there was less satisfaction regarding student-instructor (8; 32%) and student- 4 5 6 student (5; 20%) interactions. 7 8 Conclusions: Approximately one-fifth of Egyptian medical undergraduates have heard about MOOCs 9 10 11 with only about 6.5% actively enrolled in courses.However, students who actively participated showed 12 13 a positive attitude toward the experience, but better time management skills and faster internet 14 15 connection speedsFor are required. peer Further studies review are needed to address only the enrolled students for a better 16 17 18 understanding of their experience. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 33 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 6 1 2 3 4 5 6 STRENGTHS AND LIMITATIONS OF THIS STUDY 7 8 9 10 11 12 - This study is the first to assess the actual prevalence of awareness and use of MOOCs among medical 13 14 students in Egypt.in Egypt and in the medical field. 15 For peer review only 16 - This study included a large representing sample of ten Egyptian institutions covering nearly the entire 17 18 19 geographic area of Egypt. 20 21 - Data obtained from students in all six undergraduate years. 22 23 24 - There werewas relatively low returned number of participants who enrolled and who had certificates, 25 26 which makes analysis of limitations and satisfactions less reliable. 27 28 - The study results can notcannot be generaziablegeneralizable to all developing countries. 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 34 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 7 1 2 3 4 INTRODUCTION 5 6 7 8 Massive open online courses (MOOCs) have been recently proposed as a disruptive innovation with 9 10 1 11 high expectations to solve challenges facing higher education. The idea behind MOOCs is to offer 12 13 world-class education to a (massive) number of students around the globe with internet access (online) 14 15 for little or no feesFor (open). Thepeer courses consist review of prerecorded videoonly lectures, computer-graded tests and 16 17 2 18 discussion forums to discuss talk over course materials or to get help. These courses have gained 19 20 immense popularity over a short period, attracting millions of participants and crossing the barriers of 21 22 location, gender, race and social status; making 2012 the year of MOOCs according to New York 23 24 3 25 TimesNewYork journal. Coursera, the largest MOOCs provider, in its lastestlatest infographinfograph 26 27 in October 2013 showed an extraordinary growth reaching more than 100 institutional partners, more 28 29 than 500 courses and more than 5five million students.4 30 31 32 33 In medical education, the number of related MOOCs is steadily increasing. In a recent study in 2014, 34 http://bmjopen.bmj.com/ 35 it was found that 98 free courses were offered during 2013 in the fields of health and medicine with an 36 37 5 38 average length of 6.7 weeks. These courses were introduced as a possible solution that may help 39 40 solving great challenges facing medical education nowdays.6 These challenges includeincluding the 41 on September 24, 2021 by guest. Protected copyright. 42 issues of quality, costs and the ability to deliver education to enough adequate number of students who 43 44 7 45 will cover the health care system’s system`s needs. Nowdays, there are ongoing discussions aiming to 46 47 determine the role of MOOCs in medical educationThere are uprising discussions to determine which 48 49 50 roles MOOCs can play in the medical field.. However, information about how medical students 51 52 perceive such courses is still limited, especially in the developing countries where high-quality learning 53 54 is often scarce. 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 35 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 8 1 2 3 Although there is a great hope that MOOCs can play a role in solving developing countries’ lack of 4 5 6 high quality educationMOOCs are considered as a hope to provide developing countries with education 7 8 of high quality, ,the current demographic data reveal that most of the MOOCs’ MOOCs` participants 9 10 11 are from the developed countries with very low participation rates from low-income countries, 12 4 13 especially in Africa. These lLow participation rates wererate was thought to be due to various 14 15 complicated conditions,For such peer as lack of access review to digital technologies, only linguistic and cultural barriers 16 17 8 18 and low poor computer skills. In addition, lack of awareness of the presence of this newly introduced 19 20 concept may be considered as another problem. 21 22 23 24 To our knowledge, there are no available cross-sectional studies that assessed the actual prevalence 25 26 of awareness and use of MOOCs among medical communities in the developing countries, including 27 28 Egypt. Our study primarily aims to assess the prevalence of awareness and use of these courses among 29 30 31 Egyptian undergraduate medical students, as an example of a developing country. Second, the study 32 33 will assess the limitations that prevent students to enroll and completehinder students from enrolment 34 http://bmjopen.bmj.com/ 35 and completing the courses, besides assessing the satisfaction level of using MOOCs to better 36 37 38 understanding of the role these courses in medical education. 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 36 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 9 1 2 3 4 METHODOLOGY 5 6 7 8 This is a multi-center, cross-sectional study utilizing a structured, web-based, pilot-tested and self- 9 10 11 administered questionnaire. The study was ethically approved by the institutional review board at 12 13 Faculty of Medicine, Menoufia University, Egyptinstitutional review board at Faculty of Medicine, 14 15 Menoufia University,For Egypt, peer ethically approved review the study. only 16 17 18 19 Study Population and Sample 20 21 22 Our target population was undergraduate medical students in Egypt enrolled in 19 medical schools 23 24 25 for the academic year 2013/14. We randomly selected ten out of the 19 medical schools to be our study 26 27 settings using simple random sampling technique. The se weresample included Ain Shams, Al-Azhar 28 29 30 medical school in Cairo, Alexandria, Assiut, Benha, BeniSuef, Cairo, Menoufia, Suez Canal and Tanta 31 32 medical schools. 33 34 http://bmjopen.bmj.com/ 35 Students in these schools are enrolled in a six-year MBBCh program, in which the first three years 36 37 38 are called academic years and the last three years are called clinical years. According to 99% 39 40 confidence interval (CI), 3% margin of error and 50% response distribution; 1784 students were 41 42 on September 24, 2021 by guest. Protected copyright. 43 required to represent the study population. According to a confidence interval (CI) of 99%, margin of 44 45 error 3%, and response distribution of 50%; 1784 students were required to represent the study 46 47 populatioWe used a stratified simple random technique to select our sample with an equal allocation of 48 49 50 participants in each university and each study year. Accordingly, using the registered students’ 51 52 students`names lists, we randomly selected 270 students from each faculty (45 for each study year) for 53 54 a total of 2700 participants. We excluded non-Egyptians students and those who changed their 55 56 57 enrollment school at the time of data collection. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 37 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 10 1 2 3 Data collection 4 5 6 7 We invited the selected participants via e-mail and social media websites to complete take our survey 8 9 using a unique code for each participant during the period of March–April 2014. We used an online 10 11 12 survey program to administer the questionnaire (SurveyGizmoSurvey Gizmo; Boulder, Colorado, US). 13 14 Students who did not have access to the internet at the time of data collection were allowed to record 15 For peer review only 16 their responses using a self-administered paper version of the questionnaire. We sent up to five 17 18 19 reminder messages for participants to complete the survey. The participants were informed about the 20 21 study aims in the cover letter, and they voluntarily consented to participate with no incentives. 22 23 24 25 Questionnaire Development 26 27 28 The study questionnaire was developed by the research team through group discussions after an 29 30 31 extensive literature review. The draft was then reviewed by two experts in the fields of medical 32 33 education and Biostatistics. We used the final draft in a pilot testing on 175 students in all participating 34 http://bmjopen.bmj.com/ 35 medical schools. Detailed feedback about the format, clarity and completion time werefeedback about 36 37 38 the format, clarity and completion time was collected and we made minor changes in response 39 40 toparticipants’ participants`comments. We did not include the pilot responses in our analysis. 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 The questionnaire was in Arabic, the participants’ participants`native language, and it comprised 29 45 46 questions in four sections using branching logic function (Figure 1). The first section addressed study 47 48 aims, consent and participants’ participants` personal information. This section was followed by a main 49 50 51 question asking about their knowledge about MOOCs. Based on this answer, participants were directed 52 53 to different sections. Students who knew about MOOCs were asked how they heard about it and their 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 38 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 11 1 2 3 state of enrollmentenrolment. If the participant was not enrolled in any course, he/she was asked about 4 5 6 the limitations, and then the questionnaire ends. 7 8 9 Enrolled students were directed to the next section, which assessed their perspectives and experience 10 11 12 with MOOCs. For students who gained certificates, further questions were asked regarding their level 13 14 of satisfaction as well as any obstacles they might have faced. Finally, four questions were addressed to 15 For peer review only 16 assess students’ students`opinion about integration of MOOCs in the medical field. 17 18 19 20 Most of the questions were single answer multiple-choice questions. However, there were three 21 22 multi-select check-box questions. For assessment of limitations, satisfaction and opinions, a 5five-point 23 24 25 Likert scale between 1 one (strongly agree/satisfied) and 5five (strongly disagree/unsatisfied) was used. 26 27 28 Statistical analysis: 29 30 31 32 Results were presented as numbers and perecentagespercentages with confidence interval at 99%. The 33 http://bmjopen.bmj.com/ 34 significance of the association between qualitative variables of interest was analyzed using Chi-square 35 36 37 test or Fisher’s Fisher`s exact test, as indicated. In order to focus on clear opinions, the 5-five-point 38 39 Likert scale of limitations, satisfaction and opinions were collapsed into 3three categories 40 41 (agree/satisfied, neutral and disagree/unsatisfied). Class year was recoded as a dichotomous variable in 42 on September 24, 2021 by guest. Protected copyright. 43 44 order to compare results for students in academic versus clinical education. The acknowledgment of the 45 46 importance of getting a certificate before enrollment was also recoded as a dichotomous variable 47 48 (important/very important versus limited importance/not important) in order to test the significance of 49 50 51 association between the primarily reported importance of acquiring a certificate and the actual 52 53 possession of the certificate by McNemar test. All tests were bilateral and a P value of 0.01 was the 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 39 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 12 1 2 3 limit of statistical significance. Statistical analysis was performed using the IBM SPSS statistical 4 5 6 software package version 22. 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 40 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 13 1 2 3 4 RESULTS 5 6 7 8 Respondent characteristics 9 10 11 12 Of 2700 total participants, 62 (2.3%) were excluded being non-Egyptians or having changed their 13 14 enrollment school, in addition to 111 (4.1%) students’ students` whose contact information could not be 15 For peer review only 16 reached with final eligible 2527 students. During the data collection phase, 2357 (93.3%) online 17 18 19 questionnaire invitations and 170 (6.7%) paper versions were sent out. Out of these distributed 20 21 questionnaires, 2016 responses were received (response rate 83.3%). Table 1 showes participants’ 22 23 24 participants`demographics regarding school, class and gender. 25 26 27 Knowledge about MOOCs 28 29 30 We found that 456 (21.7% [99% CI, 19.4%–24%]) students had heard about MOOCs or websites 31 32 providing such courses. There was no statistically significant difference in knowledge between males 33 http://bmjopen.bmj.com/ 34 and females (43.6% vsvs. 56.4%, 99 CI, P = .8). However, clinical years` students had higher rates of 35 36 37 knowledge than students in the academic years (P< .001) (Table 1). Additionally, there was no 38 39 difference between medical schools in students’ students` knowledge about MOOCs (P=.04). 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 After clarifying the concept of MOOCs to students who did not know about itAfter informing the 44 45 students who did not know about MOOCs that this system provides scientific courses in different 46 47 disciplines by specialists from top universities worldwide for no or low fees through the internet, 1342 48 49 50 (81.3% [99% CI, 78.8%–83.8%]) students showed an interest to participate with a significant difference 51 52 among different medical schools (P< .001). 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 41 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 14 1 2 3 Table 1. Participants’ demographics and their state of knowledge, enrollment and certificate 4 5 6 attainment. 7 8 9 Knowledge about MOOCs P Enrollment in courses P Certificate Attainment P 10 value value value 11 Total (%) Yes (%) No (%) Total Yes (%) No (%) Total Yes (%) No (%) 12 (n=2106) (n=456) (n=1650) (n=456) (n=136) (n=320) (n=136) (n=25) (n=111) 13 Faculty Ain Shams 207 38 169 38 13 25 13 3 10 14 15 (9.8%) For(18.4%) (81.6peer %) review(34.2%) (65.8%) only (23.1%) (76.9%) 16 Al-Azhar 216 42 174 42 11 31 11 1 10 17 (10.3%) (19.4%) (80.6%) (26.2%) (73.8%) (9.1%) (90.9%) 18 Alexandria 222 48 174 48 19 29 19 4 15 19 (10.5%) (21.6%) (78.4%) (39.6%) (60.4%) (21.1%) (78.9%) 20 Assuit 180 33 147 33 6 27 6 2 4 21 (8.5%) (18.3%) (81.7%) (18.2%) (81.8%) (33.3%) (66.7%) 22 Benha 205 57 148 57 16 41 16 0 16 23 (9.7%) (27.8%) (72.2%) (28.1%) (71.9%) (0.0%) (100.0%) 24 Beni Suef 220 38 182 P= 38 6 32 P= 6 0 6 P= 25 (10.4%) (17.3%) (82.7%) .04 (15.8%) (84.2%) .13 (0.0%) (100.0%) .02 26 Cairo 188 39 149 39 12 27 12 2 10 27 (8.9%) (20.7%) (79.3%) (30.8%) (69.2%) (16.7%) (83.3%) 28 Menoufia 248 53 195 53 22 31 22 10 12 29 (11.8%) (21.4%) (78.6%) (41.5%) (58.5%) (45.5%) (54.5%) 30 31 Suez 199 59 140 59 20 39 20 2 18 32 Canal (9.4%) (29.6%) (70.4%) (33.9%) (66.1%) (10.0%) (90.0%) 33 Tanta 221 49 172 49 11 38 11 1 10 34 (10.5%) (22.2%) (77.8%) (22.4%) (77.6%) (9.1%) (90.9%) http://bmjopen.bmj.com/ 35Class Academic 1076 176 900 176 40 136 40 4 36 36 (51.2%) (16.4%) (82.6%) P< (22.7%) (77.3%) P= (10.0%) (90.0%) P= 37 Clinical 1024 280 744 .001 280 96 184 .01 96 21 75 .1 38 (48.8%) (27.3%) (72.7%) (34.3%) (65.7%) (21.9%) (78.1%) 39 Gender Male 926 199 730 199 71 128 71 17 54 40 41 (44.1%) (21.4%) (78.6%) P= (35.7%) (64.3%) P= (23.9%) (76.1%) P= 42 Female 1174 257 920 .83 257 65 192 .02 65 8 57 .08 on September 24, 2021 by guest. Protected copyright. 43 (55.9%) (21.8%) (78.2%) (25.3%) (74.7%) (12.3%) (87.7%) 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 42 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 15 1 2 3 Enrollment and certificate attainment 4 5 6 7 Of those who knew about MOOCs, 136 (29.8% [99% CI,24.3%–35.3%]) were enrolled in at least 8 9 one course. Most students (125; 91.9%) registered in 1–5 courses, while only 113 (83.1%) students 10 11 12 reported watching at least one video lecture. Home (109; 99%) was the first place where they watched 13 14 these videos. There was no statistically significant difference in enrollment state between males and 15 For peer review only 16 females (52.2% vsvs. 47.8%, 99% CI, P= .016). However, there was a significant difference between 17 18 19 students’ students` class and their enrollment (P=.009) (Table 1). Coursera was the most commonly 20 21 used website (99; 72.8%), followed by Edx (14; 10.3%). 22 23 24 25 Only 25 students (18.4% [99% CI, 9.8%–26.9%]) completed courses and attained one certificate or 26 27 more with 81.6% dropout rate. Interestingly, more than half of students who earned certificates (13; 28 29 52% [99% CI,26.3%–77.7%]) have verified them from the universities that proposed the courses. The 30 31 32 vast majority of enrolled students assumed that getting a certificate is important to them (32 [23.5%] 33 http://bmjopen.bmj.com/ 34 very important, 37 [27.2%] important, 50 [36.8%] important to some extent, and 17 [12.5%] not 35 36 37 important). Out of the 69 students who assumed that getting a certificate is important before enrollment 38 39 (important/very important), 17 were finally certified (24.6%); compared to only 8 certified students out 40 41 of the 67 who were not concerned with having certificates (important to some extent/not important; 42 on September 24, 2021 by guest. Protected copyright. 43 44 11.9%); P< .001. 45 46 47 Ways of knowledge and students’ students`motivations 48 49 50 To assess how students knew about MOOCs and what were their motivations, two multi-select 51 52 questions were addressed. Social media was the main way through which 206 (45.2%) students 53 54 knewstudents knew about MOOCs, while knowledge through a friend was the second (184; 40.4%). 55 56 57 Using web-search engines (87; 19.1%) was ingot the third place, followed by extracurricular activities 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 43 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 16 1 2 3 (46; 10.1%). MOOCs providers’ providers` advertisements played a very small role (27; 5.9%) in 4 5 6 reaching students as did medical schools’ schools` official websites (15; 3.3%). Notably, there was no 7 8 association between the ways through which students learned about MOOCs and their enrollment. 9 10 11 Nevertheless, students who knew through extracurricular activities were found to enroll more 12 13 frequently (P= .005). 14 15 For peer review only 16 Concerning students’ students` motives, most students reported that their main motivation was “to 17 18 19 learn new things” followed by “to help me studying medicine” (Figure 2). Interestingly, the students 20 21 who enrolled aiming to have a certificate or to help them in obtaining a future job were significantly 22 23 24 more likely to complete the courses (P= .001) and (P= .008), respectively. 25 26 27 MOOCs and Medicine 28 29 By asking the enrolled students (n=136) about their experience and attitude toward medical MOOCs, 30 31 32 103 (75.7% [99% CI,66.2, 66.2%–85.2%]) declared participation in at least one medical course. Of 33 http://bmjopen.bmj.com/ 34 them, 24 students (17.6% [99% CI, 7.9%–27.3%]) had completed medical courses and earned 35 36 37 certificates. Regarding their medical MOOCs experience, 102 (75%) students agreed that MOOCs 38 39 helped them in developing their theoretical background about the topic discussed. However, there was 40 41 less agreement (68; 50%) on the role of MOOCs in developing their practical skills. Most students (89; 42 on September 24, 2021 by guest. Protected copyright. 43 44 86.4%) agreed that MOOCs help in studying medicine, while 83 (61%) believed that MOOCs will help 45 46 them in getting a better future job opportunity. 47 48 49 50 Limitations of MOOCs 51 52 53 Our study reported two types of limitations: enrollment and completion. Students who knew about 54 55 56 MOOCs, but did not enroll in any courses (n=320) were asked about their enrollment limitations. The 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 44 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 17 1 2 3 majority of students (226; 70.4%) agreed that lack of time was the main limitation, while 147 (45.9%) 4 5 6 agreed that slow internet speed was another cause (Figure 3). Regarding completion limitations, the 7 8 enrolled students (n=136) were asked to assess the limitations that made them drop out of courses. 9 10 11 Similar to the enrollment limitations, it was obvious that lack of time (105; 77.2%) and slow internet 12 13 speed (73; 53.7%) were the main obstacles. While lack of technology access, computer literacy, 14 15 language difficultyFor and culture peer conflicts had review less agreement on theironly roles as limitations (Figure 16 17 18 3). Only 16 (11.8%) students agreed that the scientific content was difficult. In addition, 93 (68.4%) 19 20 students disagreed that “low content than expected” is to be a limitation. 21 22 23 24 For further assessment of the internet speed, we asked the enrolled students to rate their internet 25 26 speed. Sixty students (44.1%) reported that the speed was reasonable, while 55 (40.4%) reported slow 27 28 speed and only 21 (15.4%) had) had a higher connection speed. When we compared the students’ 29 30 31 students` evaluation of internet speed and if they watched video lectures or not, we did not find a 32 33 significant association (P= .69). 34 http://bmjopen.bmj.com/ 35 36 37 Students’ Students` satisfaction of MOOCs 38 39 40 The 25 students who obtained certificates were asked to report their opinions about each part of the 41 42 on September 24, 2021 by guest. Protected copyright. 43 MOOCs experience. The results showed that most students (21; 84%) were satisfied with the overall 44 45 experience, including video lectures (18; 70%), exams and assignments (16; 64%), quality of the 46 47 presented materials (21; 84%), and the technology used (20; 80%). However, there was less satisfaction 48 49 50 regarding student–student (5; 20%) and student–instructor (8; 32%) interactions (Figure 4). 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 45 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 18 1 2 3 4 DISCUSSION 5 6 7 8 Available information about MOOCs participants is data obtained from course-end demographics, 9 10 11 which usually reports heterogeneous populations of different age groups and educational levels from 12 13 different countries globally. These data show that most MOOCs’ MOOCs` users are well-educated 14 15 males with low participationFor peer from developing review countries and undergraduates. only9-11 To our knowledge, this 16 17 18 study is the first in the medical field and in one of the developing countries to use a cross-sectional 19 20 study design in a homogeneous population for assessment of prevalence and uptake of such courses 21 22 among undergraduate medical students. 23 24 25 26 Knowledge and Enrollment 27 28 29 30 Our results show a funnel-shaped participation pattern, with 22.7 % of the respondents knowing 31 32 about MOOCs and 6.5% actually enrolled. Moreover, only 5.4% watched videos and 1.2% obtained 33 http://bmjopen.bmj.com/ 34 certificates. Although there are no similar cross-sectional studies with which our results can be 35 36 37 compared, the knowledge of about one-fifth of the Egyptian medical students about MOOCs is 38 39 considered promising in a developing country that depends mainly on regular education. Additionally, 40 41 these courses are still new and there was little role of MOOCs providers’ providers`advertisements for 42 on September 24, 2021 by guest. Protected copyright. 43 44 reaching students beside that there is no any medical MOOC which is given by an Egyptian institution 45 46 till now. Social media and personal experience transfer among friends played a vital role in the 47 48 spreading of the MOOCs’ MOOCs` idea, raising students’ students` awareness to this level. This is in 49 50 51 line with the uprising role of social media websites in medical students’ students` life with more than 52 53 90% of medical students in the US using social media.12 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 46 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 19 1 2 3 Notably, it was obvious that there is a disproportion between knowledge about MOOCs and 4 5 6 enrollment with only one-third of students having the awareness registered in courses. The students 7 8 reported lack of time and low internet speed as the main limitations. From Out of these enrolled 9 10 11 students, 18.4% (23.3% for medical courses) completed courses and earned certificates. These 12 13 completion rates are higher than the reported average completion rates in the course demographics. In 14 15 2013, The ChronicleFor of Higher peer Education suggestedreview an average onlyof 7.5% completion rate 13, while a 16 17 14 18 recent study in 2014 reported a rate of about 6.5%. This may be explained by the reported importance 19 20 of certificates for students to add to their resumes hoping for better future chances. It was interesting to 21 22 note that about half of them paid money to verify their certificates, although there is no academic credit 23 24 15 25 for undergraduates for any MOOCs from any medical school in the US and Egypt tilluntil now. 26 27 28 Although there was no association between gender and students’ students` knowledge or enrollment, 29 30 31 class had a significant association. Clinical year students were found to have higher knowledge and 32 33 enrollment rates. This may be due to the high stress and pressures experienced by first yearsyears` 34 http://bmjopen.bmj.com/ 35 medical students adapting new systems with little time available for extracurricular activities.16 In 36 37 16-18 38 contrast, final year students were reported to have less stress with more attention to their career 39 40 plans by searching for new learning channels to increase their competitiveness. 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 MOOCs and Medicine 45 46 47 Of the enrolled students, 75.7% participated in at least one medical course with 23.3% completion 48 49 50 rate. They strongly agreed that these courses helped them develop theoretical background about the 51 52 topics discussed with less agreement on their role in developing their clinical skills. This raises 53 54 questions about the effectiveness of MOOCs with the current lecture-based teaching style in covering 55 56 57 the different aspects of medical education, including its clinical part, which needs student–patient 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 47 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 20 1 2 3 interaction. However, in the new evolving era of online learning, a question arises: “why to waste 4 5 6 precious class time on a lecture?” Students may watch the instructor’s instructor`s lecture remotely in 7 8 their homes and utilize class time for learning clinical skills.19 Most of the current opinions expect a 9 10 11 complementary role of MOOCs in undergraduate education with an increasing role in educating those 12 15 13 students after their graduation in continuing medical education. 14 15 MOOCs limitationsFor in Egypt peer review only 16 17 18 19 Lack of time and slow internet speed were the two main limitations reported for causing low 20 21 MOOCs enrollment and course completion rates. MOOCs, being a self-learning educational system, 22 23 24 require a considerable amount of time to choose courses, watch videos, take exams and interact through 25 26 discussions. This imposes burden on students, leading to the need of increased commitments besides 27 28 their busy regular medical education. Time management, either in the design of courses or from 29 30 31 participants, is critically needed to enhance their performance and increase completion rates. 32 33 http://bmjopen.bmj.com/ 34 Low internet speed is a commonly reported problem facing online education in developing 35 36 20 37 countries. This problem prolongs the time needed to watch high-quality videos or to download course 38 39 content, rendering students less adherent and more susceptible to dropout. The main solution to this 40 41 problem is enhancing the internet infrastructure in Egypt. Liyanagunawardena et al. suggested allowing 42 on September 24, 2021 by guest. Protected copyright. 43 44 lower resolution versions of the videos as an alternative solution to help engaging students with limited 45 46 bandwidth.8 InterestinglyIinterestingly, we did not find computer literacy, language or culture as 47 48 barriers, although it was expected that they would represent problems in Egypt, being a developing 49 50 51 country. 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 48 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 21 1 2 3 MOOCs experience satisfaction 4 5 6 7 Encouragingly, most of the participants who completed MOOCs (n= 25) were satisfied with the 8 9 overall experience. However, there was an obvious dissatisfaction regarding student–student and 10 11 12 student–instructor interactions. This problem is in pervasivecommon in online education in general, 13 14 with a lack of face-to-face interaction leading to some feelings of isolation and disconnectedness, which 15 For peer review only 16 are thought to be two main factors in dropout rates.21 Some MOOCs providers such as Coursera support 17 18 19 efforts beside the usual discussion forums for overcoming this point. These include more peer 20 21 assessments, social media groups, Google hangouts and real in-person Meetups. Despite that, more 22 23 24 involvement of participants is still needed to ensure the full psychological presence. 25 26 27 Study strengths and limitations 28 29 The strength of our study is that it included participants from all study years in 10 institutions, 30 31 32 covering nearly the entire geographic area of Egypt with high confidence interval (99%) and high 33 http://bmjopen.bmj.com/ 34 response rate (83.3%). However, our main limitation was the relatively low returned number of 35 36 37 participants who enrolled (n=136) and who had certificates (n=25), which makes analysis of limitations 38 39 and satisfactions less reliable. However, these results are important as a first start to make an 40 41 evidenceevidence about the real prevalence of MOOCs in Egypt to help the future studies to bluidbuild 42 on September 24, 2021 by guest. Protected copyright. 43 44 upon and take more representative samplessamples that are representative to the students who knew 45 46 about MOOCs for a better understanding of their experience. 47 48 49 50 Conclusions: 51 52 53 Approximately one-fifth of undergraduate medical students in Egypt have heard about MOOCs. 54 55 56 Students who actively participated showed a positive attitude toward the experience, but better time 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 49 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 22 1 2 3 management skills and faster internet connection speeds are required. FurthorFurther studies are needed 4 5 6 to address the enrolled students to assess their experience in large representative samples. In addition, 7 8 more efforts are needed to be done to raise the awareness of students of such courses as most of 9 10 11 students who did not hear about MOOCs, showed interest to participate. 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 50 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 23 1 2 3 4 STATEMENTS: 5 6 7 8 Acknowledgements: The authors deeply acknowledge HadeerAlsayed, Islam Shedeed (Menoufia 9 10 11 University), ZyadAbdelaziz,Dina Maklad, Ahmed Gebreil, Mahmoud Medhat (Alexandria university), 12 13 Mohammed Alhendy, AyaSobhy, Omar Azzam (Al-Azhar University in Cairo), Hassan AboulNour, 14 15 Sara Elganzory (TantaFor university), peer Mohamed review Eid, AyaTalaat, Mohamedonly Emad (BeniSuef university), 16 17 18 Mohamed Abdelzaheer, Ahmed Abdelhamed, Ahmed Saleh (Suez Canal university), Ahmed Zain, 19 20 KhaledGhaleb, Yossri Mohamed (Benha university), Ahmed Alaa, Mohamed Gamal (Assuit 21 22 university), Marina Nashed, Ibrahim Abdelmone'm (Ain Shams university), BassantAbdelazeim, 23 24 25 Ramadan Zaky (Cairo university) for their assistance in data collection. None of them received 26 27 compensation for their assistance. 28 29 30 31 Contributors: Aboshady, Radwan and Hassouna were responsible for the conception and design of the 32 33 study. Aboshady and Radwan coordinated the study and managed the data collection. Aboshady, 34 http://bmjopen.bmj.com/ 35 Radwan, Eltaweel, Azzam, Aboelnaga, Darwish, Hashem, Salah, Kotb, Afifi, Noaman and Salem 36 37 38 collected the data. Hassouna did the analyses, Aboshady, Radwan, Hassouna, Eltaweel, Kotb and 39 40 Aboelnaga contributed to interpretation of the findings. Aboshady, Eltaweel and Azzam wrote the first 41 on September 24, 2021 by guest. Protected copyright. 42 draft of the manuscript while Radwan, Hassouna, Aboelnaga, Kotb, Hashem, Salah, Darwish, Salem, 43 44 45 Afifi and Noaman made a critical revision of the manuscript for important intellectual content. All 46 47 authors approved the final version of the manuscript. 48 49 50 51 Funding: All funding required was provided by Aboshady and Radwan on their own expenses. 52 53 54 support for this project. 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 51 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 24 1 2 3 Competing interests: None. 4 5 6 7 Ethics approval: Institutional Review Board at Menoufia University, Faculty of Medicine, Egypt. 8 9 10 11 Data sharing statement: No additional data are available. 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 52 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 25 1 2 3 4 REFERENCES 5 6 7 8 1. Gooding I, Klaas B, Yager JD, Kanchanaraksa S. Massive Open Online Courses in Public Health. 9 10 11 Front Public Health 2013;1 doi: 10.3389/fpubh.2013.00059 12 13 14 2. Hoy MB. MOOCs 101: an introduction to massive open online courses. Med Ref Serv Q 15 For peer review only 16 2014;33(1):85-91 doi: 10.1080/02763869.2014.866490. 17 18 19 20 3. Pappano L. The Year of the MOOC.The New York Times 2013. 21 22 23 24 4. A Triple Milestone: 107 Partners, 532 Courses, 5.2 Million Students and Counting! Coursera Blog: 25 26 Coursera 2013. 27 28 29 30 5. Liyanagunawardena TR, Williams SA. Massive open online courses on health and medicine: review. 31 32 J Med Internet Res 2014;16:e191. doi:10.2196/jmir.3439 33 34 http://bmjopen.bmj.com/ 35 6. Mehta NB, Hull AL, Young JB, Stoller JK. Just imagine: new paradigms for medical education. 36 37 38 Acad Med 2013;88(10):1418-23 doi:0.1097/ACM.0b013e3182a36a07. 39 40 41

7. Cooke M, Irby DM, O'Brien O`Brien BC. Educating physicians: a call for reform of medical school on September 24, 2021 by guest. Protected copyright. 42 43 44 and residency: John Wiley & Sons, 2010; 25(2): 193–195 45 46 47 8. Liyanagunawardena T, Williams S, Adams A. The impact and reach of MOOCs:a developing 48 49 50 countries’ countries` perspective. eLearning Papers 2013(33) 51 52 53 9. Emanuel EJ. Online education: MOOCs taken by educated few. Nature 2013;503(7476):342-42 doi: 54 55 56 10.1038/503342a. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 53 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 26 1 2 3 10. Group ME. MOOCs @ Edinburgh 2013: Report #1: the University of Edinburgh, 2013. 4 5 6 7 11. Huhn C. UW‐Madison Massive Open Online Courses (MOOCs): Preliminary Participant 8 9 Demographics: Academic Planning and Institutional Research, 2013. 10 11 12 13 12. Bosslet GT, Torke AM, Hickman SE, Terry CL, Helft PR. The patient-doctor relationship and 14 15 online social networks:For results peer of a national review survey. J Gen Intern only Med 2011;26(10):1168-74 doi: 16 17 18 10.1007/s11606-011-1761-2|. 19 20 21 13. Kolowich S. The professors who make the MOOCs. The Chronicle of Higher Education 2013;25 22 23 24 25 14. Jordan K. Initial trends in enrolment and completion of massive open online courses. The 26 27 International Review of Research in Open and Distance Learning 2014;15(1) 28 29 30 31 15. Harder B. Are MOOCs the future of medical education? Bmj 2013;346:f2666 doi: 32 33 10.1136/bmj.f2666|. 34 http://bmjopen.bmj.com/ 35 36 37 16. Dahlin M, Joneborg N, Runeson B. Stress and depression among medical students: a cross-sectional 38 39 study. Med Educ 2005;39(6):594-604 doi: 10.1111/j.1365-2929.2005.02176.x. 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 17. Guthrie E, Black D, Bagalkote H, Shaw C, Campbell M, Creed F. Psychological stress and burnout 44 45 in medical students: a five-year prospective longitudinal study. J R Soc Med 1998;91(5):237-43 46 47 48 49 18. Bassols AM, Okabayashi LS, Silva AB, et al. First- and last-year medical students: is there a 50 51 difference in the prevalence and intensity of anxiety and depressive symptoms? Rev Bras Psiquiatr 52 53 (Sao Paulo, Brazil : 1999) 2014;0:0 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 54 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 27 1 2 3 19. Frehywot S, Vovides Y, Talib Z, et al. E-learning in medical education in resource constrained low- 4 5 6 and middle-income countries. Hum Resour Health 2013;11(1):4 doi: 10.1186/1478-4491-11-4. 7 8 9 20. Angelino LM, Williams FK, Natvig D. Strategies to Engage Online Students and Reduce Attrition 10 11 12 Rates. Journal of Educators Online 2007;4(2):n2 13 14 15 21. Prober CG, HeathFor C. Lecture peer halls without review lectures--a proposal only for medical education. N Engl J 16 17 18 Med 2012;366(18):1657-9 doi: 10.1056/NEJMp1202451. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 55 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 243x145mm (300 x 300 DPI) 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 57 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 186x105mm (300 x 300 DPI) 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 58 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 161x118mm (300 x 300 DPI) 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 59 of 59 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 STROBE Statement—Checklist of items that should be included in reports of cross-sectional studies 3 Item 4 No Recommendation 5 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 6 7 Perception and use of massive open online courses among medical students of a 8 developing country: multicenter cross-sectional study 9 (b) Provide in the abstract an informative and balanced summary of what was done 10 and what was found (Done) (page 4-5) 11 12 Introduction 13 Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 14 (Done) (page 7-8) 15 Objectives For3 peer State specific objectives, review including any prespecified only hypotheses (Done) (page 8, 16 17 last paragraph) 18 Methods 19 Study design 4 Present key elements of study design early in the paper (Done) (page 9, first 20 21 paragraph) 22 Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, 23 exposure, follow-up, and data collection (Done) (page 9-10) 24 Participants 6 (a) Give the eligibility criteria, and the sources and methods of selection of 25 26 participants (Done) (page 9, last paragraph) 27 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect 28 modifiers. Give diagnostic criteria, if applicable (Not applicable) 29 Data sources/ 8* For each variable of interest, give sources of data and details of methods of 30 measurement assessment (measurement). Describe comparability of assessment methods if there is 31 32 more than one group (Done) (page 11) 33 Bias 9 Describe any efforts to address potential sources of bias (Not done) http://bmjopen.bmj.com/ 34 Study size 10 Explain how the study size was arrived at (Done) (page 9) 35 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, 36 37 describe which groupings were chosen and why (Done) (page 11) 38 Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 39 (Done) (page 11) 40 (b) Describe any methods used to examine subgroups and interactions (Not 41 42 applicable) on September 24, 2021 by guest. Protected copyright. 43 (c) Explain how missing data were addressed (Not applicable) (no missing data) 44 (d) If applicable, describe analytical methods taking account of sampling strategy 45 (Done) (page 9,11) 46 e 47 ( ) Describe any sensitivity analyses (Not done) 48 Results 49 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially 50 eligible, examined for eligibility, confirmed eligible, included in the study, 51 52 completing follow-up, and analysed (Done) (page 13) 53 (b) Give reasons for non-participation at each stage (Not done) 54 (c) Consider use of a flow diagram (Done) (Figure 1) 55 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and 56 57 information on exposures and potential confounders (Done) (Table 1) 58 (b) Indicate number of participants with missing data for each variable of interest 59 (Not Done) (No missing data) 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml1 BMJ Open Page 60 of 59 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 Outcome data 15* Report numbers of outcome events or summary measures (Done) (page 13-14) 3 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and 4 their precision (eg, 95% confidence interval). Make clear which confounders were 5 6 adjusted for and why they were included (Done) (page 13-17) 7 (b) Report category boundaries when continuous variables were categorized (Not 8 applicable) 9 (c) If relevant, consider translating estimates of relative risk into absolute risk for a 10 meaningful time period (Not applicable) 11 12 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and 13 sensitivity analyses (Not applicable) 14 Discussion 15 For peer review only 16 Key results 18 Summarise key results with reference to study objectives (Done) (page 18) 17 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or 18 imprecision. Discuss both direction and magnitude of any potential bias (Done) 19 (page 21) 20 21 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, 22 multiplicity of analyses, results from similar studies, and other relevant evidence 23 (Done) (page 18-21) 24 Generalisability 21 Discuss the generalisability (external validity) of the study results (Done) (page 6, 25 21) 26 27 Other information 28 Funding 22 Give the source of funding and the role of the funders for the present study and, if 29 applicable, for the original study on which the present article is based (Not 30 31 applicable) (No external funding) 32 33 *Give information separately for exposed and unexposed groups. http://bmjopen.bmj.com/ 34 35 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 36 37 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 38 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 39 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 40 available at www.strobe-statement.org. 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml2 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

Perception and use of massive open online courses among medical students of a developing country: multicenter cross-sectional study

For peer review only Journal: BMJ Open

Manuscript ID: bmjopen-2014-006804.R2

Article Type: Research

Date Submitted by the Author: 02-Dec-2014

Primary Subject Medical education and training Heading:

Secondary Subject Heading: Medical education and training

Computer-Assisted Instruction , Medical Education , Distance Education , Keywords: MOOCs, Egypt

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 1 1 2 3 4 Title Page 5 6 7 Title 8 9 10 Perception and use of massive open online courses among medical students of a 11 12 13 developing country: multi-centre cross-sectional study 14 15 Authors For peer review only 16 17 18 - OmarA. Aboshady 19 20 6th year medical student, Faculty of Medicine, Menoufia University, Shebin El-Kom, Menoufia, Egypt. 21 22 [email protected] 23 24 25 - Ahmed E. Radwan 26 27 6th year medical student, Faculty of Medicine, Menoufia University, Shebin El-Kom, Menoufia, Egypt. 28 29 [email protected] 30 31 - Asmaa R. Eltaweel 32 33

th http://bmjopen.bmj.com/ 34 6 year medical student, Faculty of Medicine, Alexandria University, Alexandria, Egypt. 35 36 [email protected] 37 38 - Ahmed Azzam 39 40 6th year medical student, Faculty of Medicine, Al-Azhar University, Cairo, Egypt. 41 42 on September 24, 2021 by guest. Protected copyright. 43 [email protected] 44 45 - Amr A. Aboelnaga 46 47 5th year medical student, Faculty of Medicine, Tanta University, Tanta, Egypt. 48 49 [email protected] 50 51 52 - Heba A. Hashem 53 th 54 6 year medical student, Faculty of Medicine, Beni Suef University, BeniSuef, Egypt. 55 56 [email protected] 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 2 1 2 3 - Salma Y. Darwish 4

18 Assuit 180 33 147 33 6 27 6 2 4

19 (8.5%) (18.3%) (81.7%) (18.2%) (81.8%) (33.3%) (66.7%)

20 Benha 205 57 148 57 16 41 16 0 16 21 (9.7%) (27.8%) (72.2%) (28.1%) (71.9%) (0.0%) (100.0%) 22 BeniSuef 220 38 182 P= 38 6 32 P= 6 0 6 P= 23 (10.4%) (17.3%) (82.7%) .04 (15.8%) (84.2%) .13 (0.0%) (100.0%) .02 24 Cairo 188 39 149 39 12 27 12 2 10 25 (8.9%) (20.7%) (79.3%) (30.8%) (69.2%) (16.7%) (83.3%) 26 Menoufia 248 53 195 53 22 31 22 10 12 27 (11.8%) (21.4%) (78.6%) (41.5%) (58.5%) (45.5%) (54.5%) 28 Suez 199 59 140 59 20 39 20 2 18 29 Canal (9.4%) (29.6%) (70.4%) (33.9%) (66.1%) (10.0%) (90.0%) 30 Tanta 221 49 172 49 11 38 11 1 10 31 (10.5%) (22.2%) (77.8%) (22.4%) (77.6%) (9.1%) (90.9%) 32 33Class Academic 1076 176 900 176 40 136 40 4 36 34 (51.2%) (16.4%) (82.6%) P< (22.7%) (77.3%) P= (10.0%) (90.0%) P= http://bmjopen.bmj.com/ 35 Clinical 1024 280 744 .001 280 96 184 .01 96 21 75 .1 36 (48.8%) (27.3%) (72.7%) (34.3%) (65.7%) (21.9%) (78.1%) 37Gender Male 926 199 730 199 71 128 71 17 54 38 (44.1%) (21.4%) (78.6%) P= (35.7%) (64.3%) P= (23.9%) (76.1%) P= 39 Female 1174 257 920 .83 257 65 192 .02 65 8 57 .08 40 (55.9%) (21.8%) (78.2%) (25.3%) (74.7%) (12.3%) (87.7%) 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 15 1 2 3 Enrolment and Certificate Attainment 4 5 6 7 Of those who knew about MOOCs, 136 (29.8% [99% CI,24.3%–35.3%]) were enrolled in at least 8 9 one course. Most students (125; 91.9%) registered in 1–5 courses, with 113 (83.1%) students reporting 10 11 12 having watched at least one video lecture. Home (109; 99%) was the primary place where they watched 13 14 these videos. There was no statistically significant difference in enrolment between males and females 15 For peer review only 16 (52.2% vs. 47.8%, 99% CI, P= .016). However, there was a significant difference between students’ 17 18 19 class and their enrolment (P=.009) (Table 1). Coursera was the most commonly used website (99; 20 21 72.8%), followed by Edx (14; 10.3%). 22 23 24 25 Only 25 students (18.4% [99% CI, 9.8%–26.9%]) completed courses and attained one certificate or 26 27 more with an 81.6% dropout rate. Interestingly, more than half of students who earned certificates (13; 28 29 52% [99% CI,26.3%–77.7%]) have used the signature track to obtain verified certificates from the 30 31 32 universities that offered the courses. The vast majority of enrolled students stated that getting a 33 http://bmjopen.bmj.com/ 34 certificate was important to them (32 [23.5%] very important, 37 [27.2%] important, 50 [36.8%] 35 36 37 important to some extent, and 17 [12.5%] not important). Out of the 69 students who assumed that 38 39 getting a certificate is important before enrolment (important/very important), 17(24.6%) were finally 40 41 certified, as compared to only 8(11.6%) certified students out of the 67 who were not concerned to 42 on September 24, 2021 by guest. Protected copyright. 43 44 receive a certificate at the time of enrolment (important to some extent/not important; 11.9%); P< .001. 45 46 47 Ways of Knowledge and Students’ Motivations 48 49 50 To assess how students found out about MOOCs and what their motivations were, two multi- 51 52 selection questions were asked. Social media was the primary way through which 206 (45.2%) students 53 54 were introduced to MOOCs, while knowledge through a friend was the second (184; 40.4%). Web- 55 56 57 search engines (87; 19.1%) took the third place, followed by extracurricular activities (46; 10.1%). 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 16 1 2 3 MOOCs providers’ advertisements played a very small role (27; 5.9%) in reaching students as did the 4 5 6 official websites of medical schools (15; 3.3%). Notably, there was no association between the method 7 8 through which students learnt about MOOCs and their enrolment. Nevertheless, students who were 9 10 11 introduced through extracurricular activities were found to enrol more frequently (P= .005). 12 13 14 Concerning students’ motives, most students reported that their main motivation was “to learn new 15 For peer review only 16 things” followed by “to help me study medicine” (Figure 2). Interestingly, the students who enrolled 17 18 19 aiming to have a certificate or to help them in obtaining a future job were significantly more likely to 20 21 complete the courses (P= .001 and P= .008, respectively). 22 23 24 25 MOOCs and Medicine 26 27 By asking the enrolled students (n=136) about their experience and attitude toward medical MOOCs, 28 29 103 (75.7% [99% CI, 66.2%–85.2%]) declared participation in at least one medical course. Of them, 24 30 31 32 students (17.6% [99% CI, 7.9%–27.3%]) had completed medical courses and earned certificates. 33 http://bmjopen.bmj.com/ 34 Regarding their medical MOOCs experience, 102 (75%) students agreed that MOOCs helped them in 35 36 37 developing their theoretical background about the topic discussed. However, there was less agreement 38 39 (68; 50%) on the role of MOOCs in developing their practical skills. Most students (89; 86.4%) agreed 40 41 that MOOCs helped in studying medicine, while 83 (61%) believed that MOOCs will help them in 42 on September 24, 2021 by guest. Protected copyright. 43 44 securing a more desirable, better job opportunity in the future. 45 46 47 Limitations of MOOCs 48 49 50 51 Our study reported two types of limitations: enrolment and completion. Students who knew about 52 53 MOOCs, but did not enrol in any courses (n=320) were asked about their enrolment limitations. The 54 55 56 majority of students (226; 70.4%) agreed that a lack of time was the main limitation, while 147 (45.9%) 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 17 1 2 3 agreed that slow Internet speed was another cause (Figure 3). Enrolled students (n=136) were asked to 4 5 6 assess the limitations that made them drop out of courses. Similar to the enrolment limitations, lack of 7 8 time (105; 77.2%) and slow Internet speed (73; 53.7%) were the main obstacles. Lack of technology 9 10 11 access, computer literacy, language difficulty and culture conflicts were less frequently selected as a 12 13 limiting factor to completion of the course (Figure 3). Only 16 (11.8%) students agreed that the 14 15 scientific contentFor was difficult peer for them to comprehend.review In addition, only 93 (68.4%) students disagreed that 16 17 18 “lower content than expected” wasa limitation. 19 20 21 For further assessment of Internet speed, we asked the enrolled students to rate their Internet speed. 22 23 24 Sixty students (44.1%) reported that the speed was reasonable, while 55 (40.4%) reported slow speed, 25 26 and only 21 (15.4%) had a high connection speed. When we compared the students’ evaluation of 27 28 Internet speed and whetherthey watched video lectures, we did not find a significant association (P= 29 30 31 .69). 32 33 http://bmjopen.bmj.com/ 34 Students’Satisfaction of MOOCs 35 36 37 38 The 25 students who obtained certificates were asked to report their opinions about each part of the 39 40 MOOCs experience. The results showed that most students (21; 84%) were satisfied with the overall 41 42 on September 24, 2021 by guest. Protected copyright. 43 experience, including video lectures (18; 70%), exams and assignments (16; 64%), quality of the 44 45 presented materials (21; 84%) and the technology used (20; 80%). However, there was less satisfaction 46 47 regarding student–student (5; 20%) and student–instructor (8; 32%) interactions (Figure 4). 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 18 1 2 3 4 DISCUSSION 5 6 7 8 Available information regarding MOOCs participants is primarily data obtained from course-end 9 10 11 demographics, which usually demonstrate aheterogeneous population of varying age groups, 12 13 educational levels and countries globally. These data show that most MOOCs’ users are well-educated 14 15 males with low participationFor peer from developing review countries and undergraduates. only9-11 To our knowledge, this 16 17 18 study is the first in the medical field and a developing country to use a cross-sectional study design in a 19 20 homogeneous population for the assessment of prevalence and uptake of such courses among 21 22 undergraduate medical students. 23 24 25 26 Knowledge and Enrolment 27 28 29 Our results demonstratea funnel-shaped participation pattern, with 22.7 % of the respondents 30 31 32 knowing about MOOCs and 6.5% actually enrolled. Moreover, only 5.4% watched the offered video 33 http://bmjopen.bmj.com/ 34 lectures and 1.2% obtained certificates of completion. Although there are no similar cross-sectional 35 36 37 studies with which our results can be compared, the knowledge that approximately one-fifth of the 38 39 Egyptian medical students are familiar withMOOCs is considered promising in a developing country 40 41 that depends mainly on traditional education. Additionally, these courses are still new, and MOOCs 42 on September 24, 2021 by guest. Protected copyright. 43 44 providers’ advertisements had little effect in reaching students. Also, there was no medical MOOC 45 46 offered by an Egyptian institution until now. Social media and the sharing of personal experiences 47 48 among friends played a vital role in the spread of the MOOCs, raising students’ awareness to its current 49 50 51 level. This is in line with the increasing role of social media websites in medical students’ lives, with 52 53 more than 90% of medical students in the U.S. using social media.12 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 19 1 2 3 Notably, it was obvious that there was a gap between knowledge of MOOCs and enrolment in them, 4 5 6 with only one-third of students who knew about MOOCs actually registering in courses. Students 7 8 reported a lack of time and low Internet speed as the main limitations for MOOC use. Out of these 9 10 11 students, 18.4% (23.3% when looking at those enrolled in a medical course) completed the courses and 12 13 earned certificates. These completion rates are higher than the reported average completion rates in the 14 15 course demographics.For In 2013, peer The Chronicle review of Higher Education only suggested an average of 7.5% 16 17 13 14 18 completion rate , while a recent study in 2014 reported a rate of about 6.5%. This may be explained 19 20 by the importance reported by students that obtaining certificates has in terms of adding to their 21 22 resumes in the hope of improving future employment opportunities. It is interesting to note that about 23 24 25 half of them paid to verify their certificates, although there is no academic credit for undergraduates for 26 27 any MOOCs from any medical school in the U.S.15 and Egypt at this time. 28 29 30 31 Although there was no association between gender and students’ knowledge or enrolment, class year 32 33 had a significant association. Clinical years’ students were found to have higher knowledge and 34 http://bmjopen.bmj.com/ 35 enrolment rates. This may be due to the high level of stress and pressures experienced by early-year 36 37 38 medical students adapting to new academic systems with little time available for extracurricular 39 40 activities.16 In contrast, students in their final years were reported to have less stress 16-18 with more 41 on September 24, 2021 by guest. Protected copyright. 42 concern about their career plans, and searching for new learning channels to increase their 43 44 45 competitiveness. 46 47 48 MOOCs and Medicine 49 50 51 52 Of the enrolled students, 75.7% participated in at least one medical course with a 23.3% completion 53 54 rate. They strongly agreed that these courses helped them to develop theoretical backgrounds on the 55 56 57 topics discussed with less agreement on their role in developing their clinical skills. This raises 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 20 1 2 3 questions about the effectiveness of MOOCs with the current lecture-based teaching style in covering 4 5 6 the different aspects of medical education, including its clinical part, which requires student–patient 7 8 interaction. However, in the new and evolving era of online learning, the question of why waste 9 10 11 precious class time on a lecture? arises. Students may watch the instructor’s lecture remotely in their 12 19 13 homes and useclass time for learning clinical skills. Most current opinions anticipate a 14 15 complementary Forrole for MOOCs peer in undergraduate review education, with only an increasing role in educating those 16 17 15 18 students after their graduation in continuing medical education. 19 20 21 MOOCs Limitations in Egypt 22 23 24 25 Lack of time and slow Internet speed were the two main limitations reported for causing low 26 27 MOOCs enrolment and course completion rates. MOOCs, being a self-learning educational system, 28 29 requires a considerable amount of time to choose courses, watch videos, take exams and interact 30 31 32 through discussions. This imposes a significant time burden on students, leading to the need for an 33 http://bmjopen.bmj.com/ 34 increased commitment beyond their busy regular medical education. Time management, either in the 35 36 37 design of courses or from participants, is critical to the enhancement of their performance and increased 38 39 completion rates. 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 Low Internet speed is a commonly reported problem facing online education in developing 44 20 45 countries. This problem prolongs the time needed to watch high-quality videos or to download course 46 47 content, rendering students less adherent and more susceptible to dropout. The main solution to this 48 49 50 problem is enhancing the Internet infrastructure in Egypt. Liyanagunawardena et al. suggested allowing 51 52 lower resolution versions of the videos as an alternative solution to help engaging students with limited 53 54 bandwidth.8Interestingly, we did not find computer literacy, language or culture as barriers, although it 55 56 57 was expected that they would represent problems in Egypt, being a developing country. 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 21 1 2 3 MOOCs Experience Satisfaction 4 5 6 7 Encouragingly, most of the participants who completed MOOCs (n= 25) were satisfied with the 8 9 overall experience. However, there was an obvious dissatisfaction regarding student–student and 10 11 12 student–instructor interactions. This problem is common in online education in general, with a lack of 13 14 face-to-face interaction leading to some feelings of isolation and disconnectedness, which are thought 15 For peer review only 16 to be two main factors affecting dropout rates.21 Some MOOCs providers, such as Coursera, support 17 18 19 efforts beyond the usual discussion forums to help overcome this issue. These efforts include more peer 20 21 assessments, social media involvment, Google+ hangouts and real in-person meet-ups. Despite that, 22 23 24 more involvement of participants is needed to ensure the full psychological presence. 25 26 27 Study Strengths and Limitations 28 29 The strength of our study is that it included participants from all study years in ten institutions, 30 31 32 covering nearly the entire geographic area of Egypt with a high confidence interval (99%) and high 33 http://bmjopen.bmj.com/ 34 response rate (83.3%). However, our main limitation was the relatively low returned number of 35 36 37 participants who enrolled (n=136) and who had certificates (n=25), which makes the analysis of 38 39 limitations and satisfaction of MOOCs less reliable. However, these results provide an important 40 41 contribution as a first stepin gathering evidence about the prevalence of perception and use of MOOCs 42 on September 24, 2021 by guest. Protected copyright. 43 44 in Egypt. In addition, these results will facilitate the ability of future studies to build upon our findings 45 46 and select samples that are representative of students with prior knowledge of MOOCs, leading to a 47 48 better understanding of their experience. 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 22 1 2 3 Conclusions 4 5 6 7 About one-fifth of in undergraduate medical students Egypt have heard about MOOCs. Students who 8 9 actively participated showed a positive attitude towards the experience, but better time management 10 11 12 skills and faster Internet connection speeds are required. Further studies are needed involving enrolled 13 14 students in large representative samples, to assess their experiences using MOOCs. In addition, more 15 For peer review only 16 effort is needed to raise the awareness among students of such courses, as most students who had not 17 18 19 heard about MOOCs did show interest in participating once they became aware of the courses. 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 23 1 2 3 4 STATEMENTS 5 6 7 8 Acknowledgements: The authors acknowledge Hadeer Alsayed, Islam Shedeed (Menoufia 9 10 11 University), Zyad Abdelaziz, Dina Maklad, Ahmed Gebreil, Mahmoud Medhat (Alexandria 12 13 University), Mohammed Alhendy, Aya Sobhy, Omar Azzam (Al-Azhar University in Cairo), Hassan 14 15 AboulNour, SaraFor Elganzory peer (Tanta University), review Mohamed Eid, Ayaonly Talaat, Mohamed Emad (Beni 16 17 18 Suef University), Mohamed Abdelzaheer, Ahmed Abdelhamed, Ahmed Saleh (Suez Canal University), 19 20 Ahmed Zain, Khaled Ghaleb, Yossri Mohamed (Benha University), Ahmed Alaa, Mohamed Gamal 21 22 (Assuit University), Marina Nashed, Ibrahim Abdelmone'm (Ain Shams University), and Bassant 23 24 25 Abdelazeim, Ramadan Zaky (Cairo University) for their highly-valued assistance in data collection. In 26 27 addition, we acknowledge Bishoy Gouda (Canada), Susannah L. Bodman (U.S.), Melanie Haines, 28 29 Marion Mapham (Australia), Mohamed Aleskandarany (U.K.) and Moahmed Alaa (Egypt) for their 30 31 32 much-appreciated help in the English revision of our paper. None of them received compensation for 33 http://bmjopen.bmj.com/ 34 their assistance. 35 36 37 38 Contributors: Aboshady, Radwan and Hassouna were responsible for the conception and design of the 39 40 study. Aboshady and Radwan coordinated the study and managed the data collection. Aboshady, 41 on September 24, 2021 by guest. Protected copyright. 42 Radwan, Eltaweel, Azzam, Aboelnaga, Darwish, Hashem, Salah, Kotb, Afifi, Noaman and Salem 43 44 45 collected the data. Hassouna did the analyses;Aboshady, Radwan, Hassouna, Eltaweel, Kotb and 46 47 Aboelnaga contributed to interpretation of the findings. Aboshady, Eltaweel and Azzam wrote the first 48 49 50 draft of the manuscript while Radwan, Hassouna, Aboelnaga, Kotb, Hashem, Salah, Darwish, Salem, 51 52 Afifi and Noaman made a critical revision of the manuscript for important intellectual content. All 53 54 authors approved the final version of the manuscript. 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 24 1 2 3 Funding: All funding required was provided by Aboshady and Radwan on their own expenses. 4 5 6 7 Competing interests: None. 8 9 10 11 Ethics approval: Institutional Review Board at Menoufia University, Faculty of Medicine, Egypt. 12 13 14 Data sharing statement: No additional data are available. 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 25 1 2 3 4 REFERENCES 5 6 7 8 1. Gooding I, Klaas B, Yager JD, Kanchanaraksa S. "Massive Open Online Courses in Public 9 10 11 Health."Front Public Health 2013;1 doi: 10.3389/fpubh.2013.00059. 12 13 14 2. Hoy MB. "MOOCs 101: An introduction to massive open online courses."Med Ref Serv Q 15 For peer review only 16 2014;33(1):85-91 doi: 10.1080/02763869.2014.866490. 17 18 19 20 3. Pappano L. “The Year of the MOOC.” The New York Times 2013. 21 22 23 24 4. "A Triple Milestone: 107 Partners, 532 Courses, 5.2 Million Students and Counting!"Coursera Blog: 25 26 Coursera 2013. 27 28 29 30 5. Liyanagunawardena TR, Williams SA. "Massive open online courses on health and medicine: 31 32 Review."J Med Internet Res 2014;16:e191. doi:10.2196/jmir.3439. 33 34 http://bmjopen.bmj.com/ 35 6. Mehta NB, Hull AL, Young JB, Stoller JK. "Just imagine: New paradigms for medical 36 37 38 education."Acad Med 2013;88(10):1418-23 doi:0.1097/ACM.0b013e3182a36a07. 39 40 41

7. Cooke M, Irby DM, O`Brien BC. Educating physicians: A call for reform of medical school and on September 24, 2021 by guest. Protected copyright. 42 43 44 residency: John Wiley & Sons, 2010; 25(2): 193–195. 45 46 47 8. Liyanagunawardena T, Williams S, Adams A. "The impact and reach of MOOCs:a developing 48 49 50 countries` perspective."eLearning Papers 2013(33). 51 52 53 9. Emanuel EJ. "Online education: MOOCs taken by educated few."Nature 2013;503(7476):342-42 54 55 56 doi: 10.1038/503342a. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 26 1 2 3 10. Group ME. MOOCs @ Edinburgh 2013: Report #1: the University of Edinburgh, 2013. 4 5 6 7 11. Huhn C. UW‐Madison Massive Open Online Courses (MOOCs): Preliminary Participant 8 9 Demographics: Academic Planning and Institutional Research, 2013. 10 11 12 13 12. Bosslet GT, Torke AM, Hickman SE, Terry CL, Helft PR. "The patient-doctor relationship and 14 15 online social networks:For Results peer of a national review survey."J Gen Intern only Med 2011;26(10):1168-74 doi: 16 17 18 10.1007/s11606-011-1761-2|. 19 20 21 13. Kolowich S. "The professors who make the MOOCs."The Chronicle of Higher Education 2013;25. 22 23 24 25 14. Jordan K. "Initial trends in enrolment and completion of massive open online courses."The 26 27 International Review of Research in Open and Distance Learning 2014;15(1). 28 29 30 31 15. Harder B. "Are MOOCs the future of medical education?"Bmj 2013;346:f2666 doi: 32 33 10.1136/bmj.f2666|. 34 http://bmjopen.bmj.com/ 35 36 37 16. Dahlin M, Joneborg N, Runeson B. "Stress and depression among medical students: A cross- 38 39 sectional study."Med Educ 2005;39(6):594-604 doi: 10.1111/j.1365-2929.2005.02176.x. 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 17. Guthrie E, Black D, Bagalkote H, Shaw C, Campbell M, Creed F. "Psychological stress and 44 45 burnout in medical students: A five-year prospective longitudinal study."J R Soc Med 1998;91(5):237- 46 47 43. 48 49 50 51 18. Bassols AM, Okabayashi LS, Silva AB, et al. "First- and last-year medical students: Is there a 52 53 difference in the prevalence and intensity of anxiety and depressive symptoms?"Rev Bras Psiquiatr 54 55 56 (Sao Paulo, Brazil : 1999) 2014;0:0. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 27 1 2 3 19. Frehywot S, Vovides Y, Talib Z, et al. "E-learning in medical education in resource constrained 4 5 6 low- and middle-income countries."Hum Resour Health 2013;11(1):4 doi: 10.1186/1478-4491-11-4. 7 8 9 20. Angelino LM, Williams FK, Natvig D. "Strategies to Engage Online Students and Reduce Attrition 10 11 12 Rates."Journal of Educators Online 2007;4(2):n2. 13 14 15 21. Prober CG, HeathFor C. "Lecture peer halls without review lectures-a proposal only for medical education."N Engl J 16 17 18 Med 2012;366(18):1657-9 doi: 10.1056/NEJMp1202451. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 1 1 2 3 4 Title Page 5 6 7 Title 8 9 10 Perception and use of massive open online courses among medical students of a 11 12 13 developing country: multicenter cross-sectional study 14 15 Authors For peer review only 16 17 18 - Omar A. Aboshady 19 20 6th year medical student, Faculty of Medicine, Menoufia University, Shebin El-Kom, Menoufia, Egypt. 21 22 [email protected] 23 24 25 - Ahmed E. Radwan 26 27 6th year medical student, Faculty of Medicine, Menoufia University, Shebin El-Kom, Menoufia, Egypt. 28 29 [email protected] 30 31 - Asmaa R. Eltaweel 32 33

th http://bmjopen.bmj.com/ 34 6 year medical student, Faculty of Medicine, Alexandria University,Alexandria, Egypt. 35 36 [email protected] 37 38 - Ahmed Azzam 39 40 6th year medical student, Faculty of Medicine, Al-Azhar University in Cairo, Cairo, Egypt. 41 42 on September 24, 2021 by guest. Protected copyright. 43 [email protected] 44 45 - Amr A. Aboelnaga 46 47 5th year medical student, Faculty of Medicine, Tanta University, Tanta, Egypt. 48 49 [email protected] 50 51 52 - Heba A. Hashem 53 th 54 6 year medical student, Faculty of Medicine, BeniSuef University, BeniSuef, Egypt. 55 56 [email protected] 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 2 1 2 3 - Salma Y. Darwish 4

Study Population and Sample on September 24, 2021 by guest. Protected copyright. 42 43 44 45 Our target population was undergraduate medical students in Egypt enroled enrolled in 19 medical 46 47 schools for during the 2013-14 academic year 2013/14. We selected ten 10 out of the 19 medical 48 49 50 schools to be our study settings using simple random sampling technique. The sampleSelected 51 52 institutions included Ain Shams, Al-Azhar medical school in Cairo, Alexandria, Assiut, Benha, 53 54 BeniSuef, Cairo, Menoufia, Suez Canal and Tanta medical schools. 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 38 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 10 1 2 3 Students in these schools are enroledwere enrolled in a six-year MBBCh program, in which the first 4 5 6 three years are called academic years and the last three years are called clinical years. According toTo 7 8 achieve a 99% confidence interval (CI), 3% margin of error and 50% response distribution; , 1784 9 10 11 students were required to represent the study population. We used a stratified simple random technique 12 13 to select our sample with an equal allocation of participants in each university and each study year. 14 15 Accordingly, usingFor the registered peer students`names review lists, we randomly only selected 270 students from each 16 17 18 faculty (45 for each study year) for a total of 2700 participants. We excluded non-Egyptians students 19 20 and those who changed their enrolment school at the time of data collection. 21 22 23 24 25 26 27 28 29 30 31 Data collectionCollection 32 33 http://bmjopen.bmj.com/ 34 Selected participants were invited by e-mail and social media websites to participate in our survey 35 36 37 using a unique code for each participant during the period of March–April 2014. In each university, a 38 39 team of data collectors was recruited (two active members from each class), which were led by a local 40 41 study coordinator (LSC). This team received standardized training on how to approach selected 42 on September 24, 2021 by guest. Protected copyright. 43 44 students either online or offline. Each LSC was responsible for obtaining the students’ lists for each 45 46 class withthrough official channels. The two principle investigators selected the students randomly 47 48 from these lists according to the planned sampling technique. Initially, participants from two 49 50 51 universities were invited using their official emails. However, there was very low response rate as most 52 53 many students do not check their emails regularly, which is partially explained by the fact that this 54 55 56 email service was not introduced into Egyptian universities until recently. Therefore, we shifted our 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 39 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 11 1 2 3 data collection plan to the use of social media websites (mainly Facebook). The majority of Egyptian 4 5 6 medical students have Facebook accounts, and each class has a Facebook group, including all students 7 8 of this class, for study-related discussions. The two data collectors of each class were responsible for 9 10 11 obtaining the personal account of each selected student. To confirm that the collected account 12 13 edbelonges to the selected student, a personal message was sent first to this account to confirm his 14 15 personal details.For After receiving peer the confirmation, review a Facebook messageonly was sent containing a cover 16 17 18 letter with study aims, the participant’s special code and a link for the online questionnaire. The student 19 20 was to first fill out a voluntary consent form after reading the study aims and instructions. We sent up 21 22 to five reminder messages to participants to complete the survey. If we did not get responses in two to 23 24 25 three weeks, non-responders were approached in lectures’ rooms and training sessions to ask them to 26 27 complete the questionnaire. If any of them informed us of a lack of Internet access, a paper version of 28 29 the questionnaire (same questions and format as the online version) was provided for immediate 30 31 32 completion, if the respondent agreed to participate. LSC were responsible for entering the data into our 33 http://bmjopen.bmj.com/ 34 online system. We used an online survey program to administer the questionnaire (Survey Gizmo; 35 36 37 Boulder, Colorado, U.S.). 38 39 40 We invited the selected participants via e-mail and social media websites to take our survey using a 41 on September 24, 2021 by guest. Protected copyright. 42 unique code for each participant during the period of March–April 2014. We used an online survey 43 44 45 program to administer the questionnaire (Survey Gizmo; Boulder, Colorado, US). Students who did not 46 47 have access to the internet at the time of data collection were allowed to record their responses using a 48 49 50 self-administered paper version of the questionnaire. We sent up to five reminder messages for 51 52 participants to complete the survey. The participants were informed about the study aims in the cover 53 54 letter, and they voluntarily consented to participate with no incentives. 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 40 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 12 1 2 3 Questionnaire Development 4 5 6 7 The study questionnaire was developed by the research team through group discussions after an 8 9 extensive literature review. The draft was then reviewed by two experts in the fields of medical 10 11 12 education and Biostatisticsbiostatistics. The questionnaire was then piloted on 175 students, from all 13 14 participating medical schools.We used the final draft in a pilot testing on 175 students in all 15 For peer review only 16 participating medical schools. Detailed feedback about the format, clarity and completion time was 17 18 19 collected and used to make minor changes.we made minor changes in response 20 21 toparticipants`comments. We did not include the pilot responses in our analysis. 22 23 24 25 The questionnaire was in Arabic, the participants’ native language, and it comprised included 29 26 27 questions in four sections using a branching logic function (Figure 1). The first section addressed study 28 29 aims, consent and participants`’ personal information. This section was followed by a main question 30 31 32 asking if the student had heard about the new open online educational system (MOOCs) provided in 33 http://bmjopen.bmj.com/ 34 websites like Coursra, Edx, Udacity and FutureLearn, among others. about their knowledge about 35 36 37 MOOCs. Based on this his answer, the participants wasere directed to different sections. Students who 38 39 knew about MOOCs were asked how they heard about it and their state of enrolment. If the participant 40 41 was not enrolled in any course, he/sherespondents was were asked about the limitations to their use, 42 on September 24, 2021 by guest. Protected copyright. 43 44 and then the questionnaire endsended. 45 46 47 Enrolled students were directed to the next section, which assessed their perspectives and 48 49 50 experiences with MOOCs. For students who gained certificates, further questions were asked regarding 51 52 their level of satisfaction as well as any obstacles they might have faced. Finally, four questions were 53 54 addressed asked to assess students’ opinion about integration of MOOCs in the medical field. 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 41 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 13 1 2 3 Most of the questions were in a single- answer multiple-choice questionsformat. However, there 4 5 6 were three multi-selection check-box questions. For the assessment of limitations, satisfaction and 7 8 opinions, a 5five-point Likert scale between one (strongly agree/satisfied) and five (strongly 9 10 11 disagree/unsatisfied) was used. 12 13 14 Statistical analysisAnalysis: 15 For peer review only 16 17 18 Results were presented as numbers and percentages with confidence interval at 99%. The significance 19 20 of the association between qualitative variables of interest was analyzed using Chichi-square test or 21 22 Fisher`s exact tests, as indicated. In order tTo focus on clear opinions, the 5five-point Likert scale of 23 24 25 limitations, satisfaction and opinions were was collapsed into three categories (agree/satisfied, neutral 26 27 and disagree/unsatisfied). Class year was recoded as a dichotomous variable in order to compare results 28 29 for students in academic versus clinical education. The acknowledgment of the importance of getting a 30 31 32 certificate before enrolment also was also recoded as a dichotomous variable (important/very important 33 http://bmjopen.bmj.com/ 34 versus limited importance/not important). This was to in order to test the significance of association 35 36 37 between the primarily reported importance of acquiring a certificate and the actual possession of the 38 39 certificate by McNemar test. All tests were bilateral and a P value of 0.01 was used as the limit of 40 41 statistical significance. Statistical analysis was performed using the IBM SPSS statistical software 42 on September 24, 2021 by guest. Protected copyright. 43 44 package version 22. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 42 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 RESULTS 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 43 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 15 1 2 3 Respondent Characteristics 4 5 6 7 Of 2700 total participants, 62 (2.3%) were excluded for being non-Egyptians or having changed their 8 9 enrolment school, in addition to 111 (4.1%) students who could not be reached, resulting in final 10 11 12 eligible cohort of 2527 studentsin addition to 111 (4.1%) students` whose contact information could not 13 14 be reached with final eligible 2527 students. During the data collection phase, 2357 (93.3%) online 15 For peer review only 16 questionnaire invitations and 170 (6.7%) paper versions were sent out. Out of these distributed 17 18 19 questionnaires, 2016 responses were received (response rate 83.3%). Table 1 showes participants’ 20 21 `demographics regarding school, class and gender. 22 23 24 25 Knowledge about MOOCs 26 27 We found that 456 (21.7% [99% CI, 19.4%–24%]) students had heard about MOOCs or websites 28 29 providing such courses. There was no statistically significant difference in knowledge between males 30 31 32 and females (43.6% vs. 56.4%, 99 CI, P = .8). However, clinical years`’ students had higher rates of 33 http://bmjopen.bmj.com/ 34 knowledge than students in the academic years (P< .001) (Table 1). Additionally, there was no 35 36 37 difference between medical schools in students’ knowledge about MOOCs (P=.04). 38 39 40 After informing the students who did not know about MOOCs that this system provides scientific 41 42 on September 24, 2021 by guest. Protected copyright. 43 courses in different disciplines by specialists from top universities worldwide for no or low fees 44 45 through the Iinternet, 1342 (81.3% [99% CI, 78.8%–83.8%]) students showed an interest to 46 47 participatein participating with a significant difference among different medical schools (P< .001). 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 44 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 16 1 2 3 Table 1. Participants’ demographics and their state of knowledge, enrollment and certificate 4 5 6 attainment. 7 8 9 Knowledge about MOOCs P Enrollment in courses P Certificate Attainment P 10 value value value 11 Total (%) Yes (%) No (%) Total Yes (%) No (%) Total Yes (%) No (%) 12 (n=2106) (n=456) (n=1650) (n=456) (n=136) (n=320) (n=136) (n=25) (n=111) 13 Faculty Ain Shams 207 38 169 38 13 25 13 3 10 14 15 (9.8%) For(18.4%) (81.6peer %) review(34.2%) (65.8%) only (23.1%) (76.9%) 16 Al-Azhar 216 42 174 42 11 31 11 1 10 17 (10.3%) (19.4%) (80.6%) (26.2%) (73.8%) (9.1%) (90.9%) 18 Alexandria 222 48 174 48 19 29 19 4 15 19 (10.5%) (21.6%) (78.4%) (39.6%) (60.4%) (21.1%) (78.9%) 20 Assuit 180 33 147 33 6 27 6 2 4 21 (8.5%) (18.3%) (81.7%) (18.2%) (81.8%) (33.3%) (66.7%) 22 Benha 205 57 148 57 16 41 16 0 16 23 (9.7%) (27.8%) (72.2%) (28.1%) (71.9%) (0.0%) (100.0%) 24 Beni Suef 220 38 182 P= 38 6 32 P= 6 0 6 P= 25 (10.4%) (17.3%) (82.7%) .04 (15.8%) (84.2%) .13 (0.0%) (100.0%) .02 26 Cairo 188 39 149 39 12 27 12 2 10 27 (8.9%) (20.7%) (79.3%) (30.8%) (69.2%) (16.7%) (83.3%) 28 Menoufia 248 53 195 53 22 31 22 10 12 29 (11.8%) (21.4%) (78.6%) (41.5%) (58.5%) (45.5%) (54.5%) 30 31 Suez 199 59 140 59 20 39 20 2 18 32 Canal (9.4%) (29.6%) (70.4%) (33.9%) (66.1%) (10.0%) (90.0%) 33 Tanta 221 49 172 49 11 38 11 1 10 34 (10.5%) (22.2%) (77.8%) (22.4%) (77.6%) (9.1%) (90.9%) http://bmjopen.bmj.com/ 35Class Academic 1076 176 900 176 40 136 40 4 36 36 (51.2%) (16.4%) (82.6%) P< (22.7%) (77.3%) P= (10.0%) (90.0%) P= 37 Clinical 1024 280 744 .001 280 96 184 .01 96 21 75 .1 38 (48.8%) (27.3%) (72.7%) (34.3%) (65.7%) (21.9%) (78.1%) 39 Gender Male 926 199 730 199 71 128 71 17 54 40 41 (44.1%) (21.4%) (78.6%) P= (35.7%) (64.3%) P= (23.9%) (76.1%) P= 42 Female 1174 257 920 .83 257 65 192 .02 65 8 57 .08 on September 24, 2021 by guest. Protected copyright. 43 (55.9%) (21.8%) (78.2%) (25.3%) (74.7%) (12.3%) (87.7%) 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 45 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 17 1 2 3 4 5 6 7 Enrolment and certificate Certificate attainmentAttainment 8 9 10 11 Of those who knew about MOOCs, 136 (29.8% [99% CI,24.3%–35.3%]) were enroled enrolled in at 12 13 least one course. Most students (125; 91.9%) registered in 1–5 courses, while onlywith 113 (83.1%) 14 15 students reportedFor reporting havingpeer watching review watched at least one only video lecture. Home (109; 99%) was 16 17 18 the first primary place where they watched these videos. There was no statistically significant 19 20 difference in enrolment state between males and females (52.2% vs. 47.8%, 99% CI, P= .016). 21 22 However, there was a significant difference between students`’ class and their enrolment (P=.009) 23 24 25 (Table 1). Coursera was the most commonly used website (99; 72.8%), followed by Edx (14; 10.3%). 26 27 28 Only 25 students (18.4% [99% CI, 9.8%–26.9%]) completed courses and attained one certificate or 29 30 31 more with an 81.6% dropout rate. Interestingly, more than half of students who earned certificates (13; 32 33 52% [99% CI,26.3%–77.7%]) have used the signature track to get verified ied thecertificates m ffrom 34 http://bmjopen.bmj.com/ 35 the universities that proposed the courses. The vast majority of enrolled students assumed stated that 36 37 38 getting a certificate is was important to them (32 [23.5%] very important, 37 [27.2%] important, 50 39 40 [36.8%] important to some extent and 17 [12.5%] not important). Out of the 69 students who assumed 41 42 on September 24, 2021 by guest. Protected copyright. 43 that getting a certificate is important before enrolment (important/very important), 17 (24.6%) were 44 45 finally certified (24.6%), as; compared to only 8 (11.6%) certified students out of the 67 who were not 46 47 concerned with to receive a certificate at time of enrolmenthaving certificates (important to some 48 49 50 extent/not important; 11.9%); P< .001. 51 52 53 Ways of knowledge and studentsStudents`’ motivationsMotivations 54 55 56 To assess how students knew found out about MOOCs and what were their motivations were, two 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 46 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 18 1 2 3 multi-selection questions were addressedasked. Social media was the main primary way through which 4 5 6 206 (45.2%) students knew aboutwere introduced to MOOCs, while knowledge through a friend was 7 8 the second (184; 40.4%). Using wWeb-search engines (87; 19.1%) got took the third place, followed by 9 10 11 extracurricular activities (46; 10.1%). MOOCs providers`’ advertisements played a very small role (27; 12 13 5.9%) in reaching students as did medical schools`’ official websites (15; 3.3%). Notably, there was no 14 15 association betweenFor the ways peer method through review which students learnedonly about MOOCs and their 16 17 18 enrolment. Nevertheless, students who knew were introduced through extracurricular activities were 19 20 found to enrol more frequently (P= .005). 21 22 23 24 Concerning students’ motives, most students reported that their main motivation was “to learn new 25 26 things” followed by “to help me studying medicine” (Figure 2). Interestingly, the students who enrolled 27 28 aiming to have a certificate or to help them in obtaining a future job were significantly more likely to 29 30 31 complete the courses (P= .001) and (P= .008), respectively). 32 33 http://bmjopen.bmj.com/ 34 MOOCs and Medicine 35 36 37 By asking the enrolled students (n=136) about their experience and attitude toward medical MOOCs, 38 39 103 (75.7% [99% CI, 66.2%–85.2%]) declared participation in at least one medical course. Of them, 24 40 41 students (17.6% [99% CI, 7.9%–27.3%]) had completed medical courses and earned certificates. 42 on September 24, 2021 by guest. Protected copyright. 43 44 Regarding their medical MOOCs experience, 102 (75%) students agreed that MOOCs helped them in 45 46 developing their theoretical background about the topic discussed. However, there was less agreement 47 48 (68; 50%) on the role of MOOCs in developing their practical skills. Most students (89; 86.4%) agreed 49 50 51 that MOOCs help in studying medicine, while 83 (61%) believed that MOOCs will help them in 52 53 securinggetting a more desirable better job opportunity in the future job opportunity. 54 55 56 57 Limitations of MOOCs 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 47 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 19 1 2 3 Our study reported two types of limitations: enrolment and completion. Students who knew about 4 5 6 MOOCs, but did not enrol in any courses (n=320) were asked about their enrolment limitations. The 7 8 majority of students (226; 70.4%) agreed that a lack of time was the main limitation, while 147 (45.9%) 9 10 11 agreed that slow internet Internet speed was another cause (Figure 3). Regarding completion 12 13 limitations, the eEnrolled students (n=136) were asked to assess the limitations that made them drop out 14 15 of courses. SimilarFor to the enrolment peer limitations, review it was obvious thatonly lack of time (105; 77.2%) and slow 16 17 18 Iinternet speed (73; 53.7%) were the main obstacles. Lack of technology access, computer literacy, 19 20 language difficulty and culture conflicts were less frequently selected as a limiting factor to completion 21 22 of the courseWhile lack of technology access, computer literacy, language difficulty and culture 23 24 25 conflicts had less agreement on their roles as limitations (Figure 3). Only 16 (11.8%) students agreed 26 27 that the scientific content was difficult for them to comprehend. In addition, 93 (68.4%) students 28 29 disagreed that “lower content than expected” is was to be a limitation. 30 31 32 33 For further assessment of the iInternet speed, we asked the enrolled students to rate their Iinternet 34 http://bmjopen.bmj.com/ 35 speed. Sixty students (44.1%) reported that the speed was reasonable, while 55 (40.4%) reported slow 36 37 38 speed and only 21 (15.4%) had a higher connection speed. When we compared the students’ evaluation 39 40 of internet Internet speed and if whether they watched video lectures or not, we did not find a 41 on September 24, 2021 by guest. Protected copyright. 42 significant association (P= .69). 43 44 45 46 Students`’ Ssatisfaction of MOOCs 47 48 49 50 The 25 students who obtained certificates were asked to report their opinions about each part of the 51 52 MOOCs experience. The results showed that most students (21; 84%) were satisfied with the overall 53 54 experience, including video lectures (18; 70%), exams and assignments (16; 64%), quality of the 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 48 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 20 1 2 3 presented materials (21; 84%) and the technology used (20; 80%). However, there was less satisfaction 4 5 6 regarding student–student (5; 20%) and student–instructor (8; 32%) interactions (Figure 4). 7 8 9 10 11 12 13 14 15 For peer review only 16 17 DISCUSSION 18 19 20 21 Available information about regarding MOOCs participants is primarily data obtained from course- 22 23 24 end demographics, which usually reports demonstrate a heterogeneous populations of different varying 25 26 age groups and, educational levels from and different countries globally. These data show that most 27 28 MOOCs’ users are well-educated males with low participation from developing countries and 29 30 9-11 31 undergraduates. To our knowledge, this study is the first, in the medical field and in one of thea 32 33 developing countries country to use a cross-sectional study design in a homogeneous population for the 34 http://bmjopen.bmj.com/ 35 assessment of prevalence and uptake of such courses among undergraduate medical students. 36 37 38 39 Knowledge and Enrolment 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 Our results show demonstrate a funnel-shaped participation pattern, with 22.7 % of the respondents 44 45 knowing about MOOCs and 6.5% actually enrolled. Moreover, only 5.4% watched the offered videos 46 47 lectures and 1.2% obtained certificates of completion. Although there are no similar cross-sectional 48 49 50 studies with which our results can be compared, the knowledge of about that approximately one-fifth of 51 52 the Egyptian medical students about are familiar with MOOCs is considered promising in a developing 53 54 country, that depends mainly on regular traditional education. Additionally, these courses are still new, 55 56 57 and MOOCs providers’ advertisements had little effect in reaching students. Also, there is no medical 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 49 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 21 1 2 3 MOOC offered by an Egyptian institution until now. Additionally, these courses are still new and there 4 5 6 was little role of MOOCs providers`advertisements for reaching students beside that there is no any 7 8 medical MOOC which is given by an Egyptian institution till now. Social media and the sharing of 9 10 11 personal experiences personal experience transfer among friends played a vital role in the spreading of 12 13 the MOOCs` idea, raising students’ awareness to this its’ current level. This is in line with the uprising 14 15 increasing role ofFor social media peer websites in reviewmedical students’ life only, with more than 90% of medical 16 17 12 18 students in the U.S. using social media. 19 20 21 Notably, it was obvious that there is was a disproportion gap between knowledge about of MOOCs 22 23 24 and enrolment in them, with only one-third of students who knew about MOOCs having the awareness 25 26 registered registering in courses. The sStudents reported a lack of time and low internet Internet speed 27 28 as the main limitations for MOOC use. Out of these enrolled students, 18.4% (23.3% when looking at 29 30 31 those enrolled in a medical coursesfor medical courses) completed the courses and earned certificates. 32 33 These completion rates are higher than the reported average completion rates in the course 34 http://bmjopen.bmj.com/ 35 demographics. In 2013, The Chronicle of Higher Education suggested an average of 7.5% completion 36 37 13 14 38 rate , while a recent study in 2014 reported a rate of about 6.5%. This may be explained by the 39 40 reported importance reported by students that obtaining of certificates has in terms of adding for 41 on September 24, 2021 by guest. Protected copyright. 42 students to add to their resumes in hopes of improving future employment opportunitieshoping for 43 44 45 better future chances. It was is interesting to note that about half of them paid to verify their certificates, 46 47 although there is no academic credit for undergraduates for any MOOCs from any medical school in 48 49 15 50 the U.S. and Egypt until nowat this time. 51 52 53 Although there was no association between gender and students’ knowledge or enrolment, class had 54 55 56 a significant association. Clinical years’ students were found to have higher knowledge and enrolment 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 50 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 22 1 2 3 rates. This may be due to the high level of stress and pressures experienced by first years`’ medical 4 5 16 6 students adapting to new academic systems with little time available for extracurricular activities. In 7 8 contrast, final years’ students were reported to have less stress 16-18 with more attention to concern 9 10 11 towards their career plans by searching for new learning channels to increase their competitiveness. 12 13 14 MOOCs and Medicine 15 For peer review only 16 17 18 Of the enrolled students, 75.7% participated in at least one medical course with a 23.3% completion 19 20 rate. They strongly agreed that these courses helped them to develop theoretical backgrounds about the 21 22 topics discussed with less agreement on their role in developing their clinical skills. This raises 23 24 25 questions about the effectiveness of MOOCs with the current lecture-based teaching style in covering 26 27 the different aspects of medical education, including its clinical part, which needs student–patient 28 29 interaction. However, in the new and evolving era of online learning, a question arises: “why Why to 30 31 32 waste precious class time on a lecture?” Students may watch the instructor’s lecture remotely in their 33 http://bmjopen.bmj.com/ 34 homes and utilize use class time for learning clinical skills.19 Most of the current opinions expect 35 36 37 anticipate a complementary role of for MOOCs in undergraduate education, with an increasing role in 38 15 39 educating those students after their graduation in continuing medical education. 40 41 MOOCs limitations in Egypt 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 Lack of time and slow Iinternet speed were the two main limitations reported for causing low 46 47 MOOCs enrolment and course completion rates. MOOCs, being a self-learning educational system, 48 49 50 require a considerable amount of time to choose courses, watch videos, take exams and interact through 51 52 discussions. This imposes a significant time burden on students, leading to the need of for an increased 53 54 commitments besides beyond their busy regular medical education. Time management, either in the 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 51 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 23 1 2 3 design of courses or from participants, is critically needed to enhance their performance and increase 4 5 6 completion rates. 7 8 9 Low internet Internet speed is a commonly reported problem facing online education in developing 10 11 20 12 countries. This problem prolongs the time needed to watch high-quality videos or to download course 13 14 content, rendering students less adherent and more susceptible to dropout. The main solution to this 15 For peer review only 16 problem is enhancing the internet Internet infrastructure in Egypt. Liyanagunawardena et al. suggested 17 18 19 allowing lower resolution versions of the videos as an alternative solution to help engaging students 20 21 with limited bandwidth.8 Interestingly, we did not find computer literacy, language or culture as 22 23 24 barriers, although it was expected that they would represent problems in Egypt, being a developing 25 26 country. 27 28 29 30 31 32 33 MOOCs Experience Ssatisfaction 34 http://bmjopen.bmj.com/ 35 36 37 Encouragingly, most of the participants who completed MOOCs (n= 25) were satisfied with the 38 39 overall experience. However, there was an obvious dissatisfaction regarding student–student and 40 41 student–instructor interactions. This problem is common in online education in general, with a lack of 42 on September 24, 2021 by guest. Protected copyright. 43 44 face-to-face interaction leading to some feelings of isolation and disconnectedness, which are thought 45 46 to be two main factors in dropout rates.21 Some MOOCs providers, such as Coursera, support efforts 47 48 beside beyond the usual discussion forums for to help overcoming overcome this point. These efforts 49 50 51 include more peer assessments, social media groupsinvolvment, Google+ hangouts and real in-person 52 53 Meetupsmeet-ups. Despite that, more involvement of participants is still needed to ensure the full 54 55 56 psychological presence. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 52 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 24 1 2 3 Study strengths Strengths and limitationsLimitations 4 5 6 The strength of our study is that it included participants from all study years in 10 institutions, 7 8 covering nearly the entire geographic area of Egypt with a high confidence interval (99%) and high 9 10 11 response rate (83.3%). However, our main limitation was the relatively low returned number of 12 13 participants who enrolled (n=136) and who had certificates (n=25), which makes the analysis of 14 15 limitations and satisfactionFor speer of MOOCs less review reliable. However, theseonly results are provide an important 16 17 18 contribution as a first start step to makein gathering evidence about the real prevalence of perception 19 20 and use of MOOCs in Egypt. In addition, these results will facilitate the ability of future studies to build 21 22 upon our findings and select samples that are representative of students with prior knowledge of 23 24 25 MOOCs, leading to a better understanding of their experience. 26 27 28 29 30 31 32 to help the future studies to build upon and take samples that are representative to the students who 33 http://bmjopen.bmj.com/ 34 knew about MOOCs for a better understanding of their experience. 35 36 37 38 Conclusions: 39 40 41 Approximately About one-fifth of undergraduate medical students in Egypt have heard about 42 on September 24, 2021 by guest. Protected copyright. 43 44 MOOCs. Students who actively participated showed a positive attitude toward the experience, but 45 46 better time management skills and faster internet Internet connection speeds are required. Further 47 48 studies are needed involving to address the enrolled students in large representative samples, to assess 49 50 51 their experiences using MOOCs in large representative samples. In addition, more efforts is are needed 52 53 to be done to raise the awareness of among students of such courses, as most of students who had 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 53 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 25 1 2 3 notdid not heard about MOOCs, did showed interest in participating once they became aware of the 4 5 6 coursesto participate. 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 STATEMENTS: 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 54 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 26 1 2 3 Acknowledgements: The authors deeply acknowledge Hadeer Alsayed, Islam Shedeed (Menoufia 4 5 6 University), Zyad Abdelaziz, Dina Maklad, Ahmed Gebreil, Mahmoud Medhat (Alexandria university), 7 8 Mohammed Alhendy, Aya Sobhy, Omar Azzam (Al-Azhar University in Cairo), Hassan Aboul Nour, 9 10 11 Sara Elganzory (Tanta university), Mohamed Eid, Aya Talaat, Mohamed Emad (Beni Suef university), 12 13 Mohamed Abdelzaheer, Ahmed Abdelhamed, Ahmed Saleh (Suez Canal university), Ahmed Zain, 14 15 Khaled Ghaleb, ForYossri Mohamed peer (Benha university),review Ahmed Alaa, only Mohamed Gamal (Assuit 16 17 18 university), Marina Nashed, Ibrahim Abdelmone'm (Ain Shams university), Bassant Abdelazeim, 19 20 Ramadan Zaky (Cairo university) for their assistance in data collection. In addition, we deeply 21 22 acknowledge Bishoy Gouda (Canada), Susannah L. Bodman (U.S.), Mohamed Aleskandarany (U.K.) 23 24 25 and Moahmed Alaa (Egypt) for their help in English revision of our paper. None of them received 26 27 compensation for their assistance. 28 29 30 31 Contributors: Aboshady, Radwan and Hassouna were responsible for the conception and design of the 32 33 study. Aboshady and Radwan coordinated the study and managed the data collection. Aboshady, 34 http://bmjopen.bmj.com/ 35 Radwan, Eltaweel, Azzam, Aboelnaga, Darwish, Hashem, Salah, Kotb, Afifi, Noaman and Salem 36 37 38 collected the data. Hassouna did the analyses, Aboshady, Radwan, Hassouna, Eltaweel, Kotb and 39 40 Aboelnaga contributed to interpretation of the findings. Aboshady, Eltaweel and Azzam wrote the first 41 on September 24, 2021 by guest. Protected copyright. 42 draft of the manuscript while Radwan, Hassouna, Aboelnaga, Kotb, Hashem, Salah, Darwish, Salem, 43 44 45 Afifi and Noaman made a critical revision of the manuscript for important intellectual content. All 46 47 authors approved the final version of the manuscript. 48 49 50 51 Funding: All funding required was provided by Aboshady and Radwan on their own expenses. 52 53 54 support for this project. 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 55 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 27 1 2 3 Competing interests: None. 4 5 6 7 Ethics approval: Institutional Review Board at Menoufia University, Faculty of Medicine, Egypt. 8 9 10 11 Data sharing statement: No additional data are available. 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 56 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 28 1 2 3 4 REFERENCES 5 6 7 8 1. Gooding I, Klaas B, Yager JD, Kanchanaraksa S. Massive Open Online Courses in Public Health. 9 10 11 Front Public Health 2013;1 doi: 10.3389/fpubh.2013.00059 12 13 14 2. Hoy MB. MOOCs 101: an introduction to massive open online courses. Med Ref Serv Q 15 For peer review only 16 2014;33(1):85-91 doi: 10.1080/02763869.2014.866490. 17 18 19 20 3. Pappano L. The Year of the MOOC.The New York Times 2013. 21 22 23 24 4. A Triple Milestone: 107 Partners, 532 Courses, 5.2 Million Students and Counting! Coursera Blog: 25 26 Coursera 2013. 27 28 29 30 5. Liyanagunawardena TR, Williams SA. Massive open online courses on health and medicine: review. 31 32 J Med Internet Res 2014;16:e191. doi:10.2196/jmir.3439 33 34 http://bmjopen.bmj.com/ 35 6. Mehta NB, Hull AL, Young JB, Stoller JK. Just imagine: new paradigms for medical education. 36 37 38 Acad Med 2013;88(10):1418-23 doi:0.1097/ACM.0b013e3182a36a07. 39 40 41

7. Cooke M, Irby DM, O`Brien BC. Educating physicians: a call for reform of medical school and on September 24, 2021 by guest. Protected copyright. 42 43 44 residency: John Wiley & Sons, 2010; 25(2): 193–195 45 46 47 8. Liyanagunawardena T, Williams S, Adams A. The impact and reach of MOOCs:a developing 48 49 50 countries` perspective. eLearning Papers 2013(33) 51 52 53 9. Emanuel EJ. Online education: MOOCs taken by educated few. Nature 2013;503(7476):342-42 doi: 54 55 56 10.1038/503342a. 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 57 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 29 1 2 3 10. Group ME. MOOCs @ Edinburgh 2013: Report #1: the University of Edinburgh, 2013. 4 5 6 7 11. Huhn C. UW‐Madison Massive Open Online Courses (MOOCs): Preliminary Participant 8 9 Demographics: Academic Planning and Institutional Research, 2013. 10 11 12 13 12. Bosslet GT, Torke AM, Hickman SE, Terry CL, Helft PR. The patient-doctor relationship and 14 15 online social networks:For results peer of a national review survey. J Gen Intern only Med 2011;26(10):1168-74 doi: 16 17 18 10.1007/s11606-011-1761-2|. 19 20 21 13. Kolowich S. The professors who make the MOOCs. The Chronicle of Higher Education 2013;25 22 23 24 25 14. Jordan K. Initial trends in enrolment and completion of massive open online courses. The 26 27 International Review of Research in Open and Distance Learning 2014;15(1) 28 29 30 31 15. Harder B. Are MOOCs the future of medical education? Bmj 2013;346:f2666 doi: 32 33 10.1136/bmj.f2666|. 34 http://bmjopen.bmj.com/ 35 36 37 16. Dahlin M, Joneborg N, Runeson B. Stress and depression among medical students: a cross-sectional 38 39 study. Med Educ 2005;39(6):594-604 doi: 10.1111/j.1365-2929.2005.02176.x. 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 17. Guthrie E, Black D, Bagalkote H, Shaw C, Campbell M, Creed F. Psychological stress and burnout 44 45 in medical students: a five-year prospective longitudinal study. J R Soc Med 1998;91(5):237-43 46 47 48 49 18. Bassols AM, Okabayashi LS, Silva AB, et al. First- and last-year medical students: is there a 50 51 difference in the prevalence and intensity of anxiety and depressive symptoms? Rev Bras Psiquiatr 52 53 (Sao Paulo, Brazil : 1999) 2014;0:0 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 58 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from 30 1 2 3 19. Frehywot S, Vovides Y, Talib Z, et al. E-learning in medical education in resource constrained low- 4 5 6 and middle-income countries. Hum Resour Health 2013;11(1):4 doi: 10.1186/1478-4491-11-4. 7 8 9 20. Angelino LM, Williams FK, Natvig D. Strategies to Engage Online Students and Reduce Attrition 10 11 12 Rates. Journal of Educators Online 2007;4(2):n2 13 14 15 21. Prober CG, HeathFor C. Lecture peer halls without review lectures--a proposal only for medical education. N Engl J 16 17 18 Med 2012;366(18):1657-9 doi: 10.1056/NEJMp1202451. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 59 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 267x159mm (300 x 300 DPI) 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 61 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 292x165mm (300 x 300 DPI) 28 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 62 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27

28 241x146mm (300 x 300 DPI) 29 30 31 32 33 34 http://bmjopen.bmj.com/ 35 36 37 38 39 40 41 42 on September 24, 2021 by guest. Protected copyright. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 63 of 63 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from

1 2 STROBE Statement—Checklist of items that should be included in reports of cross-sectional studies 3 Item 4 No Recommendation 5 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 6 7 Perception and use of massive open online courses among medical students of a 8 developing country: multicenter cross-sectional study 9 (b) Provide in the abstract an informative and balanced summary of what was done 10 and what was found (Done) (page 4-5) 11 12 Introduction 13 Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 14 (Done) (page 7-8) 15 Objectives For3 peer State specific objectives, review including any prespecified only hypotheses (Done) (page 8, 16 17 last paragraph) 18 Methods 19 Study design 4 Present key elements of study design early in the paper (Done) (page 9, first 20 21 paragraph) 22 Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, 23 exposure, follow-up, and data collection (Done) (page 9-10) 24 Participants 6 (a) Give the eligibility criteria, and the sources and methods of selection of 25 26 participants (Done) (page 9, last paragraph) 27 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect 28 modifiers. Give diagnostic criteria, if applicable (Not applicable) 29 Data sources/ 8* For each variable of interest, give sources of data and details of methods of 30 measurement assessment (measurement). Describe comparability of assessment methods if there is 31 32 more than one group (Done) (page 11) 33 Bias 9 Describe any efforts to address potential sources of bias (Not done) http://bmjopen.bmj.com/ 34 Study size 10 Explain how the study size was arrived at (Done) (page 9) 35 Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, 36 37 describe which groupings were chosen and why (Done) (page 11) 38 Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding 39 (Done) (page 11) 40 (b) Describe any methods used to examine subgroups and interactions (Not 41 42 applicable) on September 24, 2021 by guest. Protected copyright. 43 (c) Explain how missing data were addressed (Not applicable) (no missing data) 44 (d) If applicable, describe analytical methods taking account of sampling strategy 45 (Done) (page 9,11) 46 e 47 ( ) Describe any sensitivity analyses (Not done) 48 Results 49 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially 50 eligible, examined for eligibility, confirmed eligible, included in the study, 51 52 completing follow-up, and analysed (Done) (page 13) 53 (b) Give reasons for non-participation at each stage (Not done) 54 (c) Consider use of a flow diagram (Done) (Figure 1) 55 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and 56 57 information on exposures and potential confounders (Done) (Table 1) 58 (b) Indicate number of participants with missing data for each variable of interest 59 (Not Done) (No missing data) 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml1 BMJ Open Page 64 of 63 BMJ Open: first published as 10.1136/bmjopen-2014-006804 on 5 January 2015. Downloaded from