An Investigation of the Relationship between Videogames and Occupational Health in the Age of Gaming Disorder

Työn aihe ja tarkoitus
Valtaosa suomalaisista pelaa säännöllisesti erilaisia videopelejä, kuten konsoli-, mobiili- ja pöytäkonepelejä. Pelaamisen, terveyden ja työelämän väliset jännitteet ovat nousseet pinnalle etenkin Maailman terveysjärjestön (WHO) hiljattain käyttöön ottaman diagnoosin myötä: ongelmallisesta videopelaamisesta on voinut saada mielenterveydenhäiriön diagnoosin vuodesta 2022 alkaen. Hankkeen tavoitteena oli ymmärtää, miten näitä pelaamiseen liittyviä terveysongelmia voidaan tunnistaa ja mikä on näiden vaikutus työelämään.
Aineisto ja metodi
Hankkeessa kerättiin yli 16-vuotiailta vastaajilta kyselyaineisto (n=8217), joka edustaa Suomea iän, sukupuolen ja asuinpaikan mukaan. Tutkittavaksi erotettiin edustava työväestö (n=3921), mikä sisältää työssäkäyvien lisäksi työttömiä ja lomautettuja. Tuhat vastaajaa täytti kyselyn sekä 2021 että 2022. Lisäksi haastattelimme kuusi videopelaamisen vuoksi ammattilaisen tukea hakenutta työikäistä pelaajaa ja 10 yli neljä tuntia päivässä pelaavaa työntekijää. Yhdeksän heistä haastateltiin sekä 2021 että 2022. Lopuksi haastateltiin kuusi terveysalan ammattilaista ja 10 HR- tai esihenkilöedustajaa. Kaikki päätutkimukset julkaistiin rekisteröityinä tutkimusraportteina (RR), joiden tutkimuskysymykset, hypoteesit ja menetelmät vertaisarvioitiin etukäteen. Tutkimukset julkaistiin avoimesti, vertaisarviot ovat avoimia ja kaikki aineistot tallennettiin avoimeen jatkokäyttöön Tietoarkistoon. 
Keskeiset tulokset
Ongelmallisen videopelaamisen esiintyvyys Suomen työväestössä oli kyselymittarien mukaan 0.4–7.5%. Monet erot tunnistettujen ongelmaryhmien välillä kuitenkin viittaavat siihen, että nykyiset mittarit ja mallit eivät ole luotettavia. Etenkin kliinisiä validaatiotutkimuksia tarvitaan luotettavan epidemiologisen tiedon tuottamiseen tulevaisuudessa. Samana aikaan laadullinen analyysi todentaa, miten yksilötasolla ongelmat voivat olla suuria. Pelaamiseen liittyvät ongelmat näyttävät olevan osa laajoja ongelmien verkkoja, jotka rakentuvat nuorena ja voivat johtaa syrjäytymiseen.
Johtopäätökset
Tulosten valossa pelaamisella ei toistaiseksi ole selkeitä vaikutuksia työterveyteen populaatiotasolla. Yksilötasolla syrjäytyminen, terveysongelmat ja niihin liittyvä pelaaminen voi kuitenkin katkaista koulutus- ja urakehityksen, minkä vuoksi työnantajien tulisi tehdä yhteistyötä tukipalveluja tarjoavien tahojen kanssa näiden henkilöiden auttamiseksi työelämään. Yleisesti ottaen työntekijöiden pelaamiseen tulisi suhtautua tavanomaisena harrastuksena, joka parhaimmillaan tasapainottaa ja tukee yksilön elämänkokonaisuutta.

Videogames have been a popular pastime in Finland for decades. Likely related to the global success of the Finnish game industry-led by companies such as Supercell, Rovio, and Remedy-the public views and sentiment regarding gaming have arguably been highly positive, at least in comparison to countries like China and South Korea where top-down regulations have been implemented in the past. Thus, when the World Health Organization (WHO) recently made a decision to take "gaming disorder" into effective psychiatric use as an independent mental disorder under "addictive behaviours" for the International Classification of Diseases , this also sparked change in the Finnish discourse. As the average age of people who play videogames in Finland is now 40 years (Kinnunen et al. 2022), the relationship between gaming and occupational health has become a relevant research question.
Despite the long-term global growth of the gaming industry and the increasing age of actively playing populations, there has been little research on the link between videogames and work life in general. In Finland, perhaps the only in-depth previous investigation is a report funded by the Finnish Work Environment Fund (Vahlo et al. 2015), which found 20% of their surveyed employees across various fields playing videogames during work hours. Additionally, the report examined gaming applications in occupational wellbeing and found three ways in which they operate: a) Primary benefits are experienced when gaming is considered beneficial for autonomous reasons, e.g. when gaming serves as relaxing recovery from work.
b) Secondary benefits are experienced when gaming is instrumentally applied, e.g. as a special means to motivate learning or work outcomes.

INTRODUCTION
c) Tertiary benefits are experienced when gaming indirectly leads to desired goals, e.g. when playing with colleagues improves social relationships.
However, the results implied that it was very unusual for companies to engage with these benefits or actively pursue them (Vahlo et al. 2015). Notably, at the time, the above WHO decision had not been announced yet and the scientific discourse regarding the potential negative occupational effects was not strongly present in the hours of play or other elements of gaming might also contribute to occupational problems or health issues. In brief, the goal of this project was to investigate the latter, as follows.
RQ1: How to identify gaming-related health problems in the working adult population?
RQ2: How do gaming-related health problems affect work life, especially in Finland?
Follow-up research meta-question: What is an optimal line of action in the current situation affected by "gaming disorder" ?
Conceptual issues. Research on gaming-related health problems (GRHPs) has been carried out since the 1980s (e.g., Ross et al. 1982), and the literature over these past decades has used dozens of diverse terms to refer to these problems, typically without definitions. Currently, it is still a standard practice for studies to use the term "gaming disorder" but not apply the official criteria of the ICD-11, from which the term derives. Sometimes "gaming disorder" is mixed with "internet gaming disorder", the latter of which refers to the American Psychiatric Association's proposed classification in the Diagnostic and Statistical Manual of Mental Disorders  and has entirely different symptomatology. On top of this, "gaming disorder" also tends to label most of the recent meta-reviews and meta-analysis (e.g., Stevens et al. 2021), which draw data from scales that were developed before both the DSM-5 and the ICD-11, thus referring to numerous vague ad hoc concepts that researchers developed for their own study purposes in various different conditions.
As a consequence of the above, in many studies, one might do well to more correctly speak of GRHPs instead of gaming disorder unless the specific ICD-11 cri- teria are addressed and properly tested as a diagnostic construct. On the other hand, the diagnostic practice around gaming disorder is only beginning and there are few or no clinical experts with the knowledge and will to support proper assessment. As our results later show, the currently available screening measures are generally unreliable and lack various types of validity, which makes addressing gaming disorder scientifically a difficult task. In the present report, we will speak of both GRHPs and gaming disorder by the following premise: "GRHP" is used when the nature of problems is unclear, and "gaming disorder" is used when ICD-11-based measures have been applied or for other reasons there are grounds for diagnostic language. In the next section, we outline the project strategy.
Although GRPHs have been studied for more than 40 years, little progress has been made in the field. As it was already discussed in the 1980s, some players have health problems related to their gaming habits, but the nature of these problems and their link to specific design or other elements remains unclear (for a review of the current hypotheses, see Flayelle et al. 2023). This is partially due to the constantly evolving game industry that introduces new types of play and technology applications each decade, thus changing the landscape of human-computer interactions somewhat radically in frequent periods (aka design vestigiality, see Karhulahti & Grabarczyk 2021) . For instance, the initial concerns of overspending in arcade games (1980s) were soon replaced by the immersive qualities of the home computer (1990s), which again was overcome by the concerns of the internet in the early 2000s and, once more, online gaming more specifically in the 2010s. As we are now witnessing the most recent stage of evolution in the 2020s, the worries are moving increasingly toward more general gamification features in smartphones (used in various social media applications) as well as monetisation tactics that are arguably moving some sectors of game design toward the gambling industry (e.g., Xiao & Henderson 2021).
In this historical context, it is not surprising that reaching stable research findings is hard.
Another persistent problem concerns not only the research on gaming but the revolutionising meta-scientific practices more generally. In the past two decades, especially social and medical sciences have come face what is nowadays reflected on as the "replication crisis," referring to the generally low level of research quality and, specifically, the fact that most findings published even in the best journals cannot be replicated (e.g., Chambers 2017). Various reasons have been suggested as contributors for this lack of replicability, ranging from issues in measurement and

PROJECT STRATEGY: A REGISTERED REPORTS APPROACH
validity to errors in statistical inference and, most critically, a number of questionable research practices (e.g., John et al. 2012). In brief, the current academic system has drifted into a situation where research and researchers are assessed primarily by attention value and quantity: top journals want researchers to report "shocking" discoveries that find their way into newspaper headlines and increase journal readership (i.e., profits); meanwhile, researchers are desperately driven to publish such discoveries because the academic incentives reward this quantity via university and individual metrics (see Schönbrodt et al. 2023). As a consequence, various fields are now filled with "discoveries" that in closer investigation (or replication) do not hold after all, while at the same time, null results or other findings that do not imply radical change rarely even get published-either because researchers are not incentivised to write such findings to begin with, or because journals are not interested in publishing "boring" findings (e.g., Smaldino & McElreath 2016). Ultimately, consequences of these vicious cycles are many "file drawer" problems and publication biases that undermine the trustworthiness of modern science in general (see Munafò et al. 2017). As an antidote to this state of affairs, the scientific world has recently and increasingly started adopting various "open science" practices, which collectively increase the trustworthiness of research and improve both the transparency and quality of studies in many ways (e.g., Crüwell et al. 2019). Examples of these practices include data sharing, transparent reporting standards, study pre-registration, and methodological justifications. As the most radical addition to this list of open science practices, more than 350 scientific journals, including flagship journals like Nature, have recently implemented a completely new format of publication called "registered reports" (RRs; Figure 1) where all the above best-practice details are peer reviewed before the study is carried out, thus ensuring a high quality of research from the beginning and making sure that the "shock value" of any upcoming results will not affect the prospective publication of studies (e.g., Chambers & Tzavella 2022 The workflow of the project's four key studies is presented below ( Figure 2). • The first quantitative study was designed to yield answers to RQ1. The idea was to assess the overall prevalence of GRHPs in Finland based on four unique survey measures and test their associations with health. In sum, the goal was to investigate whether it would be possible to gather reliable evidence by using survey GRHP instruments, and if so, what would that evidence be.
• The second quantitative study was designed to yield answers to RQ2. The idea was to assess the overall prevalence of GRHPs in the Finnish working population based on four unique survey measures and test their associations with work ability and recovery (with time spent gaming). In sum, the goal was to investigate what GRHP survey instruments, if applied as such, tell about the link between GRHPs and work. Additionally, an exploratory component was designed to test the stability of GRHPs over a period of one year.
• The first qualitative study was designed to further inform RQ1 and RQ2.
The idea was to investigate the phenomenological differences between treatment-seeking and actively gaming individuals in the working population by means of in-depth interviews. Experts were additionally interviewed to better contextualise the analysed experiences.
• The second qualitative study was likewise designed to further inform RQ1 and RQ2. The idea was to follow both gaming groups by a 1-year period and specifically investigate the potential phenomenological changes that such a time frame could introduce to actively gaming people in the working population. An in-depth interview approach was applied again with systemic health measures to better understand the temporal associations between gaming, health, and work.
In addition to the above four key studies, the project involved a number of side components that contributed to the research meta-question and deepened or supported the key studies. Some of these side components were planned from the start (e.g., qualitative investigation with HR people regarding the role gaming in the workplace), whereas others were included due to unexpected opportunities that opened exceptionally promising research avenues (e.g., qualitative investigation of Restart program data consisting of working age treatment-seeking adults). The findings of side components are reported separately in their own section below.
As an appendix to this report, we attach a previously unpublished study that was carried out to better understand both gaming and gambling in relation to work ability, work engagement, and work recovery. The study was written in Finnish and submitted for review first to Lääkärilehti and then Duodecim. Both journals rejected the study outright without the opportunity to revise or resubmit, for which the results have not passed peer review nor published in a scientific journal. Therefore, the results are not discussed in this report. As the project time schedule did not allow us to reconstruct the study for a third Finnish journal, these events further support the core approach undertaken by this project, namely, to publish important work as registered reports, which can a) save the limited resources available from funding bodies, b) respect the volunteer labour exerted by journal peer reviewers, c) facilitate scientific knowledge distribution by allowing all meaningful results to be published, and d) improve the mental health and working conditions of all occupationally involved academic parties by eliminating bureaucratic loops generated by content-based (not data or method based) chains of journal resubmission. For more details and research on these meta-scientific topics, we guide the reader to the Peer Community in Registered Reports platform (https://rr.peercommunityin.org/) and the Center for Open Science (https://www.cos.io/initiatives/registered-reports). This study was designed to acquire statistical knowledge of GRHPs and the instruments that are currently being used for their measurement. We chose four popular measures that represent different ontological systems: gambling-based criteria in the DSM-IV, internet gaming disorder criteria in the DSM-5, gaming disorder criteria in the ICD-11, and self-assessed frequency of problems. The applied measurement scales are the following: Game Addiction Scale (GAS7) ), Internet Gaming Disorder Test (IGDT10) (Kiraly et al. 2017), Gaming Disorder Test (GDT) (Pontes et al. 2021), and Problem Gaming Test (THL1) (Salonen & Raisamo 2015).

Study 1: Identification and Measurement of GRHPs
We then registered three hypotheses with exact statistical procedures, as follows:

H1:
We expect the ICD-11 and DSM-5 based GRHP prevalence rates to be meaningfully lower than the DSM-IV and Self-assessment based prevalence rates. H2: We expect ICD-11, DSM-5, and DSM-IV based GRHPs to overlap. Those who meet the ICD-11 criteria also meet the DSM-5 criteria, and both above additionally meet the DSM-IV based criteria. H3: We expect those with DSM-IV and Self-assessment based GRHPs to have (equally) lower health in comparison to the general population.

QUANTITATIVE FINDINGS
The hypotheses were developed by utilising pilot data (N=1000), which provided us with good initial estimates of the operability of measures and their relationships. Collecting the pilot data also enabled us to carry out longitudinal quantitative analyses in a later part of the project (see Quantitative Study 2).

Data and methodology
According to our plan, the data were collected with Bilendi, i.e. respondents were invited remotely from the company's 2. We included two control questions in the survey and Bilendi removed those responses that failed both (i.e. not included in N = 8217). Additionally, participants who reported not having played videogames within the past six months did not fill out the GRHP instruments, and they were considered not having GRHPs. The order of the screening instruments was randomised.
To test H1, we planned to calculate the prevalence rates with 95% confidence intervals for each of the four instruments. We set the H0 interval (null corroboration region) to the lower bound of 95% CI in the smallest obtained prevalence rate and the H1 interval (alternative acceptance region) twice above the upper bound of 95% CI in that prevalence rate. To test H2, we planned to assess group overlap against conditional probability P > 0.917. The null was tested against conditional probability P < 0.714. To test H3, we planned to compare the means of mental and physical health between the DSM-IV based group and the rest of the sample (one-tailed Welch t-test). This was repeated with self-assessment. Because mental and physical health were measured separately, we carried out the test twice with a corresponding alpha level 0.025 (multiple comparison correction). Equivalence testing was used in case of no evidence for effect.

Findings and discussion
For H1, we found all the measures producing significantly different prevalence rates.
This corroborated the first hypothesis (see Figure 3). For H2, which assessed the overlap of groups produced by the measures, the results were mixed ( Figure 3). The overlap between DSM-5 and ICD-11 based measures was poor, but that between DSM-IV and DSM-5 based measures was good. The overlap between ICD-11 and DSM-IV did not produce corroboration nor falsifying evidence. For H3, we found corroborating evidence for the measures to identify groups with lower mental health than the general population, but we did not find evidence for them to be different or similar in regards to health levels. Additionally, we carried out a vast number of exploratory analyses. A clear message from those analyses concerned physical health: we did not find any evidence of lower physical health in the GRHP groups in comparison to the general population, and in fact, those who self-assessed having GRHPs had a significantly higher physical health. Against previous beliefs, it appears that GRHPs are not related to low physical health, albeit confirmatory research is required to corroborate that hypothesis.
Finally, we also explored the differences between measures regarding their performance with our control questions. We found strong exploratory evidence for mischievous responding taking place among participants who report GRHP symptoms: the more problems participants reported, the more they also failed with the control question (Table 1). This finding, likewise, requires follow-up research to be confirmed, yet adds another grain of salt to the current epidemiological estimates produced by the presently available GRHP measures.
Taken together, the findings confirm two things. First, we found corroborating evidence for the measures to identify groups that have lower mental (but not physical) health than the general population. Second, we found corroborating evidence for the measures to identify groups of significantly different prevalence, and in the case of DSM-5/ICD-11 overlap, also completely different individuals. As a conclusion, there is a strong need for better and more consistent measures, which are able to identify individuals who may need support in their lives. Table 1.
Participants who failed one of the two control questions based on how much they reported having GRHPs. In this article, we zoomed into the Finnish working population in particular. The study is the first Finnish registered report in which the planning, hypotheses, and methods were also peer reviewed before the study was carried out. The idea was to explore both via confirmatory and exploratory approaches the connections between GRHPs and occupational health in working adults. The explicit research question for this study was as follows:

Study 2: GRHP Prevalence and link to work ability and recovery in Finland
How do gaming and GRHPs relate to work ability and work recovery?

Data and methodology
In this study, working ability was defined as the balance between personal (mental, physical, emotional) resources and the demands of work (Ilmarinen, 2009). Upholding a high level of working ability can increase well-being amongst the working age group, and employment in general. In terms of work recovery, existing studies (e.g. Reinecke

2009
) have reported the popularity of using gaming as a recovery method after work.
From the previously described dataset (Quantitative Study 1), we selected participants who belong to the working population, i.e. currently working, laid off, or unemployed (N = 3921), excluding retiree and student segments. For measuring working ability, the work ability index in short form was used (von Bonsdorff et al 2008). Work recovery is defined and measured based on the model developed originally in Finnish . Three hypotheses were registered:

H1:
We expected the work ability of people with GRPHs to be lower when compared to the general population. H2: We expected large amounts of play-hours (30+ hours a week) to be connected to low work ability. H3a: We expected the work recovery of people with GRPHs to be lower than in the general population. H3b: We expected large amounts of play-hours (30+ hours a week) not to be connected to low work recovery.
The time spent on gaming was measured with a question about the participants' estimated time spent gaming weekly, with an option to specify used platforms between a computer, game console, and smartphone.

Findings and discussion
To our knowledge, the present data collection was the most extensive Finnish survey related to gaming and work so far. According to Gaming Disorder Test, 0.4 % of the working population was found to meet gaming disorder criteria. Other measures provided higher estimates. The highest estimate was produced by Game Addiction Scale, which identified 7.5 % of the Finnish working population with game addiction (as an ad hoc construct).
Both H1 and H2, regarding the expectedly low work ability of individuals with GRPHs and large amounts of gaming hours, did not receive support. The work ability of the group identified by Gaming Disorder Test was not significantly lower than the general population, however, we did find statistically significant effects with other measures. The effect sizes ranged between g = .39-61, and while they did not exceed the preset clinical cutoff g = .67, some of the confidence intervals did. Notably, statistical significance was also dependent on whether ANOVA or Welch's ttest was used. H3a, in turn, expected a weaker work recovery for the GRHP groups but not for those with high gaming hours. Only the group recognized by Game Addiction Scale showed a statistically significant negative difference (g = .37), which was solitary and so small that the results supported a null hypothesis.
It should be kept in mind that statistical differences between groups cannot, however, identify the potential problems concerning individual participants. Effect heterogeneity, referring to individual variety beyond group differences (e.g. Johannes et al. 2021), should be taken into account in future study. Again, the large discrepancies between the different tools also deserve emphasis. It will be important to improve the validity of the currently available tools and ensure that they are measuring the intended phenomena.
In addition, the longitudinal data gave us an exploratory opportunity to assess the stability of GRPHs for 492 participants in a 9-month period during 2021 and 2022.
As Table 2 illustrates, temporal stability does not appear to be strong. This could be because of the possible cyclical nature of GRPHs, which can both manifest and vanish in intervals of months. Alternatively the variety of results could mirror inaccuracy with survey measurement or participant response. These issues should be addressed by designing studies with these issues in mind, combining a variety of tools and qualitative methods for comparing a variety of health effects and their stability. A critical area considering GRPHs has been the lack of clarity concerning the differences between them and healthy intensive play, the latter of which involves several gaming hours per day without related health problems. In this study, we approached the question by interpretive phenomenological analysis with interviews in two groups of highly involved videogame players: those who seek or have sought clinical help for their problems with gaming (n=6), and those who play esports more than 4 hours per day without self-reported related health problems (n=10). These data were contextualised with interviews of health experts (n=6) who have experience of working with the former group. The health experts' views were generally in line with the other data, for which we do not discuss health expert data separately in this report.

Data and methodology
The treatment-seeker data were generated with the help of our clinical collaborators, such as the Restart program and A-Clinic. The esports players were recruited with an open call in local online gaming communities and forums, starting from Kanaliiga that is an ongoing amateur esports league for working adults in Finland. The third

QUALITATIVE FINDINGS
group, clinical experts, were recruited from local hospitals, therapy centres, and other medical institutions, as well as through our collaborating institutions.

Findings and discussion
The analysis produced three subordinate themes for both gaming groups.

Escaping and solving difficulties in adolescence by gaming; current play experiences and their associations with self-image shadowed by memories
All individuals seeking treatment had experienced challenges during their childhood or teenage years, including bullying, discrimination, and family problems. These difficulties were explicitly mentioned by many of them as the reason for their excessive gaming when they were young. The relationship between these life problems and GRHPs was either direct or indirect. In some cases, gaming provided an immediate escape from these issues, while in others, it acted as a response to the anxiety, depression, and other related problems caused by these challenges. The participants did not follow consistent patterns in terms of how or what they played, or whether they played alone or with others. However, videogames, in general, provided a safe haven where they could experience feelings of autonomy, belongingness, comfort, and control, which served as a coping mechanism for their otherwise difficult life situations.

Struggling with limited means to regulate gaming in a paradoxical situation where it both satisfies basic needs and contradicts personally meaningful obligations
The individuals seeking treatment expressed that gaming fulfilled their psychological needs, and the effectiveness of gaming in meeting these needs made it challenging or even impossible for them to control their gaming habits. Throughout years or decades, they had acquired the skill of using gaming as a means to bring enjoyment and purpose into their lives. However, as they matured and their self-perception and values changed, it became difficult for them to replace gaming with alternative activities, despite their attempts to do so. To some extent, videogames had become a mediator in their interactions with the real world on a daily basis.

Feeling of guilt or regret for having wasted time on gaming over experiences that would have had "real value"
The treatment-seekers experienced feelings of guilt or regret due to the substantial amount of time they had spent on gaming, which they perceived as inferior to reallife achievements and "true values." This forms the third underlying theme among these individuals. In various ways, the participants had come to the realization that gaming was not what they truly desired to engage in to the extent they had been doing. Gaming was no longer a priority in their hierarchy of values. However, implementing practical changes in their daily behaviors, which had been established over years or decades, proved to be challenging for them due to various reasons that were partially unknown to us.

Gaming is a core component of identity and values
The early narratives of esports interviewees bear similarities to those of our treatment-seekers in several ways. Both groups had developed gaming as a hobby, which often became an integral part of their identity during adolescence. They dedicated a significant amount of time to playing games and considered gaming as a crucial aspect of their lives. However, the esports interviewees displayed the ability to engage in other activities and make conscious decisions not to play. They demonstrated active regulation of their gaming habits on a daily basis, establishing rules for themselves to pause gaming and attend to other mental, physical, or personal needs.
Despite consciously moderating their gaming, they still regarded it as a foundational element of their identity and life that they were not willing to abandon. They believed that quitting gaming would directly result in feelings of anxiety, depression, and other negative emotions.

Awareness of gaming having the potential to become excessive, which has produced successful balancing and regulation of play
Similar to the treatment-seekers, esports players also spent a significant amount of time on gaming. Many of them openly acknowledge that, in retrospect, there were instances when they should have played less. They recognize the potential negative impact of excessive gaming and have actively pursued other life goals throughout their journey. They strive to strike a balance between gaming and non-gaming aspects of their lives. Despite dedicating entire days and nights to gaming, esports players reflect upon their gaming habits, actively regulate them, and integrate gaming into their non-gaming lives. This ensures that their lives continue to progress and flourish without being hindered or stagnant due to excessive gaming.

Gaming is a high-investment meaningful routine
Several participants in this group had non-gaming competitive hobbies that they actively engaged in alongside their gaming pursuits. However, gaming tended to consume a significantly greater amount of their time. This disparity can be attributed to the accessibility of gaming, as it does not rely on the physical presence of others or necessitate the scheduling of practice sessions. Additionally, gaming is less physically demanding compared to various types of exercise. For esports players, gaming provides a sense of accomplishment and allows fulfilling basic needs; thus, they hold a strong investment in gaming, which has an integrated part of their identity.
In summary, our in-depth analysis suggested three unique themes for each group, and their differences can be summarised as follows: For treatment-seekers, the experience of disorder derives from gaming interfering with self-what one wants to be, do, and have throughout life. For esports players, the experience of balance derives from gaming being integrated into self-what one wants to be, do, and have throughout life.
Our findings are perhaps the strongest evidence so far to demonstrate the specificity and diversity of problems related to gaming, and as such, suggest that gaming disorder, as a diagnostic category, should be subject to ongoing investigation along with the accumulating evidence on the variety of related problems that people have-and how those problems are related to specific sociocultural factors and mental disorders.
Finally, it is worth reporting that 14 of our 16 participants also filled out GRHP survey scales, as applied and listed earlier. The most efficient scale for identifying treatment-seekers was the single-item self-assessment measure: three treatmentseekers self-reported to have problems at least "often" (zero esports players). The other validated instruments performed worse. In other words, the best current way to identify treatment-seekers seems to be their self-assessment of problems. In fact, during participant checking, one of the treatment-seekers told us that the scales did not measure their gaming problems due to poor verbal phrasings related to time.
Based on our qualitative findings, we suggest that one key reason for the validated instruments to be inefficient for identifying treatment-seekers is that they clearly measure multiple "problem" (and non-problem) dimensions at once; moreover, as our participants were already seeking treatment, official diagnostic criteria such as increasing interest in gaming simply did not apply. On the other hand, none of our esports players met any of the cutoffs either, which is comforting regarding the fears of false positive diagnoses. Our goal was to better understand how the experiences in both groups evolve and develop over time, and to gain exploratory knowledge of how health interacts with those changes in the working population.

Data and methodology
We recontacted the participants of the previous study and received confirmation for participation from 5 treatment seekers and 4 esports players. We used the Phenomenology of Play Follow-up (POP-UP) interview frame that is similar to the one used in the baseline study but also considers temporal phenomenological changes in the light of the interviewees' previous answers. Contextualizing survey data were collected using the same instruments as in the baseline study.

Findings and discussion
A process model (Figure 4) was created to visualise the findings, which implied adolescence and early adulthood experiences to initiate coping-gaming for both groups.
We hypothesise that coping-gaming can either be successful or unsuccessful, and it plays a part in the growing relationship the player has with game-playing. The participants with no self-identified gaming problems still often remembered using gaming as a coping mechanism for their various adverse adolescent experiences. In these cases of successful coping-gaming, it helped the players in identity-building, building supportive social networks, and developing emotionally in safe sociocultural contexts. These interviewed players held onto gaming and their largely positive memories related to it throughout their childhood, sometimes identifying gaming as resilience. The interviewed treatment-seeking participants also reported using gaming as a coping mechanism during adolescence. While these gaming-memories are also largely positive in nature, they were often retrospectively associated with negative experiences in a different manner. A major difference between these two groups is how treatment-seeking participants often described "loops", where gaming is unable to solve existing problems which subsequently keep returning in cycles.
As such, it seems that coping-gaming in adolescence shapes the meanings of play early on, and these meanings remain present in adulthood until an axiological shift where such meanings begin to lose value for the individual. Some treatmentseeking participants experienced their problematic patterns as ephemeral, able to reconstruct or replace the meanings said individual has for gaming. Others identified their problematic patterns as persistent, implying that their negotiation continues without resolution. The latter pattern in particular can be expected to hinder or com- The resilience-integration process, in turn, describes a successful version of using coping-gaming for adverse adolescence experiences, during which the hobby integrates as part of the individual's resilience. Gaming remains a stable part of their identity, without self-perceived problems. It's also able to rapidly adapt and evolve at the face of unexpected life events and continues to be an efficient coping tool for the individual, as well as one of their many meaningful domains in life. For these players, value and meaning in neither life domains nor gaming are at direct risk due to a balance between the personal goals and gaming habits of the player. As such, life-thinning is not experienced despite an active gaming hobby. For our esport players, gaming kept its meaningful and gratifying status over the years.
The study distinguishes these two processes ( Figure 5) as models to describe the paths between players who seek treatment for their gaming and players to whom gaming is an integral part of their support system and identity. To simplify, based on this model, a life-thinning path can eventually prevent a person's occupational opportunities, but other types of intensive gaming can help in building a strong social network that improves the general well-being and occupational health of the individual. The original project plan included two additional studies, which were not designed as registered reports. As discussed in the Project Strategy section earlier, one of these studies was rejected by two Finnish authoritative journals, for which it has no journal publication but is attached here as an appendix. The second planned publication-a qualitative study with HR representatives from Finnish workplaces-is discussed below. Additionally, during the project time, we encountered various opportunities to reflect on our (meta) methodological approach as well as engage with unforeseen data and analyses. These reflections and engagements are reported here as side components.

Insights from the HR (manuscript to be submitted in June)
• There was a clear divide between the interviewees from IT companies and those from other fields in regard to how much gaming was present in the work. In the IT companies the interviewees saw gaming as part of the workplaces and

FINDINGS FROM SIDE COMPONENTS
workers they represent to varying degrees. For instance, many employees play games in their free time and talk about these experiences at work. Some companies had teams participating in competitive leagues for games and in one this was used to promote company spirit and team building by inviting others to watch their teams matches. As such, gaming has become one more leisure activity supported by these companies, which they hoped will help their employees to relax and feel home.
In contrast, the interviewees from outside of IT felt that gaming was not really present in their workers professional life in any way. In fact, many were hesitant to participate in the study because they were worried if we wished to hear their experiences as they had little to nothing to give in relation to gaming. In the few instances where gaming was mentioned in workplaces, the interviewees noted that it was done in passing and rarely. For example, when employees discussed their weekend plans they might mention a plan "to relax with their gaming buddies." However, for none was gaming an important or consistent activity to such a degree that they would regularly discuss it in work.
There was also a clear divide between the IT companies and the rest in how the company get-togethers were organised during COVID-19. Traditionally these used to include things like bowling, food, drinks and sauna, but the pandemic and subsequent restrictions made most of this impossible. During the pandemic, the IT companies organised team building events where the employees would get to order food home paid by the company, get together online over a conferencing app and play games together. In contrast, in companies from other fields such events were either not organised, or they involved something else than gaming. For example in one company they had listened to a concert together. Regardless of how differently gaming was present in the daily work life of the companies, they all had similar views on gaming "as a hobby among others." All of our interviewees stated that gaming could be a dear hobby for a person that contributed positively to their general wellbeing by providing a means for relaxation and recovery after work. They did not feel that gaming would be somehow detrimental to work as such.
However, the companies also recognised that gaming can become a hindering pastime if it takes too much time. Most mentioned that a lack of sleep could become an issue, especially in the kind of work that requires one to be present early in the morning. Some also mentioned possible physical problems, from back and neck pain to bad posture and even wrist troubles. Yet gaming was not seen as a only possible hobby that could be problematic for companies: participating in competitive sports could require employees to take time off in order to travel to competitions and tournaments, and injuries caused by sports and exercise could also require long medical leaves. The general ethos of the interviews was that all hobbies, including gaming, were good in moderation and could be problematic in one way or another if taken into extremes.
Finally, when discussing how work life might change in the future in regards to gaming the interviewees were somewhat in agreement. The IT companies saw it becoming even more important for their workers and as a company sponsored or organised activity. The non-IT companies tended to believe that with younger generations who have been playing throughout their youth and childhood entering the workforce it was likely that gaming could also become a more prominent part of their work life.

Methodology and Open Science
The present project took place during the COVID-19 pandemic, which significantly affected both the developing use of technological applications as well as workplace cultures around the world. Arguably, one of the benefiting parties of these events were gaming companies, many of which during the lockdown era increased their user activity and revenues. As a response to this, numerous calls for responsibility were made and the companies were demanded to share their player behaviour data for scientific use in order to advance scientific progress regarding the potential effects of gaming on human life. Nonetheless, these calls often echoed an absurd situation where the academic community aimed to establish research collaboration with companies in the name of open science-yet itself was unable to follow the same principles, namely, systematically keeping datasets and critical methodological features (e.g., statistical code) hidden and thus unreproducible. The first side component of the project is a time-critical commentary of these events , further highlighting the foundational importance of the open science principles followed in the project.
A year later, the commentary was followed by two related publications. The first one ) reflected on the project's two qualitative studies, which were published as registered reports. Retrospectively, it appears that our project was, in fact, the first one in the world to publish in-depth qualitative analysis in the RR format-something that we started suspecting while being able to find any previous examples or guidelines that would facilitate the workflow. Another thing that soon became manifest was the journals' lack of interest in qualitative results: at the stage of initial inquiry, many journals explicitly stated that they are not interested in publishing qualitative results regarding the topic, even though their journal descriptions or aims and scope did not include such methodological exclusions. Accordingly, a proposal for journals to adopt qualitative registered reports was made  to combat this issue, which has stagnated research progress in many fields from medical to social sciences and beyond. Linked to this, a complete original article was written to explain how data sharing can (and should) be a default also in qualitative studies, which pursue best practices and cumulative scientific findings .
Two other methodological side components were published during the project.
The first one took place at the early stages of the project and was established as a research collaboration with our Slovakian colleagues . In this effort, we systematically investigated all currently used screening scales for gaming disorder (and internet gaming disorder) by means of qualitative content analysis in order to better understand the potential differences in measurement. Indeed, we found critical differences in measurement operationalization, which thus informed methods, thus providing alternative interpretations of the phenomenon. In our correspondence, we further review the constructs through a network analytic approach and celebrate the opportunity to make rapid scientific progress in transparent dialogue with colleagues.

Preparation and Triangulation
At the start of the project, two side components were published to support the launch and planning of future designs. In the first one ), ongoing data collection and analysis were utilised for the development of the Phenomenology of Play interview frame, which was eventually used in both of the project's qualitative main outcomes (Qualitative Study 1, Qualitative Study 2). In the second one , an initial review of literature was carried out to map out the existing scientific knowledge regarding gaming and work environment topics. These efforts facilitated writing all Stage 1 proposals and established a basis for maintaining a bibliography at the project website through the years.
Two more studies were carried out to further understand the results gained by the earlier quantitative and qualitative efforts. In the first one (Adamkovič et al. 2023)-in collaboration with our Slovak colleagues and supported by their parallel efforts to improve measurement practices in the psychopathological literature-we applied network analysis to a newly collected sample of actively gaming adult individuals. Having learned from our previous study that identical symptoms (usually from the DSM-5-TR or the ICD-11) tend to be measured in myriad ways by different validated measurement scales, we tested whether changing the operationalization of a single item would lead to significant differences also on the network level. Indeed, we found exploratory evidence for this hypothesis, in addition to which the results implied no network differences across different gaming groups, thus indicating that working and other populations are unlikely to differ in in this regard.
The final side component ) was initiated as a collaboration with the Restart treatment program. Although the program had collaborated with our project from the start, only later were their treatment-seeker data made available for research use. We chose to utilise this valuable opportunity and applied qualitative template analysis to open descriptions of reasons for treatment-seeking (N=110). The findings implied that Finnish adults seek treatment for many kinds of games and gaming-related problems, including but not limited to those, which are described in diagnostic manuals ( Figure 6). In fact, more than half (58 %) of the treatment-seekers did not even meet the more flexible internet gaming disorder criteria.
These figures can be at least partially explained by the learned thematic insights; for instance, the existential and financial problems expressed by the treatment-seekers are not listed in either of the dominant diagnostic manuals (DSM-5-TR, ICD-11). In conclusion, we must end with a cliche call for more research that will help us better understand the nature of problems that people have with gaming, and to what degree those problems entail clinical or diagnostic solutions. On the other hand, for occupational purposes, non-clinical and non-diagnostic instances might well be tackled with support services in a similar vein to "work ergonomics" that can improve everyday conditions and prevent future challenges without explicit intervention. The importance of this matter is emphasised by the tensions between gaming, health and working life, which have risen to the surface especially by the World Health Organization's effective decision to start diagnosing "gaming disorder" in 2022. On the other hand, during the COVID-19 pandemic, the World Health Organization also encouraged people to play videogames e.g. for social health benefits.
While the benefits obtained from gaming can reflect improved wellbeing of the working community, problems related to gaming may have an undesirable effect, for instance, on the individual's ability to work.
Based on the project outcomes, we propose a set of three recommendations, which can help both employers and employees to better construct and support gaming-work relationships in the present era.

Occupational health is a sum of all life components; gaming can serve as a component in many different ways
For most players at the working age, gaming is a hobby among others. A passion for gaming can speak of a balance between work and everyday life or an active life attitude. According to our results (Qualitative Study 1, Qualitative Study 2), employees who played for more than four hours daily had excellent work ability and they perceived gaming as an asset that supported their resilience. Gaming seems to connect to problems mainly when multiple challenges pile up on one individual over several years. Many who experience gaming in a web of problems may be at risk of being socially marginalised, and typically they do not have a long-term work contract.

RECOMMENDATIONS FOR OCCUPATIONAL STAKEHOLDERS
However, they may have skills that are in demand on the job market, such as IT and programming skills.
Recommendation. As a default, employees' gaming should be perceived as a regular hobby, which at best balances and supports an individual's overall health and life. Recommendation. To support excessively gaming individuals at the risk of social exclusion, employers' should further develop means to integrate them into work communities. Providing tasks that are challenging and rewarding in various ways can help in the process.

Problems related to gaming are rare and varied
No more than 0.4 % of our respondents in the nationally representative survey of the Finnish working population met the diagnostic gaming disorder criteria (Quantitative Study 2). It is essential to emphasise that the survey represented the entire working-age population of Finland in terms of age, geographical location and gender, without an overrepresentation of young people. Although the measures need much more development, it is evident that diagnostic-level problems concern a very small part of Finnish employees. Based on the follow-up research (Qualitative Study 2), it also seems that people report problems in cycles. For most of those who experience GRHPs, the problems had already disappeared after a year (Quantitative Study 2).
Although more research is needed to more accurately answer several open questions, the results are somewhat clear in one regard: the current prevalence of gaming is not a reason to panic from the viewpoint of Finnish working life and wellbeing. At the same time, it is good to emphasise that in clinically significant instances the life situation seems to be difficult overall, in which case the availability of support services is important. Employers can help identify and understand such instances, and serve as supportive pillars in the process.
Recommendation. Marginalisation and social exclusion, with and without related gaming, do not usually allow for conventional education and career development. Employers could cooperate with public support services, such as the Restart program in Finland, in order to bring the skills of players who have been outcast in the society into working life. Recommendation. Supporting employees who have gaming as an active hobby, e.g. by enabling occupational amateur esport teams or by involving remote workers in cooperative activities through gaming, can further improve the overall well-being and sense of community in the workplace.

Gaming is part of a wider network of technology at work and outside of it
Today, gaming is converging with many forms of technology and media. Social media, streaming services and interactive live broadcasts (live-stream) contain numerous game-like features, while videogames, in turn, contain diverse narrative and social features. In the future, it will be increasingly valuable for the employer to understand how both work and employees live and develop in these changing networks.
Recommendation. Gaming literacy is a critical part of today's media literacy. A basic understanding of how videogames and their mechanics work can help an employer better understand changes in working life, employee well-being and hidden skills.
Meanwhile, for the employees, good gaming literacy can help them reflect on their own media use as well as their role in today's technology-infused workplaces.
To our knowledge, this was so far the largest scientific project with a goal to investigate the relationship between gaming and work, and in particular, how health and wellbeing fit in that relationship. First, the project asked: (RQ1) How to identify gaming-related health problems in the working adult population? Based on the results, it is clear that the currently available survey measures cannot be used as reliable indicators. In the absence of better tools, for now, the best line of action appears to be the use of direct self-reflection-how often does one experience gaming problems?-as was originally done by the National Institute of Health in Finland (Salonen & Raisamo 2015). In both quantitative and qualitative analysis, using the self-identification of GRHPs as a means for assessing problems performed at least equally to other validated measures, in addition to which the brevity of the method makes it easy and efficient to apply in almost any context. In summary, much more research is needed to develop improved (clinically) validated measures. Meanwhile, a high frequency of self-reported problems can be used for screening individuals who may need support. We leave an extra note on the exploratory findings regarding financial problems: it appears that some individuals specifically have problems with using too much money on gaming. This phenomenon is best described as a type of impulsive shopping, thus not directly related to gaming or gambling per se. Needless to say, this new area requires more research.
Secondly, the project asked: (RQ2) How do gaming-related health problems affect work life, especially in Finland? For this question, answers split between individual and population domains. On the latter level, the prevalence of gaming disorder in the Finnish working population-when measured by the ICD-11 criteria or problems being self-reported on the highest frequency-was low: approximately 0.4 %. For instance, compared to gambling disorder that currently has a prevalence of 1.4 % in the Finnish population (Castrén 2023), gaming disorder represents but less than a third. On the other hand, as highlighted earlier, the measures remain CONCLUSIONS underdeveloped and better knowledge will accumulate once clinically and otherwise validated instruments enter public usage. For individual life contexts, however, the results are in line with the premise that gaming can be related to health problems in some vulnerable groups. Based on the presently available data, in these groups gaming appears to serve primarily as an unsuccessful coping mechanism that is used to alleviate various underlying (mental and social) health challenges or trauma. Because of this, most people with notable GRHPs seem to suffer from marginalisation and social exclusion, and thus do not have major occupational responsibilities or careers.
Finally, we asked: What is an optimal line of action in the current situation affected by "gaming disorder"? In the light of the above findings, the current evidence implies that adverse gaming cycles-which represent the most notable GRHPs (hypothetically the identified 0.4 %)-begin primarily in adolescence and early adulthood in disadvantaged social contexts. This is typically a time when young people accrue valuable educational and initial work experience, and it is specifically these early resources that can be left unharvested for those to whom gaming is among their unsuccessful coping tools. Interventions that successfully combat this issue early on can improve the work ability and wellbeing in Finnish work life in the long run. At the same time, providing supportive gaming-related activities for those employees to whom gaming is successfully integrated in their lives can improve workplace wellbeing and enrich the employer's tool set for positive social engagement.

Gaming and Gambling Related Health Problems in Finnish Workplace Wellbeing
Background: Since 2022, both gaming and gambling disorders have been part of the ICD-11. It is not known yet what is the role of such health problems in Finnish workplace wellbeing.

Methods:
We collected a nationally representative sample of the Finnish work population by age, gender, and region. Gaming and gambling based risk groups were identified by validated screening scales and ANOVA was utilised to map out their associations with anxiety, depression, loneliness, work ability, work engagement, and work recovery.
Results: Both gaming and gambling based risk groups were associated with lower health and work wellbeing scores, when compared to other respondents. Especially the gaming based risk group was negatively associated with all measured health domains with relatively large effect sizes.

Conclusions:
The results imply that problems accumulate in difficult life situations: both risk groups reported symptoms of anxiety, depression, and loneliness, while also having lower work ability, work engagement, and work recovery scores. Future research should focus on investigating casualties and aetiology.