This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
Many countries worldwide have developed mobile phone apps capable of supporting instantaneous contact tracing to control the COVID-19 pandemic. In France, a few people have downloaded and are using the StopCovid contact tracing app. Students in the health domain are of particular concern in terms of app uptake. Exploring their use and opinions about the app can inform improvements and diffusion of StopCovid among young people.
The aim of this study is to investigate health care students’ knowledge of and attitudes, beliefs, and practices (KABP) toward the StopCovid app.
A field survey was conducted among 318 students at the health sciences campus of the University of Bordeaux, France, between September 25 and October 16, 2020. A quota sampling method was used, and descriptive statistics and univariate analyses were performed.
Of the 318 respondents, 77.3% (n=246) had heard about the app, but only 11.3% (n=36) had downloaded it, and 4.7% (n=15) were still using it at the time of the survey. Among the 210 participants who had heard about the app but did not download it, the main reasons for not using the app were a belief that it was not effective given its limited diffusion (n=37, 17.6%), a lack of interest (n=37, 17.6%), and distrust in the data security and fear of being geolocated (n=33, 15.7%). Among the 72 students who had not heard of the app and were given a brief description of its functioning and confidentiality policy, 52.7% (n=38) said they would use it. Participants reported that the main solution for increasing the use of the app would be better communication about it (227/318, 71.4%).
Even among health students, the contact tracing app was poorly used. The findings suggest that improved communication about its advantages and simplicity of use as well as clarifying false beliefs about it could help improve uptake.
Nonpharmaceutical interventions have been used to contain the spread of the SARS-CoV-2 virus [
Several smartphone apps have been developed worldwide to automatically and rapidly trace contacts in real time. Across all continents more than 45 apps are currently used [
The recrudescence of the virus after the general lockdown from March to May 2020 has especially concerned young people across France and in the Bordeaux region in particular, where incidence of COVID-19 positive cases among young adults aged 20-30 years has increased to about 252/100,000 per week (weeks 41-42) [
The contact tracing app “StopCovid” was launched by the French Government on June 2, 2020 [
As of October 2020, StopCovid had been installed more than 2.7 million times since the beginning of June (about 4% of the French population, 67 million). Only 7969 users had declared being COVID-19 positive in the app, and only 472 notifications had been sent to potential at-risk contacts. The uptake was less than for apps in Germany (downloaded by 18 million people, about 21% of the German population, 84 million), England and Wales (16 million downloads in a population of 59 million, 27%), and Italy (9 million downloads in a population of 60 million, 26%). All these percentages are low considering that approximately 60% of the adult population would have to adopt the app to contain the pandemic [
The effectiveness of the StopCovid app must be framed in the specific French context: the testing strategy was and is still unclear with limited testing capacities, tracing was not always possible, and self-quarantining was voluntary and not always followed [
Recently, several researchers have investigated the acceptability and use of contact tracing apps in the context of the COVID-19 pandemic. Some studies are based on surveys assessing the uptake of these apps among different population samples. These studies mostly refer to a hypothetical app and the intention to use it [
Concerning surveys, a multicountry cross-sectional study on 1849 adults across France, Germany, Italy, the United Kingdom, and the United States [
Acceptance of COVID-19 contact tracing apps has also been explored in France. In the first survey [
Concerning critical viewpoints and opinion papers, they describe the public debate on privacy concerns due to the sensitive nature of the collected data. In particular, several researchers have argued that the adoption of contact tracing apps could lead to the economic exploitation of private data and might create a mass electronic surveillance system [
This study was conducted at the beginning of the academic year at the University of Bordeaux, France, when face-to-face education was re-established, and students could freely circulate after the first general lockdown. The StopCovid app was available and downloadable for 4 months. The aim of this study is to describe knowledge, attitudes, beliefs, and practices (KABP) about the StopCovid app among students in the health domain in the Bordeaux region. The expected impact is informing on potential improvements as well as public-oriented communication strategies and appropriate political decisions to increase the app diffusion.
This study was conducted within the framework of the larger ongoing i-Share (Internet-Based Students Health Research Enterprise [
The field survey consisted of a paper questionnaire administered face-to-face by five undergraduate students (interviewers) who had been trained to take notes, fill in the questionnaire, and describe the app to respondents. Interviewers approached their peers in the halls, canteen, courtyards, library, and study rooms at the health sciences campus of the University of Bordeaux. The collection of the data started on September 25, 2020, and ended on October 16, 2020. A sample size of 300 respondents was targeted with quotas set for the sample to be representative of the overall population of students in the health domain at the University of Bordeaux (n=16,566) in terms of sex, age (18-30 years), specific field of health-related study (medicine, dentistry, nursing, pharmacy, public health, etc), and year of study (1 to >6 years). The inclusion criteria were being older than 18 years, being a student in the health domain enrolled at the University of Bordeaux, and providing oral informed consent.
The questionnaire was co-designed and tested by a team of 14 public health researchers and operational staff following a structured survey construction method in five steps [
The questionnaire was based on the KABP scheme, which stands for the assessment of knowledge, attitudes, beliefs, and practices of populations about a specific health-related topic. This scheme is extensively used as a quantitative method (predefined questions formatted in a standardized questionnaire) that provides access to quantitative and qualitative information. Thus, items of the questionnaire were developed to capture students’ KABP about the StopCovid app. The results are discussed following the four components of this scheme.
Collected data were interpreted a posteriori through the prism of the technology acceptance model (TAM) [
All data from the paper questionnaires were entered by the student interviewers in a digital database through the EpiData software version 3.1 [
A total of 590 students were approached to complete the survey after a brief explanation of its objective; 318 completed the survey, while 272 refused to participate, creating a final participation rate of 53.9%. Reasons for not participating were lack of time or no interest in the study topic. Of the 318 participants, 65.7% (n=209) were female students, and the mean age was 20.4 (SD 2.39) years. All fields and years of study were represented. The majority (n=193, 60.7%) of the participants were medical students, which is in line with the total number of medical students at the University of Bordeaux. In accordance with university-related statistics, first-year students were also the most represented (n=129, 40.6%).
Flowchart of the study population (n=318).
Sociodemographic characteristics of the study population and comparison with all students in the health domain.
Characteristics | Study population (n=318) | Total health care student population at the University of Bordeaux (n=16,566)a | |
|
|||
|
Female | 209 (65.7) | 11,713 (70.7) |
|
Male | 109 (34.3) | 4853 (29.3) |
Age (years), mean (SD) | 20.4 (2.39) | 23.8 (—b) | |
|
|||
|
1 | 129 (40.6) | 4445 (31.2) |
|
2 | 64 (20.1) | 2341 (16.4) |
|
3 | 46 (14.5) | 2334 (16.4) |
|
4 | 23 (7.2) | 1200 (8.4) |
|
5 | 44 (13.8) | 1267 (8.9) |
|
>5 | 10 (3.1) | 2651 (18.6) |
|
Other | 2 (0.6) | — |
|
|||
|
Medicine | 193 (60.7) | 7104 (49.9) |
|
Pharmacy | 61 (19.2) | 1138 (8.0) |
|
Dentistry | 12 (3.8) | 521 (3.7) |
|
Nursing | 14 (4.4) | 3972 (27.9) |
|
Public health | 15 (4.7) | — |
|
Other | 23 (7.2) | 1503 (10.6) |
aData obtained from internal university documents.
bData not available.
The majority (n=246, 77.3%) of the participants had already heard about the app, mostly through the media (216/246, 87.8%) and secondly through family and friends (39/246, 15.9%). Concerning these variables, no statistically significant differences were found based on age (
Most of the 246 students that knew of the app correctly knew that the app was promoted by the government (n=179, 72.8%), but 25.2% (n=62) answered that they did not know who the promoter was. Male students knew significantly more than female students that the app was promoted by the government (69/85, 81.2% vs 110/161, 68.3%;
Among the 246 participants who had heard about the app, 14.6% (n=36) had actually downloaded it when it was first released in June 2020 (22/36, 61.1%) or with the new cases of COVID-19 at the beginning of the university year (6/36, 16.7%). Of these 36 students, 41.6% (n=15) were still using the app. Of the total 318 participants, 4.7% (n=15) of students were using the app at the moment of the survey, while those who uninstalled the app had used it from 1 day (6/36, 16.7%) to several weeks (6/36, 16.7%). The main reasons for uninstalling the app were that it was not useful (14/21, 66.7%), the respondent forgot to activate the Bluetooth (5/21, 23.8%), the app drained the phone battery (4/21, 19.0%), and too few people were using it thus making the app ineffective (4/21, 19.0%). Accordingly, students reported that the main fault of the app was that it seemed inefficient given its limited uptake (17/35, 48.6%; 1 missing). For 25.7% (9/35), the app presented technical problems like draining the battery, depending on Bluetooth, or occupying too much storage on the phone. Concerning all previous variables, no statistically significant differences were found based on age (
Some of these students also reported that its qualities were that it was easy to use (18/35, 51.4%) and that it was reassuring (9/35, 25.7%). Male students found the app significantly more user-friendly than female students (12/17, 70.6% vs 6/18, 33.3%;
Reasons for downloading or not downloading the app are shown in
Among the 210 participants who had heard about the app but did not download it, the main reasons for not using the app were lack of interest (n=90, 42.9%), belief that it was neither effective nor useful given its limited diffusion (n=37, 17.6%), not having time to think about it (n=37, 17.6%), and distrust in data security and fear of being geolocated (n=33, 15.7%). The majority of these students might change their mind and use the app if they had more information about it through better communication strategies (n=61, 29.0%) and if more people would use it (n=54, 25.7%). Nonetheless, 26.2% (n=55) would not change their mind and would still not download the app. On the other hand, the main reasons for downloading the app were out of curiosity (13/36, 36.1%) and to protect one’s family, others, and oneself from possible infection (13/36, 36.1%).
Reasons for downloading the StopCovid app.
Reasonsa | Yes, I have downloaded the app (n=36), n (%) | Yes, I would download the app (n=38), n (%) | |
|
|||
|
Out of curiosity | 13 (36.1) | 14 (36.8) |
|
To protect my family, others, and myself from possible infection | 13 (36.1) | 24 (63.2) |
|
Because the government advised downloading of the app | 5 (13.9) | 0 (0.0) |
|
The app could be useful to contain the spread of the virus in general | 5 (13.9) | 20 (52.6) |
|
I am afraid of the virus, and all strategies are good to avoid it | 1 (2.8) | 1 (2.6) |
|
I was reassured the app was anonymous | N/Ab | 0 (0.0) |
|
Other | 11 (30.6) | 1 (2.6) |
aMultiple answers possible.
bN/A: not applicable.
Reasons for not downloading the StopCovid app.
Reasonsa | No, I have not downloaded the app (n=210), n (%) | No, I would not download the app (n=34), n (%) | |
|
|||
|
Cannot see the interest or need | 90 (42.9) | 17 (50.0) |
|
Do not like the general idea of this app | 10 (4.8) | 1 (2.9) |
|
Do not know how it works, did not get enough information on the app | 23 (11.0) | 2 (5.9) |
|
I am suspicious of this type of app | 18 (8.6) | 1 (2.9) |
|
Do not trust, because I do not know who is offering this app | 11 (5.2) | 1 (2.9) |
|
Not sure about the security of the data, fear of geolocation | 33 (15.7) | 1 (2.9) |
|
My family and friends have discouraged me from downloading it | 1 (0.5) | 0 (0.0) |
|
No storage on my phone or it is not powerful enough to have an extra app (battery, Bluetooth) | 27 (12.9) | 6 (17.6) |
|
Do not carry my phone with me at all times | 2 (1.0) | 4 (11.8) |
|
Do not use public transportation and/or do not go out much in public places (do not come into contact with strangers) | 11 (5.2) | 4 (11.8) |
|
It does not seem to be effective (too few people use it) | 37 (17.6) | 8 (23.5) |
|
Heard negative feedback on this app | 7 (3.3) | 1 (20.9) |
|
Do not really have time to think about it | 37 (17.6) | 3 (8.8) |
|
By negligence, not concerned | 28 (13.3) | 5 (14.7) |
|
Not sure what it is all about, the principle and/or the functioning | 11 (5.2) | N/Ab |
|
Other | 15 (7.1) | 2 (5.9) |
aMultiple answers possible.
bN/A: not applicable.
The 72 students who had never heard about the app were asked to imagine its content and objective: 41.7% (n=30) reported that it was an app providing advice and information about COVID-19, 29.2% (n=21) reported that it was an app to limit the spread of the virus, 29.2% (n=21) did not know, and 15.3% (n=11) answered “other.” After a short description of the app, 52.7% (n=38) said they would download it. The reasons for downloading or not downloading the app are similar to those provided by the sample who had heard about the app. Among the 34 students who had never heard about the app and were still not willing to download it after a brief description, 32.4% (n=11) would not change their mind, 17.6% (n=6) would download it if more people used it, and 11.8% (n=4) would download it if they had a better mobile phone.
Concerning the functioning of the app, 83.3% (30/36) of the respondents said that they were able to explain it. However, when further asked about geolocation, access to contact information, and how data were transmitted and stocked, their answers were mostly incorrect. As expected, students who had not heard about the app before, but who were presented a quick description of it during the survey, provided correct answers more than their peers. Detailed results are shown in
Knowledge and beliefs about the functioning and data management of the StopCovid app.
Knowledge and beliefs | Yes, I have heard about the app (n=246), n (%) | No, I have not heard about the app (n=72), n (%) | Total (N=318), n (%) | |
|
||||
|
No (correct answer) | 78 (31.7) | 44 (61.1) | 122 (38.4) |
|
Yes | 138 (56.1) | 19 (26.4) | 157 (49.4) |
|
Not sure | 30 (12.2) | 9 (12.5) | 39 (12.3) |
|
||||
|
No (correct answer) | 178 (72.4) | 56 (77.8) | 234 (73.6) |
|
Yes | 37 (15.0) | 5 (6.9) | 42 (13.2) |
|
Not sure | 31 (12.6) | 11 (15.3) | 42 (13.2) |
|
||||
|
No (correct answer) | 124 (50.4) | 45 (62.5) | 169 (53.1) |
|
Yes | 82 (33.3) | 15 (20.8) | 97 (30.5) |
|
Not sure | 40 (16.3) | 12 (16.7) | 52 (16.4) |
|
||||
|
No (correct answer) | 191 (77.6) | 55 (76.4) | 246 (77.4) |
|
Yes | 28 (11.4) | 5 (6.9) | 33 (10.4) |
|
Not sure | 27 (11.0) | 12 (16.7) | 39 (12.3) |
Finally, all 318 participants were asked about factors for increasing the use of the app. For the majority (n=227, 71.4%), the solution was a better communication strategy. Other factors were making the app compulsory (n=45, 14.2%), registering more COVID-19 cases (n=30, 9.4%), more information and explanations about the app (n=21, 6.6%), better technical features (n=10, 3.1%), and “other” (n=64, 20.1%).
As far as knowledge is concerned, 1 out of 5 students had never heard about the StopCovid app; this rate is surprisingly high considering that students in the health domain should be informed of existing tools to limit the spread of COVID-19. Those who knew the app had heard about it mostly through the media (216/246, 87.8%). The majority of students (179/246, 72.8%) correctly knew that the app was promoted by the government. However, concerning the functioning of the app, some students did not know how the contact tracing system worked and how data was managed: percentages of errors in describing the app ranged from 10.4% (33/318) to 49.4% (157/318). For them, the app was not straightforward; in the light of the TAM, reduced ease of app use determines lower acceptance. Furthermore, as an external variable, system complexity might have mediated the perception of ease of use for the app. In general, limited information about a tool, from knowing that it exists to knowing how it works, is associated with poor use of the tool itself. Consistently, when asked how they would improve StopCovid adoption, 71.4% (227/318) of students suggested deploying better communication and information strategies for increasing knowledge about the app.
In terms of attitudes, students reported several reasons for not downloading or uninstalling the app. Their intention not to use the app was mostly due to the fact that they considered the app as neither useful nor effective (14/19, 73.7%), especially because few people were using it. As suggested by the TAM, perceived usefulness is a key determinant of acceptance of a new technology, which justifies the low adoption of StopCovid by students of our sample. Technical issues like draining the battery, use of Bluetooth, and mobile phone storage were also mentioned (9/19, 31.0%). Once again, according to the TAM, technological components are strictly related to acceptance of a digital tool. On the other hand, reasons to download the app included wanting to protect one’s family and friends: percentages ranging from 36.1% (13/36) to 63.2% (24/38). According to the PMT, the effort or cost (ie, response efficacy) of using the app was worth it to protect others from the virus. A few students (5/36, 13.9%) reported that the promotion of the app by the government motivated them to use it. This result might reflect a certain degree of confidence in political authorities.
As for beliefs, when specifically asked about the functioning and data management of the app, half of the total sample believed that the app was intrusive: it could geolocate them, track their movements, and access phone contacts. For 1 out of 6 students, fear of being tracked and that data could be collected and shared discouraged them from downloading the app. This false belief might have been nourished by the fact that students could have heard in the media that data breaches were possible, directly on their phones through Bluetooth, and that central servers could be violated. According to the PMT model, if beliefs do not support the recommended preventive behavior, probability of adopting such behavior is reduced. Furthermore, if a data breach is felt like a threat, individuals would be motivated not to download the app to protect themselves. Among those who received a clear explanation of the functioning of the app and its confidentiality policy (no geolocation and no access to personal data), 1 out of 2 students felt reassured and would finally download the app. These results confirm that beliefs, either true or false, influence behavioral intention.
Finally, concerning practices, 14.6% (36/246) of participants had actually downloaded the app, and in the whole sample, only about 4.7% (15/318) were still using it at the time of the survey, which is in line with national statistics concerning the general French population (4%). Furthermore, 26.2% (55/210) of respondents would not change their mind and still would not use the app. Possible justifications could include the fact that young people might perceive the pandemic as not dangerous for them. Epidemiological data confirm that COVID-19 is fatal mostly for people older than 60 years or who have a chronic disease [
Overall, international and French surveys (eg, [
A study exploring the real uptake of an existing app in Singapore, the TraceTogether app, had an uptake of 20% [
In our study, the three main reasons for not downloading the app were lack of interest (90/210, 42.9%), belief that it was neither effective nor useful given its limited diffusion (37/210, 17.6%), and not having time to think about it (37/210, 17.6%). No previous survey has reported these same reasons, even if in the literature the notion of contact tracing apps’ effectiveness has been widely discussed [
Distrust in data security and fear of being geolocated were mentioned by our sample as the fourth reason for not downloading the app (33/210, 15.7%). Researchers worldwide, from Europe to Asia, have emphasized the privacy controversies of contact tracing apps, presenting them as the main fault of this type of technology [
Similarly, trust in the authorities was mentioned in previous research as a factor influencing the uptake of the app: individuals who have less trust in their national government were also less supportive [
The main reasons for downloading the app were curiosity (13/36, 36.1% of those who downloaded the app and 14/38, 36.8% of those who would download the app) and wanting to protect family, others, and oneself from possible infection (13/36, 36.1% and 24/38, 63.2%, respectively). The second reason was also reported in the Luxembourg survey [
Finally, we must consider the specificity of our population compared to previous research. Although studies were conducted on the general population (mostly nationally representative samples), we presented data exclusively from university students in the health domain. Our sample might have felt less concerned by the pandemic or simply less interested in this type of digital solution, or studying in the health domain and potentially working in hospitals might have made a contact tracing app for our respondents superfluous since they could be in contact with patients who were infected. A qualitative study would be useful to analyze the motivation for not using the app in this specific population.
This was one of the first studies reporting data on students’ KABP about a contact tracing app in a pandemic context. Previous studies have explored the intention of downloading this type of app as a general idea but were not based on a developed and currently diffused app [
Limitations of this study include the relatively small sample. More than 300 students in the health domain were interviewed among a total population of 18,000 students. Findings cannot be generalized, but the sample was recruited according to quota sampling to be, as much as possible, representative of sex, age, specific field of study (from medicine to pharmacy), and year of study. This might increase the representativeness of the interviewed population group. However, it is possible that students interested in the topic were more willing to participate so that the final sample might be biased (self-selection bias). The small sample also justifies the few significant differences that were identified. This is confirmed by the low statistical powers that were obtained following performed statistical tests (<0.50).
StopCovid received several criticisms and even the French Prime Minister Jean Castex officially declared not having downloaded the app. The government considered the low uptake of the app as the main issue of StopCovid. Some weeks after the implementation of this study, on October 22, 2020, the French President Emmanuel Macron announced the launch of a new contact tracing app, TousAntiCovid. There are two main differences between the two versions of the app: embedded functionalities and promoting strategy. Concerning functionalities, they include provision of information on new cases (effective R, incidence rate, hospitalizations, etc), advice and news about COVID-19, geolocation of testing centers, and generation of the mandatory certificates for permission to be outdoors during the lockdown. This last functionality, in particular, might have increased the download and use of the app. As for promoting strategy, it was more intense but less coercive and more transparent about the app compared to StopCovid. The French President and Prime Minister were strongly engaged in the communication campaign from the beginning, whereas StopCovid had mainly been promoted by the Minister of Health and by the Digital Secretary of State [
This survey was conducted as the preliminary phase of a complex intervention aimed at promoting the uptake of the StopCovid app among students in the health domain at the University of Bordeaux. After this first appraisal of KABP about StopCovid, the next steps are to implement a series of actions at the university. Professors and lecturers have been mobilized and trained to present the contact tracing app to their students during classes. Furthermore, students will also be informed by more communication such as short videos on the university website and intranet, flyers, posts on social networks, and posters. Student ambassadors and associations will also be involved in the diffusion of the app. This complex intervention will be evaluated through a second series of random field surveys aimed at observing an increase in the number of app downloads. Depending on the results of the evaluation, the intervention will be extended to students in other fields of study at the University of Bordeaux and other universities across France.
Overall, we found broad support for app-based contact tracing, notwithstanding the low uptake of StopCovid among French students in the health domain. The results suggest that the functioning and purpose of the app were not well known and appraised among participants, especially because of the lack of factual communication. Efforts are to be taken in these terms to increase knowledge about the new TousAntiCovid app, diffuse its adoption, and consequently improve preventive behaviors among young people who represent an important target audience in the strategies to limit the transmission of COVID-19. The way the app traces contacts should be better explained so as to maximize its download and consequential use by eliminating any potential false belief. The French government should be particularly involved in providing quality, clear, appropriate, and straightforward information about TousAntiCovid.
English version of the StopCovid field survey questionnaire.
Internet-Based Students Health Research Enterprise
knowledge, attitudes, beliefs, and practices
protection motivation theory
technology acceptance model
The project is led by the University of Bordeaux in close collaboration with the University Hospital of Bordeaux, the Regional Health Authority, and the National Institute for Research in Digital Science and Technology (Inria).
The authors wish to thank the i-Share team [
All data generated or analyzed during this study are included in this manuscript. The full data set is available on reasonable request.
The i-Share team is currently supported by an unrestricted grant of the Nouvelle-Aquitaine Regional Council (Conseil Régional Nouvelle-Aquitaine; grant no. 4370420) and by the Bordeaux
IM was responsible for writing the manuscript, conducting the literature search, data collection, and interpretation. CT, RT, and NR were responsible for the study design and revision of the manuscript.
None declared.