**4. Discussion**

The usefulness of DCT apps has been a subject of intense discussion during the COVID-19 pandemic [18,25]. Most governments have struggled with low participation rates, which, in turn, have limited the effectiveness of these tools, contributing to the idea that they are useless and, thus, hindering their adoption [21]. Recently, several researchers have investigated the acceptability and use of contact tracing apps. Most studies are based on surveys assessing the uptake of DCT apps among different population subgroups with a focus on hypothetical tools or the intention to use it, but only a few collect information on the use of an existing app [21]. The majority of documents report the real uptake using data of national statistics without a scientific and theoretical background, while other studies are critical viewpoints arguing on the ethical, technical, political, and scientific impact of contact tracing apps on society [21]. In our study, we found a higher uptake rate of the IMMUNI app compared to that in the general Italian population [12], probably because our sample consisted of students attending healthcare courses, who are more likely to be committed to health prevention strategies [26]. In addition, the fact that, in our analysis, almost all students had the opportunity to download the app since they owned a smartphone, in contrast to the official data where it is more difficult to estimate the number of people eligible for the app uptake, may have contributed to such discrepancy [18]. Nevertheless, we found that use of the DCT app was relatively limited, albeit at a similar rate to uptake of comparable apps in other European nations [21]. This is a concern, however, as these students are the healthcare workforce of tomorrow, and, therefore, it is imperative to implement educational programs that further encourage the adoption of preventive strategies [27]. Moreover, it should be mentioned that the current increase in virus transmission rates due to the omicron variant, and the concomitant abolition of restrictive measures at both the national and regional level, could make it difficult to promptly identify the transmission chain using traditional methods [28]. In this scenario, a high uptake rate of the app would have some advantages, including the support to trace

contacts, but also would make the population autonomous in the timely application of the preventive measures and create an environment in which citizens are effectively engaged in maintaining their personal and community health [29].

As for determinants, IMMUNI uptake was not associated with any socio-demographic characteristic, including HL, probably because of the healthcare curricula of our students, but some attitudes towards the pandemic seemed slightly different between the two groups. Risk perception was confirmed to be a key driver, but it changed over time; thus, people may have become used to a high level of risk as the pandemic continued, consequently reducing their motivation to act and use DCT tools [18]. Such changes in risk perception could explain the app download trend in Italy, which consisted of an initial peak when the app was launched in June 2020 (up to 600,000 downloads in a single day), followed by another massive increase in downloads at the beginning of the second wave, reaching more than 200,000 per day. This then tailed off to around 2000 downloads per day until April 2021, when the number of cases was limited and the vaccination campaign was at full deployment [30]. Among other factors explored, a few studies have already documented how high levels of trust in governments and health authorities can motivate people to adhere to prevention strategies [18,19,31]. It is fundamental that institutions convey official messages clearly and coherently, and combating disinformation from other sources as much as possible [18]. In addition, good communication seems important for increasing the acceptability of the app in our study population: the strongest predictor of app uptake in our analyses was being advised to download it, while a reason for non-adoption was a lack of awareness of the app. Lastly, our participants belonged to an age group that may be characterized by sociability, the importance of self-identifying with a peer group and the influence of peers on the adoption of health behavior [32]; therefore, exploiting these social mechanisms by implementing targeted communication strategies is likely to be effective at reaching a large fraction of this population [33].

As aforementioned, at the time of the survey (April 2021), Italy was at the end of the second wave, which had been characterized by a high incidence of SARS-CoV-2 infections during the fall and winter of 2020–2021 [34]. Hence, in the low transmission risk scenario of April 2021, it was not unexpected that we found a low perception of the utility of the DCT tool, similarly to other studies [19,21]. Within this context, communication policies that help people understand the importance of such measures in safeguarding their own and community health should be devised [33]. Such campaigns are most effective when risk perception is high, because people are motivated to take action to protect themselves, which potentiates DCT acceptance [18]. In fact, it is well known that a low adoption rate is the main barrier to the effectiveness of these apps [35] and the poor uptake may be responsible for the limited number of app notifications that our students reported. However, while potential contacts mostly followed the health recommendations provided by the app, which is encouraging because it highlights their awareness of the need to adopt preventive measures promptly, a substantial proportion of our students claimed they were hampered by technical issues with the notification process. This underlines the importance of investing in technical improvements of these apps and making them easy to use for the entire population, most of whom are less digitally literate than young people [35,36]. It is, in fact, important to highlight that several technical skill challenges could occur, such as some people not knowing how to download and install an app, or how to interact with it, thus limiting its acceptance and usage [18].

Interestingly, our findings contrast with a few international studies that report how concerns about data privacy can negatively impact DCT app adoption [16,31,37,38]. This could be explained by the fact that, compared to other, similar apps, IMMUNI collects relatively few data [39]. Additionally, our cohort was composed of university students, who may be accustomed to sharing their data on the web and not be particularly concerned about privacy issues [40]. Lastly, as for incentives that might promote the adoption of DCT apps, despite their recognized importance [41], few studies have investigated this aspect and available evidence focuses only on financial incentives [42,43]. In our study, it was the

app's actual utility (or otherwise) that seemed to influence its adoption rate. This highlights how the feeling of being engaged may motivate people to participate in a DCT system and confirms the importance of investing in communication policies that point out the potential health benefits of using such technologies [44].

This study has some limitations. Firstly, the cross-sectional design hindered the opportunity to draw causal conclusions between app uptake and associated factors. Secondly, the relatively low number of participants might have limited the statistical power. Thirdly, since we investigated students enrolled in healthcare degree courses, results are not generalizable to all university students. For these reasons, further analyses should be conducted comparing students of both medical and non-medical subjects to highlight possible differences between the two groups. However, to the best of our knowledge, this is the first study that investigates how Italian students relate to IMMUNI by analyzing factors that affect its adoption. Since these factors are specific and different across population subgroups, it is fundamental to assess and monitor them over time, so that they can be addressed in the development of similar technologies. In addition, we were able to examine the experience of students that used the app and also to explore possible incentives to encourage reluctant or disinterested users. The data provided in this study may support policymakers in developing effective strategies for the promotion of app uptake and, more broadly, to facilitate engagement of people with digital health prevention measures.
