Digital technology and mental health during the COVID-19 pandemic: a narrative review with a focus on depression, anxiety, stress, and trauma

1. Introduction

The COVID-19 pandemic has had catastrophic effects on physical health and mortality with over 670 million confirmed cases and 6.8 million deaths worldwide as of March 3, 2023 (1). The consequences of the pandemic on mental health have also been devastating with increases or exacerbations seen in disorders such as depression, anxiety and post-traumatic stress disorder (PTSD). This has occurred due to the social isolation, lockdown and other governmental measures taken to stop the spread of the virus (2–4). In addition, it is now apparent that a significant proportion of these cases may be a direct consequence of viral infection or post-infection neurological sequelae (5–8).

According to the Office for National Statistics in the United Kingdom (UK), an estimated 10% of the adult population had moderate to severe depressive symptoms during the period of July 2019 to March 2020, before the COVID-19 pandemic and ensuing lockdowns and social restrictions began (9). This percentage approximately doubled to 19% by the first 3 months of the first COVID-19 wave, peaked at 21% during January to March 2021, and marginally eased to 17% by August 2021 and to 16% by October 2022 (Figure 1) (9, 10). Despite these effects on mental health, fewer people with depression and other mental disorders sought help from healthcare professionals during the first 18 months of the pandemic (11). The waiting time for psychotherapy amounted to several months even before the pandemics. According to the German Psychotherapists’ Association, the average waiting time reached 19 weeks in 2017 (12). Because many digital apps are practitioner-independent, the use of the apps provides faster help for patients seeking psychotherapy, while guided apps are more effective (13).

Depression and other depressive disorders are often managed by a general practitioner, although more severe cases may require specialist services (14). Most of the time, depression is treated with antidepressants or psychotherapy, although these approaches are not always successful. In England, the Royal College of Psychiatrists estimated that an improvement in symptoms with the first antidepressant prescribed occurs for only 50–65% of patients and unsuccessful cases may be prescribed other antidepressants in a successive manner until symptom improvements are achieved (15). Thus, a high proportion of patients can go several weeks or months without any signs of improvement (16). This can create a significant burden at the individual, healthcare and societal levels, and highlights the importance for improvement of existing approaches or development of new treatment strategies for this debilitating mental disorder. Furthermore, this situation was compounded by reduced access to both outpatient and inpatient mental healthcare during heightened periods of pandemic where lockdown and social restrictions were at their highest. Thus, efficient solutions were needed to cope with this challenge. This has led to increased use of videoconferencing (telehealth/remote consultation), internet-based support and employment of mobile phone and web-based apps as potential substitutes for face-to-face psychotherapy (17–19). Such approaches have been seen to benefit patients and healthcare workers in all areas of medicine due to the convenience of these methods, as well as the obvious associated reductions in the risk of contracting and transmitting the virus (20–23).

Although it is common for frontline medical staff and other healthcare workers to experience or witness physical and mental traumas, as well as suffering and death as part of their daily routine, such incidences increased significantly during the first waves of the pandemic. Several reviews and meta-analyses have investigated the effects of the pandemic on healthcare workers around the world, and these have identified an increased prevalence of conditions in this population, such as acute stress disorder, anxiety, burnout, depression and PTSD (24–31).

In this narrative review, we describe how digital technologies such as questionnaire-based apps, artificial intelligence (AI)-guided chatbots, online therapy software, goal-oriented physical activity apps and video games have helped to meet the challenges of the COVID-19 pandemic in delivering mental healthcare solutions (Figures 2, 3). We have focused these efforts on conditions such as depression, anxiety, stress and trauma, which markedly increased in prevalence during the pandemic (32, 33). Also, we have identified areas where the most significant progress has been made, as well as scenarios where further research is needed to achieve the most effective outcomes. Our main focus is to describe such digital approaches for use in diagnostics, patient monitoring and as a treatment option for the general public and frontline medical staff suffering with mental health issues.

2. Symptom monitoring and treatment guidance

3. Physical activity and game-based apps

Online physical activity sessions, such as virtual reality exercise (74), interactive physical-activity-based games (75), singing (76), and dance (77), have also been shown to be effective in reducing depressive and anxious symptoms.

Groenveld et al. investigated the utility of self-administered virtual reality physical, relaxation and cognitive exercises to aid physiotherapy-based recovery after individuals had been infected with the SARS-CoV-2 virus (74). All participants performed the exercises for 6 weeks and physical performance, activities, cognitive function, quality of life, and anxiety and depression symptoms were measured before and after this interval. The results showed that 75% of the patients found that the experience had a positive effect on their recovery from the viral infection, with an average use time of 30 min per session, and frequency of 3 to 4 times per week, over 3–6 weeks. Although adherence to the physical-based exercises decreased over time, the use of the relaxation and cognition-based exercises was constant throughout the study period. Importantly, the physical performance and quality of life measures showed significant improvements after the 6 weeks session, compared to the levels seen at baseline time point. Similarly, Kim et al. described a framework for use of gamification in the development of exercise-based games focussed on cardiovascular and strength training as a treatment for depression (75). The researchers are currently planning a pilot study to obtain user feedback to guide refinement of the game content. This will be followed by a randomized controlled trial to assess the efficacy and cost-effectiveness of this intervention compared existing standard methods.

Exploring other avenues, a 12 weeks randomised controlled trial assessed the effect of singing compared to usual care on quality of life, breathlessness, balance confidence, physical activity, as well as depression (PHQ-9) and anxiety (GAD-7) in 120 chronic obstructive pulmonary disease (COPD) patients (76). As nine singing and nine control patients had completed five face-to-face sessions and seven online sessions due to adjustments made for COVID-19 restrictions, the authors reported on the outcomes in this group as a pilot study to inform further trials. The results revealed improvements in depression and balance confidence scores in this group that transitioned to the online sessions, despite a general preference for face-to-face meetings. In another study, Humphries et al. found that a single 60 min session of self-selected dance resulted in an increase in positive affect and self-esteem and a decrease in negative affect and depression symptoms in healthy adults (n = 47) during the social isolation associated with the COVID-19 pandemic (77).

Tong et al. investigated several aspects regarding the use of mobile apps and fitness trackers during the pandemic using an online survey of 552 adults from June to September 2020 in Australia (78). This showed that users of fitness trackers and health-related apps were more physically active compared to non-users and more women used these health-related apps compared to men. Interestingly, most users stated that the apps did not adapt quickly enough to keep pace with the changing conditions of the pandemic. This highlights a potential problem that should be addressed quickly in future disaster situations.

A recent systematic review found that playing video games helped many people from all aspects of life to cope with the difficult life experiences that they encountered during the pandemic, particularly the times of the peak lockdown and isolation periods (79). Interestingly, this revealed that augmented reality and online multiplayer games helped to ameliorate the loneliness, stress and anxiety experienced by adolescents and young adults during these times. It is possible that the feeling of regaining control over life and social contacts, even if this is only virtual, plays a role in this. On the other hand, playing such games had detrimental addictive effects in the longer term for high-risk individuals, such as problematic gamers. A collaborative study carried out by the Centre for Addiction and Mental Health and National Research Council of Canada evaluated an online cognitive behavioural therapy-based video game called Legend of Evelys for ameliorating the increased stress due to the pandemic (80). The storyline of this fantasy game was based on the emotional status of people in a fictional setting that mirrored the negative effects of the pandemic on the population. The game allowed interactions between the avatars chosen by the users with prompts encouraging their engagement in therapeutic behavioural exercises. After a 2 weeks trial the participants gave the game a rating of “good” (Systems Usability Scale score) and the study showed a significant reduction in perceived stress compared to the baseline levels.

4. Treatment of frontline medical staff

A number of recent studies have demonstrated that frontline medical professionals and healthcare staff experienced increased stress, PTSD, anxiety, and depression at peak periods of the COVID-19 pandemic. There are many reasons why these stressors may have occurred. On the one hand, there was not enough previous experience, which meant that there were no guidelines to give support and security. On the other hand, due to the high number of patients, triage was necessary and not all patients could be offered help due to limited number of ventilators and nurses. This can stress medical staff to a great extent when there is the impression that help could potentially be given but resources are not sufficient, and the feeling of helpless overload sets in. This contradicts the professional ethos of many clinical staff members.

Intrusive memories are key feature of post-traumatic stress syndrome (PTSD) (81). Interventions within just a few hours after the initial trauma have shown some effectiveness in blocking the encoding or consolidation of these memories into long-term storage (82, 83). It is also now considered that even consolidated memories can become labile, albeit transiently, if they are recalled into working memory (84). This suggests that interventions for long-term intrusive memories may also be effective. One key question is: what sort of intervention would be most effective and could be translated easily into routine practice with minimal discomfort to the sufferers?

In 2017, Hagenaars et al. tested whether memory reactivation 4 days after participants watched an aversive film would result in fewer intrusive memories recorded in a diary if the participants performed a visuospatial or verbal task, compared to a no-task control (85). The film consisted of live footage of road traffic accidents and ensuing traumas such as the work of the emergency recovery personnel and the effects on the victims, including injuries and death. The visuospatial task consisted of playing the computer game Tetris^® (86) and the verbal task consisted of word games, each performed for 10 min. The results showed that both the visuospatial and verbal tasks were effective in the reducing the number of intrusive memories. Interestingly, the verbal task was marginally more effective in reducing intrusive memories but was rated by the participants as more difficult to perform, compared to the visuospatial task.

In 2018, the same research group extended these studies by carrying out a randomised controlled trial to investigate the effects of the visuospatial intervention for 20 min per day for 1 week on the number of intrusive memories experienced by real-life trauma victims (87). They found that there were fewer intrusive memories in the Tetris-play group compared to a control group that recorded an activity log for the same 20 min time period. The participants in the Tetris group also reported less distress from the intrusive memories at 1 week, as determined using the Impact of Event Scale-Revised (88). Members of the same research group also carried out a single case series study involving 20 longstanding PTSD patients, who were administered a similar Tetris gameplay intervention for 25 min on a weekly basis over 5 to 10 weeks in an inpatient setting (89). The results showed that intrusive memories decreased by an average 64% and 16 out of the 20 participants met the criteria for showing a favourable response to the intervention. Similar positive findings were reported using this same intervention in the case of refugees with traumatic memory intrusions in Stockholm Sweden (90) as well as for trauma-exposed women in Iceland (91).

Two studies were registered on clinicaltrials.gov to investigate the effects of applying the remotely delivered Tetris gameplay intervention to reduce the number of intrusive memories in frontline healthcare staff who experienced or witnessed traumas involving their work during the COVID-19 pandemic. The first of these aims to investigate the effects of a remotely delivered cognitive task over 4 weeks on intrusive memories and other symptoms in Swedish hospital staff (92, 93). The intervention consists of a brief memory cue of the traumatic event followed by Tetris gameplay. The control arm consisted of staff who listened to a daily podcast for the same time-period. The design includes completion of a daily intrusive memory diary and various questionnaires on day 1, week 1, week 4 and various time points up to 6 months, as possible. It is expected that the results of the study will be reported in 2023. The second study used a similar design to reduce the number of intrusive memories experienced by National Health Service intensive care unit staff who worked during the COVID-19 pandemic in the United Kingdom (94). A key difference between this and the Swedish study was the incorporation of a delayed arm in which the participants were given usual care for the first 4 weeks followed by access to the intervention for the next 4 weeks. The primary outcomes in this study were the number of intrusive memories at week 4 in both arms of the study. The secondary outcomes included the number of intrusive memories at week 8 compared to week 4 in the delayed arm. The results of this study were published recently in Molecular Psychiatry (95). The results confirmed a positive treatment effect, with the immediate arm reporting fewer intrusive memories compared to the delayed arm at week 4. The results of the delayed arm of the study will be published later this year.

5. Limitations and current challenges

As with all narrative reviews, this paper was limited by incompleteness of the literature search. Such completeness would not be possible in this case as we have focussed on a wide variety of digital healthcare technologies and apps which have either emerged or showed increased usage during the pandemic. Furthermore, the efficacy of these approaches have been investigated in a number of psychiatric disease areas, and so we opted to limit this to investigation of depression, anxiety, stress and trauma, which all increased in occurrence during the first waves of the COVID-19 infections. In addition, healthcare apps offer many benefits to users but they also have limitations and current challenges that need to be addressed. One significant challenge is user engagement, where some people may download a healthcare app but fail to consistently use it due to lack of motivation, forgetfulness, or poor user experience. One study found that although the number of app installs and daily active minutes of use may seem high, only a small portion of users actually used the apps for a long period of time (96). In particular, elderly people were reported to feel discomfort integrating health apps and wearables into their routine due to the constant reminder of their illness, according to qualitative interviews (97). Therefore, the usability and acceptability of digital health apps need to be improved for people with chronic conditions (98).

Additionally, there is a risk of over-reliance on healthcare apps, which might be racially biased due to the choice of distinct population groups used to train the relevant algorithms (99), or they may be unable to properly address crises like suicidal ideation due to inconsistencies in language across different apps (100). Also, there is the problem of the digital divide due to issues regarding access to the internet or technologies for some individuals. This is particularly difficult for those who are already socially or economically disadvantaged, the elderly population and those with cognitive disorders, which can further perpetuate inequities (101–103). To add to the problem, private and governmental organizations and stakeholders may encounter challenges in adequately validating and approving new digital health technologies. Therefore, more research is necessary before implementing a digital product in the healthcare sector (104). There are also concerns about data security and privacy risks, as highlighted by various studies, and governmental regulations are needed to enhance fairness and protection of personal data (105).

Health care apps need to be systematically assessed to ensure the safety of target users (106). With the evolution of healthcare apps, there are new risks associated with the automation of the chat interface. App creators need to report on safety and bias protection mechanisms to mitigate potential harm to their users, explain potential risks and harms to users, and regularly monitor and track these mechanisms (106).

6. What next steps are needed?

The COVID-19 pandemic has provided valuable insights into the role of digital mental healthcare and highlighted key areas that should be integrated into preparedness plans to help us cope better in case of future pandemics and other crises. The lessons learned include the importance of flexibility, data security, and the need for interdisciplinary collaboration, along with enabling access to technology for diverse populations, including those with limited access or for individuals who may face barriers due to socioeconomic and other factors (Table 3). This information can be used to inform future pandemic preparedness plans and to help ensure that mental health support is readily available during challenging times. This will require addressing key areas including the need for scalability, accessibility, adaptability, as well as instilling an increased awareness of mental health issues during crisis situations and the importance of early intervention to maximize treatment effectiveness (Table 4). This will help to make sure that digital mental health tools are safe, effective, and accessible to those who need them.

7. Conclusions and future perspectives

The evidence presented in this review supports the case that use of digital technologies such as mobile phone apps, chat bots and video games, may serve as an effective coping strategy with therapeutic potential during epidemic, pandemic or disaster situations. Several studies have now demonstrated that such approaches not only enabled patients with mental disorders to effectively maintain their visitations with their healthcare providers, in many cases they were also shown to be effective in ameliorating symptoms such as stress, anxiety and depression. For example, the use of online mood-tracking apps, questionnaires and AI-guided chat bots appears to support patients by involving them in monitoring their own symptoms, which can be used to inform the treatment process.

As further examples, several studies have demonstrated the effectiveness of online guided physical activities, such as physical fitness exercises, dancing and singing. Computer or mobile phone-based games have also proven to be a useful distraction. Along these lines, playing the computer game Tetris has been shown to alleviate the number of intrusive memories in individuals who have either witnessed or experienced traumas such as terrible or fatal injuries of road traffic accident victims. Considering the disastrous effects of the COVID-19 pandemic on mental health issues on a global level, it will important to incorporate many of digital approaches into the rapid response infrastructure to help humanity to cope more effectively at both the physical and mental levels with disaster scenarios. This includes earthquakes, floods, tornados, hurricanes and future pandemics. Nevertheless, it is crucial to continue researching ways to enhance the user experience in order to ensure the successful integration of digital mental health interventions into clinical practice and promote increased user engagement in order to further increase therapeutic effectiveness of these programs.

Author contributions

PG, JS, VV, and JE wrote and edited the manuscript. PF proofread and provided valuable input into the final version. All authors contributed to the article and approved the submitted version.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.