Dealing with information overload: a comprehensive review

Information overload is a problem that is being exacerbated by the ongoing digitalization of the world of work and the growing use of information and communication technologies. Therefore, the aim of this systematic literature review is to provide an insight into existing measures for prevention and intervention related to information overload. The methodological approach of the systematic review is based on the PRISMA standards. A keyword search in three interdisciplinary scientific databases and other more practice-oriented databases resulted in the identification of 87 studies, field reports, and conceptual papers that were included in the review. The results show that a considerable number of papers have been published on interventions on the behavioral prevention level. At the level of structural prevention, there are also many proposals on how to design work to reduce information overload. A further distinction can be made between work design approaches at the level of information and communication technology and at the level of teamwork and organizational regulations. Although the identified studies cover a wide range of possible interventions and design approaches to address information overload, the strength of the evidence from these studies is mixed.


Introduction
With the digitalization of both work and private life, information is available in large quantities in digital form; we live in an "information society" (Karvalics, 2007). It is possible to conveniently and actively access diverse information, and we also passively receive large amounts of information and messages. Despite the different channels of information, information is mostly consumed through screen displays. Bawden and Robinson (2009) refer to this phenomenon as "homogenized diversity. " Currently, the amount of information that is created every two days is roughly equivalent to the amount of information that was created between the beginning of human civilization and the year 2003 (Jackson and Farzaneh, 2012). The amount of information available has thus become excessive, but it is difficult to assess its quality. As a result, information overload has become a widespread problem. Indeed, information overload was cited as one of the most frequent stressors by 22.5% of respondents in a representative German sample (Meyer et al., 2021). The COVID-19 pandemic can be seen as a catalyst for these developments, and it seems likely that the resulting increased use of mobile working patterns, virtual meetings, and collaborative digital software will be permanent (Rigotti et al., 2021).
Empirical evidence shows that information overload is positively related to strain, burnout (Girard and Allison, 2008;Day et al., 2012;Antoni and Ellwart, 2017), and various health complaints (Junghanns and Kersten, 2020), and negatively related to job satisfaction (Hunter and Goebel, 2008). Furthermore, information overload is associated with serious performance losses, especially in connection with disruptions and interruptions (Baethge and Rigotti, 2010;Rigotti, 2016). Finally, studies show that the quality of individuals' decisions is affected by information overload (Phillips-Wren and Adya, 2020).
Given the relevance of information overload for the health and well-being of employees, as well as their performance outcomes, we provide a systematic review of preventive measures for information overload. Previous reviews on the topic of information overload aimed to identify the factors that influence information overload (e.g., Eppler and Mengis, 2004;Antoni and Ellwart, 2017) or to present the consequences of information overload (e.g., Antoni and Ellwart, 2017). Furthermore, a meta-analysis by Graf and Antoni (2020) focused on information characteristics as antecedents of information overload. Although some review articles have examined the design or intervention options to counteract information overload, these are usually specific to one professional group (mainly medicine, e.g., Khairat et al., 2018).
Firstly, the aim of this review is to systematically describe the tools and interventions that can be used to manage information overload. The various recommendations and interventions are clustered according to the levels of the causes (Eppler and Mengis, 2004) or sources (Graf and Antoni, 2020) of information overload. This clustering makes it possible to determine whether the recommendations address the amount of incoming information or, conversely, the handling of the incoming information. In addition, this study assesses the extent to which preventive measures are proposed in terms of behavioral versus design solutions.
Secondly, we aim to assess the current state of knowledge on design recommendations for information overload. We will also examine whether the guidelines and recommendations are concrete or still relatively vague and, thus identify the areas in which knowledge deficits remain.

Theoretical classification and definitions
Several papers have already addressed the theoretical grouping of the factors that contribute to information overload and the consequences of information overload (Eppler and Mengis, 2004;Graf and Antoni, 2020). These models thus provide a theoretical framework for the present review. In addition, the cognitive load theory (Atkinson and Shiffrin, 1968) and the media richness theory (Daft and Lengel, 1986) are often referred to in the literature on information overload. Therefore, we will briefly describe these theories.
In their framework model of the concept of information overload, Eppler and Mengis (2004) suggest that there are several interrelated causes of information overload: the characteristics of the person receiving the information, the characteristics of the information, tasks and processes, organizational processes, and information technology. The resulting consequences of information overload require the use of countermeasures, which, in turn, affect the causes of information overload. This process is circular, and all the aspects are interdependent (Eppler and Mengis, 2004).
Cognitive load theory suggests that the human working memory is limited to approximately seven ± two units of information (Atkinson and Shiffrin, 1968). Accordingly, information overload occurs when the amount of information exceeds the working memory of the person receiving it (Graf and Antoni, 2020). Cognitive load theory identifies three different categories of cognitive load: extraneous, intrinsic, and germane cognitive load. Extraneous cognitive load is influenced by the design of the information (Sweller, 2005). Intrinsic cognitive load results from the content of the information, such as its complexity. Finally, germane cognitive load is the favorable, learning-enhancing cognitive load that results from focused engagement with the information. Ideally, this final form of cognitive load leads to the construction of schemata and mental models (Sweller, 2005).
Media richness theory also provides a theoretical framework for studying information overload. One of the goals of using information and communication technology (ICT) is to reduce the ambiguity of information (Daft and Lengel, 1986). Therefore, information richness, which refers to the extent to which information can change understanding in a given time interval (Daft and Lengel, 1986), should be considered in relation to information overload. In addition to the information itself, the various possible communication channels may also differ in their richness. Face-to-face conversations are thought to be the richest form of communication, while letters or e-mails are less rich (Daft and Lengel, 1986;Kauffeld et al., 2016). Cognitive load theory provides a precise definition of information overload, while the media richness theory better supports the development of design interventions. Against the background of these relevant theories, the terms used in this study can be defined.
Information overload is a topic that is relevant to many disciplines, including medicine, social sciences, marketing, computer science, education, and psychology (Edmunds and Morris, 2000). This means that there is no single, universally accepted definition. In everyday language, information overload is often equated with receiving too much information (Eppler and Mengis, 2004). According to Klausegger et al. (2007), a consistent feature of the various scientific definitions of information overload is that the amount of information is initially related to better performance or better decisions but that, above a certain amount of information, the effect changes, and the amount of information leads to worse outcomes (an inverted U-shaped relationship; Eppler and Mengis, 2004;Klausegger et al., 2007). More specifically, Klapp (1986) defined information overload as an excessive amount of information that the receiver can no longer process efficiently without distraction, stress, increased errors, or other costs that reduce the efficient use of the information. Similarly, Eppler and Mengis (2004) argued that information overload occurs when the amount of information exceeds the processing capacity of the recipient.
As mentioned above, information overload is closely related to the use of ICT and, therefore, also to the concept of technostress. Technostress can be defined as the "stress experienced by individuals due to the use of ICTs" (Ragu-Nathan et al., 2008, p. 418). Information overload and constant availability are the two main stressors caused by the use of ICTs (La Torre et al., 2019). Therefore, information overload can be seen as a feature of technostress. Although this review focuses on information overload, it also includes primary studies that refer to both technostress and information overload. Five technostressors are commonly discussed in the technostress literature. These are techno-overload (leading employees to work longer and faster), techno-invasion (constant availability, including in leisure time), techno-complexity (complexity of the digital tools leading to feelings of inadequacy of computer skills), techno-insecurity (threat to the security of one's job), and techno-uncertainty (constant upgrades of hardware and software) (La Torre et al., 2019). Factors with the potential to buffer the detrimental effects of techno-stressors on employee well-being are discussed under the label of techno-inhibitors (Ragu-Nathan et al., 2008;Estrada-Muñoz et al., 2022). Examples of techno-inhibitors are literacy, participation, or innovation support (Jena, 2015;Estrada-Muñoz et al., 2022).

Search strategy
Following the recommendation by Methley et al. (2014) we used the PICO tool in order to define our keywords in both English and German (see Table 1). As the discussion on information overload comes from several fields, many synonyms for information overload are used. To cover these, we included a large number of synonyms as outcomes in our search string, linked with an OR operator. As we are interested in methods and interventions to reduce information overload, a further part of the search string refers to aspects that could be used as countermeasures against information overload. In order to also cover a wide range of potential countermeasures, we included 'software' as a keyword beneath the more psychological aspects of intervention/training/workshop or work/task design. In addition, to address individual strategies, we included the keyword 'coping strategies' . As described, the focus of this review is on information overload in the workplace, so a third set of keywords defining the population was added to ensure the relevance of the search results to the workplace. In order not to narrow down the results further than necessary, keywords in the comparison category were omitted.
In the second step, we defined the databases to be searched. We searched the scientific databases Web of Science, Ebscohost, Medline, PsycInfo, and PsycArticles using complete search strings with Boolean operators. With the aim of including the applied literature, we also searched other practical databases, including PSYNDEX Interventions, Rehadat, Arbeitssicherheit.de, the BAuA (Federal Institute for Occupational Safety and Health) publications, and publications of the Verwaltungs-Berufsgenossenschaft (VBG). As most of these platforms do not allow searches using Boolean operators, all the synonyms for information overload were used individually as keywords.
In the subsequent review process, we followed the PRISMA standards (Moher et al., 2009) and, accordingly, recorded how many results were obtained from the literature searches conducted in this way (identification, see Figure 1). Following the literature search, all duplicates were removed, and two independent raters classified the search results as relevant or irrelevant to the research question on the basis of titles and abstracts (screening). The screening process itself was blinded. If the two raters disagreed about certain identified documents, we made a joint decision about inclusion or exclusion. We then read the full texts of the documents that were deemed relevant which led to further exclusions if they did not fit with the research question (suitability).
In the final step, we coded and categorized the content of the eligible documents. We recorded the level of the intervention (person, information, tasks and processes, organizational processes, information technology), a precise description of the intervention, the type of article, the research question, the design of the study, the sample, the results of the study, and the implications in terms of recommendations for action and the person who should implement them. Excerpts of the included sources are provided in Tables 2-6).

Inclusion and exclusion criteria
Due to the rapid development of ICT, the time frame of the search was limited to the period from 2000 to 2021. We included the following: (1) evaluation studies on concrete interventions; (2) scientific studies that allow conclusions to be drawn about design options for dealing with information overload; (3) studies on information quality; (4) studies on the design of work-relevant information to make it easier to process; (5) studies on the topic of knowledge management that might prevent information overload; (6) studies that give advice on the possible content of workshops at the

AND AND
Comparison n/a n/a Outcome ("information overload" OR "flood of information" OR "information surplus" OR "information avalanche" OR "data trash" OR "data smog" OR "information input overload" OR "information burden" OR "data explosion" OR "technostress") We excluded the following types of papers: (1) papers referring exclusively to multitasking and not simultaneously to information overload, (2) papers referring to the use of ICT outside of working hours or information overload when using social media, (3) technical papers on the possibilities of storing large amounts of data as well as purely technical-methodological presentations of ICT solutions; (4) papers on the area of consumer research and communication with customers, especially those on recommendation systems, and (5) papers giving design recommendations for specific occupational groups and application areas, such as librarians, molecular biologists, autonomous driving, studying and teaching, and medical knowledge from the perspective of patients (health literacy).
A total of 1,054 papers were extracted from the scientific databases, of which 72 were identified as suitable after screening. A total of 259 papers were extracted from the practice-oriented databases, of which 15 were identified as suitable and included in the review.

Intervention and design approaches
The causes of information overload can be found on the societal level as well as on the organizational and interpersonal levels. At the societal level, aspects such as the accelerated production of information and the rapid dissemination of information via the internet are causes of information overload. These societal aspects are difficult to change through occupational health management. At the organizational and interpersonal level, information overload can be caused by five different aspects (Eppler and Mengis, 2004): (1) information, (2) person, (3) tasks and processes, (4) organizational processes, and (5) ICT.
Information characteristics relevant to information overload include the quantity, frequency, intensity, and quality of the information. In addition, other general characteristics of the information may play a role (Eppler and Mengis, 2004). Indeed, Graf and Antoni (2020) report that the complexity of the information, its degree of ambiguity, the novelty of the information, and its structure are relevant to information overload.
At the person level, the person receiving, processing, or communicating the information is a relevant factor for information overload. Specifically, the person's attitude, qualifications, or experience, including their competencies, skills, and motivation, can influence whether a given amount of information leads to information overload (Eppler and Mengis, 2004).
At the level of tasks and processes, information overload can arise from the work tasks themselves or from cooperation with other people. In particular, more routine procedures reduce the processing capacity required of the people involved, while more complex and new procedures are more likely to cause information overload (Eppler and Mengis, 2004).
Organizational processes refer to all formal and informal work structures at the organizational level (Eppler and Mengis, 2004). Changes and redesigns in organizational processes and team structures can increase the amount of information presented to the   (Eppler and Mengis, 2004). The emergence of new technologies and their use play an important role in causing information overload, as exemplified by widespread use of e-mail. It is important to take advantage of the opportunities offered by certain technologies, while at the same time minimizing the risks they pose.
These five levels of causes can serve as starting points for design options and interventions to address information overload. Accordingly, the results of this literature review are structured along these five levels. In this context, the design recommendations at the level of the person are classified as behavioral prevention measures, while the recommendations related to the other four levels can be regarded as structural prevention measures.

Level of information
According to Graf and Antoni (2020), a distinction can be made between the quantity and the quality of information, and both aspects of information are related to information overload (Graf and Antoni, 2020). Quantity is mainly understood as the amount of information (objective). However, if the subjective perception of the amount of information is also taken into account, contradictory results can be obtained. For example, empirical studies suggest that there is a positive correlation between the quantity of information and information overload, but at the same time some studies have reported a negative correlation between these two variables (see Graf and Antoni, 2020). It can therefore be assumed that the subjective assessment of the quantity of information may be influenced by the available resources and the individual's ability to manage the incoming information. The quality of information includes the various aspects that contribute to the fit of the information to the needs of the person receiving it. These aspects include, for example, the complexity or relevance of the information.
With regard to the quantity of information, it should be noted that a large number of papers have described the use of various information technologies to reduce the quantity of information. Therefore, papers that mainly describe and evaluate these technologies and thus address the problem of high information quantity are included in the section on information technology.

Quality and presentation of information
Much of the literature on the presentation of information relates to the medical field. Indeed, the condition of critically ill patients can be affected if an important piece of information is overlooked, and time pressure is often high in medicine. Beasley et al. (2011) showed that problems of complex "information chaos" are particularly relevant in the medical field. Information overload is often cited as one aspect, along with information deficits, information conflicts, and scattered or incorrect information. The following papers each address one or more of these aspects.
Based on 17 primary studies, Khairat et al. (2018) reported in their review that visualization dashboards reduce the time spent collecting data, the difficulty of the data collection process, the  cognitive load, the time to task completion, and the error rate. These visualization dashboards also improve situational awareness, adherence to evidence-based safety guidelines, ease of use, and navigation through the program. Therefore, the presentation of selected critical patient data in a clear manner can be recommended in clinical settings. Another review paper with the same target group was presented by , and their results indicated that more than half of the included peer-reviewed primary studies (12 out of 22) on the implementation of dashboards showed positive effects on outcomes such as patient health, process outcomes, efficiency, and costs. However, the authors themselves reported that the empirical evidence on the effectiveness of dashboards implementation in clinical settings is limited by the low to moderate quality of the primary studies. It should also be noted that these studies rarely considered information overload as an outcome. Three other original studies also address the clinical application context. Patapovas et al. (2013) tested a clinical decision support system, but since the study did not implement a pre-and postsurvey and did not include a controlled-randomized design, the results only permit the conclusion that the test users used the electronic patient record regularly. Ahmed et al. (2011) compared a new format for representing information with an established format. With the new format of representation (reduction, filtering, and better categorization of patient data), the test subjects showed better performance, faster processing times, and most importantly also a lower task and information load. The study suggests that the reduction, filtering, and better categorization of patient data have a positive effect on reducing the effort required to process the information. Another dashboard for the clinical context was developed and tested by Tan et al. (2013) based on Toyota's Andon Board. The dashboard extracts data from electronic patient records every minute and notifies clinicians of urgent orders (e.g., laboratory tests, medications, etc.), abnormal laboratory and radiology results, and infection alerts. Overall, the dashboard seems recommendable; in the study, it ensured efficient and safe work and was well accepted by the users. However, this empirical evidence is limited due to the lack of a control group.

Quantity of information
In a different work context (an LED factory) but still with a similar approach, Wu et al. (2016) used eye tracking to investigate three different levels of complexity in the presentation of technical information at the human-machine interface. The results showed that the time it took the subjects to fixate on the target object differed significantly depending on the complexity of the presented information. Specifically, fixation was fastest in the low complexity condition. Another finding was that the search patterns of novices were significantly more complex than those of experts. Furthermore, experts reported lower cognitive load than novices. Overall, these findings suggest that user interfaces should be designed to be as simple and clear as possible while still providing the necessary level of functionality. Furthermore, the same amount of information may have different effects depending on the professional experience of the individual. Indeed, Ries and Deml (2019) reported a similar conclusion in relation to the work of control room staff; they also proposed specific recommendations on how to optimize the presentation of information in the form of video recordings. Wnuk et al. (2016) proposed a way to visualize the content and progress of a project. This tool, called FSC+ (Feature Survival Chart), is specifically aimed at the fields of business management or project management. The tool makes it possible to visualize the scope of a project and to show the dwell time of individual project steps in the project, thus facilitating project decision making. A test of the tool with 20 sample users showed that the interviewees were positive about the tool. However, the learning curve for using the tool was longer than expected. The study showed that the tool is more useful for people who need to work with large amounts of data and information, while the tool is less useful for people who only need an overview of a limited number of project steps. The study provides weak evidence that the FSC+ tool can be used in project management and project leadership to reduce information overload. Roy et al. (2017) addressed the question of how to increase management's knowledge of processes within the organization without leading to information overload. Based on an unsystematic review, the authors proposed the following solutions at the knowledge level: the use of summaries, comments by experts, indicators, and priority setting, the distribution of information over time, and the visualization of information. Solutions based on the use of digital media were as follows: the use of search engines, blogs, dashboards, news and updates, notes and virtual bookmarks, and alerts. In addition, with the aim of providing relevant information to decision makers in companies, Dzokoto et al. (2013) proposed a "SMART Push System" based on a literature review. This system actively suggests the relevant documents or content to decision-makers, taking into account all the available information and contextual knowledge, without them having to actively request the information. However, this article was only a conceptual work. Riener and Ferscha (2008) took a different approach, initially independent of the work context. They proposed the use of tactile stimuli to relieve the visual and auditory sensory channels. Their study focused exclusively on spatial orientation and tested the use of a vibrotactile belt to inform subjects about the spatial distance to a target object. Based on the results of the study, we cautiously conclude that it may be useful to use tactile stimuli to convey information. However, the participants only showed a learning effect when they were informed about the accuracy of their distance estimations, whereas their estimation performance deteriorated when no feedback was provided. Shah et al. (2019) examined the quantity of information sent to primary care providers through the Department of Veterans Affairs in the US. They also looked at new regulations to filter the information sent, as well as staff training on how to individually enable and disable certain types of messages. The study showed that the number of daily messages could be reduced with these policies, but no concrete impact or evaluation of the staff training was reported. It should also be noted that there was no control group and that the implementation of the intervention varied between regions.

Quantity of information
Another study in the clinical work context by Pickering et al. (2013) examined the information used by intensive care unit (ICU) staff when a new patient is admitted. The aim of the study was to identify the type of information that needs to be prioritized in the ICU. To answer this question, observations and interviews were conducted over 1.5 years in three different ICUs. Out of a total of 51 different clinical information concepts, an average of 11 concepts were used when a single patient was admitted. The four most commonly used concepts were heart rate, oxygen saturation, respiration, and Frontiers in Psychology 08 frontiersin.org blood pressure, and each was used in more than half of the admissions.
The study suggests that when a patient is transferred to the ICU, certain information should be presented in a prioritized manner. Access to other information should not be prevented, but the most commonly used categories of information should be presented in a clear and concise manner. Another approach to reducing the amount of information or helping individuals remember the most important information was examined in a Cochrane systematic review by Arditi et al. (2017). They focused on computer-generated reminders provided in paper to healthcare professionals. The data from 34 individual studies showed that this form of intervention significantly improved the quality of care. There was also weak evidence that patient outcomes could be positively influenced by these reminders. However, the extent to which this intervention could reduce information overload for health professionals was not investigated.

Summary: level of information
Empirical evidence on the quantity and quality of information to reduce information overload focuses mainly on the digital representation of information. When designing software, it is important to ensure that the amount of information presented is manageable and customizable. At the same time, in the software, it should be possible to intuitively access background information that is not visible at first glance. Design principles and design laws should also be considered. Although the studies to date are of relatively moderate quality (e.g., few have used a randomized control group design) and mainly relate to the medical field, the findings are also applicable to other professional groups for whom it is essential to have important information presented as clearly as possible (e.g., employees in control rooms or managers). Based on the literature presented here, the use of dashboards appears to be recommended.

Person level
At the individual level, the literature included in this review offers some advice on how to manage information and information overload in general. Specifically, information overload can be addressed at the personal level through education. Furthermore, we present recommendations on how to deal with e-mail and the Internet in general. The following section also includes suggestions on how to actively manage the boundaries between work and leisure time and the workplace itself.

Dealing with information overload
Information overload should be recognized as early as possible (Mustapar et al., 2016) and proactively addressed at the individual level (Landale, 2007). Landale (2007) presented a three-step approach that can be applied by individuals in combination with memory training: (1) receive and evaluate the document and decide whether to read it, delegate it, or ignore it, (2) gain an initial understanding of the document and decide whether to read it in detail and, if so, how quickly and how deeply, (3) adapt the new knowledge.
Similarly, Warbington (2000) recommended that when receiving information, one should first clarify the goal one is trying to achieve, from which the importance of different pieces of information can be assessed. Accordingly, a survey of 124 managers by Farhoomand and Drury (2002) and an editorial by Kluge et al. (2020) suggest that filtering information is seen as a helpful measure against information overload. This filtering can be achieved, for example, by suppressing, deleting, or selecting irrelevant information (Kluge et al., 2020). In the process of filtering or screening, according to Farhoomand and Drury (2002), delegating work is also an option, and this delegation should be directed not only to colleagues but also, if possible, to digital agents (Kluge et al., 2020). Finally, prioritizing information can also be a helpful action to manage information overload (Farhoomand and Drury, 2002).
Mindfulness also plays a role in managing information overload, and according to an unsystematic review by Stich et al. (2018), mindfulness skills should be developed in order to manage e-mailrelated information overload. Mindfulness can be assessed and improved through questionnaires that measure and provide feedback on the level of strain experienced by information overload or attitudes toward ICT.

Training
At the individual level, previous studies have recommended attending training to reduce and prevent information overload (e.g., Drössler et al., 2018;Stadin et al., 2020), with companies in particular being encouraged to invest in such training for their employees. However, Pfaffinger et al. (2020) also suggest that individuals should proactively demand further training opportunities.
A review by Antoni and Ellwart (2017) made it clear that individuals are able to process a large amount of complex information if they have the competencies to use modern ICT. These skills should be trained through further education, for example, on the functionalities of these technologies. Furthermore, according to Nagarajah (2016), the introduction of new technologies should be accompanied by adequate training in order to prevent technostress.
Based on a case study, Cheuk (2008) also suggested that information literacy is a human resource development issue in four dimensions: first, the skillful use of information at a strategic level; second, the organization and control of information; third, the knowledge of access and tools; and fourth, the linking of finding and using information.
In addition, individuals should be taught appropriate work strategies to deal with information overload, such as how to manage large amounts of information (Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), 2006). These strategies could include improving self-and time management, as these are effective measures against information overload (Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), 2006; Drössler et al., 2018).
According to a quantitative study by Yener et al. (2020), in addition to training in time management, technological self-efficacy training could also counteract the negative effects of technostress on burnout and performance. Mahapatra and Pati (2018) highlighted that techno-invasion and techno-insecurity are also positively related to Frontiers in Psychology 09 frontiersin.org   (1) "Resources for dealing with information overload": adequate technical equipment, information culture in the company, social support, media competence, and work methodology.
(2) From the qualitative study, several lived measures and strategies emerge, including acceptance and optimization, rejection and avoidance, selection strategies.
(2) No expectation to read e-mails outside of work hours and accept different handling of the frequency of reading e-mails (against compulsive checking).
(4) Only forward of CC someone when necessary.
(5) Avoid "buck-passing": do not pass the task to someone else by forwarding an e-mail.
(6) Avoid "broadcasting": send to a large distribution list only when necessary; otherwise, send individual requests.
(7) BCC ("blind copying") should remain allowed, as there are usually good reasons for doing so.
(8) Send one well-thought-out e-mail, not multiple or periodic follow-ups.

EM
Soucek ( (2) Concrete tips to avoid information overload concerning e-mails, including precise formulation of concerns based on the subject line, communicating the further need for action, standard templates, avoiding CC and forwarding, and keeping the amount of information as small as possible.
(3) Concrete tips against information overload regarding the Internet, including precise keywords, checking for up-todateness, filing system, and direct communication to (1) On the recipient's side: automatic filtering of incoming mail, individual strategies, e.g., cross-reading (2) On the sender's side: question the necessity of the e-mail, send only to relevant recipients, and choose precise wording.

(Continued)
Frontiers in Psychology 12 frontiersin.org burnout, while the latter mediates the relationship between technocomplexity and burnout. Therefore, techno-insecurity and perceived techno-complexity should also be addressed in training measures. The study by Le Roux and Botha (2021) provides a complementary reference to these findings, as their work showed that perceived techno-complexity and techno-uncertainty increase with age. Consequently, training in this area should be provided to older employees in particular. Zhao et al. (2020) suggested further avenues for interventions to address information overload. First, problem-focused coping (e.g., performing a cognitive assessment of technostress and seeking instrumental support) is helpful in alleviating information overload. Second, training in emotion-focused coping or venting is also recommended. Based on this, Becker et al. (2021) found in a quantitative study with 3,363 knowledge workers from Germany that coping strategies reduced the negative indirect relationship between technostress and performance mediated by exhaustion. Therefore, the authors recommend that different coping strategies should be trained. Although both dysfunctional (avoidance of the problem) and active-functional (including problem-focused and emotion-focused coping) coping strategies reduce the strain caused by techno-stressors, the negative long-term consequences of dysfunctional coping should be explained to individuals, and in particular, active-functional strategies should be part of further training measures.
A cross-sectional study by Yin et al. (2018) focused on the timeliness of information processing, which refers to the perceived support for timely information provided by mobile ICT. Their study found that the timeliness of information processing reinforces the negative relationship between information overload and job satisfaction and is therefore not recommended as a coping mechanism.
Based on their qualitative and quantitative study results, Moser et al. (2002) developed an exemplary training concept to be applied at the personal level, including the following aspects: know-how, efficient handling of information overload, time management methods, and training on e-mail program. The training was highly accepted by the participants, led to an increase in knowledge, and reduced the participants' feeling of information overload.
In a quantitative intervention study, Soucek and Moser (2010) examined a training intervention to explain the use of modern ICT, and their results showed that e-mail communication could be improved by training in media literacy, individual work processes, and knowledge of the principles of e-mail communication. According to Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA) (2006) and Stich et al. (2018), training-based interventions in relation to e-mail use should also include "netiquette, " (i.e., topics such as virtual politeness, respect, and commitment). With regard to training on e-mail use, Antoni and Ellwart (2017) further noted that individuals should be made aware of the negative effects of e-mail interruptions and should be presented with concrete techniques for managing e-mails efficiently. For example, setting fixed times for processing e-mails can be helpful.
A concrete example of an intervention that has already been developed and evaluated is the training program of the initiative "New Quality of Work, " which includes modules on the organizing of incoming mail, processing and managing e-mails, and designing correspondence (Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), 2006). According to Moser and Soucek (2005), a needs analysis is required before a training intervention should be offered in a company. Ultimately, a shared e-mail culture can be developed through joint participation in the program.
De Bruin et al. (2020) showed a positive effect of mindfulness training with regard to stress (measured by the Perceived Stress Scale). However, a positive effect of mindfulness training specifically on information overload cannot be inferred from these findings. According to an interview study by Camargo (2008), other creative solutions, such as learning videos or presentations, can also contribute to the further training of employees in relation to managing information overload.

Dealing with E-mails
According to Drössler et al. (2018), managing e-mails poses a challenge for employees in terms of information overload. Indeed, both individuals and organizations should take responsibility for implementing measures to address this issue, which can be done by creating a detailed information culture in the company and clear internal company regulations, such as communication and e-mail guidelines (Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), 2006; Ramsay and Renaud, 2012;Drössler et al., 2018;Stich et al., 2018;Stadin et al., 2020). Soucek (2009) also recommended structuring and managing the inbox, scheduling e-mail communication, and using e-mail correspondence in a manner appropriate to the medium.
Clear articulation of concerns related to task objectives is considered important in e-mail communication. Furthermore, in order to avoid unnecessary communication, it should be made clear whether further action is required as a result of the e-mail sent and, if so, what form this action is expected to take (e.g., how quickly a response is expected; Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), 2011; Ramsay and Renaud, 2012). In EM/L Sources in italics were mentioned in multiple of the running text's sub-chapters at the individual level. However, these are only listed once in the tabular overview. Quant, quantitative study; Qual, qualitative study; review, review paper (systematic or unsystematic); Meta, meta-analysis; Concept, concept paper or opinion; Hand, handout, rule or information; EM, employees; L, leaders; T, teams; O, organization; EX, external support (e.g., IT staff or employees of a consultancy). Psychology  13 frontiersin.org addition, the necessity of the information should be questioned before sending an e-mail communication, and e-mails should only be sent to relevant recipients (Rack et al., 2011). Standard templates can be used to reduce the effort required to formulate e-mails (Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), 2011). Furthermore, the group of people to whom the information is disseminated to should be limited by avoiding the use of carbon copies (CC) and forwarding e-mails (Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), 2011; Ramsay and Renaud, 2012;Drössler et al., 2018), the amount of information disseminated by the e-mail should be kept as small as possible, and large attachments should be avoided (Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), 2011). So-called "buckpassing, " (i.e., passing on a task to another person by forwarding an e-mail, should be avoided (Ramsay and Renaud, 2012), but blind copying (BCC) should be retained if deemed appropriate, as there are usually good reasons for using this functionality. However, duplicate communication through multiple channels should be avoided according to Drössler et al. (2018). Individuals should refrain from regular follow-up e-mails (Ramsay and Renaud, 2012) and instead consider other methods of communication, such as face-to-face contact instead of e-mail (Camargo, 2008; Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), 2011; Drössler et al., 2018). In addition to individual strategies, such as skimming (reading quickly to get a general overview), automated filters in the inbox can also be helpful for recipients of large amounts of e-mail information (Rack et al., 2011). Indeed, developing rules and filters and integrating them into one's e-mail use can help separate important e-mails from irrelevant information, such as spam (Camargo, 2008). Ramsay and Renaud (2012) recommend ignoring or deleting potentially risky e-mails as a measure against information overload. This approach is particularly useful when virus-laden content is suspected.

Frontiers in
To counteract the accumulation of information, subscriptions to automated services such as newsletters or newsfeeds should be made selectively (Drössler et al., 2018). Similarly, individuals should also inform their colleagues directly if they do not need information sent via their e-mail distribution lists (Camargo, 2008).
To protect individuals from compulsive e-mail checking, Ramsay and Renaud (2012) recommended two strategies. First, colleagues should accept that individuals have different ways of managing e-mails in terms of how often they read them. Second, individuals should free themselves from the expectation that e-mails should be answered outside of work hours, thus allowing them to make a clear distinction between work and private life (cf. Camargo, 2008).

Dealing with the internet
A report by the Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA) (2011) provides concrete recommendations on how to manage the Internet in general. In particular, the use of meaningful and concise search terms is recommended to significantly reduce the number of search results. In addition, the topics of documents and materials from the internet should be checked immediately to eliminate information that is not relevant. The basic principle of these recommendations is to identify only as many documents on a topic as can be processed in a reasonable amount of time. According to the authors, a clear filing system can help navigate through the information. Furthermore, when working with the Internet, direct communication with colleagues about the perceived importance and relevance of various documents can also reduce one's own information overload (Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), 2011). Finally, a systematic review by Drössler et al. (2018) suggested that active information searches should be used instead of push notifications. 4.2.5. Active design of the workplace Pfaffinger et al. (2020) report that active workplace design and boundary management can be helpful against information overload. In fact, an interview study by Camargo (2008) with 17 employees in the "high technology" sector also found that separating work and private life can reduce the stress caused by e-mails.
Soucek and Moser (2010) recommended self-management techniques, such as prioritizing and sequencing tasks according to one's needs, to improve the individual workflows. For example, managing e-mails should be integrated into an individual's daily schedule at a time that makes sense for them.
Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), (2011) recommends actively creating time for breaks and reducing interruptions during these times, which can be achieved, for example, by switching off the incoming notifications for e-mails (cf. Drössler et al., 2018). Finally, to reduce technostress, no e-mails should be processed or sent outside of working hours (Gaudioso et al., 2017).

Summary: person level
At the individual level, several studies have recommended that individuals participate in training interventions that, in addition to supporting the development of general competencies such as self and time management, should improve their competencies in managing ICT and the software used in the company. However, there is a lack of methodologically sound empirical studies that test the effectiveness of the proposed intervention approaches. In fact, we identified only two evaluated interventions that reduced subjective information overload among participants, but these studies did not include any objective measures. In terms of avoiding information overload, e-mail management seems to be particularly challenging. In addition to participating in training, studies have suggested concrete actions that individuals can take to counteract information overload. An example is to organize one's workplace in a way that supports the setting and maintaining of boundaries between work and private life and creates free spaces and breaks. It should be noted that while the advice for individuals is presented at the individual level, the support of the company in the implementation of these strategies is elementary. For example, the company's regulations, the corporate culture, or human resource development policies may be important in helping employees manage information overload. Figure 2 provides an overview of the suggested content for training.

Level of tasks and processes
At the task and process level, it may be useful to standardize certain procedures to address information overload (Eppler and Mengis, 2004). Piecha and Hacker (2020) and Piecha (2021) presented a detailed report on a multi-year multi-method research project on information overload. The aim of the report was to identify approaches to managing a large amount of information transmitted through ICT at work in a non-overwhelming way. From their systematic field studies, the authors derived several recommendations. These include interviewed 20 health care managers in Sweden using the critical incident method. The negative aspects of digital communication mentioned were high workload, the invasion of privacy, and negative feelings (e.g., fear of missing something in the e-mail inbox). Promoting digital literacy, designing ICT according to needs, and redistributing work and ICT systems were mentioned as possible solutions to the negative aspects of digital communication.
Furthermore, e-mail culture, support from colleagues, individual resources (individual strategies, such as routines and structures), individual competencies (e.g., learning by doing or preparation), and organizational resources (e.g., IT support) were mentioned as important aspects in dealing with information overload. Stich et al. (2018) highlighted the negative consequences of excessive use of digital information technology and ICT: technology overload, interruptions, multitasking, work-home interference, and cyber deviance. The following process recommendations were made to address these issues: (1) establish clear policies on ICT use (e.g., blocking e-mail traffic outside working hours, switching from e-mail to an internal social network, establishing policies against cyberbullying or regarding ICT use in general); (2) implement individual interventions; (3) create individual mindfulness (e.g., measure and provide feedback on the level of technostress or attitude toward ICT through surveys); (4) create mindfulness toward colleagues (particular responsibility of managers, but colleagues should also be asked which communication channels they prefer).
Bordi et al. (2017) introduced the term information ergonomics, focusing on the workload induced by information-intensive tasks. In their study, discussions in several focus groups with a total of 36 employees (including insurance and financial service providers) were qualitatively evaluated. The following aspects were identified in the discussions: the establishment of common rules (etiquette), quiet times, training in the use of tools, and prioritization.
In their editorial, Kluge et al. (2020) discussed the filtering of information as a measure to address information overload. They also emphasized that tasks can be delegated from humans to digital agents and that management information systems and interactive assistance systems can be used to reduce information overload. Effective measures also include "organizational unlearning, " which refers to forgetting information that is no longer relevant. Okolo et al. (2018) conducted a cross-sectional study with 319 bank employees in Nigeria to examine the indirect relationship between job-related resources according to the Job Characteristics Model (i.e., task feedback, task autonomy, task significance, task identity, and skill variety) and work engagement mediated by the experience of technostress. Technostressors (overload, complexity, and invasion) could be categorized as challenging demands, defined as a demand that shows motivational effects in addition to being straining. Activity resources showed a positive indirect relationship with work engagement mediated by technostress. Concrete recommendations for action cannot be derived from this study, but the indication that experienced technostress may have the potential to act as a challenge suggests that motivational gains may be derived from technostress under certain conditions.
Oldroyd and Morris (2012) focused on a specific group of employees, the so-called high performers or stars, whose productivity can be up to 10 times higher than the average in an occupational field. Such stars tend to be very well connected socially, which means that the amount of communication and, thus, the information load they experience is very high. To maintain the productivity of such stars, the authors suggest, among other things, (1) training in the ability to distinguish between useful and superfluous information, (2) the (2) Individual interventions (e.g., e-mail training, workshops on virtual courtesy, respect, and engagement).
(3) Creating mindfulness towards oneself (e.g., measuring and providing feedback on levels of technostress or attitudes towards ICTs through questionnaires).
(4) Create mindfulness towards colleagues (special responsibility of managers, but also ask colleagues which communication channels they prefer). (1) Person: Training in the ability to distinguish between useful and superfluous information; creating schemas to support long-term memory; motivational support; scope for action.
(2) Tasks & processes: support employees for star employees (gatekeepers); reduce the width of the star network; increase the density of the star network; distribute standardized information to the entire network instead of individualized information.
(3) Organizational processes: Clarity about where to find information, cost of sharing information; using information systems to contextualize information and make it more useful to formation of schemas to support long-term memory, (3) decision latitude, (4) supportive colleagues, (5) reducing the breadth of the social network while increasing the density, and (6) distributing standardized information to the entire network instead of individualized information. The authors also made other recommendations at the organizational level: (1) clarify where information can be found, (2) increase the cost of sharing information, (3) use information systems to contextualize information and make it more useful to employees; (4) create special knowledge-sharing positions for star employees, and (5) encourage relationships with other star employees. Some of these recommendations do not appear to be limited to star employees. (1) individual attentiveness to the situation, (2) attentiveness to the situation in the team and (3) plan formulation. L Sources in italics were mentioned in multiple of the running text's sub-chapters at the task and process level. However, these are only listed once in the tabular overview. Quant, quantitative study, Qual, qualitative study; review, review paper (systematic or unsystematic); Meta, meta-analysis; Concept, concept paper or opinion; Hand, handout, rule or information; EM, employees; L, leaders; T, teams; O, organization; EX, external support (e.g., IT staff or employees of a consultancy).

Cooperation in teams
Ellwart et al., 2015 evaluated an online intervention for structuring teamwork (STROTA), which was shown to reduce information overload. In the STROTA intervention, team members are first encouraged to analyze the current situation with regard to the triggers and conditions of experienced information overload in a structured way. Then, a facilitated team discussion creates a shared mental model and situational awareness (team situation awareness), and the team develops concrete goals and plans (team adaptation).
Based on interviews with 10 emergency responders with leadership roles, Norri-Sederholm et al. (2015) also emphasized the importance of situational awareness, which depends on having relevant information available. Paul and Nazareth (2010) used a group support system to regulate the flow of information within a team, helping them to process information effectively despite high complexity. As an intervention, the groups were given access to aggregated information from the work of previous groups engaged in a similar decision situation. In terms of the results, the provision of a decision scheme cannot be unconditionally recommended, as the expected effect (a higher threshold for the occurrence of information overload) did not occur. Instead, the relationship between information complexity, time pressure, and time to decision changed fundamentally to a U-shaped relationship in the treatment group. Specifically, at high levels of information complexity, the decision time was longer when the decision scheme was available to the group. Therefore, the advantages of having a schematic summary of information are only relevant for information of medium complexity. Wang et al. (2007) investigated how an adaptive awareness system for shared virtual workspaces can be technically implemented and how its comprehensibility and usefulness can be evaluated by testers. The goal of this study was to present a system to reduce the information load of digital collaboration. However, this study provides only weak evidence that the automatic adjustment of notification rules can be useful in virtual collaboration, because information load and information overload were not considered as outcomes.
The results of a study by Bergström et al. (2010) suggest that applying the theoretical basis of the coordination of joint activities is a promising way to develop a contrasting perspective of the factors that allow teams to maintain control in escalating situations. In a simulation in this study, both lay and expert teams were confronted with an escalating emergency situation on a ship. A key finding of the work was that in such highly complex situations information must be prioritized and filtered to avoid information overload. Ferreira et al. (2011) also aimed to 'address the problem of information overload in synchronous group work' (p. 643). The authors experimentally examined team performance in a brainstorming task under the conditions "without attention management" (one team member's ideas were sent directly to all the group members) and "with attention management" (team information was sent only when the team member was not currently working individually). The use of attention management resulted in better team performance in terms of the number of ideas generated, although the generalizability to other types of team tasks is unclear. Stich et al. (2018) found that creating mindfulness toward colleagues can be helpful in reducing information overload, for example, inquiring about which communication channels colleagues prefer. The authors suggested that leaders should take responsibility for this collegial mindfulness. In meetings, it can be helpful to establish rules for the use of digital media, such as turning off notification signals during meetings (Drössler et al., 2018).

Leadership
Only a few papers have reported suggestions for leaders to reduce information overload. For example, Becker (2009) conducted interviews with 12 managers in the health care sector, and based on the results, the author suggested limiting organizational priorities, creating clarity about project responsibilities and decision-making, organizing fewer and better-structured meetings, and establishing guidelines for the use of technology. Spagnoli et al. (2020) reported that workaholism and authoritarian leadership were positively correlated with experienced technostress. Specifically, for employees who worked exclusively from home, authoritarian leadership strengthened the relationship between workaholism and technostress.
Based on her own experience as a trainer and coach, Mill (2010) presented ideas and strategies for managers to introduce training and other measures to reduce work stressors related to information overload and ICT use and to promote a positive work culture. In her view, these strategies should include effective communication, training for employee development, coaching, mentoring, and good leadership and inspiration, which can all positively influence the work culture and, thus, lead to improved health, and well-being.

Summary: level of tasks and processes
In general, measures for stress-optimized work design have been proposed for the area of tasks and processes. Information overload is often associated with other quantitative and qualitative stressors, such as time pressure, high workload, interruptions, or role ambiguity. Although the causal direction of these relationships is not clear, a reduction in the accompanying stressors seems to contribute to a reduction in perceived information overload or technostress. Similar to the personal level, many proposals at the task and process level include approaches to increase the competence of individuals in dealing with digital information technologies. Additional support services (IT support, supervision) are also seen as helpful. The establishment of common rules (etiquette in dealing with ICT) has been mentioned several times in the literature as a worthwhile approach to reduce information overload. At the team level, the establishment of situational awareness (attentiveness to the situation) and shared mental models have been suggested. In addition, assistance systems for filtering and prioritizing information have been mentioned as starting points for dealing with information overload. Recommendations regarding leadership remain relatively superficial; a supportive and constructive leadership style is considered favorable, whereas an authoritarian or destructive leadership style tends to be unfavorable with respect to technostress and information overload.

Level of the design of organizational processes
At the level of organizational process design, behavioral and action guidelines can be introduced to address information overload (Manwani et al., 2001). The strategic use of internal communication can reduce the risk of information overload (Remund and Aikat, 2012). However, corporate culture is also a relevant factor (Sumecki Frontiers in Psychology 18 frontiersin.org  (1) Coping strategies: eliminate maladaptive coping strategies and then build adaptive ones, increase awareness & reduce barriers (e.g., through training, job design, reward systems, peer pressure, and technical support).
(2) Reduce technostress (e.g., no e-mail outside work hours, improve organizational culture). O Quant, quantitative study; Qual, qualitative study; review, review paper (systematic or unsystematic); Meta, meta-analysis; Concept, concept paper or opinion; Hand, handout, rule or information; EM, employees; L, leaders; T, teams; O, organization; EX, external support (e.g., IT staff or employees of a consultancy). Psychology  19 frontiersin.org et al., 2011). Kluge and Gronau (2018) investigated the process of intentional forgetting to reduce information overload in change processes. Furthermore, Day et al. (2012) showed that the provision of up-to-date software and personal support from IT staff can mitigate the negative effects of information overload. Farhoomand and Drury (2002) interviewed 124 managers in different countries about their experiences and approaches to information overload. Among other things, the managers mentioned the following: 1) identifying the dimensions of information overload (extent, irrelevant information, time constraints, variety of information channels) to provide targeted support to affected employees; 2) in order to provide better tools and techniques to better process information from internal and external sources; and 3) ensuring flatter hierarchies in combination with intelligent IT within the organizational structure.

Frontiers in
In a cross-sectional questionnaire study, Florkowski (2019) examined the associations between organizational policies and technology-related stress experiences, uncertainty, and job satisfaction in a sample of managers from 169 human resources (HR) departments in the US. The managers highlighted the importance of expectation management in the use of technology in human resources (HR) and recommended the continuous adaptation of the organizational climate in terms of flexibility, risk-taking, evidence-based experimentation, and support for innovation. Gaudioso et al. (2017) tested a serial mediation model and showed the indirect effects of techno-invasion via work-family conflict and of techno-overload via distress and coping (adaptive vs. maladaptive) on exhaustion. However, the cross-sectional design is a major limitation of this study. From the reported findings, the authors derived their recommendations to eliminate maladaptive coping strategies and subsequently develop adaptive coping strategies. In addition, they reported that awareness of technology use should be increased and barriers to technology use should be removed (e.g., through training, work design, reward systems, peer pressure, or technical support). To reduce technostress, the authors recommended, among other things, not sending and processing e-mails outside working hours and improving the organizational culture.
Based on interviews with 26 employees from different countries, Pfaffinger et al. (2020) suggested the following to reduce information overload: invest in IT security, communicate, structure, and accompany the introduction of new technologies effectively, develop new work concepts, clarify expectations regarding flexibility of time and place, ensure compliance with work regulations, implement ergonomics in the home office, provide the latest technologies and corresponding technical support, provide a helpdesk, and training for managers, and provide training for individual employees. In addition, the study also addressed the societal level with the following recommendations: ensure participation, public training, or help desks, ensure high speed Internet for mobile phones, prohibit or regulate surveillance, protect data rights, enact penalties for violations, offer social services and security in case of job loss, ensure the humanity of new forms of work, create a legal framework for the separation of work and private life, and provide a functioning infrastructure for companies.
Soucek (2017) examined the extent to which e-mail communication agreements are perceived as useful in organizations. As a conclusion, Soucek (2017) explained that "such agreements on e-mail communication are mainly adopted company-wide and outweigh the number of intra-departmental agreements, with the intra-departmental agreements proving to be particularly helpful" (Soucek, 2017, p. 20). Stich et al. (2018) also recommended setting clear guidelines for the use of ICT. For example, they suggested that e-mail traffic should be blocked outside working hours, that switching from e-mail programs to an internal company social network may be helpful, and that policies against cyberbullying or more generally about the use of ICT should be established.

Summary: level of design of organizational processes
Recommendations at the level of organizational processes cannot always be clearly distinguished from those at the level of tasks and processes. Indeed, the (participatory) development of policies and the establishment of a corporate culture regarding the use of ICT are also addressed in the studies at this level. In addition, there are references at this level to technical assistance systems and to the promotion of competence in the use of digital technologies in the context of human resources development measures.

Information technology level
Media richness theory suggests that the choice of an appropriate technology is relevant for the transmission of information (Graf and Antoni, 2020). At the level of information technology, the implementation of filters appears in several papers. Other technologyrelated approaches can be grouped under the categories of algorithms that summarize or extract information and the automation of monitoring tasks. Monitoring tasks can be considered as a specific form of work task, which is usually characterized by a large amount of information. However, this section first presents some of the results that relate more generally to the use of ICT.

Use of information and communication technology
Based on a cross-sectional study by Ayyagari et al. (2011), it can be concluded that caution is required when introducing new ICT or changing existing ICT. Dynamic changes in ICT should be considered a source of stress. Drössler et al. (2018) warned that the introduction of a new technology should be thoroughly considered and, based on this, special care should be taken when choosing a form of ICT for the corresponding information transfer. Similarly, in line with media richness theory, Kauffeld et al. (2016) argued that the richness of the chosen medium should match the type of information to be transferred. For example, when establishing a relationship with a new colleague, face-to-face communication would be preferred, because it is the richest form of communication, whereas the coordination of an appointment could be clarified with the least rich form of information-an e-mail. Regarding the appropriate choice of ICT, Oehme et al. (2019) showed that the use of ICT in control rooms and in the work of field staff has a high potential to facilitate team processes and reduce errors. In addition, previous literature has recommended the use of filtering capabilities. Warren (2014) made other recommendations, such as the use of unified file systems, tagging, context management, and the use of semantic technology to combat information overload, facilitate context switching, and help employees integrate information.
Frontiers in Psychology 20 frontiersin.org  (2) Use of tagging in addition to traditional hierarchical folder structures.

Use of information and communication technology
(3) Control of a user's contextual information to support information retrieval and reduce overload when switching between contexts.
The system was positively evaluated by those who tested it. O/T

Medicine -
The authors derive ten suggestions for the design of data systems in medicine from their literature review: (1) Automatic integration of data (2) Collection and integration of the newly available and historical data  Tagging (tags are used to add information to content; tagging is primarily used to make information easier to find or link) was recommended by Jackson and Smith (2012). In their qualitative case study, the authors showed that traditional hierarchical file systems can lead to the retrieval of irrelevant or no information, even though the relevant information is present. The study provides an initial indication that tagging could provide an effective solution to this problem and reduce the retrieval of duplicate or irrelevant information. Demirsoy and Petersen (2018) proposed an example of semantic technology and evaluated this using an interview study. The proposed semantic knowledge management system enables automated text processing. Specifically, a basic ontology has to be created, which is then filled by automated semantic annotation in the next step. In a final step, the users are provided with a convenient and efficient keyword search (Demirsoy and Petersen, 2018) to prevent information overload caused by a large amount of information. Although the system was positively evaluated by the users in this study, information overload was not considered as an outcome.
A nonsystematic review by Celi et al. (2014) provides a variety of recommendations for the design of data systems in medical practice to prevent information overload. The authors provided ten suggestions for the design and implementation of a medical data system: 1) automatic integration of the data, 2) collection and integration of newly available and historical data, 3) automatic suggestion of diagnostic, therapeutic, prognostic, and other documentation suggestions, 4) the use of machine learning, 5) offering individual customization options, 6) ensuring aggregation of the data for research, 7) considering reports with "best practice" recommendations, 8) ensuring changeability of the system, 9) changeability also regarding to the integration of new data, and 10) prototype testing. Martignene et al. (2020) proposed an R-based tool that extracts the relevant information from patient data and presents it visually. The program is called Heimdall and was initially evaluated only in terms of its functionality. The field application of this tool is still pending. According to Fellmann et al. (2019), the implementation of stresssensitive IT systems would be useful in principle, but urgent attention must be paid to the needs of the users. For example, in their study, the users clearly preferred a two-step process, in which, initially, they were first informed about their workload and stress, while recommendations for countermeasures were presented only upon request.

Filter and decision support systems
In a vignette study, Ulfert et al. (2022) examined decision support systems with varying degrees of autonomy. The effects of the experimental conditions on participants' information overload and their intention to use the system were considered as outcomes. The study found that increasing the degree of autonomy of the decision support system (under certain conditions) led to a reduction in information load. However, respondents still reported high levels of technostress, and increasing automation had a negative effect on their intention to use the system.
A similar decision support system was proposed by Garnsworthy et al. (2004) as a prototype for the work of metro surveillance staff. In addition, according to Mack et al. (2009), decision support systems can improve clinical work in pediatric ICUs. However, the choice of system should be made with careful consideration of the potential negative effects of introducing a new technological system. Sappelli et al. (2016) proposed an algorithm that goes beyond the filtering and categorization of traditional e-mail program systems. In fact, this algorithm can extract the action performed by the sender (e.g., reply, deliver, greet, etc.), the required response (e.g., no reply required, immediate reply, etc.), the sender's implicit reason for the e-mail (e.g., internal collaboration, travel planning, etc.), and the number of tasks contained in the e-mail. In addition, Zhang and Xiao (2018) proposed another technology developed to increase the speed of text comprehension.

Automating monitoring tasks
One area where employees are exposed to too much information at once is camera surveillance. Therefore, Tung et al. (2011) developed an algorithm that detects unusual movement patterns from surveillance camera images and alerts employees of conspicuous or suspicious cases. In their study the new algorithm was tested only on sample data, without measuring any attitudinal or behavioral data. The NoDoze program (Hassan et al., 2019) can be applied to IT security. The algorithm ranks incoming security alerts according to their urgency, with potential false alerts being ranked very low. This algorithm is intended to counteract "threat alert fatigue" among IT staff. Another algorithm based on traffic monitoring has been proposed with BeAware (Baumgartner et al., 2014). As these three studies only focused on the functionality and accuracy of the algorithms, no conclusion can be drawn as to whether these algorithms reduce the information overload of (IT) security staff. McFarlane et al. (2017) tested the transfer of a military surveillance system (HAIL) to a clinical setting. Using the system, nurses responded three times faster to clinically relevant alarms, with no negative impact on their other tasks. On a subjective level, the test users indicated that they found the system useful and comfortable. Therefore, the study provides initial evidence that the use of the HAIL system in a clinical context could be a starting point for reducing the amount of information presented to clinical staff. With a similar aim, Connell et al. (2019) also tested the use of automated alarms in medical care and showed that patient care improved due to the alarms. However, effects on information overload were not reported. For the implementation of automated alarms, the authors recommended that users should be trained to ensure that the workload and the handling of the information received is well organized. Furthermore, such systems should be continuously optimized to reduce false alarms (Connell et al., 2019).

Summary: use of information and communication technology
A wide variety of technological options have been proposed in the literature to reduce the risk of information overload. Overall, user needs must be considered when introducing new technologies to minimize the likelihood of increasing technostress. Specific suggestions for reducing information overload when using ICT include the use of tagging to complement traditional folder structures, filters and decision support systems, and algorithms for automatically summarizing text to reduce the amount of information presented. In evaluating these recommendations, it is important to keep in mind that many of the technological options have not been systematically evaluated in human use. However, studies that have evaluated these interventions suggest that care must be taken to ensure that the technology does not unduly restrict the user's autonomy.

Discussion
Most empirical research in the area of information overload has focused on examining predictors and consequences or providing descriptive data on the prevalence of information overload. Some of this research can be used indirectly to derive recommendations for workplace interventions. However, the focus of this report is on concrete measures for managing large amounts of information in the work context. The following sections summarize the collected recommendations, and Figure 3 provides an overview of the key findings.

Summary of the results across the five levels
At the level of information design, several studies were found, which can be roughly divided into those focusing on the quality of information and those focusing on the quantity of information. Regarding the quality of information presentation, most studies have focused on the medical field. Specifically, visualization dashboards have been recommended to reduce information overload (cf. Khairat et al., 2018). A dashboard (or cockpit) presents essential information for a defined subject area in an intuitive and understandable way. Dashboards should be designed so that users can get an overview of the information and can derive clear recommendations from this. Empirical evidence suggests that dashboards in a medical context have a positive impact on situational awareness and cognitive load. However, there is a lack of evidence that dashboards reduce information overload. In addition, other studies highlight the benefits of well-designed electronic patient records (e.g., Ahmed et al., 2011;Tan et al., 2013).
When implementing such systems in the medical field (as well as in other application areas), attention should be paid to differences in Frontiers in Psychology 24 frontiersin.org information processing according to work experience (Wu et al., 2016;Ries and Deml, 2019) as well as to the information needs of different task areas (Wnuk et al., 2016). Overall, there are few studies on interventions to prevent information overload at the level of information design. The overarching recommendations for information design can be found in guidelines for display design or more generally in the field of software ergonomics. A relatively large number of measures to reduce information overload relate to the personal level. These measures include recommendations on how to manage e-mails or the Internet, specific training as a personnel development measure, and coping strategies to mitigate the negative effects of experienced information overload. Training approaches may include the use of software solutions and the development of more general competencies, such as digital literacy, information literacy, media and information competence, and self and time management (Drössler et al., 2018). In general, the recommendations at the individual level overlap with literacy as a technostress inhibitor (Jena, 2015). Descriptions of concrete training concepts can be found in Moser et al. (2002) and in a program of the "New Quality of Work" initiative (Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA), 2006). However, there is a lack of methodologically sound empirical studies to evaluate the effectiveness of such training approaches. Recommendations are largely derived from qualitative interview studies, and evaluation studies to date have not assessed actual reductions in information overload. Regarding the use of e-mail and the Internet, recommendations include negotiating and formulating clear organizational rules. In addition, there are several recommendations for the individual handling of e-mails. Other approaches at the personal level include the development of an active boundary between work and leisure time and the improvement of time-and self-management. At the task and process level, Piecha and Hacker (2020) developed numerous recommendations based on a multi-method research project. In the area of task and process design, however, the empirical effectiveness of the recommendations for action presented has not been comprehensively evaluated. An exception to this is the evaluation of an online intervention for structuring teamwork (STROTA; , which reported a reduction in information overload in the intervention group compared to the control group. In general, promoting situational awareness seems to be a recommendable mechanism for managing complex information. Especially in complex decision-making situations, filtering and prioritizing information may be relevant (Bergström et al., 2010). Constructive and supportive leadership behavior is generally discussed as a resource for coping with work-related stress. However, we could not identify concrete intervention approaches and training programs for managers.
Approaches at the organizational level are probably the least separable from the other levels of intervention, since the introduction of new software solutions, the development of human resources, and the design of tasks and processes should also be embedded in the overall strategy of an organization. Therefore, some of the recommendations collected at this level refer to training approaches (e.g., Moser et al., 2002). In addition, the provision of adequate technical solutions, company agreements for the management of digital communication, IT security, and technical support are also addressed in the literature. Together with the techno-inhibitors of participation and support for innovation (Jena, 2015), these factors provide Summary of the key results of the systematic review.
Frontiers in Psychology 25 frontiersin.org some guidance on how to reduce the potentially harmful effects of information overload and technostress. On the level of information technology, general recommendations can be found for the selection of suitable communication media depending on the task (e.g., Kauffeld et al., 2016). Furthermore, algorithm-based approaches for filtering as well as extracting and summarizing information are presented in the literature. A unique intervention method is the automation of monitoring tasks. Overall, many studies in this area present basic recommendations as well as algorithmic concepts. At this level, field studies that quantify the effects of interventions on information overload using an experimental control group design are still largely lacking.
The implementation of structural preventive measures represents a process of change within a company, and this implementation requires good planning and, if possible, participatory design and implementation (Rigotti et al., 2014). Ideally, according to Eppler and Mengis (2004), all levels of interventions mentioned in this review should be considered together when planning measures. Indeed, the redesign of the organizational information flow, possibly accompanied by technical assistance systems (e.g., software solutions), also requires the development of personal competencies. In our view, the recommendations for action presented here to reduce perceived information overload, which have been separated according to intervention levels, should not be considered in isolation but should be considered together and coordinated when planning and implementing appropriate measures.
Particularly in the area of information design and the development of information technology approaches, it is evident that the interventions to prevent information overload must be adapted to the concrete work requirements of the respective sector or occupational field. Therefore, no solution is equally suitable for all areas of application without adaptation, and the development of measures should be preceded by an analysis of the current focal points of stress in that area.
The contribution of this literature review is threefold. First, previous general review articles on information overload focused on psychological factors that influence the perception of information overload (Eppler and Mengis, 2004;Antoni and Ellwart, 2017). However, Antoni and Ellwart (2017) did not include technical measures to reduce information overload at all, while the article by Eppler and Mengis (2004) includes these aspects but is almost two decades old, highlighting the importance of an update. Second, a more recent meta-analysis shows the current state of research but is very narrow in its scope (Graf and Antoni, 2020). It focuses on the characteristics of information. We go beyond this meta-analysis by covering the five levels of causes of information overload. As our review shows, the information itself cannot be considered in isolation, since it is received and processed in the social context of tasks, team processes, and the organizational rules. Third, other review articles come from specific occupational fields and therefore only present countermeasures against information overload that are specific to the tasks of this occupation (e.g., Khairat et al., 2018). Although these countermeasures are partially generalizable to other occupations, these reviews omit important findings from other fields.

Strengths and limitations
The temporal limitation of this review to studies from the years 2000 to 2021 is justified by the rapid changes in digital information technologies, as the applicability of older studies to current work processes would be limited. Despite a systematic literature search following the recommendations of the PRISMA standards (Moher et al., 2009), it cannot be ruled out that relevant publications were not identified. In particular, general design knowledge, for example, from the field of software ergonomics, is not a comprehensive part of the present report due to the search strategy applied. Similarly, it may be possible to derive further design recommendations from a systematic review of studies that examined cause-effect relationships (predictors, consequences, mediators, and moderators) for the phenomenon of information overload. The findings summarized in this report should be understood as a systematic presentation of the applied combination and linkage of key terms. By using five of the most important databases for scientific publications (Web of Science, Ebscohost, Medline, PsycInfo, and PsycArticles), good coverage of the relevant literature can be assumed. We supplemented the search in the scientific databases by searches in PSYNDEX Interventions, Rehadat, and Arbeitssicherheit.de and a review of the publications of the BAuA (Federal Institute for Occupational Safety and Health) and the VBG.
All identified publications were assessed in a blinded manner by two independent raters against the defined inclusion and exclusion criteria. Conflicting assessments of specific studies were discussed by the project team. Due to the large heterogeneity of the publications and study formats, a formal assessment of the quality of the evidence base was not performed. However, references to the validity and generalizability of individual findings and recommendations for action were included in the text. Restricting this review to randomized controlled rials in the field of intervention evaluation, as suggested for the COCHRANE reviews, would not have been feasible. Overall, after reviewing and synthesizing the available studies, there is still a considerable lack of robust empirical evidence on the effectiveness of specific interventions to address information overload. Many studies are based on intuitive experience, do not allow for causal inference, or have other methodological limitations.
As a structuring framework, this review used the classification according to Eppler and Mengis (2004), which distinguishes between aspects of information, the person, tasks and processes, organizational processes, and information and communication technology. However, it was not always possible to assign interventions clearly to one of these levels. Here, we have decided to make an assignment in each intervention case, but in the sense of a socio-technical approach, an integration of measures from the different levels is considered necessary.

Conclusion
The studies included in this review cover a wide range of possible approaches to preventing or improving information overload. In terms of concrete information is concerned, it is important to clarify what information is actually relevant and to present this information in a clear and adaptive manner. The studies in this review present a variety of methods for managing large amounts of information for individual employees, some of which have been translated into concrete interventions. Clarifying and structuring team collaboration can also prevent information overload, and managers have a special role to play in this context. Decision-makers at the organizational level are responsible for selecting appropriate software, as well as for transparency and internal Frontiers in Psychology 26 frontiersin.org company rules that clarify information management. Technological support can help to reduce the amount of information present by using filtering systems or tools to extract the relevant information. It is striking that the intervention studies in the literature evaluating specific tools or training programs have rarely used a true control group design. Furthermore, it should be noted that information overload has rarely been considered as an outcome. This is particular true at the level of information and information technology. In many studies, the performance or the subjective satisfaction with a tool was often measured instead. A relatively high number of studies used qualitative methods to investigate the need for intervention or existing strategies to deal with information overload. Regarding the use of interviews with employees and asking questions about the strategies they use, it should be noted that this methodology helps to identify what individuals are already doing; therefore, it is often unclear how successful these strategies are and whether they are appropriate starting points for interventions.
Although the individual level has been the most studied in terms of the number of studies, the number of findings related to systemic prevention of information overload has been significantly higher than for behavioral prevention approaches. Thus, the literature of the past 20 years on information overload has already considered the effects of external factors on information overload, such as task characteristics, processes, and technical support, in addition to training and coping measures at the individual level (Meyer et al., 2021).

Author contributions
All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

Funding
This review is an edited and shortened version of an unpublished report delivered to the Bundesanstalt für Arbeitsschutz und Arbeitsmedizin [Federal Institute for Occupational Safety and Health], which provided funding.