There has been a longstanding debate in Norwegian work life concerning the amount of public resources spent on paid sick leave. Paid sick leave is a statutory right and employees are provided with full salary from the first day of sick leave. In the case of short-term illness, the employer is responsible for covering associated expenses (up to 16 days). However, if the illness is prolonged, the employee receives full salary for up to 1 year, in which compensation is provided by the Norwegian Labor and Welfare Administration. It is the employer who receives this compensation and pays the employee; thus, the employer has no further expenses after 16 days. To date, expenses associated with paid sick leave are considered high. Consequently, a central national and governmental objective has been to develop a strategy to reduce the general level of sickness absence (Norwegian Labor and Welfare Administration, 2012). This includes work adjustment so that employees can attend work despite having health conditions or impairments that otherwise would exclude them from working. In research, the phenomenon of employees going ill to work has been termed sickness presenteeism (e.g., Aronsson & Gustafsson, 2005).

In Norwegian work life, sickness presenteeism with adjustment opportunities could be regarded as a better option than sickness absenteeism. It could also be regarded as beneficial both for the individual and for the organization. However, this focus stands in stark contrast to a common perspective on sickness presenteeism, which regards working while ill as problematic because it can have negative consequences for individual productivity (Hemp, 2004; Schultz, Chen, & Edington, 2009). When sick, workers are not fully productive, so, being present may entail greater expenses than being absent. To our knowledge no earlier study has addressed sickness presenteeism in a context where governmental initiatives have been made to provide employees with adjustment opportunities. We aim to explore psychosocial workplace factors related to sickness presenteeism in a company participating in this governmental initiative. A second objective is to investigate whether sickness presenteeism shares antecedents with sickness absenteeism and long-term health.

Sickness presenteeism.
Sickness presenteeism commonly refers to situations in which employees, despite being ill, attend work (e.g., Aronsson & Gustafsson, 2005; Bergström, Bodin, Hagberg, Aronsson, & Josephson, 2009; Caverley, Cunningham, & MacGregor, 2007; Hansen & Andersen, 2008; Johns, 2010). This phenomenon has in recent years received increased attention in occupational health research, and a growing body of literature suggests that sickness presenteeism can impair employee health, leading to outcomes such as poor general health (Aronsson, Gustafsson, & Mellner, 2011; Bergström, Bodin, Hagberg, Lindh, et al., 2009; Gustafsson & Marklund, 2011), increased risk of heart disease (Kivimäki et al., 2005), and future sickness absenteeism (Bergström, Bodin, Hagberg, Aronsson, et al., 2009; Hansen & Andersen, 2009).

When investigating psychosocial workplace factors related to sickness presenteeism it is important to include known correlates. Studies have found that job demands such as a high level of responsibility (Gosselin, Lemyre, & Corneil, 2013), job stress (Elstad & Vabø, 2008; Leineweber et al., 2011), time pressure (Hansen & Andersen, 2008), and organizational change (Saksvik, 1996) are associated with sickness presenteeism. Perception of low control and limited support from colleagues also seem to be related to sickness presenteeism (Gosselin et al., 2013; Leineweber et al., 2011). Therefore, psychosocial workplace factors, such as job demands, job control, and social support are relevant when investigating sickness presenteeism.

Earlier research has shown that sickness presenteeism can be caused by high attendance pressure (Aronsson & Gustafsson, 2005; Kristensen, 1991; Saksvik, 1996) and identified four types of attendance pressure which induce employees to work while ill: importance pressure refers to worker indispensability and difficulties with finding a replacement in case of absence; censure pressure involves fear of being accused for shirking by management or colleagues; security pressure concerns job insecurity and fear of job loss if absent, and finally; moral pressure is a form of pressure related to employee’s own conscience (Saksvik, 1996). In line with research indicating that presenteeism is associated with poor health outcomes (Aronsson et al., 2011; Bergström, Bodin, Hagberg, Aronsson, et al., 2009; Bergström, Bodin, Hagberg, Lindh, et al., 2009; Gustafsson & Marklund, 2011; Hansen & Andersen, 2009; Kivimäki et al., 2005), it is likely that presenteeism resulting from such pressure negatively impacts employee health.

Studies focusing on sickness presenteeism resulting from more constructive or positive psychosocial workplace factors are scarce. Organizational adjustment is one such variable that are thought to be relevant to sickness presenteeism (Thun, Saksvik, Ose, Mehmetoglu, & Christensen, 2013). This assumption is based on research models summarizing variables influencing employee attendance (Johns, 2010; Steers & Rhodes, 1978). We define perceived organizational adjustment as the employee’s perception of management’s willingness to adjust the job so that it can be performed or to find other tasks that can be performed with the illness, without getting worse or, preferably, with improvements in health status (Biron & Saksvik, 2010; Thun et al., 2013). Johansson and Lundberg (2004) defined adjustment latitude as the opportunity employees have to reduce or change work effort. The findings showed that women with low adjustment latitude had more sickness absence after controlling for potential confounders including age, health, financial position, and family demands. However, adjustment latitude was not associated with sickness presenteeism (Johansson & Lundberg, 2004). Results from patients with chronic occupational musculoskeletal disorders who worked post-injury in a modified, adjusted work situation revealed that patients who were present returned to work faster than patients who had been absent, and they were more likely to resume normal work after one year (Howard, Mayer, & Gatchel, 2009). These results indicate that organizational adjustment may impact both employee presenteeism and absenteeism, as they are considered to be part of the same decision process (Aronsson & Gustafsson, 2005).

Sickness absenteeism and long-term health.
As employees who often are sickness absent also tend to be sickness present (Aronsson, Gustafsson, & Dallner, 2000; Gustafsson & Marklund, 2011), research on attendance behavior includes both phenomena (Aronsson & Gustafsson, 2005; Johns, 2010; Steers & Rhodes, 1978). A Canadian study found that employees were substituting presenteeism for absenteeism and that those who were sickness present did not suffer from different ailments and were not any less sick than those who were sickness absent (Caverley et al., 2007). It has been hypothesized that the ailments of the sickness present are milder than those resulting in sickness absenteeism (Hemp, 2004). However, the results of Caverley et al. (2007) indicate that some employees continue to work with health problems, despite the fact that full recovery normally requires sick leave. As such, low absence rates are not necessarily synonymous with low morbidity and healthy workers. This means that absence rates may be poor indicators of productivity. In general, this demonstrates the importance of including both sickness absenteeism and sickness presenteeism in research on employee health (Caverley et al., 2007).

While some employees get sick and must choose whether to work, research suggests that others seldom get sick and exhibit what has been called excellent work ability (Lindberg, Josephson, Alfredsson, & Vingard, 2006) or long-term health (Aronsson & Lindh, 2004). Employees classified as long-term healthy have attended work over a longer period without being sickness absent and without considering themselves as sickness present. Long-term healthy employees are thus considered to have generally good health (Aronsson & Lindh, 2004).

The primary aim of this study is to explore psychosocial workplace factors associated with sickness presenteeism in a Norwegian working life context. To obtain a more nuanced picture of employee attendance dynamics, we included analyses of sickness absenteeism and long-term health. By entering the same set of variables in three separate logistic regression analyses, we can identify whether the concepts share correlates. Based on previous research, we included psychosocial workplace factors, such as job demands, job control, social support, attendance pressure, perceived organizational adjustment, and restructuring experience, as predictors in the analyses.

• Hypothesis 1. Because research models of sickness presenteeism also include sickness absenteeism (Aronsson & Gustafsson, 2005; Johns, 2010; Steers & Rhodes, 1978), our first hypothesis was formulated: Factors associated with sickness presenteeism are also associated with sickness absenteeism.

• Hypothesis 2. Earlier research suggests that sickness presenteeism is positively associated with both attendance pressure (Saksvik, 1996) and perceived organizational adjustment (Howard et al., 2009; Johansson & Lundberg, 2004; Thun et al., 2013); hence, our second hypothesis was proposed: Perceived organizational adjustment and attendance pressure correlates positively with sickness presenteeism.

• Hypothesis 3. We defined long-term health as having had no sickness presenteeism and no sickness absenteeism during the last year. It is therefore reasonable to assume that long-term healthy employees have not faced situations where attendance pressure arose. This led to our third hypothesis: Attendance pressure correlates negatively with long-term health.

Method

Participants and procedures.
This study was conducted in a large industrial company involved in a government-funded program to reduce sick leave. The program was initialized in 2001 and offered Norwegian organizations economic support in providing organizational adjustment for employees with health conditions as a means of reducing sickness absenteeism (Norwegian Labor and Welfare Administration, 2012). The company’s registered sick leave had been at low levels for a long period of time. It was approximately 4%, compared to the national rate of 7.4%, in 2010. At the time of the study, the company had been involved in extensive restructuring and downsizing. Nevertheless, the sick leave rate had remained unchanged.

The company comprised roughly 740 employees, including operators, engineers, engineering scientists, and management. Operators constituted the majority of the employees and were engaged in various aspects of the production process. Their working conditions were physically demanding and involved noise, high temperatures, and heavy lifting. While most operators worked on a shift schedule involving night shifts and 12-hour shifts, engineering personnel and management tended to work during the daytime, and some had flexible work hours.

The data used were derived from a self-report questionnaire consisting of validated instruments collected in 2010. The study was conducted with the approval of the Norwegian Social Science Data Services and followed the ethical standards required. Questionnaires were distributed during working hours to employees present at the time.To involve all employees working on the various shifts, data collection was performed during 5 days. Respondents were given a brief presentation of the questionnaire and the study objective. Participation was voluntary and anonymity guaranteed. Of 686 questionnaires distributed, 477 were returned (69.5%).

Of the sample, 86.4% were male, 60.3% worked on a shift schedule, 30.5% worked during the daytime, and 8% reported working daytime with flexible hours. In all, 71.4% had worked at the company for more than 10 years, and 30.3% had management responsibility. Regarding age, 13.1% were 29 years or younger, 51.7% were between 30 and 49 years, 35.2% were 50 years or older.

Measurements.
Dependent variables. Sickness presenteeism was measured using the following question: «During the past 12 months, how many times did you go to work even though you should have taken sick leave?» Response categories were «I have not been sick the last 12 months,» «None,» «Once,» «2-5 times,» and «More than 5 times.» The measure was developed by Aronsson et al. (2000) and corresponds with current research practice (Aronsson & Gustafsson, 2005; Aronsson et al., 2011; Bergström, Bodin, Hagberg, Aronsson, et al., 2009; Bergström, Bodin, Hagberg, Lindh, et al., 2009; Claes, 2011; Gosselin et al., 2013; Gustafsson & Marklund, 2011, 2014; Hansen & Andersen, 2008, 2009; Leineweber et al., 2011). Demerouti, Le Blanc, Bakker, Schaufeli, and Hox (2009) performed test-retest reliability for this question and reported a value of 0.58 (p < .01) or greater for 6- and 12-month intervals. Sickness presenteeism was defined as reporting two or more incidents during the last 12 months,which corresponds with the cut-off used in earlier research (Aronsson & Gustafsson, 2005; Aronsson et al., 2000; Gustafsson & Marklund, 2011; Leineweber et al., 2011).

Sickness absenteeism was measured using the question, «How many days in the last 12 months have you been absent from work because of sick leave?» Response categories were «None,» «Five days or less,» «6-10 days,» «11-23 days,» and «More than 24 days.» Sickness absenteeism was operationalized as having taken sick leave for 6 days or more during the last 12 months. The cut-off corresponds with that used in earlier studies, where sickness absenteeism was operationalized as having taken sick leave for one week or more (Gustafsson & Marklund, 2011, 2014).

Long-term health was defined as a combination of low absenteeism and low presenteeism (Aronsson & Lindh, 2004). We operationalized long-term health more rigorously than those conducting earlier research, given the company’s low sick leave rates. Respondents categorized as long-term healthy had no sickness absenteeism (i.e., responded «None») in combination with no sickness presenteeism (i.e., responded «I have not been sick the last 12 months» or «None») in a 12-month period.

Independent variables. The demand, control, and support dimensions were measured using items from the job content questionnaire (Karasek, 1985), responses to which were given on a 5-point scale ranging from «very seldom» (1) to «very often» (5). Job demands were measured using five items that gauged how often respondents had to work fast and with short deadlines, among other factors (α = .74). Job control was measured using four items including how often respondents had the opportunity to influence decisions at work (α = .80). Job support was measured using three items (e.g., «How often do you get support from your co-workers?» α = .67).

Attendance pressure was measured using a 13-item scale (Saksvik, 1996) that measured four types of attendance pressure: importance pressure (four items, e.g., «I go to work despite being ill because it is hard to get hold of a substitute,» α = .76), censure pressure (three items, e.g., «I am afraid of being accused of shirking by my colleagues/management,» α = 72), security pressure (two items, e.g., «One should be glad to have a job today when so many are unemployed,» α = .61), and moral pressure (four items, e.g., «I have a guilty conscience if I stay at home,» α = .68). Responses were given on a 5-point scale ranging from «completely disagree» (1) to «completely agree» (5).

The scale used to obtain information about perceived organizational adjustment consisted of seven items and gauged respondents’ perceptions of workplace norms concerning adjustment (Hammer, Saksvik, Nytrø, Torvatn, & Bayazit, 2004; Ose et al., 2009; Thun et al., 2013) with statements such as «Employees with health problems get the help and support they need to do their job» and «In this workplace it is taken into consideration that different health problems may demand different arrangements,» (α = .82). Responses were given on a 5-point scale ranging from «completely disagree» (1) to «completely agree» (5).

The scale used to obtain information about perceived organizational adjustment consisted of seven items and gauged respondents’ perceptions of workplace norms concerning adjustment (Hammer, Saksvik, Nytrø, Torvatn, & Bayazit, 2004; Ose et al., 2009; Thun et al., 2013) with statements such as «Employees with health problems get the help and support they need to do their job» and «In this workplace it is taken into consideration that different health problems may demand different arrangements,» (α = .82). Responses were given on a 5-point scale ranging from «completely disagree» (1) to «completely agree» (5).

Perception of the restructuring was measured using a short version of the Change Process Healthiness Index (Tvedt, Saksvik, & Nytrø, 2009), which consists of 15 items concerning various aspects of reorganization (e.g., «Management has tried to get every view of this reorganization out in the open,» α = .85). Responses were given on a 5-point scale ranging from «completely disagree» (1) to «completely agree» (5).

Control variables. Control variables in the analyses were sex, age, seniority, working hours and subjective health. Sex was dichotomized (men = 1; women = 0), while age was entered as a grouped variable with the intervals «29 years or younger,» «30–49 years,» and «50 years or older.» The category «29 years or younger» was used as the reference category. Seniority was defined as having worked in the organization for more than 10 years. Working hours included three categories: «working daytime,» «working daytime with flexible hours,» and «working shift schedules», the last of these serving as the reference category. Information about subjective health was obtained using the question: «How would you generally describe your health?» Response categories ranged from «very good» (1) to «very bad» (5). The measure was dichotomized in accordance with earlier research (very good/good = 1; neither/bad/very bad = 0; Bergström, Bodin, Hagberg, Lindh, et al., 2009).

Analyses.
Scale reliability was tested using Cronbach’s alpha. Because variables were ordinal, Goodman and Kruskal’s gamma was used to estimate the correlations between them. Three separate logistic regression analyses were conducted using the same statistical model. The variables were entered into each model simultaneously, and the results were obtained as odds ratios with 95% confidence intervals.

Results

Descriptive statistics of the dependent variables are presented in Table 1. There was a 14.5% overlap between sickness presenteeism and sickness absenteeism. We found that only 33.1% of respondents classified as long-term healthy answered that they had not been sick. Accordingly, 66.9% of the long-term healthy did not report being without health problems or illness the last year.

Table 2 presents the inter-correlations for all variables in the analyses. Results from the logistic regression analyses are presented in TableDescriptive statistics of the dependent variables are presented in Table 1.3. In support of hypothesis 1, censure pressure (presenteeism OR = 1.46; absenteeism OR = 1.48) and subjective good health (presenteeism OR = 0.22; absenteeism OR = 0.40) significantly predicted both sickness presenteeism and sickness absenteeism. In addition, sickness absenteeism was associated with experiencing significantly less moral pressure (OR = 0.65), and sickness presenteeism was negatively associated with working daytime with flexible hours (OR = 0.25).

Partial support was found for hypothesis 2. Sickness presenteeism was positively associated with one type of attendance pressure which involves fear of being accused of shirking; censure pressure (OR = 1.46). The other three types of attendance pressure (important pressure, security pressure, and moral pressure) and perceived organizational adjustment were not significantly associated with sickness presenteeism.

Hypothesis 3, which states that long-term healthy employees do not experience attendance pressure, was not supported. Even though long-term health was negatively associated with censure pressure (OR = 0.70), it was positively associated with moral pressure (OR = 1.45). In addition, long-term health was associated with higher levels of perceived organizational adjustment (OR = 1.45), and with working during the daytime with flexible hours (OR = 2.72).

Discussion

The primary aim of the present study was to explore psychosocial workplace factors associated with sickness presenteeism in a Norwegian work life context, where governmental initiatives have been taken to reduce sickness absence. In addition, this study investigated if the psychosocial workplace factors associated with sickness presenteeism was also associated with sickness absenteeism and long-term health. After entering sickness presenteeism, sickness absenteeism, and long-term health as dependent variables in the same regression model, we compared the results and found all to be associated with one common psychosocial workplace factor: censure pressure.

The results partially support hypothesis 1, proposing that sickness presenteeism and sickness absenteeism share common antecedents (Aronsson & Gustafsson, 2005; Johns, 2010; Steers & Rhodes, 1978). The only common psychosocial workplace factor found was censure pressure. Contrary to hypothesis 2, employees exhibiting sickness presenteeism did not perceive more organizational adjustment than other employees. These findings suggest that sickness presenteeism was not resulting from organizational adjustment, but rather from fearing negative reactions from colleagues and management. In addition, sickness presenteeism was associated with reporting poorer subjective health.

Previous research has shown that censure pressure may be regarded as the most serious pressure factor from an individual health perspective (Saksvik, 1996), and suggested that sickness presenteeism is associated with future poor health and long-term sickness absenteeism later on (Bergström, Bodin, Hagberg, Aronsson, et al., 2009; Kivimäki et al., 2005). The combination of high censure pressure and poor health may, therefore, represent a problem for employees who often choose presenteeism. We recognize that how employees perceive attendance pressure may relate to personal characteristics and position. In this study, personality-related factors and position were not included. However, such variables may be interesting to explore in future research.

Hypothesis 3, proposing a negative association between long-term health and attendance pressure, was not supported. Those employees who were classified as long-term healthy seemed to experience an internal form of attendance pressure related to their own conscience. At the same time, they reported less censure pressure. Moreover, employees classified as long-term healthy reported higher levels of perceived organizational adjustment compared to other employees. Results also revealed that 66.9% of the long-term healthy did not report being without health problems, which indicates that, they may very well have been sick during the last year. Nevertheless, they reported neither sickness presenteeism nor sickness absenteeism. Perceiving more organizational adjustment may, therefore, alter employees’ perception of working while ill. When employees with health problems have the opportunity to either adjust their work themselves or receive help from management and colleagues to restructure the work situation, they may no longer think of themselves as going to work ill. Employees may have this perception because their health problems do not hinder their work and, more importantly, work does not influence their health problems. Accordingly, employees who perceive constructive adjustment might not report sickness presenteeism. To capture such nuances, more refined measures are needed. Furthermore, the results imply that employees do not have to be healthy in medical terms to be classified as long-term healthy. However, compared to other employees, those classified as long-term healthy were associated with reporting subjective good health.

Perceived organizational adjustment may be related to the type of health problem or illness employees’ experience. Some illnesses require medical treatment and/or rest that make it impossible to go to work, while other illnesses may have less severe health implications and thus adjustments can be more easily made. Furthermore, type of adjustments and employees’ perception of these can be associated with job position. The current sample comprised employees with various job positions involving different work tasks and work hours. However, variables in the analyses considered only work hours. Compared to employees reporting sickness presenteeism and sickness absenteeism, those classified as long-term healthy were more likely to work during the daytime with flexible hours. Consequently, this finding implies that employees with long-term health had the opportunity to leave earlier when sick and to regain lost hours later on. The fact that engineers and managers tended to work during the daytime with flexible hours and that operators tended to work in shifts, suggests that job position may be related to long-term health.

No significant association was found between the demand, control, and support dimensions and the dependent variables. This finding contradicts earlier studies (Aronsson & Gustafsson, 2005; Biron, Brun, Ivers, & Cooper, 2006; Hansen & Andersen, 2008). One explanation may be the relevance of the items in this particular sample. Bakker and Demerouti (2007) argued that each occupation has unique risk factors associated with job-related demands and resources. We included both employees working at an operational level with physically demanding work tasks and those engaged in office-related work. The scale items may therefore have been more relevant for some employees than for others. Future research may gain from including variables that are more occupational specific according to the sample.

Furthermore, no relationship was found between the restructuring experience and the dependent variables. This is in contrast with earlier research, finding that during times of insecurity, employees might go to work while ill as a result of security pressure (Saksvik, 1996). However, though the company was involved in extensive restructuring and downsizing, none of the dependent variables were associated with security pressure. A possible explanation is that in times of insecurity, various forms of pressure arise depending on the fluctuations in the labor market. For example, the unemployment rate in Norway is considered to be low in comparison with other countries (Organization for Economic Co-operation and Development, 2014). Security pressure may therefore be more relevant in countries where the unemployment rate is higher and social benefits are lower. It may be that the employees were more sensitive to what management and colleagues would think if they were absent from work (i.e., censure pressure), instead of fearing job loss (i.e., security pressure).

Limitations.
Our finding that there was an overlap between sickness presenteeism and sickness absenteeism is consistent with previous research (Bergström, Bodin, Hagberg, Aronsson, et al., 2009; Caverley et al., 2007; Hansen & Andersen, 2008), indicating that employees exhibiting sickness presenteeism also tend to exhibit sickness absenteeism. It is important to recognize this result; if we were to use mutually exclusive variables, the results would not be generalizable in real work settings. Nevertheless, overlapping variables may be regarded as a limitation because some employees then are included in two analyses. However, we argue that employees tend to engage in both, and it would be unsatisfactory to conduct research that did not allow overlap. In fact, they are mutual alternatives when an employee is sick (Aronsson & Gustafsson, 2005), so an overlap should be expected.

Moreover, the measures of sickness presenteeism and sickness absenteeism were collected via self-report; therefore recall bias may be a limitation. However, to date there are no alternatives for measuring sickness presenteeism (Claes, 2011). Our measure corresponds to that of previous research showing high test-retest reliability (>.58; Demerouti et al., 2009), and the sickness presenteeism rate of our sample (37.9%) was comparable to that of other studies with the same cut-off, whose values ranged from 37% to 53% (Aronsson & Gustafsson, 2005; Aronsson et al., 2000; Gustafsson & Marklund, 2011).

It is regarded a strength if register data for sickness absenteeism could be used on an individual level. However, we could not obtain these data from the company because of ethical restrictions. Nevertheless, self-report of sickness absenteeism has proven consistent with register data (Gustafsson & Marklund, 2011; Voss, Stark, Alfredsson, Vingard & Josephson, 2008).

In our sample, 26.4% of respondents were categorized as long-term healthy, but our conceptualization of long-term health was stricter than that used in earlier studies. However, the sample rate is comparable to earlier research. In Aronsson and Lindh (2004) 28% of respondents were classified as long-term healthy. As a consequence of the strict criterion for defining long-term health, it is important not to consider the remainder of the sample as sick. We found that employees classified as long-term healthy perceived their own health as better than other employees. The results revealed that long-term healthy employees may very well experience illness or health issues, but they also have the opportunity to adjust work according to their health condition. Arguably, the opportunity to adjust work may be among the differences between those classified as long-term healthy and those who were not.

Implications for future research.
This study contributes to the research field by emphasizing the need for more refined measures. Results revealed that not all employees without sickness presence and absence (i.e., the long-term healthy) report being without illness or health problems. This could mean that some of those who were categorized as long-term healthy have engaged in sickness presenteeism without reporting it. We argue that this may be a result of the question asked; that is, if the respondents have gone to work even though they should have taken sick leave. Furthermore, the long-term healthy reported higher levels of perceived organizational adjustment and were positively associated with working daytime with flexible hours. Sickness presenteeism, on the other hand, was not associated with perceived organizational adjustment and was negatively associated with working daytime with flexible hours. Therefore, adjustment opportunities might have affected the perception of under which circumstances employees worked while ill. Future research should, therefore, focus on further development of measures of sickness presenteeism that capture the diversity in employees’ perceptions of working through illness.

We argue that sickness presenteeism is associated with employees’ perception of organizational adjustment and attendance pressure. The phenomenon of sickness presenteeism can be more complex than previously presumed. By providing employees with adjustment opportunities in form of increased flexibility, in combination with preventing attendance pressure factors in the work environment, may actually influence employees’ perceptions of attending work while ill. As such, it is plausible that such measures may contribute to more constructive forms of sickness presenteeism that do not entail negative implications for employee health. An important future research objective is therefore to further explore factors associated with positive and negative health outcomes related to sickness presenteeism. This may enhance our understanding of sickness presenteeism and organizational factors that stimulate a healthy work environment. Norway may be regarded as somewhat unconventional, considering that the government actually supports sickness presenteeism to reduce the cost of absenteeism. Nevertheless, this makes Norway an interesting example. Perspectives regarding sickness presenteeism as productivity loss may also gain from research focusing on the distinction between constructive and detrimental sickness presenteeism by acknowledging that employees may be productive while sick if they are provided suitable organizational adjustment.

Conclusion

This study confirms that sickness presenteeism, sickness absenteeism, and long-term health are associated with common psychosocial workplace factors that include attendance pressure. However, the results also indicate that the phenomena have separate antecedents and stress the complexity of sickness presenteeism. Capturing such complexities needs more refined measures of sickness presenteeism. To contribute to further development of these measures, it is important to recognize that an important aspect of attendance dynamics seems to be how employees perceive attendance pressures and organizational adjustment. Opportunities to adjust work tasks may again be related to occupational factors and position, which constitutes an interesting area for future research.

References

Aronsson, G., & Gustafsson, K. (2005). Sickness presenteeism: Prevalence, attendance-pressure factors, and an outline of a model for research. Journal of Occupational and Environmental Medicine, 47(9), 958–966. doi: 10.1097/01.jom.0000177219.75677.17.

Aronsson, G., Gustafsson, K., & Dallner, M. (2000). Sick but yet at work. An empirical study of sickness presenteeism. Journal of Epidemiology and Community Health, 54(7), 502–509. doi: 10.1136/jech.54.7.502.

Aronsson, G., Gustafsson, K., & Mellner, C. (2011). Sickness presence, sickness absence, and self-reported health and symptoms. International Journal of Workplace Health Management, 4(3), 228–234. doi: 10.1108/17538351111172590.

Aronsson, G., & Lindh, T. (2004). Långtidsfriskas arbetsvillkor: En populationsstudie. [Work conditions among workers with good long term health : A population study] Arbeta och Hälsa, Vetenskaplig Skriftserie (Vol. 10). Retrieved from https://gupea.ub.gu.se/handle/2077/4332.

Bakker, A. B., & Demerouti, E. (2007). The Job Demands-Resources model: state of the art. Journal of Managerial Psychology, 22(3), 309–328. doi: 10.1108/02683940710733115.

Bergström, G., Bodin, L., Hagberg, J., Aronsson, G., & Josephson, M. (2009). Sickness presenteeism today, sickness absenteeism tomorrow? A prospective study on sickness presenteeism and future sickness absenteeism. Journal of Occupational and Environmental Medicine, 51(6), 629–638. doi: 10.1097/JOM.0b013e3181a8281b.

Bergström, G., Bodin, L., Hagberg, J., Lindh, T., Aronsson, G., & Josephson, M. (2009). Does sickness presenteeism have an impact on future general health? International Archives of Occupational and Environmental Health, 82(10), 1179–1190. doi: 10.1007/s00420-009-0433-6.

Biron, C., Brun, J. P., Ivers, H., & Cooper, C. L. (2006). At work but ill: psychosocial work environment and well-being determinants of presenteeism propensity. Journal of Public Mental Health, 5(4), 26–37. doi: 10.1108/17465729200600029.

Biron, C., & Saksvik, P. Ø. (2010). Sickness presenteeism and attendance pressure factors: Implications for practice. In C. L. Cooper, J. C. Quick, & M. Schabracq (Eds.), Work and Health Psychology: The Handbook (3rd Ed., pp. 77–96). London: John Wiley & Sons.

Caverley, N., Cunningham, J. B., & MacGregor, J. N. (2007). Sickness presenteeism, sickness absenteeism, and health following restructuring in a public service organization. Journal of Management Studies, 44(2), 304–319. doi: 10.1111/j.1467-6486.2007.00690.x.

Claes, R. (2011). Employee correlates of sickness presence: A study across four European countries. Work and Stress, 25(3), 224–242. doi: 10.1080/02678373.2011.605602.

Demerouti, E., Le Blanc, P. M., Bakker, A. B., Schaufeli, W. B., & Hox, J. (2009). Present but sick: a three-wave study on job demands, presenteeism and burnout. Career Development International, 14(1), 50–68. doi: 10.1108/13620430910933574.

Elstad, J. I., & Vabø, M. (2008). Job stress, sickness absence and sickness presenteeism in Nordic elderly care. Scandinavian Journal of Public Health, 36(5), 467–474. doi: 10.1177/1403494808089557.

Gosselin, E., Lemyre, L., & Corneil, W. (2013). Presenteeism and absenteeism: differentiated understanding of related phenomena. Journal of occupational health psychology, 18(1), 75–86. doi: 10.1037/a0030932.

Gustafsson, K., & Marklund, S. (2011). Consequences of sickness presence and sickness absence on health and work ability: a Swedish prospective cohort study. International Journal of Occupational Medicine and Environmental Health, 24(2), 153–165. doi: 10.2478/s13382-011-0013-3.

Gustafsson, K., & Marklund, S. (2014). Associations between health and combinations of sickness presence and absence. Occupational Medicine (London), 64(1), 49–55. doi: 10.1093/occmed/kqt141.

Hammer, T. H., Saksvik, P. Ø., Nytrø, K., Torvatn, H., & Bayazit, M. (2004). Expanding the psychosocial work environment: Workplace norms and work-family conflict as correlates of stress and health. Journal of occupational health psychology, 9(1), 83–97. doi: 10.1037/1076-8998.9.1.83.

Hansen, C. D., & Andersen, J. H. (2008). Going ill to work ­– What personal circumstances, attitudes and work-related factors are associated with sickness presenteeism? Social Science & Medicine, 67(6), 956–964. doi: 10.1016/j.socscimed.2008.05.022.

Hansen, C. D., & Andersen, J. H. (2009). Sick at work-a risk factor for long-term sickness absence at a later date? Journal of Epidemiology and Community Health, 63(5), 397–402. doi: 10.1136/jech.2008.078238.

Hemp, P. (2004). Presenteeism: At work – but out of it. Harvard Business Review, 82(10), 49–58.

Howard, K. J., Mayer, T. G., & Gatchel, R. J. (2009). Effects of presenteeism in chronic occupational musculoskeletal disorders: Stay at work is validated. Journal of Occupational and Environmental Medicine, 51(6), 724–731. doi: 10.1097/JOM.0b013e3181a297b5.

Johansson, G., & Lundberg, I. (2004). Adjustment latitude and attendance requirements as determinants of sickness absence or attendance. Empirical tests of the illness flexibility model. Social Science & Medicine, 58(10), 1857–1868. doi: 10.1016/s0277-9536(03)00407-6.

Johns, G. (2010). Presenteeism in the workplace: A review and research agenda. Journal of Organizational Behavior, 31(4), 519–542. doi: 10.1002/job.630.

Karasek, R. A. (1985). Job Content Questionnaire. Lowell: University of Massachusetts Lowell, Department of Work Environment.

Kivimäki, M., Head, J., Ferrie, J. E., Hemingway, H., Shipley, M. J., Vahtera, J., & Marmot, M. G. (2005). Working while ill as a risk factor for serious coronary events: The Whitehall II study. American Journal of Public Health, 95(1), 98–102. doi: 10.2105/ajph.2003.035873.

Kristensen, T. S. (1991). Sickness absence and work strain among danish slaughterhouse workers: An analysis of absence from work regarded as coping behavior. Social Science & Medicine, 32(1), 15–27. doi: 10.1016/0277-9536(91)90122-s.

Leineweber, C., Westerlund, H., Hagberg, J., Svedberg, P., Luokkala, M., & Alexanderson, K. (2011). Sickness presenteeism among Swedish police officers. Journal of Occupational Rehabilitation, 21(1), 17–22. doi: 10.1007/s10926-010-9249-1.

Lindberg, P., Josephson, M., Alfredsson, L., & Vingard, E. (2006). Promoting excellent work ability and preventing poor work ability: the same determinants? Results from the Swedish HAKuL study. Occupational and Environmental Medicine, 63(2), 113–120. doi: 10.1136/oem.2005.022129.

Norwegian Labor and Welfare Administration. (2012). Working in an IW enterprise. How does this affect me? Retrieved from: http://nav.no/no/Bedrift/Inkluderende+arbeidsliv/Relatert+informasjon/Informasjonsmateriell+-+Inkluderende+arbeidsliv.397078.cms.

Organization for Economic Co-operation and Development. (2014). OECD Skills strategy diagnostic report: Norway. OECD Publishing.

Ose, S. O., Bjerkan, A. M., Pettersen, I., Hem, K. G., Johnsen, A., Lippestad, J., Paulsen, B., Mo, T. O., & Saksvik, P. Ø. (2009). Evaluering av IA- avtalen [Evaluation of the IWL-agreement] (2001–2009). Trondheim: SINTEF. Retrieved from: http://seniorpolitikk.no/var/uploaded/images/Forskrappnyhet/IA-avtalen2009.pdf.

Saksvik, P. Ø. (1996). Attendance pressure during organizational change. International Journal of Stress Management, 3(1), 47–59. doi: 10.1007/bf01857888.

Schultz, A. B., Chen, C.-Y., & Edington, D. W. (2009). The cost and impact of health conditions on presenteeism to employers: a review of the literature. Pharmacoeconomics, 27(5), 365–378. doi: 10.2165/00019053-200927050-00002.

Steers, R. M., & Rhodes, S. R. (1978). Major influences on employee attendance: A process model. Journal of Applied Psychology, 63(4), 391–407. doi: 10.1037/0021-9010.63.4.391.

Thun, S., Saksvik, P. Ø., Ose, S. O., Mehmetoglu, M., & Christensen, M. (2013). The impact of supervisors’ attitudes on organizational adjustment norms and attendance pressure norms. Scandinavian Journal of Organizational Psychology, 5(2), 15–31.

Tvedt, S. D., Saksvik, P. Ø., & Nytrø, K. (2009). Does change process healthiness reduce the negative effects of organizational change on the psychosocial work environment? Work and Stress, 23(1), 80–98. doi: 10.1080/02678370902857113.

Voss, M., Stark, S., Alfredsson, L., Vingard, E., & Josephson, M. (2008). Comparisons of self-reported and register data on sickness absence among public employees in Sweden. Occupational and Environmental Medicine, 65(1), 61–67. doi: 10.1136/oem.2006.031427.