The digitalised work environment : Health, experiences and actions

(1)

The digitalised work environment:

Health, experiences and actions

Doctoral Thesis

Magdalena Stadin

Jönköping University School of Health and Welfare Dissertation Series No. 104 • 2020

(2)

The digitalised work environment:

Health, experiences and actions

Doctoral Thesis

Magdalena Stadin

Jönköping University School of Health and Welfare Dissertation Series No. 104 • 2020

(3)

Doctoral Thesis in Health and Care Sciences The digitalised work environment:

School of Health and Welfare, Jönköping University P.O. Box 1026

SE-551 11 Jönköping Tel. +46 36 10 10 00 www.ju.se

Printed by Stema Specialtryck AB 2020 ISSN 1654-3602

ISBN 978-91-88669-03-2

Trycksak 3041 0234 SVANENMÄRKET

(4)

A goal is a dream with a deadline – Napoleon Hill

(5)

Abstract

Background: The aim of this thesis was to examine the association between

technostress, operationalised as information and communication technology (ICT) demands, and indicators of work-related stress, as well as its association with self-rated health. Additional aims were to identify occupational groups at risk with regard to ICT demands, and to describe experiences of technostress and how it was handled by healthcare managers.

Methods: The thesis includes four individual papers. Papers I–III have a

quantitative (cross-sectional or prospective) study design and are based on data derived from the Swedish Longitudinal Occupational Survey of Health (SLOSH) and collected between 2006 and 2016. Data was analysed by statistical methods, such as linear and logistic regression analysis. Paper IV has a qualitative study design and is based on data from 20 semi-structured interviews with healthcare managers. The data was analysed using the critical incident technique.

Results: ICT demands were correlated with job strain and effort-reward

imbalance, especially the demands and effort dimensions of these measures. High ICT demands were associated with suboptimal self-rated health in cross-sectional analyses and in prospective analyses including repeated measurement. Managers, and particularly ‘managers in healthcare and other community services’, followed by ‘managers in education’, had the highest odds ratio of ICT demands, in comparison with both ‘non-managers’ and ‘all other managers’. Healthcare managers’ experiences of technostress could be categorised into the main areas ‘negative aspects of digital communication’, ‘poor user experience of ICTs’ and ‘needs to improve organisational resources’. The actions they took to cope with technostress were categorised into the main areas ‘culture, norms and social support’, ‘individual resources’ and ‘organisational resources’.

Conclusions: Technostress operationalised as ICT demands is associated with

suboptimal self-rated health. Occupational groups differ in their exposure to ICT demands by industry and position. Organisational efforts to ensure a sustainable and healthy digital work environment are warranted. ICT demands should be assessed against ICT resources for a comprehensive understanding of their association with health.

(6)

Original Papers

Paper I

Information and communication technology demands: the association with job strain, effort-reward imbalance and self-rated health in different socioeconomic strata

Magdalena Stadin, Maria Nordin, Anders Broström, Linda Magnusson Hansson, Hugo Westerlund & Eleonor I Fransson

Published in the International Archives of Occupational and

Environmental Health (2016) 89:1049–1058

DOI 10.1007/s00420-016-1140-8 Paper II

Repeated exposure to high ICT demands at work, and development of suboptimal self-rated health: findings from a four-year follow-up of the SLOSH study

Magdalena Stadin, Maria Nordin, Anders Broström, Linda Magnusson Hansson, Hugo Westerlund & Eleonor I Fransson

Published in the International Archives of Occupational and

Environmental Health (2019) DOI 10.1007/s00420-019-01407-6

Paper III

Technostress operationalised as information and communication technology (ICT) demands among managers and other occupational groups: results from the Swedish longitudinal occupational survey of health (SLOSH)

Magdalena Stadin, Maria Nordin, Anders Broström Linda Magnusson Hansson, Hugo Westerlund & Eleonor I Fransson

Submitted (2020) Paper IV

Healthcare managers’ experiences of technostress and the actions they take to handle it – a critical incident analysis

Magdalena Stadin, Maria Nordin, Eleonor I Fransson & Anders Broström Submitted (2020)

(7)

Abbreviations

BMI Body Mass Index

DCQ Demand-Control Questionnaire

DCSQ Demand-Control-Support Questionnaire CIT Critical Incident Technique

EMR Electronic Medical Record ENG English Language

ERI Effort-Reward Imbalance

ICT Information and Communication Technology

IKT Informations- och Kommunikationsteknologi (In Swedish) IS Information System

IT Information Technology JD-R Job Demands-Resources Model SEP Socioeconomic Position SES Socioeconomic Status

SLOSH Swedish Longitudinal Occupational Survey of Health SRH General Self-Rated Health

SV Swedish Language

(11)

1

Introduction

The digitalisation of working life, including the increased use of information and communication technology (ICT) has a substantial impact on the work environment from many points of view. The use of digital devices and ICT has improved work efficiency, facilitated information exchange and contributed to increased work flexibility (1-3). However, despite its advantages, there are many features in the digitalised work environment that can contribute to technostress, which simply refers to the experience of stress related to the use of ICT at work (4). Technostress is caused by employees’ attempts to cope with the constant evolution of ICT and its physical, psychosocial and cognitive demands upon them (4). Technostress has been described as being caused by ‘technostress creators’, which can be categorised in the five dimensions overload’, invasion’, ‘techno-complexity’, ‘techno-insecurity’ and ‘techno-uncertainty’ (4). Techno-overload is about ICT that forces the employees to work faster and longer. Techno-invasion refers to the invasive influence ICT has on employees, due to the high level of their availability via ICTs, which may blur the transition between work and leisure time (4). Techno-complexity refers to the complexity related to working with ICT, which causes a feeling of insufficient digital literacy, which forces employees to spend more time and effort to learn new ICTs (4) (Figure 1). From an occupational health perspective, it is important to recognise factors that might prevent the onset of technostress. These kinds of resources have been termed ‘technostress inhibitors’, and can be categorised in the three dimensions ‘literacy facilitation’, ‘technical support provision’, and ‘involvement facilitation’ (4) (Figure 1). In Papers I– III of this thesis, technostress operationalised as ICT demands is studied. ICT demands principally reflect aspects of the overload and techno-invasion dimensions of technostress, and some areas of techno-complexity, but neither techno-insecurity nor techno-uncertainty. In Paper IV, however, technostress is studied from a more comprehensive perspective.

(12)

2

Technostress

Technostress creators Technostress inhibitors

Techno-overload Literacy facilitation

ICT forces employees to work

faster and longer Organisational enabling and encouragement of improved

digital literacy for employees

Techno-invasion Technical support division

The invasive influence ICT has on employees due to the expectation

that they should be constantly available via ICTs which may blur the transition between work

and leisure time

Good (i.e. competent and accessible) IT support

Techno-complexity Involvement facilitation

Complexity related to work with ICT, causing a feeling of insufficient digital literacy and causing employees to spend more

time and effort on coping with ICTs

Enabling of user influence, employee preparation before

implementing new ICT, and encouragement to use new ICTs

Techno-insecurity

Situations where employees are afraid of losing their jobs, either

to people with better digital literacy, or to robotic process automation/artificial intelligence

Techno-invasion

Constant evolution of new ICT, implying that employees must continually learn and educate themselves in how to use it

Figure 1. Dimensions of technostress creators and technostress inhibitors according to Ragu-Nathan et. al (2008) (4).

(13)

3

Exposure to work-related stress, defined as job strain or effort-reward imbalance, has been associated with ill-health and sickness absence over time (9–11). In contrast with these work-related stress indicators, knowledge about the relationship between ICT demands and self-rated health is rather limited

(5). There is also lack of clarity about whether or how ICT demands relate to these indicators. Additionally, there is uncertainty about exposure to ICT

demands in different occupational groups with regard to industry and position, and it is unclear how technostress is experienced and handled by an occupational group with assumed high exposure to ICT demands. Additional research regarding the digitalised work environment has been called for, and Sandblad, Gulliksen, Lantz, Walldius & Åberg (2018) claim that:

“The questions about how to design better transformation processes, utilise the full potential of digitalisation, and create efficient work and a good digital work environment, are complex. Much new knowledge of different kinds is needed, of a theoretical as well as a practical nature. Different scientific disciplines need to cooperate and contribute to new research about this. General and industry-specific solutions need to be developed. In other words, more research and development in this area is needed.”

Digitalisation and the work environment (author’s translation from Swedish), page 271 (6)

(14)

4

Background

The work environment

Most people in Western society are a part of the workforce for several years of their adulthood, and even though work may have benefits for health, the work environment may also influence it negatively (7-9). The work environment refers to all characteristics of the workplace, including both the physical (such as work tools, air, noise, light etc.), and the psychosocial aspect, related to psychological and social aspects in the work culture or the organisation (9, 10). Occupational hazards refer to risks in the work environment that are associated with negative health-related outcomes, or accidents (7). Research into occupational hazards and their associations with health has thus altered in focus over time. Earlier, research in this area mainly reflected occupational hazards in the physical work environment (such as physical, chemical, biological and ergonomic features, typically in industrialised work), whereas in contemporary research, the focus has shifted to reflect psychosocial factors in the work environment, such as organisational or social aspects associated with work-related stress (7, 10). However, in the contemporary work environment characterised by digitalisation, occupational hazards of both a physical and a psychosocial nature have occurred (6).

The digitalised work environment

Historical development of the digitalised work environment

The contemporary digitalised work environment is characterised by the use of hardware; in terms of computers and computerised digital devices (e.g. smartphones), and software; in terms of ICT, apps, smart technology and, not least, the Internet (5, 6, 11). From a historical perspective, the very first computer was introduced 70 years ago, i.e. in the 1950s, but was at the time only used by experts in the field of computer development (12). Two decades passed before the first commercial personal computer was introduced (12). In 1989, the Worldwide Web (www) was introduced, and then released for public use in 1993 (13). A historical milestone of note was an email

(15)

5

conversation in 1994 between the Swedish prime minister, Carl Bildt, and the US president, Bill Clinton, which was the first-ever email conversation between heads of states (14). Even though computers had already been used in some organisations by the 1970s and 1980s, the computerisation of the majority of organisations took place from the end of the 1990s to the beginning of the new millennium (12), and the term ‘the paperless office’ was coined (15). The technological development and digitalisation of organisations has since continued at an enormous speed, and in Sweden today, approximately 90% of the working population use computers with access to the Internet in their daily work (16). This implies that organisations and the employees within them are now dependent on digitalisation to accomplish their work (5, 6).

Another revolutionary device that has influenced the digital work environment is the mobile phone, and in particular, its successor, the smartphone. The mobile phone had its breakthrough during the 1980s, even though mobile communication had been used beforehand, e.g. by the emergency services (17). The first pocket-sized mobile phone, the so-called ‘yuppie teddy’ (sv. ‘yuppie nalle’), was introduced in 1987 and was at the time principally used by businessmen. Using mobile phones to send text messages then became possible in 1992. During the 1990s and at the beginning of the 2000s, mobile phone use became common among the general working population. The next milestone was in 2007, when the first smartphone, which is basically a pocket-sized computer, was introduced (17). The widespread use of mobile phones and smartphones has changed the experience of work from many points of view. It has enabled flexible communication and information exchange regardless of time and space; however, it may also have blurred the boundary between work and private life (5, 6, 18).

ICT, administration and information overload

The use of ICT enables a wide range of information exchange, for example, email communication, documented information (such as in electronic medical health records) and simply information searches on the Internet (6). On the other hand, the easy access to an enormous amount of information via ICT has been identified as the biggest source of information overload (19). Simply described, this refers to ‘a state when too much information is received’, and

(16)

6

it is more traditionally described as ‘a state when the level of information is greater than the capacity to process the information’ (20). Exposure to information overload has been found to cause confusion and to impair decision-making capacity and the ability to decide on the relevant information (21). It has also been related to the experience of stress, and can consequently be a threat to health (21).

Contemporary working life is characterised not only by digitalisation, but also by administration (22). Administration can be defined as the process of collecting, creating, processing and compiling information in order to maintain, cooperate with and control organisation systems, that may contribute to information overload (22). The digitalisation of administration has facilitated increased capacity, but may also be a source of information overload and technostress if it is channelled via ICT (22-24). Consequently, increasing an employee’s ICT-related administrative job burden could be associated with increased experience of technostress.

ICT and boundaryless work versus recovery

The increased use of ICT via computers, smartphones and touchpads, along with increased access to the Internet, has changed the experience of work for a large proportion of the working population (5, 6, 18). In professions where computers or smartphones (including Internet access) are the major tools that are required to perform work, the workplace can literally be anywhere (5). Flexible work, including ‘worktime control’, may have positive features in terms of increased fit between work and private life and improved work motivation (25-27). However, flexible work could turn into boundaryless work, because if it is possible for an employee to be available on work-related ICT devices during leisure time, this might come to be expected by colleagues (28). In Sweden, about 45% of the working population use their work-related email during their holidays (16). This may contribute to unclear boundaries between work and private life (5, 29), and impaired psychological detachment from work (30).

Flexible work, in terms of time- and place-independent management, does not necessarily have a negative impact on health (31). However, boundaryless work might impact health negatively due to the absence of psychological

(17)

7

detachment from work and lack of psychological recovery (26, 29, 30, 32). Recovery during or after the official working day is essential to preserve good health (18, 26, 32). In situations where recovery during working hours (internal recovery) is difficult, recovery afterwards (external recovery) is even more important (32). Lack of recovery from work is associated with prolonged exposure to work demands and stress-related processes (e.g. rumination), which in the long-term increases the risk of poor health (32). Consequently, despite flexible working conditions, it is recommended that psychological detachment as well as recovery from work are maintained (29, 30).

User experience versus usability

In the digitalised work environment, and in the contact with ICT, the experience of the employee or the ‘user’ (i.e. the individual who uses an ICT device) is simply termed ‘user experience’. It refers to the user’s perception of and response to the use and/or anticipated use of a product, system or service (33). A holistic perspective of using a system is incorporated into the user experience concept, including psychological responses such as emotions, beliefs, preferences and perceptions, but also physical responses, during the whole process of using a system (33). Whether the user experience is perceived as good or bad depends partly on features in the ICT system, including brand image, presentation, functionality, performance and interactive capacity. It is also related to aspects of the context (e.g. the work environment) and characteristics of the user, such as digital literacy, personality, attitude and prior experience of a system (33). Many different definitions of user experience exist, but there are also separate perspectives on use and different contexts of application (e.g. in academia and in the design industry) (34). Three perspectives of user experience have been identified, including ‘user experience as a phenomenon’ (e.g. describing and identifying user experience), ‘user experience as a field of study’ (which refers to the actual experience and expectations of the experience, etc.) and ‘user experience as a practice’ (incorporating user experience design into ICT) (34). In this thesis, and in Paper IV in particular, the perspective of user experience as a field of study is mainly applied, because it is the experiences in the digitalised work environment (including user experience) that are studied.

(18)

8

User experience is sometimes used interchangeably with the term ‘usability’. Usability is defined as ‘the extent to which ICT, a product or service can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use’ (35). Even though there are several obvious similarities between user experience and usability, there are also some clear distinctions between these two concepts. While usability is target focused, and is intended to give an objective view of the usability of ICTs, user experience refers to the experience of using ICTs from a more holistic perspective, including aspects such as emotions, attitudes, digital literacy, etc. A commonly used metaphor for describing the distinction between usability and user experience is to picture usability as ‘science’ and user experience as ‘art’. Still, user experience and usability are deeply related concepts, and the usability of ICTs might be improved by considering the user experience. A good user experience, including a high level of usability, in the ICTs at work is an important factor for a sustainable digitalised work environment.

Health perspectives

Descriptions of the concept of health were already being produced in ancient Greece. Hippocrates (c.450–380 B.C.) described good health as a harmonic balance between different bodily fluids (36). Since then, definitions of health have developed over time. In 1946, for example, the World Health Organization described health as a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity (37). However, in the World Health Organizations definition of 1991, health was described as a changeable process rather than a steady state. The definition used in this thesis is in line with the 1991 definition, presented below:

“Health itself should be seen as a resource and an essential prerequisite of human life and social development rather than the ultimate aim of life. It is not a fixed end-point, a product we can acquire, but rather something ever changing, always in the process of becoming” (38).

(19)

9

When assessing health, both objective (e.g. clinical diagnoses) and subjective (based on self-ratings) dimensions of health should be considered, since they complement each other and contribute to a holistic view (38, 39). However, one complication when assessing health is that objective and subjective health dimensions do not necessarily go hand in hand. Health status can be perceived as good even though an individual has a clinical diagnosis (e.g. diabetes), or, conversely, an individual might rate their health as poor even in the absence of a clinical diagnosis (39). The subjective perspective of health will be applied in this thesis, by analysing the experience of stress and self-rated health. The focus will be on the working population, and an occupational health perspective will thus be applied. The World Health Organization defines occupational health as follows:

“Occupational health deals with all aspects of health and safety in the workplace and has a strong focus on primary prevention of hazards. The health of the workers has several determinants, including risk factors at the workplace leading to cancers, accidents, musculoskeletal diseases, respiratory diseases, hearing loss, circulatory diseases, stress-related disorders and communicable diseases and others.”

World Health Organization, webpage http://www.who.int/topics/occupational_health/en/

Work, stress and health

Work-related stress, sex and socioeconomic position

The concept of stress from a general point of view is both comprehensive and complex, and there are a number of definitions of stress. However, common to many definitions and theories is that it involves a demand or stressor and a stress response, and that the stress response has both psychological and physiological characteristics (16–19). In everyday use however, ‘stress’ often refers to the stress response. In this thesis, the term ‘work-related stress’ (sometimes termed occupational stress) is used, which refers to an imbalance between job demands (e.g. to work fast and hard) and lack of resources in the work environment (40-42). Work-related stress is commonly operationalised as ‘job strain’ (high job demands and low control), ‘iso-strain’ (high job demands, low control and low social support), or ‘effort-reward imbalance’

(20)

10

(imbalance between job effort and reward) (7, 43). Additional factors that influence the psychosocial work environment are workplace bullying, shift-work, boundaryless work and work-family conflict (i.e. work inflicts negatively on family life) (18, 44-46). In this thesis, work-related stress is partly seen as contextual, and the thesis mainly reflects work-related stress in the digitalised work environment, by analysing technostress and ICT demands (4, 47).

Psychosocial working conditions differ with regard to sex and socioeconomic position. The labour market has been described as sex-segregated, from a horizontal as well as a vertical point of view (48). Horizontal segregation in terms of the uneven distribution of men and women in different occupational industries. Vertical segregation in terms of the unequal balance between men and women with regard to hierarchical work positions in the organisation, and the typically higher proportion of men in higher positions (48). Women have, in general, higher job demands (e.g. psychological, cognitive and emotional demands) placed on them, and lower resources than men, and are consequently exposed to work-related stress to a higher degree than men. The sex difference in job demands is, however, larger than the sex difference with regard to resources. The sex difference in job demands can be partly explained by higher rates of job demands in female-dominated industries, such as healthcare and other community services, and the education industry (48). This suggests that occupations that include a great deal of contact with people (e.g. patients and students) might have higher job demands.

Health inequalities with regard to socioeconomic position are well investigated (49, 50). The relationship between socioeconomic position and health is modified by work-related stress operationalised as job strain and effort-reward imbalance (51, 52). Employees with low socioeconomic position have in general a higher risk of exposure to work-related stress (51, 52). Additionally, the relationship between job strain or effort-reward imbalance and health-related outcomes is stronger in groups with low socioeconomic position than in groups with high socioeconomic position. This is partly explained by an uneven distribution of resources with respect to socioeconomic position, but also to inequalities in health behaviours (51, 52).

(21)

11

Work-related stress and health

Work-related stress causes major societal costs, but also costs for organisations as well as individuals, due to its association with poor health-related outcomes and sickness absence (53-55). Sweden and countries nearby have a long-term history of studying work-related stress and its association with health (53, 56). Work-related stress operationalised as job strain or effort-reward imbalance is associated with a number of public health diseases (57-65). These include hypertension, atrial fibrillation, coronary heart disease, myocardial infarction, metabolic syndrome, Type 2 diabetes, depression and sleep disturbances, etc. (57-65). However, the influence of resources in the work environment can balance the negative impact job demands have on health (17, 53). These resources may be general and include features such as control, reward (e.g. good salary and career prospects) and social support from co-workers and executives (17, 53). They can also be contextual and reflect resources directed to specific work environments, such as the digitalised work environment (5, 34). Subsequently, strengthening organisational resources is a key factor for improving occupational health.

Self-rated health in the working population

General self-rated health is in its simplest form measured by the one-item question, “How would you rate your general state of health?” (66). This simple question is a useful predictor of future health status, as well as mortality (67). In Sweden, the majority of the working population report good general health, but about 20% of men and 18% of women report suboptimal (i.e. poor, not optimal) self-rated health (68). However, in many other Western countries, a higher proportion of women report suboptimal self-rated health than men. (69). The minor sex difference in the Swedish example is in part explained by differences in health behaviours, e.g. men take less physical exercise, eat less healthy food and drink more alcohol in general (68), but additional reasons for this result have not been fully determined. As with other health-related outcomes, there are also socioeconomic gradients with regard to self-rated health (70, 71). This health-related inequality can partly be explained by health behaviours (71), but also by occupational differences with respect to job control, heavy physical workload and psychological or chemical exposure (70).

(22)

12

Technostress and health

Even though there is some knowledge about the association between technostress and health, many studies of technostress focus on the stress experience itself, rather than on its association with other health-related outcomes. However, acute situations in digitalised work, such as computer breakdowns, have been related to a temporary increase in physiological responses such as cortisol levels (72, 73)1_{. Technostress operationalised as}

ICT demands has also been associated with cognitive complaints, such as disturbances of concentration, memory and decision-making, and ability to think clearly (47). Technostress operationalised as ‘telepressure’ (defined as thinking about ICT messages and feeling an overwhelming urge to respond) has been related to burnout, sickness absenteeism and sleep disruption (74). Associations between technostress and burnout have also been observed in a nursing context (75). Additionally, working in front of a computer has been found to be related to visual complaints as well as pain in different parts of the body, including neck, shoulders, low back, hands, wrists and elbows (76). However, less is known about the association between technostress and other health-related outcomes, such as suboptimal self-rated health. Consequently, additional research about the association between technostress and health is needed for a more comprehensive understanding.

1_{Further information about technostress and similar concepts is presented under the}

(23)

13

Theoretical framework

Job Demands-Resources model

The Job-Demands-Resources model is a conceptual model that implies that every occupation has certain risk factors associated with the development of work-related stress. These factors are classified under the categories of job demands and job resources (42). Job demands refer to different aspects that require sustained efforts or skills, and job resources refer to aspects related to efficacy in achieving work goals, reduction in job demands, and stimulation of personal growth, learning, development. These aspects may be of physical, psychological, social, or organisational character. The Job-Demands-Resources model can be applied in a variety of different working contexts, since the specific characteristics in the job demands and job resources dimensions are changeable (42). On the other hand, in other conceptual models, the demands and resources are perceived as more specific, and have been operationalised into demands, control and reward, etc. (7, 43).

Job strain and Iso-strain

Job strain is defined by the demand-control model, a conceptual model including the dimensions ‘job demand’ and ‘job control’. A modified version of the demand-control model includes social support from colleagues and superiors, and is referred to as the demand-control-support model (7). Job demand refers to psychologically demanding aspects in the work environment, such as workload, and job control refers to the individual’s influence over the work situation. A combination of high job demand, and low job control (regardless of social support) is termed job strain and increases the risk of contracting poor health. Iso-strain (i.e. a combination of high job demand, low control and low social support) also increases the risk of contracting poor health. The demand-control model and the demand-control-support model can also be used to identify ‘healthy work’, defined as a combination of moderate to high job demand and high job control, and preferably high social support (7) (Figure 2).

(24)

14

Figure 2. Author’s own illustration of the job-demands-support model according to Karasek & Theorell (7).

Job demand

Control

Social support

High Low Low High Weak Strong Iso-strain Active work Job strain

(25)

15

Effort-reward imbalance

‘Effort-reward imbalance’ refers to high job effort combined with low reward, which is an indicator of work-related stress (43). High effort comprises high workload, pressure, overtime, etc. Reward is a latent factor of three types of occupational reward, including financial reward, esteem reward, and reward related to career prospects and job security (77). Balance between effort and reward is, however, considered as a healthy state of work (43) (Figure 3).

Insufficient financial reward, esteem reward, career prospects or job security

High workload, pressure, responsibilities, overtime etc.

Figure 3. Author’s own illustration of the effort-reward imbalance model according to Siegrist (49).

Low reward High effort

(26)

16

Technostress and ICT demands

Stress in the digital work environment is not a new phenomenon. The concept of ‘technostress’, which has been described as a state of psychological and physiological arousal as a result of the imbalance between work demands and the skill of mastering computers (78), was first introduced at the beginning of the 1980s (79). Since then, the meaning of technostress has been extended to reflect additional potentially stressful aspects of the digitalised work environment (4, 72) (Figure 1; Figure 5). Parallel with the technostress concept, other concepts that refer to stress in a digital working context have been introduced, e.g. ‘ICT demands’, ‘telepressure’, and ‘stress related to information systems’ (24, 47, 74, 80). These concepts can most often be perceived as operationalisations of different sub dimensions of technostress as described above (4). ICT demands, analysed in Papers I–III of this thesis, fit mainly under the techno-overload and techno-invasion dimensions (4) (Figure 5), while Paper IV describes technostress from a more comprehensive perspective. It should also be mentioned that in the media and in the world outside academia, technostress is sometimes referred to as ‘digital stress’ or ‘IT stress’.

Resources in the digitalised work environment

It has so far been declared that many features in the digitalised work environment might cause technostress, and from an occupational health perspective, it is important to recognise resources that might prevent the onset of technostress. These kinds of resources have been termed ‘technostress inhibitors’ or ‘ICT resources’ (4, 24). Technostress inhibitors are defined as organisational or managerial factors that might prevent technostress. They consist of the three dimensions ‘literacy facilitation’, ‘technical support provision’, and ‘involvement facilitation’ (4). Literacy facilitation refers to organisational facilitating and encouragement to improve employees’ digital literacy, which is the ability to use ICTs to find, evaluate, create, and communicate information. Technical support provision refers to good (e.g. competent and accessible) IT support, which prevents technostress, since it may solve acute problems in the digitalised work environment. Involvement facilitation is related to enabling user influence, employee preparation before implementing new ICTs, and encouraging the use of new ICTs. It may prevent technostress, as it facilitates increased control. While technostress creators

(27)

17

have been found to diminish organisational commitment, as a result of decreased job satisfaction, technostress inhibitors rather increase job satisfaction as well as organisational commitment (4).

The definition of ICT resources is similar to that of technostress inhibitors, and refers to buffering resources with respect to technostress operationalised as ICT demands (24). ICT resources include the dimensions ‘ICT personal assistance’ (referring to good IT support) and ‘ICT resources/upgrades’ (appropriate and updated ICTs) (24). ICT resources have also been observed to moderate the relationship between ICT demands and the experience of stress. Consequently, organisational efforts to ensure a healthy and sustainable digital work environment, by incorporating resources in terms of technostress inhibitors and ICT resources, might be beneficial for the organisation from many points of view.

(28)

18

Rationale

Even though the digital work environment has been evolving for several decades, knowledge about the relationship between technostress and health is limited. Work-related stress, operationalised as job strain or effort-reward imbalance, its existence in different occupational groups with regard to sex, socioeconomic position and industry, and its association with health is well investigated. However, since technostress only reflects the digitalised work environment, its prevalence in different occupational groups and its association with health might be somewhat different from that of general work-related stress.

Experiences of technostress have been described in occupational groups in specific industries previously (72), but there is a lack of studies that compare the exposure to technostress between different occupational groups. Additionally, knowledge about how technostress is handled in certain occupational groups is insufficient. In order to obtain a comprehensive understanding of contemporary digitalised working life, and to identify ongoing challenges for occupational health services, supplementary research in this area is required. However, a challenge in this research area is to keep pace with the rapid transformation of the digitalised work environment, which includes the constant evolution of new ICTs. This thesis is an attempt to fill a minor part of the major gap in this research field.

(29)

19

Aim

The overall aim of this thesis was to examine the association between technostress operationalised as ICT demands and indicators of work-related stress, as well as its association with general self-rated health. An additional aim was to identify occupational groups at risk of exposure with regard to ICT demands, and to describe experiences of technostress and how it was handled by healthcare managers. The specific aims in each paper are presented in Table 1.

Table 1. Overview of the aims in the papers

Paper I

✓ To examine the association between ICT demands and job strain and effort-reward imbalance

✓ To evaluate the association between ICT demands and self-rated health

✓_{To evaluate the association between ICT demands and self-rated} health in socioeconomic groups

Paper II

✓ To examine the prospective association (including repeated measurement of exposure) between ICT demands and self-rated health

✓ To determine whether the prospective association between ICT demands and self-rated health is modified by sex or socioeconomic position

Paper III

✓ To compare the exposure to ICT demands in different occupational groups with regard to industry and position

Paper IV

✓ Describe healthcare managers’ experience of technostress and their actions for handling it

(30)

20

Methods and Material

Study design

The studies included in this thesis represent both quantitative and qualitative study designs (Table 2). Papers I and III have a quantitative, cross-sectional design and Paper II has a quantitative prospective design. Paper IV has a qualitative study design, using an inductive approach (Table 2).

Table 2. Overview of the study design in each paper

Paper Study design Study sample Analyses

I Quantitative,

cross-sectional SLOSH respondents (n=14 757) who

participated in 2014

Multiple binary logistic regression analysis, Pearson’s correlation coefficient analysis, ANOVA, Chi-square test

II Quantitative,

prospective SLOSH respondents (n= 4 468) who

participated in 2008, 2010 and 2012 or in 2010, 2012 and 2014

Multiple binary logistic regression analysis, Chi-square test

III Quantitative,

cross-sectional SLOSH respondents (n=13 572) who

participated in 2016

Multiple binary regression analysis, multiple linear regression analysis, ANOVA, Chi-square test

IV Qualitative,

(31)

21

Sampling and Participants

Sampling and Participants in the Swedish Longitudinal Occupational

Survey of Health

Papers I–III are based on data derived from the Swedish Longitudinal Occupational Survey of Health (SLOSH) and collected between 2006 and 2016 (Figure 4). SLOSH is a nationally representative survey of the working population in Sweden, with the overall aim of investigating associations between work organisation, work environment, labour force participation, retirement, and health and well-being in the working population in Sweden (81). It is an ongoing study using biennial questionnaires. The respondents of SLOSH were initially sampled by previous respondents of the Swedish Work Environment Surveys (SWES), who in turn were sampled from the Labour Force Survey (LFS), conducted by Statistics Sweden (9). Only respondents who filled in the ‘workers questionnaire’ (i.e. respondents occupied in paid work for at least 30% of the time during the last three months) in SLOSH have been included in the statistical analyses. The overall participation rate for SLOSH is 57.3%, but a slight decrease in the response rate between SLOSH 2006 (65.4%) and 2016 (50.9%) has been observed (81). A more comprehensive description of the cohort profile of SLOSH has been published elsewhere (81).

(32)

22 Figur e 4. Use of SL O SH da ta i n Pape rs I –III SL O SH 2006 SL O SH 2008 SL O SH 2020 SL O SH 2012 SL O SH 2014 SL O SH 2016 Paper II Paper II Paper II Paper II Paper I Paper II I Sam ple A: Baselin e Sam ple A: 2 nd m ea su rem en t Sam ple A: Fo llo w -up Sam ple B : Baselin e Sam ple B : 2 nd m ea su rem en t Sam ple B : Fo llo w -up

(33)

23

Occupational groups regarding socioeconomic position and industry

In Papers I and II, the exposure to ICT demands in different socioeconomic groups was examined. Additionally, the regression analyses in Papers I and II were either stratified by socioeconomic position or adjusted for socioeconomic position. Socioeconomic position was divided into the categories ‘low socioeconomic position’ (unskilled, semiskilled and skilled workers), ‘intermediate socioeconomic position’ (assistant and intermediate non-manual workers) and high socioeconomic position (employed and self-employed professionals, higher civil servants and executives). These categories were classified in line with Statistics Sweden’s manual of the socioeconomic index (82).

The exposure of different occupational groups with regard to industry and position was analysed in Paper III. Occupational group was categorised in line with Statistics Sweden’s standard classification of occupations from 2012 (83). This standard classification contains 10 major occupational groups, including ‘managers’, ‘occupations requiring advanced level of higher education’, ‘occupations requiring higher education qualifications or equivalent’, ‘administration and customer service clerks’, ‘service, care and shop sales workers’, ‘agricultural, horticultural, forestry and fishery workers’, ‘building and manufacturing workers’, ‘mechanical manufacturing and transport workers, etc.’, ‘elementary occupations’ and ‘armed forces occupations’. ‘Non-managers’ in Paper IV refers to all these occupational groups merged into one category (except for the manager category itself, obviously).

Managers as an occupational group

Managers’ exposure to ICT demands was focused on in Paper III. ‘Managers’ in this context refers to managers from different industries and in different hierarchical positions who participated in SLOSH. This occupational group includes seven subcategories of managers, divided by industry, including ‘politicians, managing directors and senior officials’, ‘managers in economics, HR, marketing, sales and other administration’, ‘managers in IT, logistics, R&D, real estate companies, construction, engineering, and manufacturing’, ‘managers in education’, ‘managers in healthcare and other community services’, ‘managers in banking, finance and insurance’, and ‘managers in

(34)

24

other service industries’. Additional information about these manager categories are found in Statistics Sweden’s standard classification of occupations from 2012 (83).

Sampling and participants in Paper IV

Paper IV is based on data derived from 20 semi-structured interviews with healthcare managers (10 men, 10 women) in various organisational positions, recruited from four hospitals spread over two county councils in Sweden. Strategic sampling was used to incorporate a rich variation of experiences and actions with regard to technostress among healthcare managers. The strategic sample included representatives with different professional backgrounds, hierarchical manager positions, control spans, experience in the management position, and sex (Paper IV, Table 1). Initially, potential informants were identified by four researchers who have a professional clinical background and/or previous cooperation with hospitals from which the participants were recruited. The author of this thesis invited the potential informants by email, including an information letter and a folder for informed consent, which were compulsory for participation. In the next step, based on the strategical selection criteria, new names were gathered either from the informants, or by contacts with persons in management positions at the in hospitals in the county councils. Healthcare managers who were interested in participating replied to this email. This process continued until 20 informants had been interviewed. The times and places of the interviews were booked about 1–3 months before they were to take place, and a few days before the interview, all informants received a reminder including short information and suggestions of how to prepare for the interview. All interviews were conducted by the author. Most of them (18 interviews) were conducted at the workplaces of the informants, one was conducted at the author’s institution, and one was conducted via telephone, at the request of the participant. The mean duration of the interviews was approximately 39 minutes (range 25–70 minutes).

(35)

25

Material

Measures

ICT Demands Index

The ICT demands at work index in SLOSH is mainly based on previous work by Johansson-Hidén et al. (84). This index can be viewed as an operationalisation of the techno-overload and techno-invasion dimensions, and partly of the techno-complexity dimension, in the broader concept of technostress (4) (Figure 5). The index is introduced as follows: ‘New technology and more flexible working conditions have changed working life for many people. Technology can be very helpful but is also conducive to new types of stress. Estimate to what extent you are stressed by…’. There follow six items2_{: ‘…too many calls and emails’, ‘…demands to be available on}

related issues during work hours’, ‘…demands to be available on work-related issues during leisure time’, ‘…demands to give immediate answers to emails and telephone calls that require a lot of work’, ‘…constant interruptions by the telephone and email’, and ‘…computers and other equipment that fails to work properly’ (See Appendix 1 in Paper III). Cronbach’s alpha of the ICT demands index was 0.79 at the 2016 data collection. The response options were rated on a Likert scale from 1 (very much) to 5 (I do not have access to this at work). The reversed median score of the full ICT demands index (2.83) was used as the cut-off value for high and low ICT demands (high ICT demands were defined as above the median) in Papers I–III. The ICT demands index was also analysed as a continuous measure in Paper I and Paper III.

2_{In SLOSH 2006, the ICT demands index consisted of only five items (‘demands to}

be available on work-related issues’ included both working hours and leisure time in one item) but in later data collections, these were split into two items.

(36)

26

Figure 5. Author’s view of the conceptual association between ICT demands and technostress

ICT demands

Technostress

Items in the ICT demands index

in SLOSH Dimensions of technostress creators

Techno-overload Stress related to too many telephone

calls and emails ICT that forces the employees to work faster and longer

Techno-invasion Stress related to demands to be

available on work-related issues during work hours

The invasive influence ICT has on employees due to the expectation of their high availability via ICTs, which may blur the transition between work and leisure time

Techno-complexity Stress related to demands to be

available on work-related issues during leisure time

The complexity related to work with ICT, causing a feeling of insufficient digital literacy and forcing employees to spend more time and effort learning new ICTs

Techno-insecurity Stress related to demands to give

immediate answers to emails and telephone calls that require a lot of

work

Situations where employees are afraid of losing their jobs, either to

people with better digital literacy, or to robotic process automation/artificial intelligence

Techno-uncertainty Stress related to constant

interruptions by telephone and email Constant evolution of new ICT implies that employees must

continually learn and educate themselves in this new technology Stress related to computers and

other equipment that fails to work properly

(37)

27

Self-rated health

General self-rated health was analysed as an outcome in Papers I and II and was measured by the one-item question “How would you rate your general state of health?” (66). The response options were rated on a Likert scale from 1 (very good) to 5 (very bad). In this thesis, responses of ‘very bad’, ‘rather bad’, or ‘neither good nor bad’ to the question were seen as ‘suboptimal’ self-rated health, and responses of ‘very good’ or ‘rather good’ were seen as ‘good’ self-rated health. Besides its usefulness as an indicator of current health status, self-rated health is also useful as a predictor of future health status and mortality (67).

Job strain

Job strain was measured by the demand-control questionnaire, comprising the dimensions of demand and control (85). The dimension of demand was covered by five items, e.g., ‘Do you have to work very intensively?’ and control was covered by six items, e.g., ‘Do you have a choice in deciding how you do your work?’ The response options were rated on a Likert scale from 1 (no, hardly ever) to 4 (yes, often). The population mean response scores for the demand and control dimensions were computed and used as cut-off values for high or low scores of demand and control (‘low demand’ was defined as equal to or below the median; and ‘high demand’ was defined as above the median, ‘low control’ was defined as below the median and ‘high control’ was defined as equal to or above the median) ‘Job strain’ was defined as a combination of high demand and low control, and was compared with all other combinations of the demand and control dimensions. The control and demand dimensions were analysed as continuous variables in Paper I, based on the individual mean values, in the respective dimensions.

Effort-reward imbalance

Effort-reward imbalance was measured by the short version of the effort-reward imbalance questionnaire, which comprised the dimensions of effort and reward (43). Effort was covered by three items, e.g. ‘I have constant time pressure due to a heavy workload.’ Reward was covered by seven items, e.g., ‘Considering all my efforts and achievements, I receive the respect and prestige I deserve at work.’ The response options were rated on a Likert scale from 1 (strongly disagree) to 4 (strongly agree). Effort-reward imbalance was

(38)

28

defined as a combination of high effort and low reward, calculated by dividing the effort dimension by the reward dimension and using the ratio of 1 as a cut-off value, with ratios strictly above 1 considered to be an effort-reward imbalance and ratios equal to or below 1 considered to show no imbalance (43). The effort and reward dimensions were also analysed as continuous variables in Paper I, based on the mean in the respective dimensions.

Interview Guide Paper IV

A short interview guide for Paper IV was developed in accordance with an interview guide applicable to the critical incident technique (86). The interview guide was tested in a pilot interview and then slightly modified after feedback from the pilot informant. It included the principal questions ‘Please describe in what way you are working with digital devices in your daily work’ (1), ‘Please tell me about different incidents and examples of when the work with digital devices and ICT has been demanding, problematic or frustrating in some way’ (2), ‘Thinking about each incident or situation separately, tell me more about how you handled them’ (3). Additional follow-up questions were asked in relation to the responses of the informant, in order to obtain a complete description of how the critical incidents were experienced and what actions were used to handle them. These follow-up questions included ‘Why was this experience critical for you?’ and ‘What were your thoughts during and after this experience?’ (86). The transcription of the data material was conducted simultaneously with the data collection, to enable evaluation and potential improvement of the interview technique. An illustration of the interview guide that influenced the interview guide in Paper IV is presented in Table 3.

(39)

29

Table 3. Author’s own illustration of an interview guide to the critical incident technique, according to Fridlund, Henricson & Mårtensson (86)

✓ Where did the critical incident happen?

✓ Can you give a detailed description of what happened? ✓ Why was the incident critical for you?

✓ What did you do in connection with the incident? ✓ What was your mindset during the incident?

✓ What were your thoughts during and after the incident? ✓ What were your feelings during and after the incident?

✓ What did you find was the most demanding aspect of the incident? ✓ What has this incident meant to you since?

Analyses

Statistical analyses

The data in the quantitative papers (Papers I–III) were analysed using statistical methods, including multiple binary logistic regression analysis, multiple linear regression analysis, Pearson’s correlation coefficient analysis, Chi-square test, independent t-test, and ANOVA. The regression models in Papers I and II were stratified by socioeconomic position, and/or sex. The association between ICT demands and suboptimal self-rated health (Papers I and II) was calculated using multiple binary logistic regression analysis. The association between occupational group and ICT demands was calculated with logistic regression analysis, and with multiple linear regression. All regression models were first calculated as crude analyses and with different combinations

(40)

30

of adjustments for potential confounders or modifying factors (see detailed description in each quantitative paper). Pearson’s correlation coefficient was used to calculate the correlation between ICT demands and the dimensions of the job strain and effort-reward imbalance models to determine potential conceptual overlap between the models. Differences in proportions or mean values with regard to socioeconomic position, sex or occupational group were calculated with Chi-square tests, independent t-test or ANOVA. The internal validity for the main measures was calculated by Cronbach’s alpha. Alpha values of <0.05 were considered statistically significant. IBM SPSS Statistics were used to calculate the results in Papers I–III.

Potential confounding or modifying factors

All regression analyses in Papers I-III were conducted as crude, unadjusted analyses, but also using models adjusted for different potential modifying or confounding factors. Many definitions of a confounding factor have been used in the literature, but common to the definition is that a confounding factor is correlated to both the independent and dependent variable (predictor and outcome variable in a regression model) and introduces a systematic bias in the result (87, 88). A modifying factor, on the other hand, is a factor that modifies the path between the independent and dependent variable (88). It is most often impossible to eliminate all potential confounding from other factors in an observational study (87), but in this thesis, efforts have been made to adjust the analyses to reduce the potential effect of confounders on the results. Adjustments have been made for sex, socioeconomic position, health behaviours (defined as smoking habits and physical exercise), BMI, job strain, effort-reward imbalance and social support. Of these variables, age, smoking, health behaviour BMI, job strain, effort-reward imbalance and social support were considered as potential confounding factors, since they correlated both with the dependent and independent variable, but also because of previously observed associations with self-rated health (52, 89-91). Moreover, job strain and effort-reward imbalance were included, to test interaction effects between ICT demands and job strain/effort-reward imbalance with regard to self-rated health. Sex and socioeconomic position were considered to be potentially modifying factors with regard to previous observations, indicating that the association between work-related stress and self-rated health is modified by sex as well as socioeconomic position (51, 52,

(41)

31

92). Further details about the adjustment for potential confounders or modifying factors are presented in each paper.

The critical incident technique

Paper IV was analysed using the critical incident technique, including an inductive approach. The technique aims to identify critical experiences and actions relevant for a specific purpose (86, 93, 94). Historically, the critical incident technique was developed in the US Air Force aviation program, during World War II, to reduce critical mistakes made by Air Force pilots (93). Since then the technique has been modified to fit additional contexts and also to incorporate psychosocial factors, such as in health and caring science and human-computer interaction (86, 94, 95). In its current use, the critical incident technique has been described as a flexible qualitative method, useful in solving practical problems (86, 94).

The transcribed data material from the interviews in Paper IV consisted of 185 pages (Times New Roman, 12 points, single line spacing). The data material was firstly read through several times by the author of this thesis, to obtain a sense of the whole. The co-authors of Paper IV then discussed the content in the data material with regard to the aim. Critical incidents involving experiences of technostress, and actions relating to how these incidents were handled, were identified and carefully compared, abstracted and labelled into subcategories, categories and main areas, in line with previous suggestions for the use of the critical incident technique (86). The data analysis process took place simultaneously with the data collection, and all co-authors of Paper IV were involved in the process and contributed complementary input to all parts of it.

(42)

32

Ethical considerations

Confidentiality and secrecy

All procedures performed in relation to this thesis were in accordance with the ethical standards of the Helsinki Declaration (96) or comparable ethical standards. One of the most important principles of this declaration is to conduct research in a way that protects the life, health, dignity, integrity, right to self-determination, privacy, and confidentiality of personal information of the research participants (96). In this thesis, all personal data were treated confidentially, and in line with the Swedish Personal Data Act (1998:204) (97) and GDPR ((EU) 2016/679) (98). The SLOSH data are stored in a large non-identifiable data file, and the security numbers are stored separately in safety cabinets (99). The personal data in Paper IV is stored in a safety cabinet and on a password and protected data domain with high security. The analyses in all papers included were conducted on non-identifiable data material, and all results in Papers I–III are presented on a group level. The quotes that illustrate the results in Paper IV are also non-identifiable, and the informants are referred to by numbers, e.g. ‘Informant 1’ and so on.

Approval(s) by the Swedish Ethical Review Authority

In line with Swedish law, those carrying out research that involves humans are obligated to apply for approval by the Swedish Ethical Review Authority. SLOSH has been approved by the Regional Ethical Review Board in Stockholm. Papers I–III have also received complementary approval by the Regional Ethical Review Board in Linköping #2014/355-31 and the Swedish Ethical Review Authority #2019-05767. Paper IV was approved by the Regional Ethical Review Board in Linköping #2017/597-31.

Accordance with medical-ethical principles

An overall principle in the Helsinki Declaration is to conduct research in which the benefits and importance clearly outweigh the risks (96). This principle is in accordance with medical-ethical principles, which state that the

(43)

33

research must respect ‘autonomy’ (i.e. autonomous behaviour in the frame of the law and social norms), ‘beneficence’ (i.e. altruistic behaviour), ‘non-maleficence’ (i.e. not inflict harm on others), and ‘justice’ (i.e. fair entitlement). These medical-ethical principles were first introduced in the Belmont Report in 1979 (100), and later modified from three to four principles (non-maleficence was introduced later) (101). The participants’ autonomy in this thesis was ensured by their voluntary participation (without economic compensation). Informed consent was compulsory for participation, and the participants were informed about the right to withdraw their participation without requirement to declare the reason. Since the research in the thesis is associated with very few risks for the participants (as they were required to answer rather conventional questions about the work environment), there is no imminent ethical dilemma that collides with the beneficence principle or its counterpart, non-maleficence (101). At the same time, there was a slight risk that negative feelings associated with the work environment or health could arise when the participants responded to the questions in the SLOSH questionnaire or in the interviews, which would have been in conflict with the non-maleficence principle (101). However, the potential societal benefits for the research in the thesis are considered to clearly outweigh the possible risks for the participants. The justice principle in the research context is often a matter of ‘distributive justice’ (i.e. equitable distribution of benefits, resources, privileges) (101). In this study, there is no specific benefit for the individual participants, other than possibly the feeling that someone is interested in their thoughts on and experiences of the digitalised work environment, or a positive feeling due to his/her contribution to research. However, since SLOSH is originally based on a randomisation, there is a just likelihood to be invited to participate in the study.

(44)

34

Results

Associations between ICT demands, job strain and

effort-reward imbalance

Based on calculation of the total study sample in SLOSH 2014 (n=14 757), the continuous measure of ICT demands was correlated with the dimensions of job strain and effort-reward imbalance. This was most pronounced in relation to the effort (r = 0.51 p ≤ 0.001) and job demand (r = 0.42; p ≤ 0.001) dimensions (Paper I, Table 2). Statistically significant differences in the proportions of the dichotomised measure of ICT demands and job strain and ICT demands and ERI were observed (Table 4). In the total study sample, 59.7% of those who reported high ICT demands also reported job strain, and 62.9% of those with high ICT demands reported effort-reward imbalance. Among those who reported low ICT demands, 40.3% reported job strain and 37.1% reported effort-reward imbalance. These proportions were different when the analysis was divided by socioeconomic position, and among those with high ICT demands, reports of job strain and effort-reward imbalance in the occupational groups with high and intermediate socioeconomic position were more prevalent than in the group with low socioeconomic position (Table 4).

(45)

35 N ot e. The co lum n pe rce nt age i s pr ese nt ed in the p ar en thes is for ca tego rica l var iab les , e.g. in t he t ot al st udy sa m pl e, ou t of tho se w ho repo rted h igh I CT de m ands , 59 .7% al so r epor ted job st ra in Tabl e 4. Chi -squ ar e t est for com par ison of p ropo rti ons of IC T dem ands in rel at ion to j ob s tra in a nd e ffo rt-reward i m bal anc e in t he t ot al st udy sa m pl e a nd i n pa rti ci pa nt s w ith low , i nt er m edi at e and h igh soc io ec ono m ic pos iti on ICT dem and s Jo b str ain E ffo rt -re wa rd im ba la nce To tal s tu dy sam ple, n (%) No str ain Jo b s train P-valu e* N o E RI ERI P-valu e* Lo w ICT d em and s 6346 (54. 0) 1186 (40. 3) <0 .001 4741 (65. 7) 2754 (37. 1) <0 .001 Hig h ICT d em and s 5408 (46. 0) 1757 (59. 7) 2474 (34. 3) 4673 (62. 9) Lo w SEP, n ( %) Lo w ICT d em and s 2124 (74. 0) 764 (63. 7) <0 .001 1641 (82. 2) 1235 (60. 0) <0 .001 Hig h ICT d em and s 747 (26. 0) 436 (36. 3) 355 (17. 8) 825 (40. 0) In ter m ed iate SEP , n (%) Lo w ICT d em and s 2785 (49. 9) 328 (24. 8) <0 .001 1978 (61. 2) 1125 (30. 8) <0 .001 Hig h ICT d em and s 2797 (50. 1) 993 (75. 2) 1255 (38. 8) 3530 (69. 2) Hig h SEP , n (%) Lo w ICT d em and s 1249 (42. 7) 69 (19. 3) <0 .001 972 (55. 6) 338 (22. 2) <0 .001 Hig h ICT d em and s 1677 (57. 3) 288 (80. 7) 775 (44. 4) 1187 (77. 8)

The digitalised work environment : Health, experiences and actions

The digitalised work environment:

Health, experiences and actions

The digitalised work environment:

Health, experiences and actions

Abstract

Original Papers

Table of Contents

Abbreviations

Introduction

Technostress

Background

The work environment

The digitalised work environment

Historical development of the digitalised work environment

ICT, administration and information overload

ICT and boundaryless work versus recovery

User experience versus usability

Health perspectives

Work, stress and health

Work-related stress, sex and socioeconomic position

Work-related stress and health

Self-rated health in the working population

Technostress and health

Theoretical framework

Job Demands-Resources model

Job strain and Iso-strain

Job demand

Control

Social support

Effort-reward imbalance

Technostress and ICT demands

Resources in the digitalised work environment

Rationale

Aim

Methods and Material

Study design

Sampling and Participants

Sampling and Participants in the Swedish Longitudinal Occupational

Survey of Health

Occupational groups regarding socioeconomic position and industry

Managers as an occupational group

Sampling and participants in Paper IV

Material

Measures

ICT Demands Index

ICT demands

Technostress

Self-rated health

Job strain

Effort-reward imbalance

Interview Guide Paper IV

Analyses

Statistical analyses

Potential confounding or modifying factors

The critical incident technique

Ethical considerations

Confidentiality and secrecy

Approval(s) by the Swedish Ethical Review Authority

Accordance with medical-ethical principles

Results

Associations between ICT demands, job strain and

effort-reward imbalance