Associations between COVID-19 related media consumption and symptoms of anxiety, depression and COVID-19 related fear in the general population in Germany

During the last months, the COVID-19 pandemic has grown to the most serious international health problem of the last decades and therefore represents a substantial challenge for communities all over the world. In addition to the immediate threat posed by the virus, the current situation also comes along with significant distress as well as substantial fears and worries related to its social and economic consequences for a significant number of people [1‐3]. The psychological strain on the general population, however, may be further increased by the requirements of quarantine or the so-called “social distancing” that is supposed not only to contain the spreading of the infection but also to profoundly impede the interpersonal communication.

In this context, the role of the media has become quite essential on different levels. First, they are important sources of information with regard to topics (in)directly related to the virus (e.g., infection rates, measures of the government, recommendations of the public health authorities, or the economic and social situation). Several studies demonstrated the meaning of media coverage during previous pandemics since the type, as well as the frequency of news reporting, was shown to significantly influence individuals’ health-related attitudes and behavior [4‐6]. In addition, particularly social media may further hold the potential to (partly) bridge the problems resulting from the restrictions of the face-to-face contact. In these times platforms like Facebook, Twitter, Instagram or specific internet forums are essential ways to share opinions, personal experiences, moments of happiness, worries or fears for many people as reflected by a dramatically increase of COVID-19-related terms up to several million mentions on these channels until March 2020 [7]. This way of communication therefore may have an important function for psychological wellbeing and applying social media for staying in contact with significant persons is decidedly recommended by official sites in Germany [8].

But these positive aspects might be just one side of the medal. The exploitation of media may also bear a certain risk for the mental health state due to elevated stress levels, particularly once a critical mass of predominantly negative news or unfiltered information is reached. Being aware of this, the World Health Organization (WHO) currently created the term “infodemic” in the context of the present crisis for an “over-abundance of information—some accurate and some not—that makes it hard for people to find trustworthy sources and reliable guidance when they need it” [9]. To “help minimize fears” the WHO, but also other international operating organizations therefore published distinct mental health considerations. Common central points of these recommendations encompass the restriction of (distress-inducing) media usage with respect to its duration and frequency as well as being kept informed using official websites of (inter)national health authorities to “distinguish facts from rumours” [10‐12]. These reservations regarding media consumption related to a macrosocial disaster are backed up by existing data in this field. Former trials reported about significant psychological strain or distinct mental conditions resulting from crisis-related media coverage [13‐15] and individual vulnerability was suspected to account significantly for this risk [16]. In the context of the current pandemic, up-to-date data also suggest that topic-specific media reporting results in a high level of stress in the vast majority of people [17]. However, we are aware of only one study that addressed the influence of COVID-19 related media exposure on stress-related mental symptoms so far. Including more than 4000 adult citizens, this web-based approach found the frequency of social media exposure to be significantly positively correlated to symptoms of depression and anxiety within the general population in China [18].

To further elucidate the relationship between media consumption and symptoms of depression, unspecific anxiety, as well as COVID-19 specific fear in western communities, we also performed a web-based investigation within the general population in Germany. In our analyses, we firmly considered not only the impact of the frequency and duration of media exploitation per day but also the meaning of specific types of media on these stress-related domains. Since it should be quite important at a practical level to define risky media usage as precise as possible, we further thought to determine concrete numeric values with respect to the daily frequency/duration of distress-inducing media exposure. Finally, we investigated whether there is a specific risk of psychological distress related to media consumption in individuals who already suffered from health-related fears to gain an insight into susceptible vulnerability factors.

First, our results of a positive correlation between COVID-19 related media exposure and the severity of unspecific anxiety, depression and topic-specific fear substantially agree with up-to-date findings of a web-based study from China. Using a comparable web-based cross-sectional design and language-validated versions of the seven-item generalized anxiety disorder scale (GAD-7) and the WHO-Five Well-Being Index (WHO-5), Gao and colleagues also found the severity of anxiety and depression to be significantly positively correlated with the frequency (determined ass “less”, “sometimes” or “frequently”) of self-reported media usage regarding COVID-19 within more than 4000 citizens [18]. In addition, the present study further demonstrates that not only the frequency but also—and even stronger—the daily duration as well as the diversity of media exposure are significantly positively correlated with global symptom severity and therefore emphasize the meaning of mental distress associated with COVID-19 media coverage, particularly once a “critical load” has been reached. This conclusion is supported by both a recent study that turned out media coverage as one of the strongest emotional stressors in the context of the current pandemic [17] and the data of the present trial. Although it is not clear if psychological strain is rather a consequence or a cause of more immoderate media exploitation, we found the frequency and the duration of COVID-19 related media exposure to be positively correlated with the subjective need for its reduction due to a substantial amount of topic-induced stress. This result may be interpreted in a way that a significant number of individuals are aware of “overconsumption” of information as a risk factor for mental wellbeing in the present crisis.

Therefore, a precise determination of risky media exploitation should be warranted to refine already existing recommendations in this field. We therefore ascertained concrete numeric values with respect to the frequency and the duration of media usage that are associated with at least mild to moderate symptom severity on the PHQ-4 and the DSM-5 Specific-Phobia-Scale. Our results indicate that the frequency with a possible hazardous association with mental health might be somewhat higher than the maximum of twice a day recommended by the WHO [10]. However, it seems to be important to note that we are able to identify COVID-19 related media usage of seven times or 2.5 h a day to be associated not only with critical values for unspecific anxiety and depression but also with COVID-19 specific fear. These results may provide some evidence for a distinct threshold of media consumption that is associated with an increased risk for a range of mental health complaints within a significant number of people.

Moreover, our results indicate that the amount of mental distress depends on the type of media primarily used for gaining information whereby the usage of social media was associated with a significantly higher degree of unspecific anxiety and depression than keeping informed by official sources. This finding supports the considerations of the WHO to primarily rely on information of health authorities for information about COVID-19 [10] and provide some evidence for a problematic role of social media in the context of mental health, probably due to an elevated stress level induced by a flood of unfiltered information and a more emotional style of communication. Interestingly, with regard to COVID-19 specific fear the usage of social media and official websites were both associated with significantly higher levels of fear compared to participants not using those media types. Although it cannot be ruled out that the favored sources of information may not be a cause but rather a result of psychological strain and/or other differences between the user groups, this finding may provide some evidence for a problematic role of both sources in case of prominent COVID-19 related fears.

Finally, we further assessed the impact of pre-existing health-related fears on the risk of psychological strain due to COVID-19 related media exposure and found both the frequency and the duration of media usage to be significantly stronger correlated to anxiety and depression in individuals with this characteristic than in those without. Although research regarding specific vulnerability for the development of psychopathology due to crisis-related media exposure is rare, some trials already focused on this topic in the context of posttraumatic stress. Results showed that the quality of emotional reaction during media exposure [15] or prior direct exposure to traumatic events [13, 14] were significantly associated with the risk for (acute) posttraumatic stress after topic-specific media consumption related to bombings or the attacks of September 11th, 2001. Our findings therefore provide further evidence that vulnerability, particularly those with relevance for the specific content of a disaster, lead to an increased probability of experiencing significant mental symptoms after crisis-related media exposure. Interestingly, participants with pre-existing health-related fears demonstrated a significantly higher frequency and duration of media exploitation as well as a substantially higher severity of (un)specific anxiety and depression than those without. This might be interpreted in a way that, probably related to dysfunctional checking behavior, vulnerable individuals slightly exceed the critical threshold of 2.5 h and 7 times daily that in turn leads to significant increased stress-related symptoms in these individuals. However, another explanation might be that this subgroup suffered from more severe anxiety and depression per se and the higher frequency/duration of media exposure is due to a more frequent safety behavior. Although causal relations cannot be proven due to the cross-sectional nature of this study, these findings may indicate that patients with health-related fears should pay special attention to their COVID-19 related media exposure to avoid extraordinary mental strain. Besides a pre-existing vulnerability for worrying about physical illness, a vulnerability for addictive behavior, respectively addictive internet/media use could be a potentially relevant factor regarding media consumption and COVID-19 related psychological parameters. Therefore it should be taken into account in future studies with validated and standardized questionnaires.

Several limitations of the present study have to be considered that may limit the generalization of our findings. Data sampling via social media might result in a major obstacle for some individuals to participate (e.g., for those who have limited access to or are not familiar with the use of social media such as the elderly). Addressing a convenience sample further might lead to a bias in a way that individuals experience a relatively high level of COVID-19 related psychosocial strain are more likely to take part in our study. Indeed, some evidence for sampling bias results from the fact that in relation to the general population an above-average portion of the participants was female and reported to work in the medical field. For reasons of feasibility, we used rather short questionnaires to assess for (un)specific anxiety and depression. The application of an unvalidated questionnaire to run for COVID-19 specific fear, however, represents a major limitation of this study. Finally, due to the cross-sectional nature of this trial, no causal conclusions can be drawn on this basis. However, due to the emerging situation, it should be important to gain an insight into the potential associations of media consumption and pandemic-related mental health as well as to compare available considerations of official authorities in this area with original data. As a consequence of our findings—especially if they will be confirmed in longitudinal studies—practical interventions for dealing with the current COVID-19 pandemic as well as with other possible crises in the future should be derived. For example, online intervention tools for people with critical media exploitation or a pronounced vulnerability for maladaptive psychological outcomes should be envisaged. Moreover, different media formats could include screening instruments for detecting a potentially problematical consumption of media and provide recommendations for a more adaptive usage of media reporting.

Summing up, our findings provide some evidence for a problematical role of media exposure in the context of the COVID-19 pandemic. The frequency/duration of media exploitation, type of media, the diversity of media usage as well as individual vulnerability might play an important role in the development of mental distress or might be a result of mental distress and therefore relevant in the perpetuation of mental distress. Since our data point to a distinct threshold of topic-specific media exposure that should be considered to avoid psychological strain, our results therefore further may serve as an orientation not only for the general population but also for professionals in the (mental) health sector to provide recommendations in this context. Future studies should try to clarify the causal relationship of our findings to create a reliable basis for considerations in this field.