Social partners and OSH: a multilevel and cross-country exploration in the hospital and social services sectors

: The article examines the role of social partners (trade unions and employers) in the field of occupational safety and health (OSH). The Covid-19 pandemic directed attention to the importance of greater national and European level policy coordination to protect and promote healthy, safe, and well-adapted work environments. On the basis of two sectors, hospitals and social services with a focus on elder

care, the article asks how OSH policy interventions at the national level and the European level are interlinked.To explore interlinkages, the article focuses on OSH policy coordination between national social partners and European level sectoral social dialogue.The analysis is informed by actor-centred institutionalist and multilevel governance approaches and uses qualitative data.The article explores patterns of interlinkages between the national and European levels in two sectors and six countries and discusses the limitations of applying a cross-national and multilevel conceptual approach.The analysis shows that the pandemic has promoted some coordination at national and between national and EU levels, yet still rather modest interlinkages and degrees of 'vertical' coordination can be observed.The paper concludes that there continues to be a degree of 'in-the-making' to the multilevel governance of employment relations, even in the presence of common challenges.

Introduction
This article focuses on national sectoral social partner organisations and explores the 'interlinkages' with European level social partners in the field of OSH.The Covid-19 pandemic has directed policy-makers' and social partners' attention to the importance of greater national and European level policy coordination to protect and promote healthy, safe, and well-adapted work environments (Degryse 2021;EC 2021b).The European Framework Directive (89/391/EEC) guarantees minimum health and safety (H&S) requirements for employees.Member states are expected to maintain those levels and are allowed to establish more stringent measures.However exceptional, the pandemic highlighted significant common gaps in the provision of OSH in most member states.The key role of social partners at various levels for H&S was prominent during the health crisis and was later also highlighted in the European Commission's (EC) strategic framework on H&S at work (EC 2021b).
The focus of this paper is on two sectors and six countries.The two sectors investigated are the hospital1 sector and the social services2 sector (with particular attention to elder care), whose workforce reported high work-related H&S risks in times of the pandemic (Eurofound 2022).Although both sectors are characterised by similar risks, they differ in terms of structural characteristics and the social partners involved.At the European Union (EU) level, we consider the sectoral social dialogue committee (SSDC) for the hospital sector and the role and actions of the (until July 2023 informal) social dialogue in the social services sector.For the national level, we focus on social partners in Germany, Italy, Lithuania, Poland, Sweden, and the UK and how they are interlinked with the European sectoral level.The six countries represent different systems of industrial relations and H&S representation structures.
We therefore apply a multilevel approach to investigate social partner interlinkages between the national and the European level.How are national social partners interlinked and how do they coordinate policy interventions with the European sectoral level to manage H&S risks?Such interlinkages constitute an important prerequisite to identifying effective ways to coordinate and provide better H&S protection.We argue that coordinated responses by social partners at the national and European level are needed to develop OSH policies and institutional practices to counter emerging health risks and in particular physical and psychosocial threats to the H&S of the workforce.The paper aims to identify inter-linkages and coordination between the levels and to investigate the role of different structural and institutional settings and approaches to OSH in two sectors and six countries.The paper, therefore, aims to contribute to the two following questions: What interlinkages and 'vertical' coordination exist between social partners at the national and European level?To what extent are sectoral and institutional characteristics contributing to the coordination (or lack thereof) between national and European levels?
The paper proceeds as follows.In section 2, we present the analytical framework and the methodology.Section 3 gives an overview of European sectoral social dialogue and raises issues related to national systems of industrial relations.Section 4 illustrates the provision and governance of OSH and section 5 presents empirical evidence on patterns of interlinkages and 'vertical' coordination between national social partners and the European level.Section 6 concludes and summarises.

Analytical framework and methodology
The provision and governance of OSH policy interventions involve multiple levels, and the important role of social partners and social dialogue at the national and European level has been emphasised (EC 2021a).To investigate the extent of coordination between the levels we explore the role of sectoral and institutional settings and approaches to OSH.

Analytical framework
We draw on an actor-centred institutional approach that takes institutional structures and policy intervention processes into account (Kaufman 2004).This enables us to establish theoretical links between (i) industrial relations systems and H&S representations and (ii) practices of strategically operating actors -limiting or enabling -effective ways of H&S protection.Accordingly, we establish a link between different institutional and contextual settings in the six countries and two sectors that shape coordination efforts between the national sectoral and European sectoral level.We approach European social dialogue as a multilevel institutional system, characterised by potential interlinkages between the social partners at different levels.The multilevel governance concept is theoretically valuable in our case for its evolving and dynamic nature, and as it captures "'bottom-up' as well as 'top-down'; cross-national (horizontal) influences mix with national (vertical)" continuous developments (Marginson/Sisson 2004: 25).
'Vertical' coordination by social partners in the field of OSH, therefore, ranges from the European (sectoral) level, to the national (sectoral) level, regional level, and workplace level, and back.Coordination efforts across levels may include gathering and sharing information, developing joint positions, conducting joint projects, monitoring developments at other levels, or negotiating and implementing joint texts (cf.Table 2).Thus, active participation at the European level in social dialogue activities by national actors is a potential indicator for 'vertical' coordination.'Vertical coordination' will be supported if 'linkages' between levels exist.For instance, the mere affiliation of a national employer organisation to the European umbrella social partner organisation constitutes such a 'linkage', however, this could be strengthened by active participation in coordination efforts.
For European social dialogue, the question of multilevel action and implementation at the national level is crucial (Keller/Weber 2011).It is particularly so, we argue, in the context of OSH, as a founding purpose of Directive 89/391/EEC -from which several pieces of legislation 'descend' -and which refers to the preventive and workers-participatory dimensions of H&S at work.The data we collected contributes to a further understanding of how the European and national levels of H&S regulation are interlinked and (if at all) vertically coordinated.The challenge of the pandemic common to all social partners represented a test of their ability to coordinate (Degryse 2021).

Case selection
The countries investigated are Germany, Italy, Lithuania, Poland, Sweden, and the UK3 which represent different systems of national industrial relations in Europe.The rationale behind the country selection was to cover the main models, with Germany representing 'social partnership', Italy representing the 'state-centred' model, Lithuania and Poland representing a 'mixed' or 'transitional' institutional setting, Sweden representing 'organised corporatism', and the UK representing the 'liberal' model (EC 2009).Industrial relations regimes differ qualitatively with regard to e. g. union membership, principal level of collective bargaining, coverage, and employee representation (cf.Table A1 in the Appendix).Models of industrial relations regimes inform expectations concerning the relationship between OSH policies and practices of social partners in different countries.In Sweden ('cor-poratist') and Germany ('social partnership') we expect a greater preference for autonomous agreements and collective bargaining, while in state-centred regimes (Italy) we expect sponsored agreements and legal standards rather than collectively agreed policies, and in liberal employer-oriented regimes (Lithuania, Poland, UK) we expect less binding standard setting and collective bargaining, and more benchmarking and guidelines on OSH.
The countries investigated provide different models of H&S representation systems (Fulton 2018;cf. Table A1).In dual systems of employee representation such as in Germany, OSH issues are primarily dealt with through the existing representational structure (works council), while in union-based systems union representatives with their own rights exist (Italy, Lithuania, Poland, Sweden, UK).In countries with low union density and/or non-unionised workplaces (Lithuania, Poland, UK), employee H&S representatives are elected by the workforce.In all countries, except Italy, joint employee/employer H&S committees are used (Fulton 2018).Union-employee H&S representation systems are expected to promote OSH standard setting and enforcement (Gunningham 2008).
The two sectors investigated, hospital and social services (with a focus on elder care), are characterised by a predominantly female and ageing workforce (Crawford et al. 2016;cf. Table A2).Care workers in both sectors are exposed to occupational risk factors for musculoskeletal disorders (MSDs) and psychosocial risks -most commonly repetitive movements, lifting or moving people, demanding interactions with service recipients, time pressure -which relate to the way work is designed, organised, and managed, and impact the H&S of workers (EU-OSHA 2020; 2022).Furthermore, both sectors have faced reductions in public expenditure and the privatisation of healthcare provision promoting new approaches to health management affecting work and employment practices (Bach 2001;Greer et al. 2013).These developments have weakened the regulation of employment and facilitated a shift towards decentralisation of collective bargaining and a decline in union membership, although hospitals largely still operate under public provision in most countries analysed (Eurofound 2021;Grimshaw et al. 2007;Galetto et al. 2014;Kahancová/Szabó 2015).Privatisation, by contrast, of social services is much more common, and the sector is characterised by a substantial presence of private care providers (Eurofound 2009;2017;2021).Understaffing and its possible effects on OSH is an issue in both sectors, but even more so in social services which can be linked to cost pressure in the latter case (Eurofound 2017;2020b).Employment concentration in the hospital sector is rather high compared to the social services sector, which is characterised by small and medium-sized establishments (Eurofound 2017).The dominance of private and small employers in this sector has implications for trade unions and employee H&S representation since union membership is generally significantly higher in large, public-sector organisations (Eurofound 2009).Accordingly, industrial relations systems and workers' protection in the two sectors vary.Health workers in the hospital sector, which is characterised by high union density and collective bargaining coverage (Eurofound 2009;2020a), experience better protection compared to the social services sector with low union density and collective bargaining coverage (Eurofound 2021).

Data collection and analysis
Our analysis is informed by the literature on the sector-specific industrial relations as well as on the role of social partners for H&S and empirically based on the data collected through 52 semi-structured interviews conducted mainly in 2022.Table 1 gives an overview of the sample.In addition, we draw on secondary data such as documents produced by the social partners.(European Hospital and Healthcare Employers' Association).EPSU is also active in the social dialogue within the social services sector whereas the employer side is represented by Social Employers (Federation of European Social Employers).All three organisations have been identified as the sectors' representative organisations (Eurofound 2009;2020a;2021).Our sampling strategy therefore primarily aimed to include the national member organisations of EPSU, HOSPEEM, and Social Employers in the six countries and sought interviewees familiar with OSH in the two sectors investigated.Furthermore, where appropriate, we interviewed experts in European sectoral social dialogue.Moreover, the sample was enriched by interviews with representatives from organisations without affiliation to the three European umbrella organisations, but with high relevance for the national sectoral level.In alignment with our sampling strategy to be able to investigate (potential) interlinkages in the field of OSH, our sample comprises a mix of different groups of national social partner respondents to grasp the different relevant views and experiences.The interviewees represent various functions, from high level officials to sector experts, OSH experts, H&S workplace representatives, and European relations experts.It is important to note that these categories are not necessarily mutually exclusive (cf.Section 5).Most of our social partner interviewees (34) are engaged in general policy and collective bargaining at national and/or sectoral level, often in a senior position, and have a good general overview.The second group represent their organisation at the EU level (16; no respondents from Italy).The third group are respondents (from Poland and the UK) who are responsible for H&S policy (13).The two smallest groups (11 each) are made up of H&S experts who have a deep knowledge in the field of OSH, and of H&S workplace representatives or managers/employers, i. e. representing the workplace level perspective.
An interview guideline was used to ensure the comparability of our data and organised along the following main themes: characteristics and main challenges of the sector, industrial relations, OSH and workplace representation, interlinkages, and coordination between levels (e. g. local, sectoral, national, European).The interviews were conducted in the local language, recorded with the interviewee's consent, fully transcribed, and coded according to a thematic coding frame, designed to identify the nature of interlinkages or coordination efforts.The themes of the coding frame were informed by ex-ante formulated propositions theoretically and empirically informed by the potential effects of national industrial relations systems (e. g. main level of bargaining), workplace representation models (e. g. union representation), institutional and contextual settings of the sectors (e. g. non-standard employment), and awareness and use of European social dialogue interventions.Accordingly, the interview data were evaluated based on these ex-ante considerations.However, we also foresaw the possibility of including additional aspects emerging from the data.Overall, the sample includes trade union representatives (33), employer representatives ( 16), and representatives of other institutions in the field of H&S where appropriate (3) (cf.Table 1).Our primary qualitative data, therefore, reflect national actors' views and experiences in relation to the 'vertical' coordination between national and European levels.

Social dialogue at EU level and national systems of industrial relations
Institutionalised social dialogue at the EU level between trade unions and employers takes place either as a cross-sectoral social dialogue or in 44 'formal' SSDCs that have been officially recognised and are financially and logistically supported by the EC (EC 2021a At the sectoral level, all SSDCs together conclude on average 40 joint texts per year (EC 2021a), mainly joint positions (Degryse 2015).With the Covid-19 pandemic, the number of joint texts rose to over 80, more than half of them dealing with the pandemic and its implications (EC 2021a).Again, most of these texts concluded between March 2020 and March 2021 were joint statements (Degryse 2021).According to the EC's database,4 129 of the overall 1,206 joint texts  address H&S, which means slightly more than 10 percent of all texts.H&S issues were discussed in both the hospital SSDC and the informal social services social dialogue5 before, during, and after the pandemic (cf.Tables A3 and  A4).Whereas the hospital SSDC can use the whole range of EU social dialogue instruments -from framework agreements implemented by Council Directive to tools and recommendations addressed to their affiliates (cf.Table 2) -to tackle OSH-specific issues, the until July 2023 informal social dialogue in the social services sector was until then able to intervene mainly via joint projects and joint statements (cf.Tables A3 and A4).This different range of instruments exemplifies the potential for varying vertical coordination between European and national levels of regulation in the respective sectors.
At the national level, social partner activities include collective agreements jointly negotiated between them, but also providing their members with advice, training, and information on H&S at work.It is important to note that sectors and countries differ in terms of collective bargaining coverage (cf.Table 3).

Provision and governance of OSH
We argue that coordinated responses at the national and European level are needed to develop robust and resilient H&S policies and institutional practices to counter existing and emerging health risks.While we saw a concerted EU effort during the pandemic to tackle the containment of the virus and collaboration to find a vaccine, different social partners adopted different approaches to H&S strategies.This common health challenge, for example, saw a varying set of responses from trade unions regarding how to include the vaccine in their H&S strategy (Thomas et al. 2022).Since contextual and employment factors interact and affect the work environment and processes, and impact H&S (EU-OSHA 2020), effective ways to protect workers need to consider those factors, such as OSH risks, the institutional setting of industrial relations and existing H&S regulations at the national level.By the same token, national social partners may benefit from European level social partner activities and resources.H&S at work represents an important EU policy area (Article 153 TFEU).The Framework Directive on OSH at Work (89/391/EEC) obliges employers to take appropriate preventive measures to make work safer and enable employee involvement and consultation on H&S issues.The Directive requires member states to ensure that employees are informed and consulted about H&S matters but allows them to make their own proposals for improvements and changes.Thus, institutional differences in employee representation in OSH reflect national developments in H&S legislation and overall national structures of employee representation (COWI 2015; James /Kyprianou 2000).
Whilst similar in terms of OSH risks (EU-OSHA 2022), the hospital and social services sectors differ markedly regarding ownership and industrial relations systems at the country level.While healthcare provided by hospitals is mainly publicly funded and provided in Lithuania, Poland, and the UK, during recent decades public provision has increasingly been supplemented by the private sector, like in Germany and, to a lesser extent, in Sweden (Eurofound 2021).Regions have the main responsibility for public hospitals in Sweden and Italy (Galetto 2017).Although privatisation of social services provision is more common (Eurofound 2017), in Lithuania, Poland, and Sweden long-term residential care is mostly public while in Germany and the UK private, for-profit providers dominate.In Italy and Germany and, to a lesser extent, in the UK, non-profit organisations are prominent providers (Eurofound 2021;cf.Furthermore, atypical working times, shift work, and irregular working patterns are more likely found in social services (Eurofound 2020b).Long-term care professions do not always require formal qualifications (Eurofound 2020b) and are often associated with low pay, minimum wages (cf.Table A2) and precarious employment practices which, in turn, are linked to adverse OSH outcomes.Research shows, for example, how temporary agency workers experience a higher incidence of workplace injury (Underhill/Quinlan 2011).At the company level, H&S representatives together with management are responsible for OSH.Models of H&S workplace representation (Fulton 2018) differ regarding structure and representation patterns.H&S representatives can either be works councillors, union representatives, or elected workforce members.Further variation exists in relation to the way representatives are chosen, the number of employees required before an H&S representative must be appointed, and before a joint employer and employee H&S committee must be set up (cf.Table A1).For the hospital and social services sectors, data indicate that the density of H&S representatives is higher compared to all sectors and that issues of OSH are more often discussed at the company level between employee representatives and management.The frequency of discussions differs between countries, with Sweden and the UK above the EU average, Germany and Italy near the EU average, and Lithuania and Poland below (EU-OSHA 2022).
Although knowledge about emerging risks has increased (EU-OSHA 2019), the rise in non-standard employment and new forms of work have added to the complexity of OSH risk management (Gallagher/Underhill 2012).Atypical forms of employment and the dominance of private and small employers have an impact on effective H&S representation.Generally, the growth of more flexible work arrangements, subcontracting and the decline in union membership have undermined both the coverage and effectiveness of OSH provisions (Johnstone et al. 2005).
Care workers in both sectors are exposed to occupational risk factors for MSDs and to psychosocial risks which relate to the way work is designed, organised, and managed, as well as to the social context of work (Eurofound 2020b;Franklin/ Gkiouleka 2021).A recent review also points to the association of both risk factors, i. e. increased risks of MSDs due to psychosocial risks and therefore calls for a holistic approach to H&S in these sectors (EU-OSHA 2021).However, countries differ in terms of regular risk assessment in the two sectors, with the UK, Sweden, Italy, and Poland above the EU average, and Germany and Lithuania below (EU-OSHA 2022).

National social partners in the field of OSH -'vertical' coordination with the EU level?
The different models of H&S representation (Fulton 2018) and the different characteristics of industrial relations systems in the EU (EC 2009), which also vary between sectors (Bechter et al. 2012) are expected to affect the provision and governance of H&S protection in the country as well as the 'vertical' coordination with the EU level, including national social partners' views on and use of European sectoral social dialogue.OSH issues have been discussed both in the hospital SSDC and in the social services sector in the context of the informal social dialogue (cf.Tables A3 and A4).During the Covid-19 pandemic, social partners in both sectors issued joint statements and letters to the EC concerning the pandemic's H&S impacts (EPSU/HOSPEEM 2020; cf.Table A4).Social partners in the SSDC hospital have been working on H&S for many years, being prominent in the SSDC's joint work programmes.The so-called 'needlestick Directive' is believed to have had the greatest impact so far in prevent-ing health worker injuries, being an instance of a social partner agreement that was subsequently implemented via Directive (Bechter et al. 2021).6Recent activities of EU level social partners in the social services sector include e. g. a webinar/thematic meeting in June 2020 in the context of their joint project DialogueS (EPSU 2020; Social Employers 2020), or a joint webinar on MSDs in autumn 2021 in the framework of an EU-OSHA campaign.
We now turn to assess what our data say about 'vertical' coordination of social partners in the six countries with the EU level in the field of OSH.

Structural aspects and (potential) interlinkage
First, an evaluation of the structure of our sample (cf.Table 1), e. g. in terms of affiliations, will provide information about (potential) linkages and 'vertical' coordination with the EU level:7 (a) The number of interviews with trade unionists (33) more than doubles that with employer representatives ( 16).(b) Most interviews pertain to the social services sector (19), followed by the hospital sector (17), and interviews where respondents represented both sectors or an intersectoral view ( 16).(c) On the trade union side, there are 14 interviews covering hospitals, followed by social services (10), and interviews with representatives for both sectors or intersectoral (9).Of the total 33 interviews with trade unionists, 21 represented an organisation affiliated to EPSU.(d) On the employers' side, more interviews were conducted in the social services sector (8), followed by the hospital sector (6) and both/intersectoral (2).Country-wise, the fewest interviews with employers were conducted in Poland (1) and Lithuania (2).Of the overall 16 employer interviews, more are affiliated to Social Employers (5) than to HOSPEEM (4).(e) In sum, in 30 of the 52 interviews, interviewees represented a national social partner organisation affiliated with one of the European social partner organ-6 Article 3 par.2 of the Directive 2010/32/EU obliges member states to communicate to the EC only the main provisions of national law adopted but not to report on the practical implementation periodically (as e. g. foreseen in Directive 89/391/EEC, Article 17a, par.1).Responsibility for implementation review lays mainly on the shoulders of the social partners, who have issued two reports (cf.Table A3), and call on the EC to provide a comprehensive report (DE-3).7 It is important to note that the number of interviews does not equal the number of organisations interviewed, e. g. several interviewees from the same trade union, but representing different sectors.
isations in the two sectors, namely EPSU and HOSPEEM for the hospital sector, and EPSU and Social Employers for the social services sector.Another five interviewees (in Lithuania and Poland) represented organisations affiliated with another recognised EU social partner organisation.Furthermore, affiliations to other EU level organisations existed in Lithuania.
Therefore, the numbers show that the (potential) 'vertical' interlinkage of our sample to the sectoral social dialogue at the EU level is already restricted due to missing affiliation to one of the EU level social partner organisations.This is connected to the fact that (sectoral) structures, especially on the employer side, do not exist.Whereas EPSU has affiliated trade unions in all six countries of our sample, the more recently established employer organisations (HOSPEEM in 2005, Social Employers in 2017) do not: HOSPEEM is lacking affiliates in Poland, and Social Employers in Sweden and the UK.
Besides existing or missing affiliations, we can evaluate the potential interlinkages within national organisations in our sample.Of the 30 interviews with organisations affiliated to EPSU, HOSPEEM, or Social Employers, in 16 interviews, the interviewed persons did represent the organisation at EU level (no respondents from Italy).Almost half are also active in other policy fields.Most of them are responsible either for general policy and collective bargaining or (sectoral) H&S policy within the organisation, one person represents the workplace level.This in turn means that potentially links exist here from EU issues to other policy areas and levels within the organisation, through these persons.
A second theme are sector (in)congruences between the EU and the national level.One observation is the high number of trade unions (and in the case of Sweden also employers) that represent both sectors (cf.Table 1).This reflects the situation at the EU level, where EPSU is active in both social dialogues.At the national level, such a 'sector demarcation' -hospital vs. social services -can sometimes be difficult to identify.Furthermore, there are issues of 'incongruence' between the European sectoral level and the national level (Eurofound 2020a;2021), as two examples on the employer side illustrate.An Italian employer respondent in the hospital sector would like to see the regions represented in HOSPEEM, given their central role in delivering healthcare in the country: "Regions all have an office in Brussels […] they could contribute to the discussion more effectively, as they are the first that will have to implement any regulation" (IT-38).German respondents pointed to the fact that private employers, which make up a considerable share in both sectors in the country, are missing at European sectoral level .Our data confirm issues of incongruence may create practical problems in terms of 'top down' implementation, as well as 'bottom up' feeding of relevant information.

Faraway and blurred, but a closer EU when it comes to an OSH-specific agreement
A third theme relates to engagement in and knowledge of EU level activities.Being a member of one of the EU social partner organisations is a prerequisite to engaging in their activities, including sectoral social dialogue.However, we find that engagement in and knowledge about the European sectoral social dialogue is the exception rather than the norm among our respondents.For the most part, interviewees did not have detailed knowledge and were thus often unable to give examples of 'vertical' coordination in the field of OSH (e. g.SE-6).This seems related to the fact that we tried, where possible, to recruit national respondents with a sector (hospital, social services) and/or thematic focus (OSH) (cf.Section 2).More generally, the European level was regarded as being rather "far away" (DE-2) and "from above" (IT-6), not so relevant but gaining in importance (DE-4; DE-6; SE-2), not relevant (IT-5), or was even rather negatively referred to (DE-5).So the 'European sectoral dimension' within national affiliates seems to be worked on rather separately, for instance in German and Swedish trade unions: "Well, I don't dare say […] it's possible that [my superior] knows that sort of thing, but I don't know" (SE-3) and "if [colleague] brings certain [European] topics into our team meetings, then you notice that a bit, but it is more a brief input from time to time" (DE-5).
A respondent who has been engaged at the EU level for more than ten years also pointed to the peculiarity of that work: You can't believe it, but I think that these European level [processes], it's so it's super complex, so it's taken a long, long time.I've had a lot of different jobs in authorities and so on before, but I have to say, I think this has been one of the jobs that has taken the longest to learn.And I'm still not there, with all these layers.And it's also because all countries are so different and differently organised.
[…] So it takes an incredibly long time actually to get into it, I still think.That there is much to learn.(SE-4) An Italian interviewee underlined "If I had to organise assemblies to tell my members what is being done at this level, I would struggle" (IT-8).Other respondents, e. g. in Italy and Germany, mentioned scarce resources that were better invested in the daily national business due to the unclear benefit of European social dialogue.Whereas small teams have been brought forward (DE-2; DE-6), another respondent declared that "we have had instructions from the top to leave this activity aside and concentrate [the organisation's] resources on other priorities" (IT-5).
A fourth theme are examples of 'vertical' OSH coordination with the European level, including other bodies than EU level sectoral social dialogue, the so-called needlestick Directive, and activities such as projects and events.These examples of 'vertical' OSH coordination were mainly provided by trade union respondents.This might be connected to the fact that on the employer side OSH is often not seen as an issue for collective bargaining or social dialogue, but for national legislation (Lithuania) or the company/workplace level and occupational insurance (Germany) (LT-1; DE-2; DE-6).
Especially for coordination in the field of OSH, more reference has been made to the EC (particularly Lithuania, Poland) and also to other bodies such as the European Economic and Social Committee or EU-OSHA (DE-3; IT-1; LT-5; PL-5; PL-8) -although sectoral social partners in the two sectors previously and currently work on OSH-related activities in the respective sectoral social dialogue.A Polish respondent noted on the EU's strategic framework on H&S at work (EC 2021b) that it is good, because it emphasises the role of social dialogue in the field of H&S at work, that the EC clearly notes that the representation of employees at many levels -whether European, company, sectoral or some kind of cross-sectoral in the field of H&S -is weak.Trade union respondents in all six countries referred to one OSH-specific agreement reached in European sectoral social dialogue.More precisely, references have been made to Directive 2010/32/EU on sharp injuries, albeit not necessarily with a strong reference to the sectoral social dialogue.The directive implemented the 'needlestick agreement' reached between EPSU and HOSPEEM in the SSDC hospitals (cf.Table A3) and received very positive assessments from respondents.Interviewees highlighted that the directive demands that "unified standards should be followed in all workplaces" (LT-3) and the positive effect on awareness and training employees in preventing needlestick injuries, which has been "successful because of this EU directive, which was introduced in 2010" (PL-3).A Swedish respondent acknowledged that "certain things [from European social dialogue] [… are a] gain for us as well, where we are not progressing as much in Sweden.[…] I think the needlestick Directive […] it's something like that" (SE-4).Other positive assessments included respondents from Germany and the UK (e. g.DE-3; UK-5).
A Lithuanian interviewee reflected on the 'vertical' coordination in the making of the needlestick agreement, stating that at the European Federation […] various position statements are being discussed [and] formulated, and this is also a channel for participation in joint [initiatives], in the drafting of joint documents.There is also an exchange of information, for example on the Sharps Injuries Prevention Directive, when it was being drafted, there was quite a large flow of information, and we were certainly able to make comments or suggestions and finally to monitor the implementation of the Directive in our country.In addition, beyond sectoral social dialogue, references were made to projects and events with sectoral social partner organisations and other European level associations (e. g.LT-1; LT-3; LT-5; LT-8; PL-8; UK-9).This could be connected to the fact that some organisations interviewed in Lithuania and Poland were not affiliated with the sectoral EU organisations.In a similar vein, respondents articulated their general interest in exchanging and sharing ideas and practices with other countries and influencing the debate: "I think it's really interesting to hear what's happening in other countries as well.And what, you know, how they're implementing European directives" (UK-9).Another UK respondent underlined that "[we] can still be involved with some of the work that the trade unions across Europe are doing.[…] And I think H&S is an area that we can continue to benefit from actually" (UK-4).A Lithuanian interviewee explained that it was the problem of workload that our organisation raised at the European level and discussions took place in internal committees, because it was very important to us.We participated in that discussion and presented our experience.(LT-3)

Country-specific factors ground social partners to the national level
A fifth set of findings is connected to national contextual and institutional factors, i. e. also influencing within-country interlinkage of levels.Due to the weaknesses of their national industrial relations and social dialogue, Polish trade union respondents pointed at European social dialogue as an important factor to influence and improve national legislation, including H&S issues.One interviewee was sure: "It is there [at EU level] where the progress is actually made, and this is translated into Polish regulations, and thanks to this we actually have quite … we have some protection" (PL-8).
A similar pattern of a 'legal approach' can be found in Lithuania, especially for the field of OSH (LT-3), whereas 'legal impacting' is rejected by Swedish respondents, profiting from an established national system of social partners' collective agreements: For [our organisation] and our members, it is important and crucial that the European regulation that exists is designed in a general way, so that there is sufficient space and flexibility […] There must be room to handle the issues through the Swedish model (collective agreements, cooperation between the parties) with consideration for local and regional democracy and self-government.For countries with decentralised systems of industrial relations or weak industrial relations structures, it is especially challenging to work with (softer) sectoral social dialogue results, to implement and use them practically (Lithuania, Poland, UK).Respondents elaborated on country-specific factors: I would like to see more concreteness and more, well, strictness […] Our institutions, as I say, always tend to avoid, to make it easier, so, well, perhaps a stricter version would be more effective […] when there is a recommendation, it usually remains just a 'recommendation'… (LT-3) Therefore, enforceable regulations are the only effective tool especially in a country like Poland, where this dialogue on H&S is basically … undeveloped, one could say.[…] In such countries, where this H&S culture does not exist, which is the case in our country, only hard regulations are an effective tool, unfortunately.(PL-8) We have got [a charter], I think we have got around between 50 and 60 councils that have adopted it.[…] And then obviously, there's a question of whether councils that have adopted it are actually ensuring that it is implemented and are actually enforcing the standards that exist within it […].But again, because the workforce often is not unionised, it can be hard, for those standards to actually be enforced.(UK-7) Finally, our interview data revealed the important role of further institutions at the sectoral national level in the field of OSH (e. g. occupational health insurance in Germany where social partners are represented, joint social partner arenas in Sweden, tripartite structures in Poland), but where no concrete interlinkages to the European sectoral social dialogue have been reported .A Swedish respondent referred to the practical output of one of these joint social partner organisations in the country: [Organisation] makes tools based on research that will facilitate work environment work in the workplace.They make, yes, hands-on tools that we should be able to get out to the respective workplace, that you should be able to work with there.(SE-3) In Poland, the (post-)pandemic situation led to the setup of a healthcare working group within an existing tripartite structure which was influential for the workplace level (PL-5).However, while the institutions in Germany and Sweden are well-established, the new tripartite healthcare group in Poland soon lost pace and influence (PL-5).Thus, in some countries developed institutions could potentially be integrated and therefore support 'vertical' OSH coordination from the EU to the workplace level in the sector, whereas other countries lack such institutions.
To summarise, bold fonts in Figure 1 indicate the initial focus applied in this paper to explore the (potential) 'vertical' coordination and social partner interlinkages between the national and the European sectoral level.The empirical evidence from our interviews indicates that these interlinkages seem rather modest.The two levels where social partners (potentially bipartitely) are involved and could connect, seem not to be well-integrated but rather to simply coexist.In the context of the needlestick Directive, one of our respondents reflected on practical approaches to make the European (sectoral) level more visible: It is interesting to note that while in some interviews the respondents explained that they would rather invest their scarce resources into their daily national activities than into EU level activities, others underlined the importance of engagement at EU level and some would prefer stronger interlinkages with the EU level.We interpret the prioritisation of national activities in terms of financial difficulty, rather than as a general lack of interest.However, while adequate resources may foster engagement at EU level, prior research has also pointed to the relevance of a person's individual desire to engage, even despite scarce resources (Weber 2010).
Apart from the underlying factors described above, further possible explanations can be considered.Especially for Polish and Lithuanian interviewees, 'social dialogue' could imply all kinds of (trade union) involvement, e. g. consultation and involvement in advisory bodies at EU level.It has been noted earlier that terms like 'social partners', which implies a cooperative concept of interaction between management and labour, may not correspond to national wordings and concepts (Hyman 2021).Post Brexit UK is in a special situation regarding engagement in European social dialogue: the UK's unions no longer participate in social dialogue meetings but do so in other EPSU activities (UK-4; UK-5; UK-9).By contrast, UK employer representatives are still actively engaged in the SSDC hospital (UK-10; DE-3).

Conclusion and outlook
This paper investigates the following two questions: What interlinkages and 'vertical' coordination are there between social partners at the national and European level?To what extent are sectoral and institutional characteristics contributing to the coordination (or lack thereof) between national and European levels?For the first question, our overall conclusion based on our sample of national respondents is that there continues to be a degree of 'in-the-making' to the multilevel governance of employment relations, even in the presence of common challenges.The pandemic has promoted some coordination at national and between national and EU levels, yet we observed still rather modest interlinkages and degrees of 'vertical' coordination.We find that one European sectoral social dialogue result has again proven to be highly appreciated, which is the 'needlestick agreement' reached in the SSDC hospitals and implemented by Directive 2010/32/EU.However, the fact that it is based on a social partner agreement seems not to be entirely clear for some respondents at the national level.As one of the trade union interviewees who is involved in European sectoral social dialogue put it, there should be much more effort in proactively 'selling' such achievements 'at home'.While in the literature the possibilities of (legally binding) social partner regulation are often the focal point (e.g.Sørensen et al. 2022), others have called for attention to other assets that the European social dialogue may provide to national social partners (Larsson et al. 2020;Weber 2013).Based on our data with a focus on the OSH area, we conclude that there is clear evidence for the importance and perceived impact of legally binding regulation.However, knowledge and practice exchange can also be found as rationales behind the 'interlinking' of national social partners with the European sectoral level.
For the second question, clearly, for countries or sectors without a 'sectoral' level or missing affiliation structures to the EU sectoral level, it is difficult to establish any interlinkage.Our finding is that European relations are still not, or only to a small degree integrated into national social partners' daily business or sector policies (Weber 2010), but that different developments, especially in the social services sector (official launch of the SSDC, increasing role of multinational companies) may lead to greater visibility and therefore 'vertical' coordination in a broader sense.Currently, our data suggest that the degree of coordination remains rather minimal, despite or even when there are dedicated staff within organisations that could communicate and channel information between EU and national levels.Moreover, what became apparent in our interviews is that OSH-related issues are often regarded as best solved at the local/workplace level, particularly on the employer side.Examples of OSH coordination with the EU level were mainly presented by trade union respondents.Trade unions might also be more willing and able to establish a link between OSH and working conditions, and therefore to 'untighten' it from the mere local/workplace level (e. g. leave entitlements can be framed as relevant for OSH and regulated via (sectoral) collective agreements or other higher-level bipartite or tripartite bodies).
Interestingly, issues of concern in the countries investigated such as staff shortages, migration of care staff and associated challenges (e. g. training, skills recognition, skills drain), PPE availability, working time and effects on wellbeing and retention, are all also discussed at the EU level.Following on from our research on the currently weak interlinkages and vertical coordination between the national and European level, we therefore argue that OSH is indeed an area where national social partners in the two sectors would benefit from a supra-national form of dialogue, if not strict coordination.
Finally, the article also points to the limitations of cross-national and multilevel research on social partners, be it in the field of OSH or others, such as incongruent sector demarcations between the European sectoral and the national level.Besides differing institutional settings, 'softer' aspects such as the perceived relevance and different concepts of OSH must be considered in further research.

Tab. 1: Sample
-Number of interviews conducted in the six countries.Own compilation.

(54) 33 (34) 16 (17) 3 17 19 16 (18)
The corresponding 'sectors' at EU level, i. e.European sectoral social partner fora, are the SSDC hospitals and the (until July 2023 informal, cf.Section 3) social dialogue in the social services sector.The European umbrella social partner organisations in the hospital SSDC are EPSU (European Public Service Union) and HOSPEEM For the interplay between European socialWeber 2010and national industrial relations, the question of multilevel action and implementation at the national level is crucial(Keller/Weber 2011).The challenges for implementation -e.g.actors' diverging interests, issues of representativeness and coverage (ibid.)-and the question of potential and actual effects in the multilevel context of European social dialogues have been discussed first and foremost in the context of autonomous agreements which have to be implemented by the social partners themselves, but also regarding other 'softer' joint outcomes(Keller/Weber 2011; Keune/Marginson 2013;Weber 2010).It has previously been noted that the extent to which such a multilevel industrial relations framework is able to produce tangible effects for European workers often remains uncertain(Marginson/Sisson 2004).
Note: NACE 87 -Residential care activities; NACE 88 -Social work activities without accommodation, except child day care activities and other social care activities without accommodation.
It would be good to have practical evidence for workers what the EU does for them.[…] Once I gave one [example] of the safe needles to a colleague, in the context of the needlestick Directive, to make it transparent, to show it can be experienced […].And [the needlestick Directive] is a really good instrument.[…] But we do not talk enough about it, also because of the complexity.(DE-1)