Important collaborative conditions for successful economic outcomes of work disability management: A mixed methods multiple case study

Abstract

BACKGROUND:

Work disability management (WDM) interventions have usually focused on a single factor and its impact on outcomes such as employee health or work disability costs. Research on company-level WDM activities and their economic impact is scarce.

OBJECTIVE:

The aim of this study was to explain the change in company-level work disability costs and their relation to WDM practices, and to identify the mechanisms of impact through which the successful economic outcome emerged.

METHODS:

The study design was a convergent mixed methods design with a multiple case study. The data from 14 business units concerned business context, personnel, investments in and processes of WDM, and the costs of work disability in 2010-2013. The data were constructed into case descriptions which were analysed using qualitative comparative analysis. The economic analysis was carried out from the employers’ perspective.

RESULTS:

Five business units gained net benefits of approximately 1.5-2.5% of the payroll sum from their investments in WDM. These benefits were characterised by a combination of four strategic processes: i) dismantling barriers to co-operation, ii) improving the visibility of the strategic goals of work ability management in everyday practice, iii) WDM actions targeting the company’s main work disability risks, and iv) the facilitation of multi-actor co-operation through co-ordination and flow of information.

CONCLUSION:

Strategic processes to support the effectiveness of WDM were found. When aiming for economic success in work disability management, in addition to measuring and managing disability costs, it is also essential to maintain collaborative operations in everyday practice.

Keywords

Work disability management occupational health qualitative comparative analysis economic evaluation

1 Introduction

The demand for skilled labour is growing. Hence, to achieve their business goals and to create value for their shareholders, companies need to invest in human capital. These investments are also expected to be economically attractive. At the same time, the demand for companies’ social responsibility is also increasing [1]. The competition for skilled labour makes well-being at work, the personnel’s work ability, decent working conditions, and well-functioning work organisation essential components of a good corporate image to attract employees.

Many types of interventions have been used to reduce absenteeism and work disability: health promotion and disease prevention programmes, fitness and exercise programmes, occupational health and safety risk management programmes, disease management programmes, employee assistance and behavioural programmes, worksite medical clinics, and disability management programmes [e.g., 2–9]. Although the results have often been conflicting, some studies have reported successful economic outcomes [10 –14]. Quantifying the impact of choices and approaches to productivity change is methodologically uncertain [15]. The country, social security, and the health care context also play a role in measuring the economic impacts of workplace-based programmes [12]. Work disability management (WDM) comprises a wide variety of activities: workplace injury prevention programmes, early intervention and rehabilitation programmes, and early and safe return to work after sickness or injury [7 , 16]. Research in the field has focused on three organisational challenges for employers: i) facilitating return to work, ii) enabling staying at work, and iii) providing work accommodation and support for workers with disabilities [17]. Research evidence shows that active workplace involvement is a fundamental component of successful and sustainable return to work and staying at work [18]. However, due to a limited number of high-quality studies, knowledge on the effectiveness of WDM programmes is scant [7, 8]. To our knowledge, no research has previously compared company-level WDM activities and their economic impact.

To an employer with a total payroll sum of EUR 100 million, sound investment in WDM is potentially worth millions of euros. There is a gap in the current literature concerning this: most interventions have focused on a single factor and its impact on work disability costs rather than comparing how the total investments in prevention affect work disability net costs. To fill this gap, an economic analysis from the employer’s perspective was carried out, using multi-perspective data on WDM activities. The objective was to explain the relation of WDM practices carried out in workplaces to the change in work disability costs in 2010–2013, and to identify the mechanisms of impact through which the successful economic outcomes emerged.

2 Materials and methods

2.1 Study design

The context of the study was the Finnish business environment and the comprehensive Finnish social security system [19]. The study design was a quantitative data dominant mixed methods study [20 –22] which used a convergent design [20], combined with a multiple case study [23 –27]. The mixed methods design was applied because it enabled to combine quantitative and qualitative research approaches, especially inference techniques, to obtain an in-depth understanding of the corroboration [21] and integrative strategies in the analysis [28]. Triangulation [22] and inter-method mixing [29], in turn, was used to gain an in-depth understanding of the complex [30], multi-faceted nature of the WDM processes in different settings, and their outcomes (Fig. 1).

Fig. 1

Mixed methods study design: convergent design combined with case study.

In the within-case analysis phase, the data were constructed into case descriptions that included all the stakeholders’ multifaceted points of view. For the across-case analysis phase, crisp-set qualitative comparative analysis (cs-QCA) [31] was used to explore the relations between the variables of interest. The cs-QCA technique is based on Boolean algebra, and uses only binary data, which means that variables or conditions of interest are dichotomised on the basis of some meaningful dichotomisation of each variable. This allowed to study how the factors combined into configurations of necessary and sufficient conditions for the outcome. In the across-case economic analysis, cost-benefit analysis was carried out using monetary values presented as a share of the total payroll sum. The economic outcome analysis was conducted from the employers’ perspective, but rather than isolated interventions, it covered the entire WDM process and its related costs.

2.2 Cases

A total number of 100 companies were contacted by phone or by mail. Sixty-four companies had taken part in a previous benchmark survey [32] and the rest were taken from the registers of a private health service provider. Ten companies with 600–11 500 employees agreed to participate. Nine of these had taken part in the previous benchmark survey. They provided retrospective data on finances and the human resource (HR) processes of WDM, and the related costs. These companies represented different business sectors, organisational models, and personnel profiles. In the ten companies, 20 business units were defined as cases for the current study. Complete datasets were available from 14 units.

2.3 Outcome

The change in average work disability costs from 2010–2011 to 2012–2013 was used as main outcome. These costs included accident insurance costs, the experience rating of disability insurance premiums, and the (direct) costs of sickness absence valued at average salary. In the within-case analysis, the costs were presented as ‘per full-time equivalent’ measures. In the across-case analysis they were presented as the percentage share of total payroll, to make them more comparable across companies.

2.4 Datasets

Both quantitative and qualitative data were collected from the companies. Figure 2 presents the datasets.

Fig. 2

Data and data integrating processes.

2.4.1 Data on work disability management investments and work disability costs

The disability prevention investments, i.e., investments in WDM, considered in the current study included health promotion, occupational health services (OHS), work safety and accident prevention, health and safety training, management of work disability training, relevant ICT systems, voluntary insurance, and other funding. The work disability costs included direct costs of sickness absence, work disability pension costs, and work accident costs. For nine companies, the data were based on a previous WDM study [32]. For the remaining five, the data were obtained directly from the companies’ human resources management (HRM). These data included information on personnel, business activities, investments in work disability prevention, and the costs of work disability. Sickness absence was valued using the human capital method, multiplying the days of absence by average wage. All monetary measures used were in euros, discounted to the 2013 value using a discount factor. Both the work disability costs and the WDM investments were presented as per full-time equivalent values in the within case analysis. In the across-case economic analysis, the percentage share of total payroll sum was used to make the measures comparable across the different business units.

2.4.2 Data on work disability prevention processes

The data on the practices, structures, resources, and processes of WDM, as well as internal and external co-operation, OHS in particular with, were collected from the company records and through questionnaires and focus group interviews of the top management, HRM, supervisors, and employee representatives. The data on business context, personnel, and well-being activities in the companies were also collected from public sources such as the companies’ annual reports.

2.4.3 Questionnaires

The data on WDM policies, structures, and resources were collected via a questionnaire, which was filled in by the person responsible for WDM. These data were verified and supplemented later by interviews and emails if required.

The data on work management processes, practices, and collaboration both inside and outside the company were gathered using a questionnaire, in line with previous studies.

The questionnaire contained the seven following themes:

Realisation of WDM operational activities in the company (13 propositions)

Realisation of WDM by senior management (4 propositions)

Realisation of WDM by supervisors (8 propositions)

Realisation of WDM from the employees’ perspective (7 propositions)

Implementation of WDM (9 propositions)

Implementation of work safety management (11 propositions)

Realisation of external collaboration in WDM (10 propositions)

The aim was to obtain the best knowledge on WDM from the stakeholders identified by the HR expert in each company. In practice, this meant contacting the management, HRM, supervisors, safety managers, safety representatives, and shop stewards in each company. The questionnaire was sent to 385 people, of whom 239 (62%) responded. The number of potential respondents in the companies varied from 13 to 143, depending on the size and organisational structure of the company. The response rate varied from 46% to 85% between the companies.

2.4.4 Focus group interviews

The semi-structured focus group interviews [33] of the main stakeholders in the WDM processes were conducted in all the companies, with two researchers present in each interview. The interviewed stakeholders were managers, HR managers and specialists, OHS managers, supervisors, employee representatives, and OHS professionals appointed by the research partner at each company. The combination of interviewees varied, and the size of the interviewees’ group varied from four to nine. The interviews, based on written consent, were recorded and transcribed verbatim, forming altogether 216 single-spaced pages. The most discussed themes in the interviews were the concept of WDM, WDM collaboration and using knowledge management as a part of WDM. Before the interviews, the documents, the first outcomes of the investments and costs analysis, and the survey questionnaire responses were analysed to form the preliminary overviews of the case and redefine the themes for the interviews. The first phase of the interview data analysis was to discuss how the mutually formed preconception changed during the interview [34]. The analysis was driven by the research questions.

2.5 Analysis

2.5.1 Analysis of quantitative and qualitative data

Data on the companies’ WDM processes, costs and related outcomes were used to form a company-level case description. Descriptive statistics were used i) to summarise and plot a graph of the development of the outcomes in 2010–2013 and ii) to summarise Questionnaires 1 and 2. These data were used to select the themes for the focus group interviews, and for the analysis carried out with cs-QCA.

The WDM processes were studied in depth, each case within its context. In addition, it was further discussed whether they had more variables of interest than the other datasets [27]. The qualitative data were gathered and analysed in accordance with the mixed methods strategy [20]. The interview data were analysed both jointly with multiple data, and as their own dataset. In preparation for the data collection, a pre-interview summary of each case was constructed using the multiple sources of data gathered earlier in the study. This also refined the interview themes. After the interview, an analysis was carried out of how the understanding of the context, processes, and implementation of the WDM process had changed according to the interview data. If required, a new summary was constructed. Second, the data for the purpose of cross-case analysis were analysed, first according to the themes of each case description and second as one dataset in the qualitative comparative analysis process.

2.5.2 Within-case analysis: Case descriptions

In accordance with the mixed methods convergent design [20], the quantitative data (investments in employee health, safety and work ability, work disability costs, and questionnaire data) and qualitative data (interviews and documents) were merged. The data were constructed as case descriptions of the WDM process and economic outcome of each company; a complete dataset of the cases was available in the case descriptions, which were submitted as a case report to each participating company. In practice, the data on the companies were condensed into case descriptions under the following themes: data on business and personnel; changes in strategic work disability practices on the managerial, operational and practical levels; case-by-case results on work disability costs; investments in WDM; WDM measures; and conclusions with practical implications. Each case description consisted of about 40 pages. The data analysis procedure integrated the within case and across-case analysis approaches [35]. The case descriptions were sent to the companies for discussion and validation.

2.5.3 Across-case analysis: Qualitative comparative analysis

The empirical part of this study employed the cs-QCA [31] on a sample of 14 companies. The aim of the across-case analysis was to distinguish the factors that promote or inhibit the WDM process in relation to the net economic outcome. In addition to being a convenient tool for across-case analysis, cs-QCA also takes account of within-case complexity, thus allowing the examination of the company-level factors that affect the success of disability management.

The conditions used in cs-QCA were composed according to the literature, research questions and early findings of the multiple data. Table 1 presents the conditions, outcomes, and their descriptions. In this study, the outcome and conditions were operationalised as follows: the main outcome ‘Did the average work disability costs decrease from 2010–2011 to 2012–2013?’ was operationalised as 1 if the costs decreased, and 0 if costs increased. If the condition was fulfilled, the condition for the case was coded as 1. If not, the condition was coded as 0. The coding of conditions was based on thorough case knowledge and the researchers’ joint decision-making in a series of multidisciplinary team meetings of four researchers from the fields of health science, occupational health, rehabilitation, and economics. Each researcher knew certain datasets more specifically and brought case-specific data justification for the extract at hand to the discussion. In the transformation of the multiple data into a single numerical value, the databased extracts for the coding of conditions on which the decisions were based were recorded in a memorandum.

Table 1
Conditions, outcome and their descriptions

Conditions Description

C1 Were the obstacles to WDM co-operation due to the nature of work or the organisational model dismantled? Changes in the business environment, the way in which work was organised, and how various work characteristics, such as business mergers and acquisitions, labour demand, staff turnover, multi-location work, and shift work were considered in the implementation of WDM. Well-being at work and occupational safety management were integrated.

C2 Were the strategic goals of WDM reflected in its practice? WDM was part of the organisation’s strategy and day-to-day operations. It had concrete goals and indicators. The implementation of objectives and measures was monitored regularly. There was commitment to WDM.

C3 Did WDM measures comprehensively target the company’s key disability risks? The organisation had a comprehensive and up-to-date view of the disability risks and the factors that affect them. WDM was an ongoing activity. Key risks were prioritised in the allocation of measures. The effects of the measures were monitored and evaluated.

C4 Were the supervisors actively involved in the implementation of WDM and OHS activities? Supervisors had sufficient WDM know-how. They participated in setting WDM goals and in planning, implementing, and evaluating measures. They informed their subordinates and provided feedback to them on WDM.

C5 Did the coordination of WDM and the flow of information support multi-stakeholder co-operation? WDM roles and responsibilities were clear. The named coordinator was responsible for the operationalisation of the disability management. Communication about WDM was continuous and transparent. Co-operation with OHS and the pension insurance company was regular.

C6 Did the employees participate actively in the implementation of WDM and OHS activities? The employees participated in setting WDM-related goals, in providing information on matters endangering well-being at work and work safety, and in planning and implementing measures to improve well-being at work and safety.

C7 Were the indicators systematically used to monitor and develop the effectiveness of the WDM and OHS activities? Concrete effectiveness targets and indicators were set for the management of disability risk, and, were monitored regularly. The results were utilised in the development of operations.

Outcome Did the average work disability costs decrease from 2010–2011 to 2012–2013? The work disability costs included accident insurance costs, experience ratings of the disability insurance premium and the (direct) costs of sickness absence valued as average salary.

Conditions	Description
C1	Were the obstacles to WDM co-operation due to the nature of work or the organisational model dismantled?	Changes in the business environment, the way in which work was organised, and how various work characteristics, such as business mergers and acquisitions, labour demand, staff turnover, multi-location work, and shift work were considered in the implementation of WDM. Well-being at work and occupational safety management were integrated.
C2	Were the strategic goals of WDM reflected in its practice?	WDM was part of the organisation’s strategy and day-to-day operations. It had concrete goals and indicators. The implementation of objectives and measures was monitored regularly. There was commitment to WDM.
C3	Did WDM measures comprehensively target the company’s key disability risks?	The organisation had a comprehensive and up-to-date view of the disability risks and the factors that affect them. WDM was an ongoing activity. Key risks were prioritised in the allocation of measures. The effects of the measures were monitored and evaluated.
C4	Were the supervisors actively involved in the implementation of WDM and OHS activities?	Supervisors had sufficient WDM know-how. They participated in setting WDM goals and in planning, implementing, and evaluating measures. They informed their subordinates and provided feedback to them on WDM.
C5	Did the coordination of WDM and the flow of information support multi-stakeholder co-operation?	WDM roles and responsibilities were clear. The named coordinator was responsible for the operationalisation of the disability management. Communication about WDM was continuous and transparent. Co-operation with OHS and the pension insurance company was regular.
C6	Did the employees participate actively in the implementation of WDM and OHS activities?	The employees participated in setting WDM-related goals, in providing information on matters endangering well-being at work and work safety, and in planning and implementing measures to improve well-being at work and safety.
C7	Were the indicators systematically used to monitor and develop the effectiveness of the WDM and OHS activities?	Concrete effectiveness targets and indicators were set for the management of disability risk, and, were monitored regularly. The results were utilised in the development of operations.
Outcome	Did the average work disability costs decrease from 2010–2011 to 2012–2013?	The work disability costs included accident insurance costs, experience ratings of the disability insurance premium and the (direct) costs of sickness absence valued as average salary.

Using scoring, all the data were condensed and transformed into binary data according to cs-QCA [31, 36]. Multiple data were jointly analysed through the researchers’ collaboration, and the level of integration of the data was merged into a joint display [24 , 37]. In practice, the scope lay in realisation and non-realisation of the conditions and the outcome. The scores were adopted in two phases: first, each researcher made their own scorings, which were then jointly reflected on in the multidisciplinary team, the aim being joint understanding of the conclusions. During the revision of the truth table, the condition was coded 0 in the case of any contradictory databased extracts on the condition.

The necessary and sufficient conditions for the presence of the outcome were analysed ahead of the minimisation of the configurations, based on Boolean simplification. Only configurations for which cases were available were determined.

3 Results

3.1 Costs and investments in the within-case analysis

Supplementary Table 1 in the Appendix presents the descriptive statistics of the 14 companies. Most of the companies operated in many locations. They represented various industries: transport and storage; the food industry; metal production; the paper industry; administrative and support services; and wholesale and retail. Their average turnover varied from tens of millions of euros to billions of euros. The number of person-years varied from 600 to 11 500. The share of women was high in the wholesale and retail industries but lower in some industrial companies. The share of older personnel was higher in the industrial, transport and storage companies than in the other industries.

The work disability costs of sickness absences, accident insurance, and disability pension experience rating amounted to an average of EUR 2600 per person-year (EUR 1045–4700, median EUR 2574) during the period 2010–2013 (Fig. 1). Work disability costs per person-year decreased in about half of the participating organisations and their units. The most significant cost item was the direct cost of sickness absences, which varied from 50%–85% of work disability costs. The amount invested in WDM in 2010–2013 was on average EUR 900 (EUR 71–3400, median EUR 710) per person-year (Fig. 2). The companies increased their investments in WDM during follow-up by approximately 0.25%–1.5% of the payroll. Work disability costs per person-year decreased in about half of the participating companies and their units. In relation to payroll, the change in work disability costs in the studied organisations was between a decrease of 2% and an increase of 1.5%. From 2010 to 2013, nine companies obtained a net benefit from their investments in work disability prevention.

3.2 Across-analysis using QCA

Table 2 reports the truth table for the cs-QCA of successful disability management, measured as a decrease in average work disability costs between 2010–2011 and 2012–2013. Eleven different configurations were supported by 14 cases. A decrease in the average work disability costs was observed in nine cases and one contradictory configuration was observed. The configurations related to the reduction in total work disability costs were i) the dismantling of barriers to co-operation, ii) showing the strategic goals of work ability management in practice, iii) the way in which the WDM focused on the main work disability risks, and iv) the facilitation of multi-actor co-operation through co-ordination and flow of information.

Table 2
Presence of conditions and outcome in case companies

Cases

Conditions and outcome F, G, I A K J E H N L M B C, D

C1: Were the obstacles to co-operation in WDM due to the nature of the work or the organisational model dismantled? 1 1 1 1 1 1 0 0 0 0 0

C2: Were the strategic goals of WDM reflected in the WDM practice? 1 1 1 1 1 1 1 1 1 0 0

C3: Did the WDP measures comprehensively target the company’s key disability risks? 1 1 1 1 1 1 0 0 0 1 0

C4: Were the supervisors actively involved in the implementation of WDM and OHS activities? 1 1 1 0 0 0 1 0 0 0 0

C5: Did the co-ordination of WDM and the flow of information support multi-stakeholder co-operation? 1 1 1 1 1 1 1 0 0 0 0

C6: Did the employees actively participate in the implementation of WDM and OHS activities? 1 0 0 1 0 0 0 1 1 0 0

C7: Were the indicators systematically used to monitor and develop the effectiveness of WDM and OHS activities? 1 1 1 0 1 0 0 1 0 0 0

Outcome: Did the average work disability costs decrease from 2010–2011 to 2012–2013? 1 1 0 1 1 1 1 0 1 0 0

	Cases
C1: Were the obstacles to co-operation in WDM due to the nature of the work or the organisational model dismantled?	1	1	1	1	1	1	0	0	0	0
C2: Were the strategic goals of WDM reflected in the WDM practice?	1	1	1	1	1	1	1	1	1	0
C3: Did the WDP measures comprehensively target the company’s key disability risks?	1	1	1	1	1	1	0	0	0	1
C4: Were the supervisors actively involved in the implementation of WDM and OHS activities?	1	1	1	0	0	0	1	0	0	0
C5: Did the co-ordination of WDM and the flow of information support multi-stakeholder co-operation?	1	1	1	1	1	1	1	0	0	0
C6: Did the employees actively participate in the implementation of WDM and OHS activities?	1	0	0	1	0	0	0	1	1	0
C7: Were the indicators systematically used to monitor and develop the effectiveness of WDM and OHS activities?	1	1	1	0	1	0	0	1	0	0
Outcome: Did the average work disability costs decrease from 2010–2011 to 2012–2013?	1	1	0	1	1	1	1	0	1	0

In more detail, cases A and K had the same configuration of conditions (Table 2), but the work disability costs in Case A decreased, whereas in Case K they increased. In most configurations of the conditions leading to a decrease in work disability costs, several conditions were fulfilled simultaneously. In the ‘necessary and sufficient’ analysis, the condition of the strategic goals of work ability management being reflected in the practice was not necessarily a sufficient condition for a decrease in work disability costs. In another case however, L (Table 2), work disability costs increased despite the condition of the strategic goal of work ability management being reflected in practice being fulfilled.

The minimisation of configurations (Table 3) is presented next. The purpose of this analysis was to determine the conditions that emerged from a decrease in work disability costs. Model ambiguity was found. Two descriptive formulas led to a decrease in the average work disability costs between 2010–2011 and 2012–2013. Both models included four shared configurations. These formulae differed in only the last two configurations.

Table 3

Minimisation of configurations leading to decrease in average work disability costs from 2010–2011 to 2012–2013. Note: Conditions C1–C7 are described in Table 1. *indicates a membership in condition. ∼indicates a non-membership in condition

Outcome: Did the average work disability costs decrease from 2010–2011 to 2012–2013?
Configuration of conditions	Raw coverage	Unique coverage	M1	M2	Consistency	Cases
C1C2C3∼C4C5∼C7	0.222	0.111	0.111	0.222	1.00	H, J
C1C2C3C4C5C7	0.444	0.333	0.444	0.333	0.800	A, K, F, G, I
∼C1C2∼C3∼C4∼C5C6∼C7	0.111	0.111	0.111	0.111	1.00	M
∼C1C2∼C3C4*C5∼C6∼C7	0.111	0.111	0.111	0.111	1.00	N
C1C2C3∼C4C5∼C6	0.222	0.000	0.111		1.00	H, E
C1C2C3C5∼C6*C7	0.222	0.000		0.111	0.667	E, A, K

The first minimised configuration from the first descriptive formula was observed in Cases H and J. The consistency of 1.00 showed that work disability costs decreased in both cases. The first configuration included the following conditions: the obstacles to co-operation in WDM being dismantled (*C1, where the asterisk indicates that there was a membership in this condition), the strategic goals of WDM being reflected in its practice (*C2), WDM targeting the key risks (*C3), and the co-ordination of WDM and flow of information supporting multi-stakeholder co-operation (*C5). The lack of supervisors’ active involvement in the implementation of WDM and OHS (∼C4, where the tilde indicates that there was a non-membership in this condition) and the systematic use of indicators to monitor and develop WDM and OHS activities (∼C7) were included in the configuration. The raw coverage value of 0.222 suggests that the configuration represented the second most empirical importance of all the configurations (Table 3, row 1).

The second minimised configuration was observed in Cases A, K, F, G and I. The consistency of 0.800 indicates that in one case (K), costs increased despite the configuration conditions being fulfilled. In these cases, the obstacles to co-operation in WDM were dismantled (*C1), the strategic goals of WDM were reflected in its practice (*C2), WDM targeted the key risks (*C3), supervisors were actively involved in the implementation of WDM and OHS (*C4), the co-ordination of WDM and flow of information supported multi-stakeholder co-operation (*C5), and indicators were systematically used to monitor and develop WDM and OHS activities (*C7). Configuration was empirically supported by a raw coverage value of 0.444. (Table 3, row 2). The exception in this configuration was Case K, which experienced a slight increase in work disability costs. This increase in costs was due to an increase in the experience rating of the disability insurance premium. Case K had a relatively high number of work disability pension cases in 2010, which affected the premium paid by the company in 2012 and 2013.

The third configuration was observed in Case M, with a coverage value of 0.111 and consistency of 1.00. The configuration included the following conditions: the strategic goals of WMD were reflected in its practice (*C2) and the employees actively participated in the implementation of WDM and OHS activities (*C6). The obstacles to co-operation in WDM were not dismantled (∼C1), WDM did not comprehensively target the key risks (∼C3), supervisors were not actively involved in the implementation of WDM and OHS activities (∼C4), the co-ordination of WDM and the flow of information did not support multi-stakeholder operation (∼C5), and indicators were not systematically used to monitor and develop WDM and OHS activities (∼C7). Case M represented an exception in the configuration of conditions. The decrease in work disability costs was due to exceptionally high accident insurance costs, caused by a major accident in the company’s production facilities in 2009. Because of this, the minimisation of configurations was repeated without Case M. By excluding this third configuration reduced the total number of configurations to five (Table 3, row 3).

The fourth configuration was only observed in Case N, with a coverage value of 0.111 and consistency of 1.00. It included the following conditions: the strategic goals of WDM were reflected in its practice (*C2), supervisors were actively involved in the implementation of WMD and OHS activities (*C4) and the co-ordination of WMD and the flow of information supported multi-stakeholder co-operation (*C5). The obstacles to co-operation in WDM were not dismantled (∼C1), WDM did not target the key risks (∼C3), employees did not actively participate in the implementation of WDM and OHS activities (∼C6), and indicators were not systematically used to monitor and develop WDM and OHS activities (∼C7). (Table 3, row 4).

The fifth and sixth configurations did not have unique coverage. The configurations were included in the cases (H and E in the fifth; and E, A and K in the sixth) that already represented other configurations. However, they both had empirical relevance, as shown in the raw coverage of 0.222 per configuration. The consistency value of 0.667 for the sixth configuration shows that in one of the cases (K), work disability costs increased despite the fulfilment of conditions. Case K faced a slight increase in work disability costs. The configurations included the following conditions: the obstacles to co-operation in WDM being dismantled (*C1), the strategic goals of WDM being reflected in its practice (*C2), WDM targeting the key risks (*C3) and the co-ordination of WDM and flow of information supporting multi-stakeholder operation (*C5). In the fifth configuration, supervisors (*C4), and employees (∼C6) were not actively involved in WDM and OHS activities. In the sixth configuration, the employees were not actively involved in WDM and OHS activities, yet indicators were systematically used to monitor and develop WDM and OHS activities (*C7). (Table 3, rows 5 and 6).

The horizontal axis in Fig. 3 shows changes in WDM investments, and the vertical axis shows changes in work disability costs between 2010 and 2013 as a percentage share of the total payroll sum. The number next to the case index shows how many conditions were fulfilled.

Fig. 3

Change in work disability costs and investments in work disability management from 2010 to 2013 as a percentual share of payroll sum. The number after the business unit index refers to how many conditions were fulfilled in the QCA. For example, K6, Company K fulfilled 6 conditions.

One case (K) did not increase WDM investments. From 2010 to 2013, Case K saw a small increase in work disability costs, while WDM investments decreased by about 0.75% of the payroll sum. It gained net benefits due to the small cost increase in work disability costs and a larger decrease in WDM investments.

Four cases (B, C, N, and L) had an increase in work disability costs and WDM investments also increased. Only a few conditions, ranging from none to three, were fulfilled in these cases.

The rest of the cases (A, D, E, F, G, H, I, J, and M) faced a decrease in costs from 2010 to 2013, while WDM investments increased. Four to seven conditions were fulfilled in Cases A, E, F, G, H, I, and J. The remaining cases, D and M, were exceptions in this group: only two (M) or no (D) conditions were fulfilled. The change in work disability costs in Case D was close to zero percent. Work disability costs were exceptionally high in Case M in 2010. This was due to substantial accident insurance costs in the same year, which resulted in a significant decrease in work disability costs as a share of the payroll sum in 2010–2013.

Net economic benefits, measured as a share of total payroll between 2010 and 2013, were gained in six cases: A, G, I, J, K and M. The decrease in work disability costs outweighed the increase in investments in Cases G, I, J, K and M. In Case K, the decrease in investments outweighed the increase in costs. Case M was an exception, but in Cases A, G, I, J and K, five to seven conditions were fulfilled.

4 Discussion

The objective of this study was to explain the relation of WDM practices carried out in workplaces to the change in work disability costs in 2010–2013, and to identify the mechanisms of impact through which the successful economic outcome of work disability costs emerged. Results showed that it was not only the money invested in, but the processes targeted at the recognised work disability risks that brought about net benefits.

Four conditions were found as important for achieving a successful economic outcome in different business units: i) the dismantling of barriers to co-operation, ii) the visibility of strategic work ability management in everyday practice, iii) the targeting of WDM actions at the main risks, and iv) the facilitation of multi-actor co-operation through co-ordination and support of the flow of information. Consensus on the efficacy of the WDM components is still limited. However, common elements for successful WDM interventions have been identified (e.g., 2, 7, 17, 38), and these elements were also part of the processes that supported the successful economic outcome of WDM in this study. Co-operation and good communication among stakeholders have often been mentioned as essential factors for successful WDM [2 , 39]. However, continuous changes in the work environment may challenge these if WDM processes are poorly described and duties and responsibilities are unclear. In this study, the companies that had taken care of their WDM processes managed better in such situations. Successful WDM requires co-operation among several stakeholders [2 , 39]. In this study, the organisation of WDM varied; for example, the responsibilities of supervisors and HRM were organised differently in different business units. In both models of organising WDM, the main success factor was co-ordination. According to the findings of this study, the business units that co-ordinated the implementation of the key elements of WDM processes were more likely to achieve favourable results. In practice, this meant recognised, jointly agreed targets and the assignment of responsibilities. Regarding the contextual factors, the business units that succeeded in maintaining the WDM structures, activities, and co-operation in the midst of business change and structural reorganisations achieved successful economic outcomes. This study also indicated that linking WDM to business objectives and effectively targeting high-risk individuals were successful, an observation also made in previous research [38].

The research strategy aimed to tackle some of the methodological limitations and uncertainties identified in previous WDM studies. First, the methodological approach of using cost-benefit estimation of WDM on the company level is seldom used. Interestingly, no previous studies have attempted to identify total investments in WDM and its impact on the total costs of work disability. A recent systematic review of economic evaluations of occupational safety and health interventions found that organisational-level economic evaluations from the employer perspective exist but are uncommon [10]. The methodological standardisation of the quantification of economic benefits is the very foundation of the attractiveness of conducting investments in occupational health and safety [15]. In this study, the quantification of the factors and measures used in economic analysis were carried out in the large-scale, company-wide operation of WDM without an existing standardised methodological approach. The comprehensive case-specific data collection included investments, costs, interventions, and processes and enabled the detailed estimation of the net economic benefit of WDM. Presenting economic measures as a share of total payroll helped make costs and investments more comparable across the companies.

The use of QCA enabled to compare the detailed multi-data descriptions and to analyse the common outcome in different business sectors. Although the companies differed from each other in business field, operational model, organisational structure, and composition of personnel, it was possible to identify the conditions that can be considered a vital part of WDM in any business sector or company. Material, method, and researcher triangulation in a multidisciplinary research team improves reliability.

4.1 Limitations

The companies had difficulties obtaining detailed personnel data, and moreover, accurate data on certain WDM investments, such as occupational health and safety, WDM training, and information technology and software investments. Often, these data could not be identified in detail in the accounting information. The structural safety investments in some of the participating companies were substantial. However, these investments were not included, as they prevented an equal comparison between the companies.

It was difficult to obtain all the required data from the companies and thus be involved in such an in-depth investigation. This caused selection bias. The companies that participated in this research project had interests in studying and improving their WDM processes. Although QCA offers a way to identify a common explanation for the outcome with a small number of cases, these findings merit confirmation with a larger, more generalisable sample of companies. Research on work disability costs can be conducted using representative register data on companies’ costs. However, detailed context- and case-specific knowledge related to investments in WDM is difficult to obtain without a serious effort in gathering the information. This limits the opportunities to confirm these findings using a larger sample of companies in the future. However, there is a demand for research in the field of investments in intangible capital and the related effects on economic outcomes.

Some economically relevant outcomes related to WDM were outside the scope of this study. These include the cost of staff turnover, the potential effects of employee satisfaction, presenteeism and the actual productivity effects of WDM. Including these outcomes would make the economic analysis more thorough. For example, a recent study [40] found that on average, 53% of lost productivity due to sickness absence was attributable to presenteeism. However, these concepts are beyond the focus of this study because the data are difficult to obtain from companies, and in the case of presenteeism, for example, difficult to measure objectively. It is also worth noting that the ways in which health care and social security are arranged affects the generalisability of these results. These system features should be taken carefully into account in economic evaluation, according to each case’s context.

4.2 Practical and scientific impact of the study

If the hypothesis that WDM processes can lead to financial effectiveness holds true, then in addition to the conditions used in this study, other conditions related to either implementation or mechanisms in the operating environments may not have emerged due to the limits of our data. These factors include trust [41] and commitment to the company; corporate social responsibility such as job retention; personal reasons such as other employment opportunities in the area; and community in common practices such as co-operation and interaction between employees and management; and employees’ mutual flexibility in working arrangements.

5 Conclusion

Evidence of the economic benefits of WDM is important for finding ways to conduct WDM in everyday operations. Companies cannot effectively control all labour costs. By managing work disability costs, i.e., costs arising from sickness absences, accident insurance costs, and other disability costs such as the experience rating in disability pensions used in Finland, a company can improve its productivity and potentially, its profitability. In a successful WDM, according to our findings, a change of -1.5% to -2.5% of the total payroll sum equals savings of several million euros per year.

Ethical approval

The study was approved by the Ethics Committee of the Finnish Institute of Occupational Health, no. ETR 6/2014.

Informed consent

Not applicable.

Conflicts of interest

None to report.

Footnotes

Acknowledgments

The authors would like to thank all companies involved in the study and Terveystalo for their co-operation.

Funding

This research received funding from the Finnish Work Environment Fund (grant no. 113307/17.12.2013).

Supplementary materials

Supplementary Table 1 is available from .

References

Braune

, Charosky

, Hikkerova

. Corporate social responsibility, financial performance and risk in times of economic instability. Journal of Management and Governance. 2019;23:1007–21. 10.1007/s10997-019-09476-y

Cancelliere

, Cassidy

, Ammendolia

, Cote

. Are workplace health promotion programs effective at improving presenteeism in workers: a systematic review and best evidence synthesis of literature. BMC Public Health. 2016;11:395. 10.1186/1471-2458-11-395

Chapman

. Meta-evaluation of worksite health promotion economic return studies: 2012 Update. American Journal of Health Promotion. 2012;26(4):1–12. 10.4278/ajh26.4.tahp

Cullen

, Irvin

, Collie

, Clay

, Gensby

, Jennings

, Hogg-Johnson

, Kristman

, Laberge

, McKenzie

, Newnam

, Palagyi

, Ruseckaite

, Sheppard

, Shourie

, Steenstra

, Van Eerd

, Amick

. Effectiveness of workplace interventions in return-to work of musculoskeletal, pain-related and mental health conditions: an update of the evidence and messages for practitioners. Journal of Occupational Rehabilitation. 2018;28(1):1–15. 10.1007/s10926-016-9690-x

Etuknwa

, Daniels

, Eib

. Sustainable return to work: a systematic review focusing on personal and social factors. Journal of Occupational Rehabilitation. 2019;29(4):679–700. 10.1007/s10926-019-09832-7

Gensby

, Labriola

, Irvin

, Amick III

, Lund

. Classification of components of workplace disability management programs: results from a systematic review. Journal of Occupational Rehabilitation. 2014;24:220–41. 10.1007/s10926-013-9437-x

Lefever

, Decuman

, Perl

, Braeckman

, Van de Velde

. The efficacy and efficiency of disability management in job-retention and job-reintegration. A systematic review. Work. 2018;59(4):501–34. 10.3233/WOR-182709

Mikkelsen Bjornskov

, Rosholm

. Systematic review and meta-analysis of interventions aimed at enhancing return to work for sick-listed workers with common mental disorders, stress-related disorders, somatoform disorders and personality disorders. Occupational and Environmental Medicine. 2018;75(9):675–86. 10.1136/oemed-2018-105073

Robson

, Clarke

, Cullen

, Bielecky

, Severin

, Bigelow

, Irvin

, Culyer

, Mahood

. The effectiveness of occupational health and safety management system interventions: A systematic review. Safety Science. 2007;45(3):329–53. 10.1016/j.ssci.2006.07.003

10.

Grimani

, Bergström

, Casallas

MIR

, Aboagye

, Jensen

, Lohela-Karlsson

. Economic evaluation of occupational safety and health interventions from the employer perspective. A systematic review. Journal of Occupational and Environmental Medicine. 2018;60(2):147–66. 10.1097/JOM.0000000000001224

11.

Baicker

, Cutler

, Song

. Workplace wellness programs can generate savings. Health Affairs. 2010;29(2):304–11. 10.1377/hlthaff.2009.0626

12.

Jacobs

, Yaquian

, Burke

, Rouse

, Zaric

. The economic impact of workplace wellness programmes in Canada. Occupational Medicine. 2017;67(6):429–34. 10.1093/occmed/kqx075

13.

Lerner

, Rodday

, Cohen

, Rogers

. A systematic review of the evidence concerning the economic impact of employee-focused health promotion and wellness programs. Journal of Occupational and Environmental Medicine. 2013;55(2):209–22. 10.1097/JOM.0b013e3182728d3c

14.

Reiman

, Ahonen

, Juvonen-Posti

, Heusala

, Takala

E-P

, Joensuu

. Economic impacts of workplace disability management in a public enterprise. International Journal of Public Sector Performance Management. 2017;3(3):297–310. 10.1504/IJPSPM.2017.085668

15.

Steel

, Godderis

, Luytten

. Productivity estimation in economic evaluations of occupational health and safety interventions: a systematic review. Scandinavian Journal of Work, Environment & Health. 2018;44(5):458–74. 10.5271/sjweh.3715

16.

Williams

, Westmorland

. Perspectives on workplace disability management: a review of the literature. Work. 2002;19(1):87–93.

17.

Shaw

, Walker

, Hogue

. The art and science of team: Enacting a transdisciplinary approach in work rehabilitation. Work. 2008;30(3):297–306.

18.

Williams-Whitt

, Bültmann

, Amick

, Munir

, Tveito

, Anema

. Workplace interventions to prevent disability from both the scientific and practice perspectives: A comparison of scientific literature, grey literature and stakeholder observations. Journal of Occupational Rehabilitation. 2016;26(4):417–33. 10.1007/s10926-016-9664-z

19.

MacEachen

. Part 3 Comprehensive Social Security Systems. In: MacEachen

. (Ed.). The Science and Politics of Work Disability Prevention. New York: Routledge. 2019;123–257.

20.

Creswell

, Plano Clark

. Designing and conducting mixed methods research. 3rd edition. Thousand Oaks: Sage. 2018.

21.

Johnson

, Onwuegbuzie

, Turner

. Toward a definition of mixed methods research. Journal of Mixed Methods Research. 2017;1(2):112–33. 10.1177/1558689806298224

22.

Teddlie

, Tashakkori

. Foundations of mixed methods research. integrating quantitative and qualitative approaches in the social and behavioral sciences. Thousand Oaks: Sage. 2009.

23.

Flyvberg

. Case Study. In: Denzin

, Lincoln

. (Eds.). The Sage Handbook of Qualitative research, 4th edition. Thousand Oaks: Sage. 2011;301–16.

24.

Kitchenham

. Mixed Methods in Case Study Research. In: Mills

, Durepos

, Wiebe

. (Eds). Encyclopedia of Case Study Research. Thousand Oaks: Sage Publications. 2010;562–64.

25.

Stake

. Case Studies. In: Denzin

, Lincoln

, editors. Strategies of qualitative inquiry. Thousand Oaks: Sage; 1995;86–109.

26.

Yin

. Validity and generalizations in future case study evaluations. Evaluation. 2013;19(3):321–332. 10.1177/1356389013497081

27.

Yin

. Case study research and applications. Design and methods. 6^th edition. Thousand Oaks: Sage. 2018.

28.

Åkerblad

, Seppänen-Järvelä

, Haapakoski

. Integrative strategies in mixed methods research. Journal of Mixed Methods Research. 2020;15(2):152–70. 10.1177/1558689820957125

29.

Johnson

, Turner

. Data collection strategies in mixed methods research. In: Tashakkori

, Teddlie

. (Eds.) Handbook of mixed methods in social and behavioral research. Thousand Oaks: Sage. 2003;297–319.

30.

Byrne

. Evaluating complex social interventions in a complex world. Evaluation. 2013;19(3):217–228. 10.1177/1356389013495617

31.

Ragin

. The Comparative Method – Moving beyond Qualitative and Quantitative Strategies. Berkeley: University of California Press. 1987.

32.

Terveystalo. Dialogi terveydestä. Tekemättömän työn kustannukset 2015. Cited 2022 Dec 19. Available from: https://www.terveystalo.com/Global/tth/Tekemättömän%20työn%20kustannukset%15%20White%20Paper.pdf 2015.

33.

Bloor

, Frankland

, Thomas

, Robson

. Focus Groups in Social Research. London: Sage. 2001.

34.

Boje

. Theme analysis. In: Boje,

. Narrative Methods for Organizational & Communication Research. New Mexico State University: Sage. 2001;122–37.

35.

Ayres

, Kavanaugh

, Knafi

. Within-case and across-case approaches to qualitative data analysis. Qualitative Health Research. 2003;13(6):871–883. 10.1177/1049732303013006008

36.

Schneider

, Wagemann

. Set-Theoretic Methods for the Social Sciences – A Guide to Qualitative Comparative Analysis. Cambridge: Cambridge University Press. 2012.

37.

Fetters

, Curry

, Creswell

. Achieving integration in mixed method design – principles and practices. Health Services Research. 2013;48(6 Pt 2):2134–56. 10.1111/1475-6773.12117

38.

Pelletier

. A review and analysis of the clinical and cost-effectiveness studies of comprehensive health promotion and disease management programs at the worksite: Update VIII 2008 to 2010. Journal of Occupational and Environmental Medicine. 2011;53(11):1310–31. 10.1097/JOM.0b013e3182337748

39.

Hoefsmith

, Houkes

, Nijhuis

FJN

. Intervention characteristics that facilitate return to work after sickness absence: A systematic literature review. Journal of Occupational Rehabilitation. 2012;22(4):462–77. 10.1007/s10926-012-9359-z

40.

Brenfort

, Persson

, Linderoth

, Ekberg

. Supervisor rating of productivity loss associated with presenteeism and sick leave due to musculoskeletal disorders and common mental disorders in Sweden. Work. 2021;68(4):1091–1100. 10.3233/WOR-213439

41.

Ståhl

, Svensson

, Petersson

, Ekberg

. A matter of trust? a study of coordination of swedish stakeholders in return-to-work. Journal of Occupational Rehabilitation. 2010;20(3):299–310. 10.1007/s10926-009-9205-0

	Cases
Conditions and outcome	F, G, I	A	K	J	E	H	N	L	M	B	C, D
C1: Were the obstacles to co-operation in WDM due to the nature of the work or the organisational model dismantled?	1	1	1	1	1	1	0	0	0	0	0
C2: Were the strategic goals of WDM reflected in the WDM practice?	1	1	1	1	1	1	1	1	1	0	0
C3: Did the WDP measures comprehensively target the company’s key disability risks?	1	1	1	1	1	1	0	0	0	1	0
C4: Were the supervisors actively involved in the implementation of WDM and OHS activities?	1	1	1	0	0	0	1	0	0	0	0
C5: Did the co-ordination of WDM and the flow of information support multi-stakeholder co-operation?	1	1	1	1	1	1	1	0	0	0	0
C6: Did the employees actively participate in the implementation of WDM and OHS activities?	1	0	0	1	0	0	0	1	1	0	0
C7: Were the indicators systematically used to monitor and develop the effectiveness of WDM and OHS activities?	1	1	1	0	1	0	0	1	0	0	0
Outcome: Did the average work disability costs decrease from 2010–2011 to 2012–2013?	1	1	0	1	1	1	1	0	1	0	0