Demographics
Four hundred seventy-six (476) university students studying in 7 academic disciplines of health-related sciences at the same university participated in the study. The participants were comprised of 77.10% females and 22.90% males. 43.07% of the participants were from the Nursing department, and the rate of the other six departments changed between 5.75% and 12.94%. The participants were from 6 different classes. Class 5 and 6 students were studying in the School of Medicine, which fills about 10% of the subjects. The highest number of students was in class 1 (31.72%). The other students were evenly distributed between classes 2-4.
Most participants (71.22%) believed digital literacy education was essential for a health-based student. In comparison, 25.63% stated that digital literacy training was necessary but optional for a health-based student, and only 15 students (3.15%) thought digital literacy education was unnecessary.
Almost 63% of participants had never received any type of coding programming/coding education. 15.76% learned coding at university, 17.22% were already exposed to any coding training before university, and 3.78 % have learned to code independently (Table 2).
The exposure to coding education differed significantly between genders. The percentage of females (67.85%) who did not take any coding education at all was significantly higher than males (47.71%) (p: 0.0002). The percentage of males starting coding at high school and the university was higher than females (p: 0.0411 and 0.0127, respectively). There was no significant difference between self-learned proportions between genders. Also, no notable difference was found between genders in their views of digital literacy education in health-related departments (p values > 0.88) (Table 3).
Questionnaire
Scores and comparisons of domains in digital literacy
The mean scores in each field of digital literacy lay between 3 and 4 on a 5-point Likert scale. The “network” and “A.I.” scores were the highest (mean 3.87±0.61 and 3.82±0.66, respectively), and the “interest-knowledge” scores were the lowest at a mean of 3.49±0.66.
When scores were compared between genders, there were no considerable differences in the “network,” “ethics,” and “A.I.” areas. In contrast, male scores were significantly higher in the “software and multimedia,” “hardware,” “security,” and “interest-knowledge” domains and total scores. The most apparent difference between median scores of genders was observed in the “security” area, where the difference was 0.50 points out of a 1-5 scale (p: 0.0288). (Table 4).
Comparisons of the scores between classes-
Spearman correlation test showed that there were weak positive correlations between the classes and the “software and multimedia”(p: 0.0008, rho: 0.1539), “ethics” (p: 0.0001, rho: 0.1830), “interest and knowledge” (p: 0.0041, rho: 0.1315) domains and the average score (p: 0.0230, rho: 0.1046). No significant correlation was detected between the classes and the “hardware,” “network,” “security,” and “A.I.” fields (p values > 0.1865) (Supplementary Table 1). Figure 1 shows the mean scores and standard deviations of each class in all fields (Fig.1.)
The “software and multimedia” scores were lowest in class 1 (3.39±0.58), compared to class 2 (3.57±0.61), class 3 (3.54±0.57), class 4 (3.63±0.69) and class 5 (3.67±0.72) scores (p < 0.020). No significant difference was found between classes regarding “hardware,” “security,” and “A.I.” areas. Class 6 (3.6±0.59) scored lower than class 4 (3.92±0.64) and class 5 (4.03±0.53) (p: 0.0440 and 0.0233 respectively) for the domain “network.”
In “ethics,” the lowest score was obtained in class 1 (3.56±0.65), relatively lower than all classes with p-values lower than 0.0202 excluding class 5 (3.75±0.70). For the same domain, class 6 (4.08±0.40) was also found to be significantly higher than class 3 (3.75±0.62) and 5 (3.75±0.70) (p: 0.0098 and 0.0420, respectively).
The field “interest and knowledge” reached the highest score in class 5 students compared to other classes (3.90±0.68, p-values < 0.0021).
The only significant difference in overall scores was observed between classes 1 and 4 (p: 0.0102). Class 4 score (3.74±0.54) was about 4% higher than class 1 (Data not shown).
Comparison between departments-
In order to find out how much the scores of each department deviated from the general trend, the scores of each section were compared with the total scores of the participants from the other departments.
The scores of the Molecular Biology and Genetics, Physiotherapy, Rehabilitation, and Psychology departments did not differ from the general outlook of all scores. Biomedical Engineering displayed the highest scores in all fields except “ethics,” with at least a 6% difference. School of Medicine scores was ranked second and showed higher scores in “hardware” and “ethics” by more than 4% and in “interest-knowledge” by 8.7%.
The Department of Nutrition and Dietetics scores were remarkably lower than the others, showing a score difference between 6.65% and 9.34% in all domains except “network,” “security,” and “A.I.” Even in these fields, the scores were lower but not statistically significant. The second lowest scores were obtained from the Department of Nursing in “hardware,” “interest-knowledge,” and “network,” with -3.01%, -3.94%, -4.56% difference, respectively. No statistically significant difference was found in other areas (Fig. 2.).
The same questionnaire was also directed to Computer Engineering students at the same university, who were trained as the best equipped in digital literacy. Thirty-nine computer engineering students agreed to participate in this study, and their scores were compared with the departments that received health-based training. The Computer Engineering students had very high scores in the “software and multimedia,” “hardware,” and “security” fields (p-values < 0.006). On the other hand, computer engineers got surprisingly ~11% lower mean scores than the rest of the participants in ethics (Supplementary Table 2).
Interview
The first interview lasted 1:22:27 hours, and the second interview 1:31:32 hours. All participants completed the first interview, and two dropouts were in the second. Table 5 shows the themes, subthemes, and example quotes of the participants of both interviews.
Participants thought they had the basic knowledge and competence to fulfill their professional responsibilities in digital literacy. However, a need for development was defined due to different applications and evolutions in this field. One participant stated, “Having digital resources and being able to use them is not sufficient to be digitally literate. We must be able to use, understand and produce information (P.7).”
The students welcomed new technological developments such as robotic surgery, decision support software, and prostheses. One participant said, “If we talk about robotic surgery in the operating room, serious errors can even be reduced to zero. (P.15)” This idea was supported by many participants that problems caused by human errors can be reduced with advanced technology. One participant expresses the possible contribution of decision support systems as follows. “Decision support software takes data many times more than the number of patients physicians can see in their lifetime and draws conclusions that need to be learned. For this reason, this system can see things that a human cannot see (P.3)” Many participants supported the contribution of robotic surgery in solving problems caused by human physical inadequacy and the possibility of decision support systems to reduce errors.
It was observed that the students were concerned that the development of new technologies would negatively affect human relations and interaction. A comment from a student summarized the negative reflection of technology on humanity, “I think interactions involving human relations rather than technology are important in treating, diagnosing, or delivering patient care for most diseases. So maybe we can count this as the difficulty of new technologies (P.15)” A majority of participants expressed their opinion that developments in the field of telemedicine will negatively affect human contact.
Participants perceived the use of new technologies as an area of growth. Some students seemed willing to generate new technologies and said they could quickly adapt to developments. However, costs and limitations in access were seen as obstacles. One participant stated, “New technologies, especially the newest technologies, are very costly in the first place. For this reason, I can't reach them easily in my private life or school education (P.12).”
The contribution of artificial intelligence in the health field was generally considered positive. The most critical concern in this regard was the changes it would bring in professional practices and the decreased need for labor in some areas. It was stated that people should be trained for newly developing professions. One student emphasized this situation: “Professions may evolve, and new careers will emerge, and as a result, we will need people who have been trained to be able to do these new works (P.12).”
Cybersecurity has been discussed with the most significant concern and participation by the students. The participants voiced their concerns in striking sentences “Nothing in the internet world is that extra safe. (P.10)” and “I feel helpless about the law regulated by the Personal Data Protection Authority (P.14).”