key: cord-0697695-lqmqnuor
authors: Ose, Solveig Osborg; Kaspersen, Silje Lill; Leinonen, Taina; Verstappen, Suzanne; de Rijk, Angelique; Spasova, Slavina; Hultqvist, Sara; Nørup, Iben; Pálsson, Jón R.; Blume, Andreas; Paternoga, Mike; Kalseth, Jorid
title: Follow-up regimes for sick-listed employees: a comparison of nine north-western European countries
date: 2022-05-10
journal: Health Policy
DOI: 10.1016/j.healthpol.2022.05.002
sha: 4aa107815d740904563e67713793d36e85e4cef0
doc_id: 697695
cord_uid: lqmqnuor

The Covid-19 pandemic has revealed the importance of social protection systems, including income security, when health problems arise. The aims of this study are to compare the follow-up regimes for sick-listed employees across nine European countries, and to conduct a qualitative assessment of the differences with respect to burden and responsibility sharing between the social protection system, employers and employees. The tendency highlighted is that countries with shorter employer periods of sick-pay typically have stricter follow-up responsibility for employers because, in practice, they become gatekeepers of the public sickness benefit scheme. In Germany and the UK, employers have few requirements for follow-up compared with the Nordic countries because they bear most of the costs of sickness absence themselves. The same applies in Iceland, where employers carry most of the costs and have no obligation to follow up sick-listed employees. The situation in the Netherlands is paradoxical: employers have strict obligations in the follow-up regime even though they cover all the costs of the sick-leave themselves. During the pandemic, the majority of countries have adjusted their sick-pay system and increased coverage to reduce the risk of spreading Covid-19 because employees are going to work sick or when they should self-quarantine, except for the Netherlands and Belgium, which considered that the current schemes were already sufficient to reduce that risk.

The vast majority of Organisation for Economic Co-operation and Development (OECD) countries have statutory paid sick-leave systems (Raub et al., 2018, p. 3). However, national sick-pay policies vary enormously across countries and much could be learned from comparative research. Comparing sick-leave rates is a complex task that requires a clear understanding of the interplay between statutory, corporate and private forms of income protection during sickness absence (Clasen, 2017 (Clasen, , p. 1101 ) Furthermore, international comparisons of sickness absence rates have long been extremely difficult because there are substantial differences in the legitimation of work incapacity, level of sick-pay and criteria for transfer to invalidity insurance (Prins & De Graaf, 1986, p. 536) .

Some early studies that used data from the Luxembourg Employment Study are available. For example, it was reported that many of the differences in the total rates of sickness absence between 11 European countries were not explained by the distribution of the characteristics of sex, age and skill level (Bliksvaer & Helliesen, 1997, p. 26) . It was also found that the sex difference in absence rates resulted from differences in the age structures of the male and female workforce and their marital status (Barmby, Ercolani, & Treble, 2002, p. 325) .

Studies using data from the European Working Conditions Surveys provide additional results. One of the first studies noted that before any clear comparisons can be drawn, data from EU countries must first be made comparable in terms of the definition of sickness absence (harmonising), which should be established by changes in sickness absence legislation in the different countries (Gimeno, Benavides, Benach, & Amick, 2004, p. 869) . Elsewhere, it was noted that the number of absence days was not influenced by employment protection, absenteeism was increased by sickness benefits and the impact of the institutional framework was lower than that of some individual worker characteristics (Frick & Malo, 2008, p. 505) .

Conditions provide general findings about the different welfare state regimes.

One study compared 26 European countries and found lower absolute and relative social inequalities in sickness in comprehensive welfare states and more favourable general rates of non-employment ( Van der Wel, Dahl, & Thielen, 2011) . Thus, welfare resources appear to be more important than welfare disincentives with respect to sickness. Moreover, the welfare regime in Scandinavian countries is more able than others to protect against nonemployment when individuals are faced with illness; this is particularly true for those with low educational levels ( Van der Wel, Dahl, & Thielen, 2012) . In general, systems that activate rather than focus only on income protection are found to have better gatekeeping towards disability pensions (Anema, Prinz, & Prins, 2013; Anema et al., 2009) .

The European economic literature on sickness absence stresses the importance of employment and adequate working conditions for the health of workers and labour market policies that focus on job sustainability and job satisfaction (Barnay, 2016, p. 693) . A recent review of the literature and analysis of EU Labour Force Survey data from 13 European countries concluded that the essential problem of sick-pay insurance is one of moral hazard, given that health is largely continuous and non-observable (Palme & Persson, 2020, p. 104) .

A key report from the EU Commission shows that sick-pay schemes have changed in almost all EU countries over the last 20 years; most have changed in the direction of fewer rights for employees in the form of waiting periods, lower coverage and shorter periods that provide sick-pay rights (Spasova, Bouget, & Vanhercke, 2016 ).

Thus, cross-national comparative and collaborative research on sickness absence systems and statistics in Europe is desirable not only to advance knowledge about return-to-work policies and practices, but also to improve them (Gimeno et al., 2014, p. 4) . Detailed understanding of sickness benefit and sick-pay schemes is needed to elucidate cross-country differences in sick-leave rates. The schemes also involve legal obligations for different actors related to following up sick-listed employees, although these obligations are often overlooked in comparative studies, which hampers insight in how the burden of sickness is shared in different systems (Barmby et al., 2002; Bryson & Dale-Olsen, 2019; Heymann, Rho, Schmitt, & Earle, 2010) .

In the same way that patients diagnosed with a disease receive follow-up care, sick-listed employees are followed up by their employer, the health services and their insurance provider. However, the legislation for this follow-up differs across countries. The aim herein was to analyse these differences and similarities to gain knowledge about burden sharing between different system actors.

Here, we define each country's system-the regulations and formal obligations imposed on the involved actors-as the ‗follow-up regime' for sick-listed employees. Although sickness absence management is a term that has previously been used for employers, it does not cover the obligations of the health services and insurance actors.

In recent years, several of the Nordic countries (Sweden, Denmark, Finland, Norway and Iceland) have implemented initiatives based on a close follow-up of sick-listed employees; our goal was to compare these regimes with those in other European countries. An examination of the follow-up regimes for sicklisted employees is warranted because the differences between these regimes can elucidate important explanations for the differences observed in the levels of sick-leave rates across countries. Initially, comparisons should focus on eligible salaried workers, who make up the largest proportion of the workforce (Anema et al., 2009; Gimeno et al., 2014) . We follow this recommendation because it would be much more complex to include the variation in regulations with respect to coverage eligibility for self-employed workers and unemployed persons.

The aims of this study are to present comparable components of the follow-up regime for sick-listed employees and to conduct a qualitative assessment of the differences with respect to burden and responsibility sharing between the social protection system, employers and employees. Sick-leave implies the right to be absent from work when sick, and return to work after recovery. In our analysis, we include both sickness benefits provided by the social protection system and sick-pay provided by the employer. We also describe the changes made in each country's system due to the Covid-19 pandemic.

We describe sickness follow-up regimes in a group of high-income countries located in north-western Europe. This is a necessary first step towards an empirical study of a comparison of sickness absence rates and development of prediction models because ‗predictions cannot be made without well-founded theories; theories cannot be made without proper classification; and classification cannot be made without good description' (Landman, 2002, p. 21) .

Because there are few comparative analyses of sickness absence rates other than the statistics provided by the OECD, qualitative descriptions of the different country regimes are required. We chose to focus on a small number of countries because it allows us to use less abstract concepts that are more grounded in the specific context under scrutiny (Mair, 1996) . Including relatively few countries also allows in-depth descriptions (Thomann, 2020).

The selection of countries was based on the aim of comparing countries located in north-western Europe that share some cultural and labour market similarities, but that nevertheless differ with respect to their sickness absence policies.

Instead of focusing on describing all the differences between countries, we follow other social science scholars who have turned their attention to the study of parts of rather than the whole of society (Chilcote, 2018; Nelson et al., 2020) .

The relevant ANT phases in this study are: i) Identify the stakeholders, ii) Investigate the stakeholders, iii) Identify stakeholder interactions to explore the level og influence, iv) Construct an actor-network model (Carroll, Richardson, & Whelan, 2012, p. 60) . Instead of taking welfare regime theory as the starting point, we take a more direct approach by studying the main actors typically involved in a sickness absence incident: the employee, the employer and the health services, which is represented by the sickness certifier and insurance provider or actor (Figure 1 ).

[ Figure 1 about here] These actors co-operate to some degree in the follow-up of sick-listed employees. This actor perspective is used when we compare regimes across countries, following actor-network theory (ANT), where actors can be individuals, groups of individuals, institutions, boundary protocols, regulations and technical artefacts (Latour, 1996; Law, 1992; Law & Hassard, 1999) . The employee may be the individual sick-listed person, the unions or other interest groups representing the sick-listed employee. The employer may be represented by management or the supervisor of the sick-listed employee or the employer organisations as interest groups. The health services may be represented by the general practitioner (GP), the occupational doctor or by a clinician in specialist health services or the health authorities. The insurance provider may be represented by a consultant from a private insurance company or by a counsellor from public services, depending on the system set-up. These actors are linked to each other in a system or network in which their relationships, incentives, beliefs, ethics and morals are interwoven in the functionality of the system. The aim of this study is not to model all the incentives embedded in the system or the behaviour of all these actors, but rather, to take a first step in studying their roles across different countries. The analysis was inspired by ANT, in that we look at how actors and actants-that is, the different human and non-human components of a sickness follow-up regime-work towards the goal of reducing unwarranted sickness absence by considering the strength of the incentives (inscriptions), or the force of the constituent non-human parts, towards achieving it. In this setting, the employee, the employer, the insurance actor and the sickness certifier operate in a non-hierarchical, complex network that produces or prevents unwarranted sickness absence via the follow-up regime. Inscriptions are the procedures or system requirements that indicate how the network of actors in the follow-up regime should operate (e.g. sick listing practices, employee contracts, documentation requirements).

The starting point of the description of the sickness follow-up regimes is sickpay and sickness benefit schemes, which can be described by the criteria for qualification, namely the duration and economic compensation in the follow-up period of the employer and social insurance, respectively, and also by what happens if the person is still ill after the benefit is terminated.

Likewise, the sick-leave follow-up system can be described by the demands and roles of the actors during the follow-up, related to, for example, sick-leave certificates and formal contacts, including rules for partial sick-listing and reassessment, duties and responsibilities, support, sanctions and dismissals, and finally, the role of occupational health services (OHS) and others.

We included nine European countries in our study, representing a diversity of welfare regimes: Sweden, Denmark and Norway (the Scandinavian countries) and Iceland may be classified as pure-type social democratic welfare states, Finland as a social democracy with a Christian Democratic component, Belgium as a Christian Democratic welfare state with social democratic elements, Germany as a pure-type Christian Democratic welfare state, the UK as a liberal welfare state and the Netherlands as an unclassified hybrid welfare state (Ferragina & Seeleib-Kaiser, 2011 ).

The Norwegian researchers initiated the study, invited and established the multidisciplinary team and developed 51 questions that were answered by the collaborators/experts/co-authors from each country. The topics were related to sickness benefits, sick-pay and the role and requirements for employers, employees, GPs, OHS and insurance providers in the follow-up regime.

There were many rounds back-and-forth to ensure that answers were comparable between countries, and all authors from each country contributed to ensuring that the tables containing the inscriptions were as accurate as possible.

The answers produce both quantitative data (e.g. number of waiting days, maximum duration) and qualitative data (e.g. duties for the sick-listed employee and their employer).

The first author facilitated a collaborative research process and all collaborators/co-authors contributed with input to each of the ANT-phases. A qualitative assessment of the follow-up regimes for each country was conducted. The components of the regimes are described and evaluated according to the strength of embedded incentives related to preventing and reducing the duration of sick-leave. We assess both the actors' overall burden (economic and otherwise) and unintended consequences using abstract reasoning in which the context and bigger picture surrounding a sickness absence incident are examined. The aim of the analyses was to explore and identify discernible patterns based on the systematic overview of the detailed inscriptions collected (Landman, 2002, p. 13 ).

The results are summarised in Table 1 , which provides information on sickleave benefits and sick-pay schemes. Tables 2-5 compare the features of the follow-up regimes across the nine countries. These tables constitute the ANTbased inscriptions. Table 1 shows that four countries have different rules for different groups of employees (Denmark, Belgium, the Netherlands and the UK). In Denmark and Belgium, blue-collar workers are compensated more poorly than white-collar workers. In the Netherlands, the sickness benefits depend on the type of employment contract. In the UK, Statutory Sick Pay (SSP) is the same for all employees, while company sick-pay varies from employer to employer.

[ Table 1 about here]

The employer period varies from 10 working days in Finland, 14 calendar days in Sweden and 16 in Norway to 2 years in the Netherlands. Finland, Germany, Norway, Iceland and Belgium provide 100% compensation in the employer period. In Denmark and the UK, a fixed amount per week is paid to all sicklisted employees, in Sweden, the sick-pay is 80% of the salary with a maximum amount per day, and in the Netherlands, 70% compensation is the minimum level.

There is considerable variation in compensation after the employer period, from 0% in the UK (no sickness benefits provided by the social protection system) to 100% in Norway for up to 52 weeks. The maximum period is longest in Sweden, where no upper limit is set in the case of severe illness.

Only the UK has waiting days in its scheme, while in the Netherlands, it is up to the employer or dependent on the social agreements with trade unions. In Sweden, the waiting day was replaced with a 20% deduction of sick-pay during an average work week on 1 January 2019.

Better conditions in individual contracts or collective agreements are common in all nine countries, while the qualification period varies from no period (rights are granted from the first day) in some of the countries to 4 weeks in Germany, Norway and Iceland.

The total tax wedge, that is, the income tax and social security contribution from employers and employees, is a measure of how much the government receives as a result of taxing the labour force. The tax wedge varies from 30.9% in the UK to 52.7% in Belgium. The total tax wedge is also high in Germany (49.5%), Sweden (43.1%) and Finland (42.3%). Denmark does not have social security contributions from employers and employees, and the income tax is therefore equal to the total tax wedge (35.7%). Table 2 provides a comparison of sick-leave certification, partial sick-leave legislation and elements of the follow-up regimes. While the employer decides if a medical sick-leave certificate is needed in Denmark and Iceland, this is demanded after 3 days in Germany, 3-8 days in Norway and after 7 days in the UK. In the Netherlands, there is no requirement for sickness absence certification from a doctor at all, only self-certification where the employee reports sick to the employer and provides a reason for the sickness to the occupational physician. In most of the countries, the GP issues sick-leave certification. However, in Germany and Norway, chiropractors and physiotherapists can also issue sick notes, and in Sweden, also dentists. In Germany, sick-leave certification depends on the contract of the National Association of Statutory Health Insurance Physicians. In every German state, a dentist can issue a sick-leave certificate; in some states, this can be done by physiotherapists. None of the countries have a requirement concerning the qualifications about working life or labour market for the sick-leave certifiers.

[ Table 2 about here] Table 3 shows that the duties for employees are similar across the nine countries, and the employee must inform the employer about the sickness as soon as possible and deliver a sick-leave certificate if required. Four of the schemes also demand that the employee try out work-related activities as soon as possible (Denmark, Germany, Norway and Sweden). In Finland, the sicklisted employee must co-operate when the OHS assess the remaining work ability and the possibilities of returning to work, but there is no demand for the employee to try to return to work during the certified sick-leave period. The obligations for employers are especially detailed in Norway, Sweden and the Netherlands.

[ Table 3 about here]

In the Netherlands, an extensive follow-up regime has been implemented, and the employer has considerable responsibility and covers all the costs during the first 2 years. Within 6 weeks of sick-leave, the employer is obliged to pay for a problem analysis of the sickness absence by a certified company doctor.

Employers have contracts with occupational physicians or OHS take care of this duty. Within 8 weeks, the employer and employee must prepare a plan for the employee to return to work, and the plan can include work accommodations, interventions paid by the employer (e.g. counselling) and an updated return-towork plan. If the employee is still on sick-leave after 1 year, the employer is allowed to try to offer the sick person another job at another organisation if their reinstatement in the original organisation is not possible; this is typically done by hiring a reintegration consultant. If the employee is not back to work after 2 years, a disability benefit can be applied for from the public social security system; this is only granted if the theoretical loss of earning capacity due to the disability is more than 35% of the former income.

Denmark has chosen a completely different model, in which municipal job centres are responsible for following up sick-listed employees. The employer still has obligations and must invite the sick person to a meeting to discuss the situation, and they must work out a plan for returning to work (reintegration).

After 8 weeks, the employee can ask the employer for a plan to stay employed at the workplace. Also after 8 weeks, the job centre requires a medical statement regardless of whether the employer received one earlier. To be entitled to sickness benefits, the sick person must try out work-related activities as soon as possible, as proposed by the job centre. Employers in Denmark are expected to be in contact with the job centre from the municipality where the employee lives; in this way, each employer is likely to be in contact with several job centres.

From Table 4 , we see that dismissal of sick-listed employees is strictly forbidden only in Norway and the Netherlands. In Belgium, dismissal is possible if it can be documented that the employee will not get well, and only then after it is confirmed by the occupational doctor that the employee would not be able to carry out any job in the company. In Germany, the employer has to go through a statutory process that determines the incapacity for work before dismissal. In Sweden, sick-leave alone is not reason enough for dismissal, but if the employee cannot do any work at all for his/her employer, the employer can terminate the contract. In Norway and the Netherlands, dismissal of sick-listed employees is only possible after 1 and 2 years of sick-leave, respectively.

[ Table 4 about here] Furthermore, all schemes seem to have some built-in sanctions. In Finland, sickness benefits can be stopped if the sick-listed employee fails to deliver the 90-day evaluation by the OHS, made in collaboration with the sick-listed employee and the employer. In Sweden, the employer can be reported to the labour authorities if the reintegration plan is of poor quality or not being followed. The strongest sanction of employees involves possible dismissal, as in Iceland, Denmark and the UK. A softer approach is taken in Norway, where the employer can dispute a sick note by writing a letter to the local Labour and Welfare Office, who will inspect the disputed sick-leave incident. [ Table 5 about here] From the employer's perspective, the incentives to prevent sickness absence are especially strong in the Netherlands and the UK, where employers bear the whole burden of sickness absence. In the Netherlands, employers might additionally be punished by the social insurance system if they have not been compliant with sickness absence guidance legislation during the sickness absence period. The incentives are weakest in Norway and Sweden, which have the shortest employer periods.

The results of this study highlight the different systems for following up sicklisted employees in the nine north-western European countries included. We are unaware of any other studies that have compared the follow-up regime for sicklisted employees between countries. Table 6 offers a rough categorisation of the follow-up regimes based on a qualitative assessment and conceptual reasoning of the results from the ANT inspired research process. The system requirements and the strength of incentives they provide for employees, employers and social insurance systems in each of the nine countries under study, constitue the ANT inscriptions assessed in the analysis.

[ Table 6 about here]

The employee burden is based on a combination of factors such as the degree of compensation and the requirements for return-to-work activities. This burden is high in Denmark, Sweden, the Netherlands and the UK because of high requirements when sick-listed; because they receive less income compensation when sick, a higher burden is imposed on blue-collar than on white-collar workers in Denmark and Belgium. In the Netherlands, all workers receive 70%, or the minimum wage if 70% is lower than the minimum wage.

The employer burden depends on the length of the employer period and financial and other requirements for employers during the sick-leave period; these requirements are high in Sweden and the Netherlands. The length of the employer period is especially long in the Netherlands and the UK, of medium length in Germany and Iceland and relatively short in the other countries. The total employer burden is low in Denmark, high in the Netherlands and UK and at a medium level in the other countries.

The burden of sick-leave on the social insurance system is difficult to assess. If we do not consider differences in sick-leave rates, the burden on the social insurance system is lowest in the UK and the Netherlands because of the long employer period. Moreover, in the Netherlands, there is no social protection for the self-employed (12% of the working population), and only 20% of the selfemployed have arranged private insurance (de Rijk, 2018) . However, it can be argued that high sick-leave rates, as in Norway, place a higher cumulative burden on the social insurance system.

We must also consider the utility or value of the current system for the working population, employers and social insurance system. Full income compensation may be more valuable for both the working population and employers who sell goods and services because income loss during a sickness absence is compensated, thus maintain purchasing power. The social insurance system might also benefit from full income compensation if it means that the shared risk implies higher employment and more taxpayers to share the costs.

It is recommended that appropriate rehabilitation and retaining measures, as well as a comprehensive prevention agenda, should underpin sickness benefit policies (Spasova et al., 2016, p. 4) . Further, we suggest that the follow-up regimes should be included in future discussion of sickness benefit policies because components of these regimes can potentially be developed further and contribute to effective incentives for preventing sick-leave.

The ANT framework has been used in the field of disability studies (Galis, 2011) ; sick-leave could be studied similarly. Disability is not a property of individuals only, because the degree of disability depends on the environment.

and what is involved in that which is happening' (Söderström, 2016, p. 98).

Sick-leave and return to work depend on the actions and behaviours of all participants (the sick-listed employee and his or her family, the employer, the insurance actor and the sickness certifier), and these interconnected complexes depend on the orientation of the current sickness benefit scheme and the social, historical and cultural context in which the sick-leave occurs (de Rijk, van Raak, & van der Made, 2007; van Raak, de Rijk, & Morsa, 2005) . Thus, if a theoretical approach were to guide the research, comparing sick-leave rates and sick-leave behaviour across countries is a much more complex task than simply comparing statistics.

International variation in sick-leave rates was previously found to be caused by factors such as ‗the generosity of granting sick-leave, the strictness of employment protection and the employment of older persons' (Osterkamp & Röhn, 2007, p. 97 ) and the extent to which activation measures supporting reintegration into work are incorporated (Anema et al., 2009) . Using data from the 2010 European Working Conditions Survey, one study found that the most significant factor for cross-country differences in the probability of absence is whether employers are required to continue paying full wages in case of illness, given that absences are significantly higher in countries where this rule applies (Chaupain-Guillot & Guillot, 2017) . Another study compared short-term sickness absences in 24 European countries and found a negative relationship between union density and sickness absence (Sjöberg, 2017) . It was argued that a stronger collective employee voice would give rise to a lower short-term sickness absence, and that this perspective could provide new insights into why countries experience different trends in sickness absence over time. A study comparing cross-national sickness absence systems and statistics identified common elements in sick-leave in Spain, Sweden and the Netherlands (Gimeno et al., 2014) , thereby concluding that basic and useful sick-leave indicators can be constructed to facilitate cross-country comparisons. Although we agree that it is possible to facilitate cross-country comparisons, we suggest that the system components in the different countries need to be understood in more detail, and that the share of economic burden must be included in comparisons, as this factor likely provides the strongest incentive for preventing sickness absence. It is not obvious that countries with the highest employer burden-and thus low sick-leave rates-have the best system if the main economic goal is to maximise employment rates.

Public schemes for sickness benefits often attract criticism because of the lack of incentive that they provide to employers for preventive and reintegration activities. Although employer incentives appear to lower sick-leave rates, the downside is that workers with poor health have fewer employment opportunities (Koning, 2016) or are sorted into temporary employment, as seen in the Netherlands (Koning & Lindeboom, 2015) (see Supplementary Figures S1-1 and S1-2).

The relationship between the generosity of paid sick-leave and three economic indicators (per capital gross domestic product (GDP), unemployment rates and competitiveness) has also been examined. No significant relationship was found between the generosity of a country's sick-leave policy and these macroeconomic indicators (Schliwen, Earle, Hayes, & Heymann, 2011) .

Employment is a key concept as a measure of economic activity because GDP is a product of employment and productivity. Employment rates are typically measured as the proportion of working-age people who have worked for at least 1 hour in the reference week. However, information about the actual time worked by the employed during 1 year is lost in this simple head count and ignores the large variation in working time arrangements and job contract durations (Brandolini & Viviano, 2016) . The full-time equivalent (FTE) employment rate is equivalent to the ratio of total actual hours worked if employed full time (Brandolini & Viviano, 2018) . While Iceland has a high FTE employment rate among both men and women, the Netherlands and Belgium have low FTE employment rates, especially among women (see Supplementary Figure S2 -1), and Denmark, Norway and Finland also have low FTE employment rates among men (see Supplementary Figure S2 -2).

According to the International Labour Organization (ILO), the Covid-19 crisis has revealed important gaps in the coverage of entitlements to social protection in case of sickness (ILO brief, May 2020). In March 2020, Sweden abolished the qualifying day of sickness because of Covid-19. In May 2020, the Danish Parliament passed a bill to amend the Sickness Benefit Act that would allow employees at higher risk of infection with Covid-19, as well as employees with a relative in a higher risk group, to receive sickness benefits. Similarly, the bill allows employers paying salaries during such absence to receive sickness benefit reimbursement from the first day of absence. In Finland, sickness allowance on account of an infectious disease provides loss-of-income compensation for persons placed in quarantine and isolation. Norway has waived the requirement for personal turn-out at the doctor's office to receive a sick note and reduced the employer period from 16 days to 4 days if the sickleave incident is due to Covid-19 (infected, suspected infection, quarantine and isolation). Iceland also adjusted their system to slow the spread of Covid-19 by ensuring that individuals can follow official instructions to enter quarantine without having to worry about their personal finances. Germany also waived the requirement for personal turn-out to receive a sick note from the doctor. In the Netherlands, there have been no measures or amendments with respect to sickpay, and the employer still has a statutory obligation to pay employees on sickleave for up to 2 years. Belgium has followed the same strategy as the Netherlands, and the normal rules apply. The UK launched the Coronavirus Statutory Sick Pay Rebate Scheme, which targets employers with fewer than 250 employees. The rebates cover up to 2 weeks of SSP and are payable if an employee is unable to work for the following reasons: they have Covid-19 symptoms; they are self-isolating because a cohabitant has symptoms; they are self-isolating after being informed by the National Health Service (NHS) or public health bodies that they have come into contact with a person with Covid-19 or a shielded person, and they have a letter from the NHS or a GP instructing them to remain at home for a period of at least 12 weeks.

One limitation of this study is its qualitative design, which does not allow us to generalise the results to other countries. A strength of this work is that it generates new hypotheses that can be tested in future studies. For instance, it would be interesting to study the distribution of the economic burden between employees, employers and social security systems, the taxing of the labour force and the consequences for employment and public health systems.

We cannot rule out the possibility that more knowledge could have been gained if we had collected and analysed information from other, or additional, countries. However, we believe that most of the inscriptions for the networks of actors were included for the countries studied.

The ANT perspective should be explored further in sick-leave research to develop better theoretical models for comparative analyses of sick-leave.

Important contextual factors, such as labour force participation, temporary employment, unionisation and other social, historical and cultural perspectives, should also be considered in future research, especially in studies that compare sick-leave across countries.

Paid sick leave is viewed as a key component in combating health and social inequalities according to the ILO Conventions and the Decent Work Agenda (Scheil-Adlung & Sandner, 2010) . A better understanding of risk-and burdensharing between employees, employers and other tax payers in future research could contribute to improving the schemes.

This assessment of the differences in burden and responsibility sharing between the social protection system, employers and employees shows that countries with shorter employer sick-pay periods also have stricter employer follow-up responsibilities because, in practice, they become gatekeepers for the public sickness benefit scheme. In Germany and the UK, employers have few followup requirements compared with the Nordic countries because the former bear most of the sickness absence costs. This is also true in Iceland, where employers carry most of the costs and have no obligation to follow up on sick-listed employees. The situation in the Netherlands is paradoxical: employers have strict obligations in the follow-up regime, despite covering all the sick-leave costs.

Appropriate rehabilitation and retaining measures, as well as a comprehensive prevention agenda, should underpin sickness benefit policies. The results show that the incentives for preventing sickness absence from the workplace are weak 29 in most of the countries studied. Further research is needed to identify the optimal burden sharing of the cost of sickness absence between employees, employers and tax-based social insurance programmes. Important contextual factors such as labour force participation, temporary employment, unionisation and other social, epidemiological, historical and cultural perspectives should be considered in future sick-leave research, especially in studies that compare sickleave across countries.

The Covid-19 pandemic has revealed the importance of social protection systems, including income security, when health problems arise. During the pandemic, most countries adjusted their sick-pay system and increased coverage to reduce the risk of spreading Covid-19 by employees who would otherwise work while sick or when they should self-quarantine. Söderström, S. (2016) . Socio-material practices in classrooms that lead to the social participation or social isolation of disabled pupils. a: In practice, more blue-collar workers, freelancers and employees in precarious jobs do not have contracts that entitle them to full compensation. Some workers have full salary during sickness leave (in this case, the employer receives the benefit instead), whereas others receive benefits. The benefits are flat rate, but with the exception that the maximum pay is 85% of the full-time salary, meaning that many receive less than 85%. White-collar workers mostly have contracts with 100% salary during sickness leave; this is also the case for some blue-collar workers. The majority of workers on contracts without salary during sickness leave are blue-collar workers. a: If the employee is not reintegrated after 1 year, the employer is obliged to offer a suitable job in another organisation (in practice, this is often facilitated by a reintegration agency).

b: After 8 weeks, the employee can ask for a plan to remain at his/her job with the employer.

c: The purpose of such a meeting is to discuss the way in which the workplace has influenced the absence of the employee and to determine whether the employer can make any changes and help to improve the employee's health.

48 No, no differentiation between ‗social risk' and ‗occupational risk'. Occupational accidents are compensated from separate insurance a: Dismissal of an employee with a permanent medical health condition will only be allowed if the occupational doctor confirms that the employee will never be able to perform any job within the company.

b: The employer has to go through a statutory process to determine the incapacity for work. 52 d: Large employers also have an incentive to prevent work disability because the amount of their pension insurance payments is determined by previous rates of disability retirement in the company.

e: Sick-pay from the employer was introduced at the beginning of the 1990s to provide incentives to prevent sick-leave.

f: In addition, through the mandatory involvement of occupational physicians in long-term sick-leave incidents.

g: VIRK is a rehabilitation fund established by employers. Table 6 Categorisation of the follow-up regimes in the nine countries

Sickness and disability policy interventions

Can cross country differences in return-to-work after chronic occupational back pain be explained? An exploratory analysis on disability policies in a six country cohort study

Sickness Absence: An International Comparison

Health, work and working conditions: a review of the European economic literature

Sickness absence: a comparative study of 11 countries in the Luxembourg employment study (LES): Norsk institutt for forskning om oppvekst

Behind and beyond the (head count) employment rate

Measuring employment and unemployment

The role of employer-provided sick pay in Britain and Norway

Service science: An actornetwork theory approach

Sickness benefit rules and work absence: An empirical study based on European data. Revue d'économie politique

Theories of comparative politics: the search for a paradigm reconsidered: Routledge

Income Security During Sickness Absence

Arbeidsre-integratie blijft mensenwerk

A new theoretical model for cooperation in public health settings: the RDIC model

Welfare Regime Debate: Past, Present

Labor market institutions and individual absenteeism in the European Union: the relative importance of sickness benefit systems and employment protection legislation

Enacting disability: how can science and technology studies inform disability studies?

Distribution of sickness absence in the European Union countries

Cross-national comparisons of sickness absence systems and statistics: towards common indicators

Ensuring a healthy and productive workforce: comparing the generosity of paid sick day and sick leave policies in 22 countries

Privatizing sick pay: Does it work? IZA World of Labor

The rise and fall of disability insurance enrollment in the Netherlands

Issues and methods in comparative politics: an introduction: Routledge

On actor-network theory: A few clarifications

Notes on the theory of the actor-network: Ordering, strategy, and heterogeneity. Systems practice

Actor network theory and after

Comparative politics: an overview. A new handbook of political science

The social policy indicators (SPIN) database

Being on sick leave: possible explanations for differences of sick-leave days across countries

Sick Pay Insurance and Sickness Absence: Some European Cross-Country Observations and a Review of Previous Research

Comparison of sickness absence in Belgian, German, and Dutch firms

Paid leave for personal illness: A detailed look at approaches across OECD countries

Evidence on paid sick leave: Observations in times of crisis

The administration and financing of paid sick leave

Employee collective voice and short-term sickness absence in Europe