DISABUJTY MANAGEMENT STRATEGIES: A PRELIMINARY INVESTIGATION OF PERCEPTIONS, POLICIES AND RETURNTO-WORK OUTCOMES, by Gabrielle McHugh B.A., University College Dublin, 1988 THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in DISABUJTY MANAGEMENT THE UNIVERSITY OF NORTHERN BRITISH COLUMBIA April, 2005 © Gabrielle McHugh, 2005 1^1 Library and Archives Canada Bibliothèque et Archives Canada Published Heritage Branch Direction du Patrimoine de l'édition 395 W ellington Street Ottawa ON K 1A 0N 4 Canada 395, rue W ellington Ottawa ON K 1A 0N 4 Canada Your file Votre référence ISBN: 0-494-04642-2 Our file Notre référence ISBN: 0-494-04642-2 NOTICE: The author has granted a non­ exclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or non­ commercial purposes, in microform, paper, electronic and/or any other formats. AVIS: L'auteur a accordé une licence non exclusive permettant à la Bibliothèque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par télécommunication ou par l'Internet, prêter, distribuer et vendre des thèses partout dans le monde, à des fins commerciales ou autres, sur support microforme, papier, électronique et/ou autres formats. The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. L'auteur conserve la propriété du droit d'auteur et des droits moraux qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation. In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis. Conformément à la loi canadienne sur la protection de la vie privée, quelques formulaires secondaires ont été enlevés de cette thèse. While these forms may be included in the document page count, their removal does not represent any loss of content from the thesis. Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant. Canada Abstract The objectives of this retrospective study were to (a) examine the predictive value of DM perceptions and DM policies in retum-to-work outcomes for the injured worker; (b) examine factors that influence DM perceptions, DM policies, and Retum-to-work outcomes; and (c) examine the relationship between demographic factors of the individual and return to work outcomes. Employers from a Northern BC community were randomly selected to participate. Analysis of the data supported the hypothesis that company perception towards DM is a significant contributing factor to the presence of DM policies in the workplace. The study draws particular attention to the need for collaboration with WCB to supplement claims data reported by employers. Interestingly, the instrament constmcted for the study shows potential. With further development the tool may be valuable in identifying incongraent attitudes towards DM between different levels of an organization. As a future direction this would merit additional investigation. 11 TABLE OF CONTENTS Abstract ii Table of Contents iii List of Tables iv List of Figures V List of Appendices vi Acknowledgement vii Introduction 1 The principles of Disability Management 2 The context of Disability Management 3 Factors influencing Disability Management 7 Organizational Attitudes 9 Organization Policy and Procedures 14 Worker Demographics 20 Conclusion 23 Method 30 Results 37 Discussion 52 Conclusion 64 References 68 Appendices 79 111 List of Tables Table 1 Descriptive Statistics for DM policy and DM perception Measures. 38 Table 2 Rotated Factor Matrix DM Policies. 40 Table 3 Rotated Factor Matrix DM Perceptions. 41 Table 4 Summary of Regression Analyses. 45 Table 5 Summary of Multivariate Analysis of Variance for Objective (b) 48 Table 6 Summary of Univariate Analyses for Objective (b). 49 Table 7 Summary of Analyses for Factors Contributing to Retum-to-work outcomes. 51 IV List of Figures Figure 1 Diagram representing the three steps to achieve Objective (a). Figure 2 Scatterplot depicting the relationship between DM policy and DM perceptions. 29 43 List of Appendices Appendix A Letter of Introduction 80 Appendix B Sample Questions from Questionnaire 82 Appendix C DM Policy Histogram 84 Appendix D DM Perception Histogram 86 Appendix E Descriptive Data 88 VI Acknowledgement My appreciation to the members of my committee; Dr. Henry Harder, supervisor, and Drs. S. Wagner and D. Voaklander. I have thoroughly enjoyed and thrived on the experience. Leaving the best until last, I thank my husband for relieving me of all household responsibilities to concentrate on attaining this degree. He is, as my father fondly describes, the last gentleman on earth. Mo gArd, mo chrof mo MufocAos. Vll Disability Management Strategies 1 Disability Management Strategies: A Preliminary Investigation of Perception, Policy and Retum-to-work Outcomes. Many employers have implemented Disability Management (DM) strategies in the workplace as a means of containing their disability costs. Measuring the success of DM in terms of cost containment is well documented throughout the literature. While this economic benefit is a valuable incentive for employers to initiate DM, the primary focus of DM is the individual worker and his employability when disability has occurred. Changing the outcome criteria from cost containment to employability poses a greater challenge to measuring the success of DM (Butler, Baldwin & Johnson, 1995). In looking at retum-towork rates, these authors found that injured workers did not sustain their employment over time. From the initial retum-to-work rate of 85% for injured Canadian workers, the rate dropped to 35% when maintenance of employment was monitored. These findings are based on the retum-to-work rates for injured workers during 1989 to 1990. At that time Canadian involvement in DM was in its infancy. It is only within the past decade that an increasing amount of Canadian employers are utilizing DM strategies to retum injured workers to the workforce. The benefit of DM is based on the assumption that workplace accommodations mitigate disability associated with impairments facilitating continued employment for that worker. To test this assumption this study examines disability management policies, perceptions and retum-to-work outcomes. Employers in a Northern B.C. community were surveyed and organizational policies and perceptions towards DM were examined. The aim was to measure the ability of policies and perceptions to predict retum-to-work outcomes for employees who sustained a compensable injury covered by Workers' Compensation Board of British Columbia (WCB). The literature review provides an account of the context that Disability Management Strategies 2 led to the emergence of disability management and the organizational and individual factors that influence retum to work outcomes. Following the presentation of the methodological framework for this study, data analyses are presented in the Results section. The final section presents the discussion concerning the implications of the Gndings and future direction for research and disability management. 77i6 /nanaggTMgnt DM is described by Shrey (1996) as a paradigm shift in the world of rehabilitation. This shift is primarily due to its workplace focus. Generally, DM is defined as a proactive, employer-based prevention and remediation strategy developed to (a) prevent the occurrence of accidents and disability, (b) to mitigate the effects through early intervention and retumto-work options when disability does occur, and (c) decrease lost time by maintaining the occupational bond and availing of optimum timing for retum to work (Habeck, Leahy, Hunt, Chan, & Welch, 1991; Shrey, 1998; Westmorland & Buys, 2002). DM is based on the empirically supported premise that an early retum-to-work facilitates the continued relationship between worker and employer (Habeck, 1999; Habeck, Hunt, & VanTol, 1998; Lacerte & Wright, 1992). DM has a broader perspective of the rehabilitation process that promotes disability prevention strategies, rehabilitation treatment concepts, and retum-towork programmes aimed at preserving the working relationship. The effectiveness of any programme for managing disability rehes on a combination of philosophical and operational components that involves the interplay of employers' goals and resources with employees' needs and expectations. The role of the workplace and the interaction between the employee and the work environment, therefore, represent a very significant part in the outcome of the work disability process (Habeck, Hunt, & VanTol, 1998). Disability Management Strategies 3 The escalating economic cost associated with workplace disability was the impetus for research into the advantages of disability management. Concurrent with rising costs, pohcy initiatives such as the Employment Equity Act (1984), the Duty to Accommodate enshrined in the Canadian Human Rights Code, and the "reasonable accommodations" provision of the Americans with Disabilities Act (ADA, 1990), were making it expedient for employers to explore new solutions to mitigate the consequences of disability (Howe & Johnson, 2000; Traver, 1990). A companion factor for stimulating research in DM was the changing nature and direction of service provision within the field of rehabilitation. In the United States, for example, the 1979 amendments to the Vocational Rehabilitation Act (1973) resulted in a demarcation of service provision. Individuals with more severe and/or congenital disabilities became the service responsibility of government funded agencies whereas acquired and/or less severe disabilities were serviced by private rehabilitation (McMahon, 1983). Another influence to permeate the development of disability management was legislative change that placed responsibility for injured workers with the employer. The workers' compensation model utilized in Australia, for example, obliges employers to provide retum-to-work assistance for injured workers (Kenny, 1995). Regulations such as the Occupational Health and Safety Act (1978) in Canada entrusted responsibility to employers and employees for the creation of joint committees to control and improve workplace hazards (Canadian Centre for Occupational Health and Safety, 1999). This inducement for co-operation laid the necessary foundation of participatory, working relationships between employers and employees to aid the development of disability management. Disability Management Strategies 4 Further facilitating the development of DM, in addition to legislative changes, was the review o f the medical model and its role in disability. The medical model is based on the management of disease whereas DM is based on the management of disability. The return to work process of DM aims at taking advantage of residual functional capabilities whereas the medical model aims at maximum medical recovery. To Lacerte & Wright (1992) the traditional medical and pathological orientation to determine work disability hinders the retum to work process. Unless the impairment is assessed in relation to the actual functional capabilities of the individual the disparity between the medical model and the orientation of DM model may delay or jeopardize a retum to work (Greenwood, 1984). The limitations of the medical model are also reviewed by Loisel et al. in their research on back pain ( Losiel, Dumad, Bertelette, Vézina, Baril, Gagnon, et al., 2001). These authors posit that disability is a multi-faceted problem requiring a broadening of the medical model. Proposing that the determinants of work disability incorporate medical, personal, and environmental aspects successful intervention, therefore, implies the adoption of a rehabilitation approach that addresses the various facets of disability and links interventions to the workplace. The concept of disability being a multi-faceted problem was initially introduced by the World Health Organization (WHO) Expert Committee in their 1976 and 1981 Reports. The WHO promoted the concept of disability as dimensional and outlined the following operational definitions and processes of disability: 1. Impairment - any loss or abnormality of psychological, physiological or anatomical structure or function. 2. Disability - any restriction or lack of ability to perform an activity in the manner or within the range considered normal. 3. Handicap - a disadvantage for a given individual, resulting from an impairment or disability, that limits or prevents the fulfillment of a role that is normal for that individual (p.8). Disability Management Strategies 5 WHO also clarified three levels of prevention aimed at reducing the impact of disability: (1) reducing the occurrence of impairments; (2) limiting or reversing disability caused by impairment; and (3) preventing the transition of disability into handicap (p.9). Influenced by this theoretical approach, the early concept of 'preventative rehabilitation' posited by Jarvikoski & Tuunainen (1978) is cited throughout the literature as the origin of disability management. The authors introduced work-based interventions aimed at containing withdrawal from the workplace if functional abilities became impaired (Jarvikoski & Lahelma, 1980, 1981; Jarvikoski & Tuunainen, 1978). Workplace based interventions were seen as fundamental in circumventing permanent exclusion from the workforce. Thus, the burgeoning of the private rehabilitation sector, legislative initiatives and the growing supposition that rehabilitation was not merely restorative in nature were important antecedents to employers becoming active agents in the whole rehabilitation process. The context of rehabilitation also changed from the clinic to the workplace. Shrey (1998) describes the basic principles of disability management that emphasize the direct involvement of the employer to secure job retention and accommodations for the injured worker. Shrey (1998) argues that "worksite-based disability management interventions capitalize on the therapeutic value of working while the worker is recovering from an injury or illness" (p.390). Recovery from injury or illness should not be an isolated event - "the focus is on enabling the worker to get well while working rather than waiting for the worker to get well in order to retum to work" (Shrey, 1998, p.391). As research demonstrates, disability is a complex issue that extends beyond the medical condition and pathological findings. Yelin, Henke, & Epstein (1986) examined the differences that distinguished workers and non workers with musculoskeletal disease and reported that symptoms of Disability Management Strategies 6 disease were poor predictors of whether work disability would occur. Early intervention and strength of occupational bond emerge as important factors in the process of managing disability (Bruyere & Shrey, 1991; Lacerte & Wiight, 1992); particularly as it is now recognized that it is more costly and difRcult to place workers after disability into new employment than it is to retum workers to the pre-injury employer (Habeck et al., 1991; Schwartz, Watson, Galvin & Lipoff, 1989). The interdependence of employer and employee interactions and organizational structures for successful management of disability is also well documented throughout the literature (Friesen, Yassi, & Cooper, 2001; Habeck & Shrey, 1991). Also, facilitating successful disability management is the recognition that worker and workplace co-exist in a reciprocal relationship that is influenced by both the characteristics of the worker and the workplace (Kenny, 1995). Moreover, the workplace is the context of service delivery for disability management (Shrey & Olshesky, 1992). The workplace supports DM activity as a coordinated process characterized by individualizing solutions to each worker and working situation, regardless of the cause of injury or illness. DM brought to the fore the importance of the employer’s role in early intervention, in preventing work loss and in maintaining the worker’s connectedness to the workplace. Thus, disability management strategies alleviate the human cost of disability. The extent to which DM has permeated the workplace, however, would not be realized if DM did not address a business need of the employer; namely containment of disability related costs. Salkever, Shinogle, & Purushotamam (2000) demonstrate that the calculation of expected financial costs and benefits plays an important role in employers’ decisions to implement DM programmes (see also Habeck et al., 1991). DM enables the employer to exert control over the cost of individual disability outcomes and also, to assume Disability Management Strategies 7 responsibility for the retum-to-work planning and co-ordination of its injured workers. The significance and importance of disability management in addressing the mutual needs of employers and employees can only be understood when compared to the size of the problem. Canadian employers paid an estimated annual $11.3 billion in absence-related costs, annual health benefit payments of $12.5 billion and total lost time WCB costs of $5.7 biUion for 2001 (Scott & Curtis, 2003). In British Columbian figures this represents workers' compensation costs of $930,289,332 or lost days from work in the amount of 3,004,318 during 2002 due to occupational disease or injury (WCB of B.C., 2004). Absence costs alone are estimated at over 4.2% of Canadian payroll (Thompson, 2001) which may be an underestimation when compared to the figures from the United States which suggest costs run in the range of 8 - 15% of payroll (Salkever et al., 2000). The most forceful statistic, however, is that reported by Mobley, Linz, Shukla, Breslin & Deng (2000) whereby 70% of employers surveyed in the United States identified workers’ compensation costs as a threat to the company’s profitability. Factors /n/Zwcacmg Düa^iZity Ma/iagcmcMt Empirical research supports that occupational disability and retum-to-work potential following an injury or illness is a process influenced by a variety of organizational and individual factors (Amick, Habeck, Hunt, Fossel, Chapin, Keller et al., 2000; Freisen et al., 2001; Habeck et al., 1991; Habeck, ScuUy, VanTol & Hunt, 1998; Westmorland & Williams, 2002). Regardless of the worker’s physical recovery process, an effective return to work is jeopardized without a facilitative and receptive work environment (Habeck, Hunt, et al., 1998). Galvin and Schwartz (1986) similarly cite the literature on the advantage of employer based disability management for early identiGcation and timeliness of rehabilitation Disability Management Strategies 8 intervention. The consensus of opinion is that neglected disability is compounded when delays in rehabilitation occur. Davis (1983) for example, indicates that the biggest single factor affecting the outcome of any case is the timeliness of rehabilitation intervention. This is particularly important when it has been found that the longer individuals are absent from the workplace, the less likely they are to retum (Aronoff & Feldman, 2000; Cunningham & James, 2000). Akabas, Fine, & Yasser (1982) report that one-third of workers disabled each year do not retum to work. Moreover, they sever their connections to the workplace and are accordingly, disadvantaged if they attempt to re-enter the labour market. Burkhauser, Butler, & Kim (1995) reported that the risk of exiting the workforce due to a work limiting health condition is highest during the first two years following onset. After two years less than 50% of employees with work impairments continue in the job; after six years only 25 % remain with the firm. The literature, however, does not infer that the implementation of disability management programmes in and of itself is the recipe for success. As Burkhauser et al. (1995) illustrate job accommodation is a critical variable for increasing the retention rate for those with disabilities. Despite the increase, the overall expected job duration of workers with health impairments remains less than that of the average worker. What is also important to improving retention rates is for the employer to develop an infrastructure that creates opportunities for temporary accommodations and alternative, productive work options while the injured worker is completing the physical recovery process (Freisen et al., 2001; Mobley et al., 2000). Co-ordinated policies and programmes are necessary but not sufficient to ensure successful outcomes for disabled workers. The intent of this study was to demonstrate that disability management policy is the mediating factor between positive employer attitudes towards disability management and favourable Disability Management Strategies 9 outcomes for the worker with disability. The current research will focus on internal company factors cited in the literature as under organizational control and pertinent to the successful implementation of disability management programmes. Employer attitudes towards disability will be examined first. A review of the accepted policy and procedural norms for implementing disability management programmes will foUow and finally, individual demographic factors relevant to the survey instrument will be reviewed. OrgowzatmnaZ aftifudes. The literature cites the difficulty and barriers experienced by workers limited by disability entering or returning to the labour market. The consequence of disability is continued low employment participation rates of this population when compared to working age adults without disabilities (Akabas, 1994; Bruyere, Erickson, & VanLooy, 2004; Gilbride, 2000; Hernandez, 2000; Petty & Fussell, 1997; Shrey & Bangs, 1991). As a result of legislative initiatives such as the ADA (1990) in the United States research finds that employers are expressing more positive global attitudes towards workers with disabilities, thereby, affording job opportunity to people with disabilities (Unger, 1999). Hernandez (2000) argues that this may be a veneer of political correctness as the improved attitudes towards disability issues do not match the real hiring practices of individuals with disabilities. The benefit of disability management however, is the goal of job retention for the disabled worker. The success of disability management would suggest that employers are more predisposed to continue the employment of their own workers with whom they have established a work history. The barrier to continued employment may be somewhat diminished for these workers as companies recognize the need, not only to comply with regulations but also to maintain the employability of a workforce that demographically, is both aging and dwindling (Akabas & Gates, 1990; Bruyere & Shrey, 1991; Tate, Habeck & Disability Management Strategies 10 Galvin, 1986). The extent of the challenges posed by the demographic changes in the population is discussed by McDonald and Harder (2003). The significance of the discussion is its emphasis on the demographic changes projected for the Canadian population and the subsequent impact on the workforce. The importance of the DM perspective in preventing disability and mitigating its effects when it does occur is particularly relevant when viewed in the context of population changes. Projections indicate that the proportion of workers aged 25-34 wiU steadily decrease as birthrate decreases while the proportion of workers in the 4564 age groups will rapidly increase. The projections estimate that workers aged 45-64 will account for approximately 70% of the net increase in the working population by the year 2010. The future direction of these projections suggests that it is economically more viable for employers to mitigate the effects of disability and maintain the employability of its workforce if they are to remain competitive. Successful disability management in the workplace also has a cultural context (Amick et al., 2000). The organization’s role is to provide a positive corporate culture of acceptance for the re-integration of the worker with disabilities into a meaningful occupation. Corporate culture embodies a set of beliefs and values collectively held and socialized by the organization. A definitive explanation of the complexities that delineate organizational climate or corporate culture, however, is rather elusive. For the purpose of this research Schein's (1992) definition will be followed which suggests that corporate culture is the pattern of basic assumptions in the organization that develop from the process of resolving internal and external problems and that these assumptions have worked well enough to be considered vahd. Schein suggests that culture provides structural stability to the group on the basis of accumulated shared learning covering behavioural, emotional and cognitive elements Disability Management Strategies 11 that form into the pattern of norms, values and behaviours that is called culture. This definition suggests that for consistency, these patterns of assumptions are shared across all units of the organization. Furthermore, corporate culture may act as a driving force in shaping acceptance of the disability management concept or inadvertently act as a restraining force rejecting the concept. Shoemaker, Robin & Robin (1992) found a significant association between corporate cultural factors such as employee value, corporate flexibility, corporate endorsement of the principle of retaining and retraining personnel, corporate social responsibility and the degree of acceptance of early retum to work programmes. Visible corporate commitment from top management is cited as a primary internal factor contributing to the successful development of disability management programmes (Pati, 1985; Shrey & Olsheski, 1992; Tate, Habeck & Galvin, 1986). Accordingly, in this research, each company is asked to indicate the degree to which they believe top management, line supervisors and union support the concept of disability management (questions 11, 12, and 13) and the extent to which they commit resources (questions 4 and 6). The economics of disability is cited as the predominant motivator for the emergence of disability management programmes. Success is frequently measured in economic terms of cost containment and benefit-cost ratios (Pati, 1985). While this is an important employer incentive, the dividends of disability management extend beyond cost ratios. Schwartz et al. (1989) in reporting on company rationale for instituting disability management noted that 77% of employers surveyed by the Human Resources Center cited concern over employee welfare; 54% cited increased use of benehts; 52% cited insurance premium increases and 39% were concerned over the decrease in worker productivity. The hospitality industry, for Disability Management Strategies 12 example, became committed to disability management as it resolved their very high tum-over rates. Turn over rates, reportedly decreased by 30 to 50 percent following the implementation of DM strategies (Pati, 1985). As a consequence of implementing DM strategies, the Polaroid Corporation reported an increase in productivity from workers with job performance issues when referred to counseling (Galvin, 1986). Prestigious corporations such as Xerox, Kodak, MacMillan Bloedel, Chrysler Canada, British Columbia Hydro to name but a few, reported decreased rates of absenteeism and improved W orkers’ Compensation experience ratings (Dyck, 2003; Shrey, 1998; Silversides, 1998) as a dividend to implementing disability management strategies. Dividends are also reported in terms of positive morale and commitment among employees who succeed in maintaining their employment (Galvin, 1986). This, however, remains a contentious issue and the findings are too inconsistent to support a positive relationship between employee commitment to the employer and variables such as absenteeism, turnover and job performance (Wright & Bonett, 2002). For individuals with disability their relationship to the employer may be more complex and commitment may be moderated by the fact that they have received accommodations and therefore view their employability to other organizations as compromised. Foremost, disability management programmes are seen as tangible evidence of the company’s regard for its employees and compliments a people-oriented culture that values the contribution of the employee (Beaudway, 1986). Although positive testimonials of cost containment and mutual employer and employee benefits from disability management policies abound, not all employers are aware of the real cost of absenteeism (Mitchell, 2002). These include direct cost of premiums, beneAt payments as well as the indirect cost of lost productivity and worker retraining Disability Management Strategies 13 associated with disability. Employers who overlook the cost of absenteeism have less incentive to adopt the concept of disability management within their workplace (National Institute of Disability Management and Research, 2003). Moreover, not all corporate cultures are receptive to disability management initiatives. Shoemaker (1989) investigated factors of resistance to early retum-to-work polices within corporations. Corporations that held beliefs commonly referred to as the insurance myth, the dependability myth, the productivity myth and the accommodation myth were resistant to early retum-to-work initiatives (Shoemaker, 1989). Nothwithstanding the empirical evidence to refute such factors, the insurance myth refers to a belief that retaining or hiring workers with disabilities will result in increased insurance and workers’ compensation costs and accident liabilities (see Krause, Dasinger, & Neuhauser, 1998). Contrary to the myth, Mobley et al. (2000) showed evidence of a 75% decrease in workers’ compensation leave rate when a disability management programme was instituted at an automotive manufacturing plant (also see Schwartz, G., 1984). The dependability myth intimates that workers with disabilities are not dependable despite research to the contrary (see Mobley et al., 2000; Pati, 1985). The productivity myth alludes to quality and quantity reductions contrary to findings. Tate et al. (1986), for example, report increased productivity and improved absenteeism rates from General Motors rehabilitation efforts (see also Cohen, Parrinello, & Kelliher, 1990; O’Brien, 2003). The accommodation myth envisions huge expense for workplace modifications which is the converse to employer reports (see Guinter, Eagels, Harringer & Trusewych, 1995; Nickel & Yangouyian, 1996; Votel, 1993). Unger (1999) discusses formal and informal accommodations the m ^oiity of which are tmobtrusive, inexpensive, or draw on existing employer resources. Additionally, there are employer reports that some Disability Management Strategies 14 accommodations were useful and effective in addressing employment-related issues among individuals with no known disability but made general good sense for all employees. The survey instrument used in this research will explore the degree to which employers continue to subscribe to these various attitudes to measure companies overall perceptions of DM. Statements are presented to measure the degree to which each employer agrees or disagrees with the insurance myth (question 8); the dependability myth and turnover (question 10 and 14); the productivity myth (question 7); the accommodation myth (questions 3 and 9); the therapeutic value of work (question 2); and re-integrating employees with restrictions (question 5). Organization policies and procedures. As previously mentioned, positive cultural context is not sufficient in itself for the successful implementation of disability management programmes. This attitude must translate into cohesive policies and well managed services. Guidelines presented by several authors (e.g. Akabas & Gates, 1990; Schwartz et al., 1989; Shrey, 1996; Tate, Habeck & Schwartz, 1986) exemplify a multidisciplinary team approach for the implementation of disability management. The following is a summary of the primary strategies identified that facilitate employee placement in jobs and improve labour support for DM practices: 1. More exphcit language included in union contracts that describes the various steps for implementing work retum transitions for disabled workers. 2. Educational efforts implemented to promote awareness among coworkers, with respect to the nature and needs of disabled or injured workers. 3. Establishing a committee of first-line supervisors, union representatives and placement coordinators to generate creative strategies designed to place restricted workers. 4. Development of a job analysis bank so that workers could be selectively placed in jobs in which the demands were congruent with their physical capacities and work restrictions. Disability Management Strategies 15 5. Labonr-management supported work site accommodations so that on-site disability management programmes could be effective in promoting work retum and worker retention. Research underscores the importance of coordinated policies that affirm and support retum-to-work commitment (Westmorland & Williams, 2002). Bmyere & Shrey (1991) state that policies and procedures jointly promulgated by labour and management are critical to reducing adverse reaction to the implementation and success of disability management programmes. Jarvikoski & Lahelma (1980, 1981) concluded that success of employer based disability management programmes ultimately depend on the joint commitment, support and active participation of labour and management. Approval of the general principles of early intervention by the employer, the different levels of the line organization as well as the employees and their labour unions is a prerequisite condition (see also Pransky, Shaw, & McLellan, 2001; Shaw, Robertson, Pransky, & McLellan, 2003). Mitchell (2002) further supports the necessity for a clear statement of expected cooperation with the retum-to-work process from all employees. Additionally, retum-towork language should be an integral part of the blueprint for an effective programme whether or not the establishment is unionized. For unions, however, the acceptance of and co­ operation with disability management programmes may be considered a shift in their traditional direction. Historically, unions have a collective focus and advance the protection of the collective group. Restrictive language clauses included in collective bargaining agreements such as job classifications, seniority rights, and transfer rights present as obstacles to the retum-to-work process (Baiil & Berthelette, 2000). Stock, Deguire, Baiil, & Durand (1999) cite seniority as one of the most difficult problems to resolve when facilitating modihed work for injured workers, particularly when Disability Management Strategies 16 higher seniority gives priority access to less physically demanding jobs. Bruyere & Shrey (1991) cite that poor labour relations translate into lengthy disability claim duration. Cohen, Parrinello & Kelliher (1990) report initial union resistance to the introduction of disability management strategies at Long Island Railroad. Described as near sabotage, resistance continued until collective bargaining agreements were renegotiated and the disability management programme was revamped through collaborative agreement. Consequently in view of past findings, questions 2 and 3 of the survey instrument for this research will measure how well current companies adhere to the ‘best practice’ guidelines and include language within the employment contract or bargaining agreement delineating disability management and early retum to work policies. Clear policy, explicit language, flexibility and open communication between the interested parties are cited as the basic elements of successful practice (Westmorland & Williams, 2002). The effectiveness of these basic elements is realized through the Akabas & Gates (1990) findings. These authors found that most disabled workers wanted to retum to work before they were able to perform full duties but up to 80% reported they perceived their employers were inflexible regarding accommodations despite case managers being available to assist with early retum-to-work planning. In the instances studied, policies were not broadly communicated to employees and procedures were not clearly identified making for lost opportunity for both disabled workers and employers. Similarly, Krause et al. (1998) report that of over 4,CXX) disability insurance beneficiaries only 20% had any knowledge that they were eligible for a work trial programme, and that those who were aware of the programme were twice as likely to retum to work. Friesen et al. (2001) further corroborated the importance of effective communication in their study into stakeholder perspectives on Disability Management Strategies 17 barriers and facilitators for retum-to-work. Investigating the interplay between the worker and the workplace structure within the regulatory insurance context, delays in the process or delivery of information or treatment and ineffective communication were found to be detrimental to the retum-to-work process (see also Mobley et al., 2000). Primary among their findings was the need for rapid and effective communication among all stakeholders which includes compensation provider and physician, not only employer and employee. The summation of their findings, similar to the contention of this thesis, is that successful retumto-work is not the result of positive relationships between parties nor formal policies alone but an interplay of organizational structures and human interactions. Several aspects of communication are measured in this research; Question 1 tests the importance of educating and informing the employees about retum-to-work options; Question 5 explores the level of cooperation between the various parties to review retum-to-work polices in joint committee; while Question 8 seeks to capture the level of co-operation within the workplace in reassigning workers with disabilities; and Question 14 gamers information regarding regular contact with injured workers to facilitate early re-entry to the workplace. Many worksites continue to prohibit the injured worker's early retum to work through restrictive policies of not returning to work until they are 1(X)% fit for duty (Aronoff & Feldman, 2000; Shrey & Olsheski, 1992). This presents as an impractical barrier for individuals who would benefit from transitional work options to condition themselves back into their regular job. It may also be exclusionary for individuals with residual limitations following maximum medical recovery that need accommodation to enable them to perform their duties. It is also inequitable to insist upon 1(X)% productive capacity for recovering workers while many of their co-workers without disabilities continue working at less then Disability Management Strategies 18 100% efGciency. This negates the therapeutic value of the workplace (Durand & Loisel, 2001; Iserhagen, 2000; Aronoff & Feldman, 2000). This research examines the degree to which employers require full recovery before permitting a return to work (question 6). The empirical research advocates a systematic approach to worksite accommodations. The findings demonstrate that the more effective strategies have moved beyond the designation of light-duty jobs and developed more flexible, individualized responses to retum-to-work accommodation needs that are transitional in nature (Habeck et al., 1998). Further supporting these findings, Iserhagen (2000) proposed that the focus of a successful integrated disability management system should be the ability to match the worker and the work. Problems typically encountered with traditional light duties pertain to the compatibility of generic tasks to the worker’s specific functional capacity; the unlimited timeframes placed on these assignments, sometimes resulting in permanent placement; and labour relation conflict particularly in unionized environments where light-duty positions are considered an earned reward based on seniority (Shrey, 1998). Transitional work is focused on meaningful work activities allowing for physical reconditioning, re-education on safe work practices, and progressive upgrading of work activities to the acquisition of regular duty in the pre-injury job. Transitional work programmes minimize the loss of job fitness and physical conditioning effects which can be detrimental to the recuperation process. Work return transition options combine the beneût of continuity of work routine and work activity by matching the functional abilities and endurance level of the worker to assigned duties for a temporary, short-term period as the worker progressively transitions back to the original full duty, full shift job (Btuyere & Shrey, 1991). Durand and Loisel (2001) showed the success of linking a work rehabilitation programme to the workplace for individuals with Disability Management Strategies 19 chronic back pain. Krause et al. (1998) reviewed the literature relevant to worksite based modiAed work progammes and found the evidence to strongly support the principle of gradual return to work as a means of successful rehabilitation for temporarily and permanently disabled workers. Primarily, the research demonstrates that employees with access to modiAed duty return to work after a disabling injury about twice as often as employees without access to any form of modified duty. Modified work programmes reduced the number of lost work days per disabling injury by 50%, supporting that such programmes are effective. There is encouraging evidence to suggest that time lost in back pain claims can be reduced by appropriately modifying duties. Such measures have reportedly realized up to a 51% reduction in claims costs and address one of the main causes of workers’ compensation claims and high costs (Frank, Sinclair, Hogg-Johnson, Shannon, Bombardier, Beaton & Cole, 1998). Similarly, Infante-Rivard & Lortie (1996) concluded from their investigation of first compensated episodes of back pain that duration of absence is not limited to the disease and pathological recovery process but is also influenced by social and work factors; namely work history and flexibility of accommodations provided. Anema, Cuelenaere, van der Beek, Knol, de Vet, & van Mechelen (2004) invesAgated a mulAnational cohort of 1,631 workers sick-listed for 3-4 months due to low-back pain and concluded that adaptarion of workplace, job tasks and working hours was effecAve in returning long-term sick-hsted to work. Failure to implement ergonomically focused workplace accommodaAons may be a contributing factor to delayed funcAonal recovery and relapse (Loisel, Gossehn, Durand, Lemaire, Poitras, & Abenhaim, 2001; Shaw, Feuerstein, Miller, Lincoln, Berger, & Wood, 2000). Salkever, et al. (2000) report the most common type of accommodaAons provided by employers is Disability Management Strategies 20 modified or partial job duties and reduced or flexible work schedules (76.93% and 73.99% respectively). Accordingly, the current research will investigate the type of accommodations most commonly offered to employees. Using a five-point Likert scale employers are asked whether they can match the abilities of the worker to the demands of the job (Question 7); accommodate modified duties (Question 9), modified schedules (Question 10), alternative duties (Question 11), modified work environments (Question 13) and/or ergonomic devices (Question 13) Worker demographics. Recent research is applying a bio-psycho-social approach to predict individual return to work outcomes as a conceptual model that integrates physical, psychological, and social components of the disability process (Aronoff & Feldman, 2000; Karjalainen, Malmivaara, van Tulder, Roine, Jauhiainen, Hurri & Koes, 2001; Schultz, Crook, Berkowitz, Meloche, Milner, Zuberbier, & Meloche, 2002; Shaw & Polatajko, 2002). Such a model is practical for multi-method research that elicits the interrelationship between the person, the environment and the occupation on the retum-to-work process. This current research, utilizing a single method approach, is concentrated on employers and investigates their attitudes and policies towards workers sustaining compensable injuries. The study parameters, therefore, focus only on worker demographic factors that are available through the employer personnel records. The following review discusses demographic factors based on empirical findings that influence retum-to-work outcomes. Typically, research indicates that size of company, age, gender, tenure, union presence, educational level, pre-injury payrate, nature and location of injury are relevant to retum-to-work outcomes. Principally, retum-to-work policies are associated with financially healthy, larger companies (Baril, Berthelette & Massicotte, 2003; Hester & DeceUes, 1990). Disability Management Strategies 21 The study conducted by Baril & Berthelette (2000) further reveals that access to and provision of early return to work measures among claimants of Quebec's Workers' Compensation Board (CSST) was dependent on the worker's age, gender, occupation, and type of injury. Similarly, Blackwell, Leierer, Haupt, & Kampitsis (2003) found the most significant individual predictors for returning to work were education, age, mandated rehabilitation, time of injury to referral for rehabilitation, and lawyer involvement. Likewise, Butler, Johnson, & Baldwin (1995) observed from a cohort of Ontario workers with permanent partial impairments that age, education and gender are characteristics that influence return to work outcomes. Shaw & Polatajko (2002) cite age, severity of illness, and gender as the three most frequently studied individual factors with age cited as the strongest but not singular predictor of work outcomes. Baril & Berthelette (2000) also found that among CSST claimants retum-to-work provisions were statistically more likely to be applied towards female workers aged between 30-39 years in the health and social service sector. The differential application of return to work measures was based on the belief by CSST and workplace representatives that older workers take longer to recover and were more likely to be laid off; younger workers had quicker recovery times but their limited work experience hindered the facilitation of reassigned duties. An association between site and type of iryury was also identifred in this study. Early retum-to-work measures were statistically more likely to be applied in cases of inflammation to upper limbs and shoulders. The explanation provided suggests that CSST and the larger employers of the health and social service sector had greater ability to accommodate this type of injury which was also associated with short duration of absences. Butler et al. (1995) reviewed the three most frequent work-related injuries of back Disability Management Strategies 22 sprains/strains, other sprains/strains, and fractures. The authors report the probability of keeping ones original job is greater for workers suffering other sprains/strains. Gender differences noted in the results are considered a reflection of the different physical demands of typical male and female jobs. Female workers were most at risk for unstable employment patterns following an injury. The probability of not returning to work increases for every year of age among older workers and among women workers. Women were more likely to experience multiple spells of absence and unsuccessful returns to work (see also Crook, Milner, Schultz, & Stringer, 2002). Salkever, Shinogle, Purushothaman (2001) analyzed long-term disability (LTD) rates and report the incidence rate to be 1.7 times greater for female workers than male workers. In contrast, the statistics reported by the WCB indicate that for the 41,275 male workers injured during 2003, the corresponding female rate of injury was 17,560, approximately 23% of total injuries (WCB, 2004). Baril et al. (2003) similarly found that females sustained lower rates of work injuries (35.4%) than males. Higher educated and more experienced workers were better able to compensate for their limitations by modifying the job demands and work schedules than were less experienced, less educated workers (see also Johnson & Ondrich, 1990). Schechter (1999) found a clear association between educational attainment and return to work. A work history of short-term jobs and a brief period on the job prior to injury reduced the probabihty of returning to work (Aronoff & Feldman, 2000). Length of tenure is also associated with incidence rate of reported accidents. Habeck, Hunt et al. (1998) noted that less than one year's tenure was significantly associated with higher accident rates. In the same study, the presence of unions was also found to be associated with a 17% higher accident rate. Similar findings are reported by Hirsch, MacPherson & Dumond (1997) citing the availabihty of Disability Management Strategies 23 claim processing information as the primary reason. Butler, Baldwin, & Johnson (2001) indicate that union presence assists in the return to work process, mainly by securing the preinjury position for the worker but that union membership has little effect on promoting stable employment following the return to work. Johnson & Ondrich (1990) cite union presence as a factor influencing duration of work absence, particularly within unions that have retained a lawyer, and when absence from the workplace is considered a strategic maneuver. These authors also found that the higher the pre-injury wage the shorter the absence from work (see also Baldwin, et al., 1995; Butler, et al., 2001; Habeck, Hunt et al., 1998). Overall, the research suggests that demographic factors are valuable predictors of return to work when considered in reference to workplace attitudes, practices and job demands. The current research will obtain demographic information relating to age, gender, tenure, union, education, pay rate, nature and location of injury to examine the relationship between these factors and employee outcomes measured as return to regular duty, post-injury pay, relapse, accommodation and continued tenure. The original intent of this study was to use structural equation modeling to test the hypothesis that outcomes for the employee are positive when company attitudes towards disability are mediated by disability management practices. Due to the inadequacy of the data modeling could not be conducted, however, the revised objectives use regression analysis to examine the predictive value of company DM perceptions and company DM policies on return to work outcomes for the injured worker. Recently, the research focus has turned to the question of durability. Researchers are examining how durable the benefits are from disability management strategies in terms of the long-term employment retention rate of the injured worker. The success of disability Disability Management Strategies 24 management programmes has consistently been measured in terms of cost containment for the employer and reduction in claim duration for the employee. As reviewed by Krause et al. (1998) modified work and return to work literature concludes (a) that iigured workers who are offered modified work return to work about twice as often as those who are not; and (b) that modified work programmes cut the number of lost work days in half. It is accepted throughout the hterature that there is a window of opportunity where return to work is an attainable outcome in the early stages of disability claims and that the probability of returning to work decreases the longer the duration of claim. Frymoyer & Nachemson (1991) found that duration of absence has a substantial influence on return to work outcomes. From their literature review, they determined that low back incidences have a return to work rate of 89% within 3 months, a 40% return to work rate after six months absence, a 20% return to work rate after one year and a return to work rate approaching zero after two years of absence. The Zabrodski (1998) literature review found that there was a workers’ compensation/ disability cost containment bias in assessing outcomes and less research assessing the employee’s long-term outcome after returning to work. Zabrodski suggests that an early return to work by itself does not guarantee a favourable long-term outcome for the worker. Cater & Smith (1999) suggests the initial retum-to-work may be an insufficient measure of the successful re-integration of the disabled worker. Research conducted by Butler et al. (1995, 2001), Johnson et al. (1995) and Baldwin et al. (1996) suggest that workers returning from WCB claims to their pre-injury employer have a higher rate of termination and lower rate of stable employment than non-injured workers over the course of time. Butler et al. (1995) demonstrated that measuring the Brst return to work is an over estimation of the successful re­ integration of WCB claimants in a population of workers sustaining permanent partial Disability Management Strategies 25 impairment. Their analyses, based on data from the Onfano q/" wrfA jpg/TMangnr f/Mpairmgnrf, infers that 85% of injured workers return to work, however, when evaluated over a period of years the rate dropped to 50% of workers able to maintain employment. Identifying post-injury patterns revealed four specific patterns; (a) single absence, successful return, (b) single absence, unsuccessful return, (c) multiple absences, successful return, and (d) multiple absences, unsuccessful return. The authors deduced from their study that better educated employees incurred greater financial investment from firms which in turn provided the incentive to accommodate the injured employee. The accommodations provided resulted in higher retention rates for these employees. Less likely to return to work were older, non-union workers with less education. Union membership had the advantage of protecting pre-injury job and initial return to work outcome but showed less success with maintaining employment status over time. No specific analysis was undertaken to illicit the impact of disability management intervention on the patterns identified. The Butler et al. (2001) analysis was based on a survey population of Ontario workers examined for permanent partial disability by WCB physicians between June 1989 and June 1990. The Canadian experience with disability management was still in its infancy during that timeframe. A 1997 Watson Wyatt report found that 39% of Canadian employers surveyed supported some form of coordinated disability management for employees (Steeves & Smithies, 1998). Harder & Voaklander (2003) found the support for disability management to have improved with 82% of responding employers reporting that they offered some form of DM services, however, only 11% of these employers had dedicated staff coordinating 'in house' services. The Butler findings, therefore, indicating that only 35% of Disability Management Strategies 26 injured workers sustained their employment after first return to work may not represent the current employment retention rates of injured workers. With Canadian employers becoming more proactive in managing disability, the underlying assumption of the current research is that workplace accommodations mitigate disabihty associated with impairments. Adapting the required job demands to accommodate the functional abilities of an injured employee mitigates the impact of a disability. Mitigating the impact of disability facilitates continued employment for that employee. To test the assumption this research measured company perceptions, policies and employment status of WCB claimants five years after claim closure. To this end, the information available from the literature was used to devise a questionnaire to measure company policies, perceptions and retum-to-work outcomes. The true outcome of interest was the mediation value of the perception and policy variables in predicting positive retum-to-work outcomes for the injured worker in terms of length of retention, resumption of regular duties, maintenance of earnings, access to accommodation and lack of relapse. The available data, however, did not permit the creation of a path diagram to investigate the hypothesis that the best fit model for predicting outcomes for injured workers is mediated by positive company policies within workplaces that have positive perceptions of disability management. Consequently, the revised measures evaluated the predictive value of DM perceptions and policies. A summary of the literature cited thus far, suggests an effective DM programme relies on a work culture that combines the philosophical and operational principles of DM (Habeck et al., 1998; Unger, 1999). The philosophical principle promotes a work culture that expresses positive attitudes towards individuals with disabilities and the operational principle promotes the policies that support individuals with disabilities in the Disability Management Strategies 27 workplace. Rather than the more frequently used outcome of cost-containment, the outcome criterion measured was Retum-to-work linking this study more directly to the objectives of DM as cited by Tate et al. (1986). Based on the literature, therefore, Objective (a), examined the predictive value of DM perceptions (philosophical principle) and DM policies (operational principle) in retum-to-work outcomes for the injured worker. Dividing the objective into steps as presented in Figure 1, the expectation for Objective (a) was that DM perceptions make a significant contribution to the presence of DM policies in the workplace (Tate et ah, 1986); that DM policies make a significant contribution to retum-to-work outcomes for the injured worker (Tate et al., 1986); but that DM perception without the support of policy is not a significant contributing factor to outcomes (Tate et ah, 1986). The second objective. Objective (b), examined factors that influence DM perceptions, DM policies, and Retum-to-work outcomes. Employer characteristics that are empirically supported (Habeck, Hunt et al., 1998) to influence retum-to-work outcomes for injured workers were used to achieve this objective. The hypothesis was that there are differences in employer characteristics of Industry type, Unionization, Hourly paid workers, and Company size that are influential factors in DM perceptions, DM policies and retum-to-work outcomes for the injured worker. In an exploratory vein, the third objective of this study. Objective (c), was to examine the relationship between outcomes for the injured worker and various factors such as age, gender, years of service, union membership, educational level, pre-injury pay rate, nature of injury and location of injury. The research supports that the variables selected are factors known to influence retum-to-work outcomes for the injured worker (refer to section Wbriker p20). The hypothesis expected that differences in age, higher educational Disability Management Strategies 28 level, and higher pre-injury pay rate would he related to outcomes for the injured worker. No a priori expectations, however, were made for gender, union membership, nature or location of injury. Disability Management Strategies 29 Disability Disability Management Management Retumtowork Perceptions Policies Outcomes Step 1; Perceptions predict Policies Step 2: Policies Predict Retum-to-work Outcomes Step 3: Perceptions are no predictor for Retum-towork Outcomes Figure 7. Diagram representing the three steps to achieve Objective (a). Disability Management Strategies 30 Method The participants were employers from a Northern B.C. community. From a listing of all companies (total 186,000) covered under the jurisdiction of the Workers' Compensation Board of British Columbia (WCB), a localized sampling frame was extracted using postal codes (V2J through to V2N) to identify companies in Prince George and surrounding area. This sampling frame of 5,736 companies was further reduced using payroll information provided from the WCB listing. The criterion of payroll over $100,000 yielded a total sampling frame of 1,559 companies. To control for inherent differences between smaller and larger companies, the sampling frame was ranked by payroll and stratified into four groups: Group 1 with payroll under $200,000 (total 676 companies); Group 2 with payroll between $200,000 and $599,0OO (total 598 companies); Group 3 with payroll between $600,000 and $999,000 (total 125 companies); and Group 4 with payroll over 1 million dollars (total 160 companies). To have an even representation of company size, random selection of 25 companies was conducted separately for each of the four groups using the point entry method (point of entry 70). Each 30^ company was selected to a total of 100 companies. This is a retrospective study using employers' Workers' Compensation Board of British Columbia claims information. The data for the study was collected using a self­ administered questionnaire. The questionnaire gathered information on company demographics, company disability management policies, company perceptions of disability management as well as employee demographic and retum-to-work information. Disability Management Strategies 31 Ethics approval from the University of Northern British Columbia and employers' informed consent was obtained prior to the commencement o f data collection. Telephone contact was made with the 100 randomly selected companies to identify the most appropriate official responsible for personnel within each organization. The questionnaire was faxed to each official. Follow-up telephone contact was made with each recipient to secure their participation in the survey. From the initial contact, 42 replacements were required for companies that indicated they were unable to participate. Replacement selection followed the same procedure as the initial selection. Questionnaires were faxed to these most appropriate officials identified by telephone contact. A further 21 replacements were selected as a consequence of refusals to participate. Telephone contacts alone (maximum 4) secured a response from 37 companies, 29 companies required on-site visits and 6 questionnaires were completed by telephone interview providing a total of 72 participants. The research sample size was estimated using multiple regression analyses and three predictors (company size, company policies, company attitudes) requiring a total of 76 participants when power =.80, a = .05 and medium effect size, i.e. =.15 (Cohen, 1992). The 72 respondents comprised the sample population. Companies were asked to complete a questionnaire that consisted of a company demographic section, a DM policy section, a section covering the company's perception towards disability management and a section advising of retum-to-work outcomes for selected injured employees. Companies were also required to identify and provide information on one injured employee who met the following criteria: (a) Minimum of 6 weeks paid disability compensation by WCB. Disability Management Strategies 32 (b) Claim closure in 1999. (c) Claim closed for reasons of retum-to-work. Employers were advised that they may use the WCB Claim History Report, personnel record computer system, or benefits computer system to identify the employee. The 14 item section of DM policies was designed to capture the company’s behaviour relating to disability management. Each item was a statement about policy and procedures currently enforced within the organization. The participant was required to indicate their level of agreement with each statement presented, for example, item 1 of the DM policy sections stated “We regularly educate and inform all employees about our early return to work options.” All items were scored on a 5 point Likert scale (l=disagree, 2= somewhat disagree, 3= neutral, 4= somewhat agree, 5= agree). The sum of the scores was defined as DM policy with higher scores indicating positive policies and lower scores indicating less positive policies regarding disability management. The 14 item section of DM perceptions was designed to capture the company’s opinion towards disability management. Items were similarly presented as statements that measured opinions on matters relating to the accommodation of workers with restrictions, e.g. item 2, DM perceptions stated “An employee returning to work with workplace restrictions hinders the recovery process of that worker." Likewise, belief in the benefit of disability management was measured e.g. item 9, DM perceptions stated “It costs less to accommodate returning employees who need workplace accommodations than to train new employees." All items were scored on a 5 point Likert scale (l=disagree, 2= somewhat disagree, 3= neutral, 4= somewhat agree, 5= agree). The sum of the scores was defined as Disability Management Strategies 33 DM perceptions with higher scores indicating positive perceptions and lower scores indicating less positive perceptions of disability management. The section, Injured Worker Retum-to-work Information comprised 5 statements to capture the status of ii^ured employees in terms of resumption of duties, income, relapse, accommodation, and retention. All items were scored on a 5 point Likert scale (l=disagree, 2= somewhat disagree, 3= neutral, 4= somewhat agree, 5= agree). The sum of the scores was defined as Retum-to-work outcomes with higher scores indicating positive outcomes and lower scores indicating less positive outcomes. The order of the questionnaire was counterbalanced to control for response bias (Ray, W.J., 1997). The cover letter and sample questions from the questionnaire are presented in Appendices A and B. The overall response rate to the questionnaire was 44%. The sample consists of 15 companies from Group 1 representing companies with payroll under $200,000 and a response rate of 60%; 18 companies from Group 2 representing companies with payroll between $200,000 and $599,000 and a response rate of 72%; 18 companies from Group 3 representing companies with payroll between $600,(X)0 and $999,000 and a response rate of 72%; and 21 companies from Group 4 representing companies with payroll over 1 million dollars and a response rate of 84%. Of the 72 responses received only 21 (29%) reported claims data. Meofure.; All statistical analyses were conducted using SPSS for Windows, version 12.0. AH data entry items, reverse scoring and coding were double-checked for accuracy. Factor analyses were conducted to ensure the variables used for the constructs of company disability management policy and company perceptions towards disability management were Disability Management Strategies 34 appropriate measures. As the dimensions identified are inter-related components of DM perception and DM policy, the sum of response scores was measured from their respective sections in the survey. D6/gcr;vg faj - to examina tAe prarActiva vaZwa parcaptio/w amf DMpoZicias in ramm-io-work owicomaj^ ybr iZia Zrÿwrad wortar. For this objective, Hi stated that DM perceptions contribute to the presence of DM policies. Hz stated that DM policies contribute to retum-to-work outcomes for the injured worker. H 3 stated that DM perception is not a direct influence on retum-to-work outcomes for the injured worker. To test these hypotheses three regression analyses were conducted. In the first regression, DM perception was tested using DM policy as the criterion variable. The second regression tested the predictive value of DM policy using Retum-to-work outcomes as the criterion variable. To evaluate the criterion variable of Return-to-work outcomes, measured as the sum of responses to Injured Worker Retum-to-work Information sheet, logarithmic score transformation (Tabachnick & Fidell, 2001) was required due to the substantial negative skewness (-.661). The third regression tested the predictive value of DM perceptions for the criterion variable, Retum-towork outcomes. DZÿecrivg (6) - to rgiMm-to-worA: iZzoi DM DM poZZcZe.y, and The overall hypothesis. Hi, for this objective stated that there is a difference between employer characteristics of (a) industry type, (b) unionization, (c) hourly paid workers, and (d) company size that are more likely to contribute to (1) positive perceptions of DM, (2) the presence of DM policies in the workplace, and (3) outcomes for the injured worker. To test this hypothesis the independent variables were Industry type, Unionization, and Hourly-paid workers and Company size. Pearson's correlation revealed Disability Management Strategies 35 that the three dependent variables of DM perception, DM policies and Return to work Outcomes were related (r = .58 between Perception and Policy; r = .38 between Policy and Retum-to-work; and r = .24 between Perception and Retum-to-work), therefore, multivariate analysis of variance (MANOVA) was conducted. As this is a preliminary study, follow-up analyses of variances (ANOVAs) were conducted to assess whether there were differences among the factors for certain of the dependent variables. Using the Bonferroni procedure (Green, Salkind, & Akey, 2000) the ANOVAs were tested at the .01 level. Bonferroni Post Hoc Tests were performed if a significant difference was found and if the assumption of homogeneity of variance was met, if not, Dunnett’s C was used. Industry type was categorized into five groups based on the reported company operation. Type 1 consisted of 18 companies involved in manufacture, construction or repair and accounted for 25% of the sample. Type 2 comprised 18 companies involved with retail, hospitality, sales and transportation service (25% of the sample). Type 3 comprised of 14 companies involved in the forestry, lumber, or trucking business and accounted for 19.4% of the sample. There were 11 organizations in Type 4 involved with health care and education (15.3%). Type 5 companies were involved with consulting (15.3%). Unionization was measured as a dichotomous variable with 0 representing no bargaining units and 1 indicating one or more bargaining units within the workplace. The variable. Hourly-paid workers, was grouped into 3 categories based on the percentage of workers paid on an hourly basis. Category 1 represented the companies reporting 40% or less of its workers were paid on an hourly basis (23.6%); Category 2 comprised companies with percentage of hourly paid workers between 41 and 80% (27.7%), and Category 3 comprised companies with percentage Disability Management Strategies 36 o f hourly paid workers between 81 and 100% (48.7%). Company size was a measure of the payroll categories used in the selection of companies for the survey. (c) - m examine the r e la t io n b e t w e e n demogr^hic^zetora^ q/^the individual and return to wort outcomes. The general statement for the exploratory hypothesis, Hi, stated that differences between demographic factors such as (a) age, (b) gender, (c) years of service, (d) union membership, (e) educational level, (f) pre-injury pay rate, (g) nature of injury and (h) location of injury are related to Retum-to-work outcomes for the injured worker. Separate ANOVAs were conducted on the component variables. The within-group variable measured was outcomes for injured worker; the between-group variables were age, gender, years of service, union membership, educational level, pre-injury pay rate, nature of injury and location of injury. The source of this data was the questionnaire section titled Demographic Information Regarding Injured Worker. Eight ANOVAs were conducted. Multiple comparison tests were completed if the factors proved to be significant (p < .01). Bonferroni Post Hoc Tests were used if assumptions of homogeneity of variance were met, if not, Dunnett’s C was used. Disability Management Strategies 37 Results Preliminary Analysis The appropriate items from the DM Pohcy measure and the DM Perception measure were reverse-scaled. The item means and standard deviations for both measures are presented in Table 1. Histograms of company means are presented in Appendices D and E. Factor analysis was conducted for the 14 items from the DM Policy measure and the 14 items from the DM Perceptions measure. All variables were screened for outliers and skewness. The distribution of the DM Policy and DM Perception scores displayed a low negative skew (-.3). Subjecting the data to square root transformations (Tabachnick & Fidell, 2 0 0 1 ), however, overcompensated the skewness therefore the untransformed data was used in the analysis. The dimensions of the 14 items from the DM Policy measure were analyzed using maximum likelihood factor analysis. The criteria used to determine the number of factors to rotate were the scree test and the interpretability of the factor solution. Four factors were rotated using a Varimax rotation procedure. Using a cut-point loading rating of 0.3 (Tabachnick & Fidell, 2001) the rotated solution, as shown in Table 2, yielded four interpretable factors labeled: (a) company DM strategy; (b) consensus for DM; (c) company DM practice; and (d) company dissemination of DM. Company DM strategy accounted for 20.49% of the item variance. This factor comprises items 9 ,1 0 ,1 1 , and 12 which focus on accommodation strategies available to employees. Consensus accounted for 12.82% of the item variance. This dimension converges on labour-management agreement and comprises items 2 and 3. The third factor Disability Management Strategies 38 Table 1 DM foZicy owf DM Perception Meaywre^ (N=72) DM Policy Mean SD DM Perception Mean SD Item 1 2.76 1.524 3.96 1.238 Item 2 2.50 1.520 4.08 1.242 Item 3 2.25 1.319 2.76 1.316 Item 4 3.78 1.436 3.94 1.331 Item 5 3.18 1.532 2.26 1.374 Item 6 2.57 1.564 4.29 1.106 Item 7 3.00 1.520 2.94 1.509 Item 8 2.88 1.661 3.49 1.353 Item 9 3.68 1.422 3.65 1.280 Item 10 3.78 1.503 3.58 1.219 Item 11 3.56 1.537 4.10 1.023 Item 1 2 3.07 1.550 4.06 0.854 Item 13 3.22 1.484 3.26 1.138 Item 14 3.85 1.296 3.40 1.109 Disability Management Strategies 39 accounted for 11.72% of the item variance and encompasses a dimension of DM practice from items 5 ,6 , 7, and 8. Factor 4, accounting for 9.92% of item variance, comprises items 1,4, and 14. This dimension converged on items relevant to the dissemination of DM among employees. Item 13 had a low loading across aU factors and was excluded from subsequent analysis. The factor analysis procedure was repeated for the 14 items from the DM Perception measure using the same cut-point loading rating of 0.3 (Tabachnick & Fidell, 2001). The rotated solution yielded three interpretable factors and is presented in Table 3. Based on the content of the three sets of items the factors identified are: (a) workplace accommodation; (b) company commitment; and (c) impact. Workplace accommodation accounted for 26.61% of the item variance. The dimension converged on items 1, 2, 3, 4, 5, 7, 8 , and 9 centering on the employer’s perception of the injured returning to duty with workplace restrictions. Factor 2, company commitment, converged on items 10 and 14, accounting for 16.28% of the item variance. The impact dimension accounted for 13.78% and comprised items 6 , 11, and 12. Item 13 had a low loading across all factors and was excluded from subsequent analysis. Disability Management Strategies 40 Table 2 Factor DM Fo/icicf Factor 1 2 3 4 Policy 1 .206 .427 .267 .712 Policy 2 .051 .754 .167 .285 Policy 3 .129 .885 .096 .142 Policy 4 .150 .273 .096 .468 Policy 5 .190 -.053 .422 .133 Policy 6 .170 .091 .515 -.467 Policy 7 .099 .155 .547 -.050 Policy 8 .144 .178 .767 .202 Policy 9 .860 .265 .167 .071 Policy 10 .662 .039 .311 .060 Policy 11 .901 .152 .236 .101 Policy 12 .721 .069 .105 .130 Policy 13 .301 -.047 .031 .152 Policy 14 .300 .134 -.011 .473 Extraction Method: Maximum Likelihood. Rotation Method: Varimax with Kaiser Normalization, a Rotation converged in 6 iterations. Disability Management Strategies 41 Table 3 Factor Matrùc(aj DM Perceptions 1 Factor 2 3 Perception 1 .726 -.191 .280 Perception 2 .739 -.064 .015 Perception 3 .673 .002 .106 Perception 4 .786 -.022 .229 Perception 5 .626 .050 .177 Perception 6 .085 -.216 .589 Perception 7 .818 .152 -.133 Perception 8 .526 .161 -.137 Perception 9 .423 .283 .158 Perception 10 .072 .741 .116 Perception 11 .100 .472 .774 Perception 12 .114 .429 .824 Perception 13 .098 -.672 .101 Perception 14 .144 .809 .209 Extraction Method: Maximum Likelihood. Rotation Method: Varimax with Kaiser Normalization. a Rotation converged In 4 Iterations. Disability Management Strategies 42 Primary Analysis Ofÿectivg (a). Structural equation modeling using Amos 4.0 was the procedure initially planned to analyze modeling hypotheses testing the best At for predicting positive outcomes for the injured worker. It was not feasible to construct the path diagram from the data provided on the injured worker. Of the 72 completed questionnaires, only 21 participants’ claims data met the inclusion criteria which were; (a) Minimum of 6 weeks paid disability compensation by WCB. (b) Claim closure in 1999. (c) Claim closed for reasons of retum-to-work. The hypotheses to achieve Objective (a) were, therefore, revised. Three hypotheses statements were used. H, stated that DM perceptions predict the presence of DM policies. Hz stated that DM policies predict retum-to-work outcomes for the injured worker. H 3 stated that DM perception is not a direct predictor of retum-to-work outcomes for the injured worker. These hypotheses allowed for three separate bi van ate linear regression analyses to be conducted. The results of the regression analyses show H, was significant (p<.05). The scatterplot, as shown in Figure 2, indicates that there is a positive linear relationship between DM perceptions and policies suggesting that as DM perceptions are more positive so too are DM policies. The strength of the relationship (r) was .58 indicating the overall accuracy in predicting the presence of company DM policy from company perception towards DM was high in this study. Approximately 34% of the variance of DM policies was accounted for by its linear relationship with DM perceptions. As hypothesized, positive DM perceptions result in positive DM policies in the workplace. Disability Management Strategies 43 80 — 60- ## # # 20- # R Sq Linear = 0.341 0- 30 40 50 60 Sum of Perception Figwrg 2. Scatterplot depicting the relationship between DM policy and DM perceptions. Disability Management Strategies 44 Hz looked at the predictive value of DM policies in Retum-to-work outcomes. The reported frequency of claims (N=21) to evaluate the criterion variable of Retum-to-work outcomes was less than expected. A more liberal approach was taken, therefore, when evaluating the data for this regression analysis by setting the alpha level at 0.1. The regression procedure to evaluate the prediction of Retum-to-work outcomes from DM pohcies proved to be significant (p=.084). The strength of the relationship (r) was .39 indicating the overall accuracy in predicting retum-to-work outcomes was moderate. Approximately 15% of the variance of Retum-to-work was accounted for by its linear relationship with DM policies. As hypothesized, positive DM policies are predictive of positive retum-to-work outcomes for the injured worker. Hg looked at the predictive value of DM perceptions in Retum-to-work outcomes. The procedure to evaluate the prediction of Retum-to-work outcomes from DM perceptions was non-significant ip> .\) as hypothesized. Table 4 presents a summary of the findings. The descriptive statistics presented in Appendix E will be reviewed further in the Discussion section to describe the pattems that have emerged from the analyses. Disability Management Strategies 45 Table 4 Predictor Variable Criterion Variable df R R^ F Sig. Perception Policy 1/70 .58 .34 36.28 .0 0 0 * Perception Retum-to-work 1/19 .24 .06 1.18 .290 Policy Retum-to-work 1/19 .39 .15 3.33 .084** *p<.0005 Disability Management Strategies 46 06/gcrivg The overall hypothesis, Hi, for this objective stated that there is a difference between employer characteristics of (a) industry type, (b) unionization, (c) hourly paid workers, and (d) company size that are more likely to contribute to (1) positive perceptions of DM, (2) the presence of DM policies in the workplace, and (3) outcomes for the injured worker. To examine factors that influence the dependent variables of DM perceptions, DM policies and Retum-to-work outcomes, multivariate analysis of variance (MANOVA) was conducted. The independent variables were employer characteristics of Industry Type, Unionization, and Hourly-paid workers and Company size. No significant differences were found (p>.05). Table 5 reports a summary of the results. In this study, the data for the dependent variable Retum-to-work outcomes (N=21) was not consistent with the remaining dependent variables (A=72) and in the MANOVA procedure a participant’s data is excluded if scores are missing on any of the dependent variables. Therefore, despite the non-significant findings of the MANOVA and as this was a preliminary study, separate analyses of variances (ANOVA) were also conducted. Using the Bonferroni procedure (Green, Salkind, & Akey, 2000), each ANOVA was tested at the .01 level. The results of the ANOVAs revealed no significant differences (p>.01) among company characteristics to perceptions towards DM, the presence of DM policies in the workplaces or in Retum-to-work outcomes for the injured worker. The summary results for the ANOVAs conducted are presented in Table 6. The descriptive statistics presented in Appendix E are further reviewed in the Discussion section to describe the pattems that have emerged from the analyses. 06/echvg (c). The general statement for the exploratory hypothesis, Hi, stated that differences between demographic factors such as (a) age, (b) gender, (c) years of service, (d) Disability Management Strategies 47 union membership, (e) educational level, (f) pre-injury pay rate, (g) nature of injury and (h) location of injury are related to Retum-to-work outcomes for the injured worker. To evaluate the relationship between these demographic factors of the individual and Return to work outcomes eight separate analyses of variance were conducted. No significant difference was found between the variables (p>.01) to support the hypothesis. F values and significance levels are presented in Table 7. The descriptive statistics presented in Appendix E will be reviewed further in the Discussion section to describe the pattems that have emerged from the analyses. Disability Management Strategies 48 Table 5 .ÿwm/Mary o/'Afw/tivan'atg Independent Factor q/"Variance /br 06/ectrve F Partial Eta Square Industry Type W ilks'A .23 df 12/21 1.32 Sig .28 Unionization .85 3/8 .48 .70 .15 Hourly-paid .66 6/16 .62 .71 .19 Company size .47 9/20 .77 .64 .22 p<.05 .39 Disability Management Strategies 49 Table 6 qyfTm'vanafg 06/gctrvg Dependent variable: DM perceptions Independent Factor df MS F Sig. Partial Eta Square Industry type 4/61 53.83 .64 .636 .040 Unionization 1/61 25.42 .30 .584 .005 Hourly-paid 2/61 1.90 .0 2 .978 .0 0 1 Company size 3/61 17.16 .21 .893 .0 1 0 Note = .088, adjusted = .061 *p< . 0 1 Dependent variable: Sum of Policy Independent Factor df MS F Sig. Partial Eta Square Industry type 4/61 287.04 2.36 .063 .134 Unionization 1/61 69.44 .57 .452 .009 Hourly-paid 2/61 51.92 .43 .654 .014 Company size 3/61 51.91 .43 .734 .0 2 1 Note *p<.01 = .179, adjusted = .044 Disability Management Strategies 50 Table 6 cont. Swrn/wzfy q/"[//livanofg OZÿgcfivg (6). Dependent variable: Retum-to-work outcomes Independent Factor df MS F Sig. Partial Eta Square Industry type 4/10 .017 1.72 .221 .408 Unionization 1/10 .010 .61 .454 .057 Hourly-paid 2/10 .015 1.56 .258 .237 Company size 3/10 .019 1.92 .191 .365 Note *p<.01 = .552, adjusted = .105 Disability Management Strategies 51 Table 7 SMmwwfy /or Factora^ Co»trZ6wri»g to Fatwrn to Wort Owtco/ngr Independent Factor df MS F Sig. Age 1/19 .003 .220 .644 Gender 1/19 .005 .443 .514 Tenure 1/19 .011 .097 .759 Union Member 1/19 .002 .166 .688 Education 4/16 .009 .834 .523 Hourly Wage 4/16 .008 .656 .631 Nature of Injury 3/17 .013 1.195 .341 Location of Injury 4/16 .012 1.184 .355 Note Dependent variable: Sum of Return to Work Outcomes *p<.01 Disability Management Strategies 52 Discussion The objectives of this study were (a) to examine the predictive value of DM perceptions and DM pohcies in Retum-to-work outcomes for the injured worker; (b) to examine factors that influence DM perceptions, DM pohcies, and Retum-to-work outcomes for the injured worker; and (c) to examine the relationship between demographic factors of the individual and return to work outcomes. Research supports that there are multi­ dimensional individual factors as well as employer factors that contribute to the successful re-employment of injured workers. Research efforts for the current study, however, confined the investigation to the organizational level. Accordingly, organizational attitudes towards DM were measured. The factors used for the individual were demographic information that was available from personnel records. Thus, only organizations were sampled from within a Northern B.C. community for their policies towards and perceptions of disability management. The instrument employed to achieve the objectives of the study was designed to capture the employer’s behaviour and opinions towards DM. The factor analysis performed on the instrument demonstrated that the items loaded on clear dimensions of DM. The items selected to measure DM pohcy yielded four interpretable factors labeled (a) company DM strategy, (b) consensus for DM, (c) company DM practice and (d) company dissemination of DM. Similarly, the items selected to meastue DM perceptions yielded three clear dimensions that were labeled (a) workplace accommodation, (b) company commitment, and (c) impact. With further development, this instrument has potential for measuring behaviours and attitudes towards DM. In particular, the value of the instrument would lie in assessing resistance towards DM and/or incongruence of attitudes towards DM across different levels Disability Management Strategies 53 of an organization. The availability of an instrument that identifies areas of conflict such as resistance or incongruence may have high utility for the employer at the implementation or evaluation phases of DM programmes. (a) voZug pgrcgpfionj wni DM poZicigj; in The literature cited thus far, (Habeck et al., 1998; Unger, 1999) suggests that effective DM strategies rely on a work culture that combines a receptive and facilitative environment to assist the injured worker return to the workplace. To assess factors that predict outcomes for the injured worker, indices of DM perceptions, DM policies and Retum-to-work outcomes were used. Mimicking the original modeling hypothesis, the general hypothesis for this objective was that DM perceptions predict the presence of DM policies; DM policies predict Retum-to-work outcomes for the injured worker; but that DM perceptions alone are not a significant predictor of Retum-to-work outcomes. Using regression analyses the value of DM perceptions and DM policies in predicting Retum-to-work outcomes was analyzed. D M perceptions. Consistent with expectations, the findings from the present study support the hypothesis that there is a relationship between DM perceptions and the presence of DM policies in the workplace (p<.05). The indicators revealed that this relationship is positive and strong (r =.58), therefore, it is reasonable to conclude that positive perceptions towards DM are a good predictor of the presence o f DM policies in the workplace. Previous research supports that the effectiveness of any DM programme relies on a facilitative and receptive work environment (Amick et al., 2000). Research also reveals that people-oriented work cultures foster DM policies that are successful at re-integrating injured workers to the workplace (Habeck et al., 1991). Moreover, the incidence rate of disability is reported as an Disability Management Strategies 54 indicator of the type of culture within an organization (Habeck, Hunt, et al, 1998). Low incidence rates are associated with cultures that foster positive perceptions of disability management strategies further highlighting the importance of a receptive and facilitative enviromnent in the whole disability management process. In a similar vein, the Gndings from this study support that positive perceptions of DM are significant predictors of positive DM policies in the workplace. The regression analysis to evaluate the predictive value of Retum-to-work outcomes from DM perceptions was not significant ip> .l) despite the liberal approach taken in consequence to the lower response rate (7V=21) for Retum-to-work outcomes. This study hypothesized that DM perceptions would not be a contributing factor to Retum-to-work outcomes on the basis that DM perceptions alone are not singularly sufficient to impact Retum-to-work outcomes without DM policies. A receptive environment is not sufficient without a facilitative environment. Policy development and programme implementation are important processes in promoting disability management (NIDMAR, 2003). Despite the inadequacy of the claims data (A=2 1 ) to find a statistical significance, the effect size (R^=.06) which accoimts for 6% of the variance, would also support the speculation that DM perceptions alone are not a contributing factor to improving Retum-to-work outcomes. DM This study hypothesized that DM policies are a significant contributor to Retum-to-work outcomes. As the response rate (N=21) for Retum-to-work outcomes did not correspondent to the response rate of DM policies (A=72) a liberal approach was taken with this regression analysis, thereby, setting the alpha level at p< l. The findings of the regression analysis did support the contention (p < .l) that the existence of more positive DM policies is a predictor of positive outcomes for the injured worker. With such a liberal Disability Management Strategies 55 approach these findings are tentative, however, the results are in agreement with existing research (Aronoff & Feldman, 2000; Habeck, Hunt, et al., 1998). It is reasonable to speculate that a larger sample size would provide stronger support for the importance of DM policies in predicting retum-to-work outcomes for the injured worker. Extrapolating from the data presented in Appendix E further supports the hypothesis that companies are favourably disposed to managing disability in the workplace. In this study, 6 6 % of injured workers continued working for the same organization fives years after the initial return to work. Previous studies have suggested a lower rate of job maintenance when employment status is monitored over time. Butler et al. (1995), for example, reported that only 35% of injured workers studied during the timeframe 1989-1990 maintained employment status. That 6 6 % of injured workers maintained employment in this study may be indicative of greater employer awareness to the benefits that can be derived from managing disability within its workforce. Employers reported that all necessary accommodations were provided in 85% of cases and 25% of injured workers had subsequent relapses. Sixty percent of injured workers returned to their regular pre-injury duties. Subsequent reduction in earnings was reported in 20% of cases. This pattern may begin to refute the unwritten rule that the hkehhood of negative outcomes is markedly increased following an injury claim as described by Aronoff & Feldman (2000). In summary, the findings of this study support that more positive perceptions towards DM are a good predictor for the presence of DM policies in the workplace. The findings of the study also supports that DM policies predict Retum-to-work outcomes. Positive perceptions are not good singular predictors of Retum-to-work outcomes. The overall findings, however, do suggest a pattem of positive outcomes for the injured worker. The Disability Management Strategies 56 tentative support for the predictive value of DM policies may be attributable to the small sample size. Participation rates in the claims portion of the questionnaire (7V=21) did not correspond to overall participation rates (A^=72) and required an adjustment to a less stringent alpha level. Future investigation with a higher response rate may tap the importance of DM policies more successfully than the current study. OZÿgcrivg Examine /ocror.; that ir^wence DM perception.;, DM poiicie.;, and Eemm-to- wort ontcome;. The role of the work environment represents a very significant part in the outcome of the disability process (Habeck, Hunt, et al., 1998). The implementation and success of disability management strategies is cited as being dependent on labour-management commitment, support and active participation (Jarvikoski & Lahelma, 1980). Certain characteristics of the workplace, union presence for example, have posed challenges to the acceptance of and co-operation with disability management strategies (Baril & Berthelette, 2000). Accordingly, the hypothesis for this objective stated that there are differences between employer characteristics of (a) Industry type, (b) Unionization, (c) Hourly-paid workers, and (d) Company size that are more hkely to contribute to (1) positive perceptions of DM, (2) the presence of DM pohcies, and (3) outcomes of the iigured worker. Contrary to expectations, the analyses conducted fail to reject the null hypothesis suggesting that differences in the characteristics of Company type. Unionization, and Hourly-paid workers, and Company size are not contributing factors to perceptions towards DM, the presence of DM pohcies or retum-to-work outcomes. Despite the lack of findings, an interesting set of pattems have emerged from the descriptive data presented in Appendix E that are discussed further. Disability Management Strategies 57 /WMffry The conclusion from this study is that differences among industry type were not a factor in DM perception, the presence of DM policies or retum-to-work outcomes for the injured worker. The pattem derived from the data, however, in regard to Industry type warrants further comment. The m^oiity of claims reported were from industries where it can be speculated that the working environment and/or demands of the job pose a high risk of injury. Companies involved in aspects of the forestry/lumber industry, manufacture and constmction accounted for 47.6% of claims reported, and 60% maintained employment 5 years post injury. In comparison, speculating that companies involved in consulting, education or retail, sales and service for example, pose less risk to injury, these companies comprised 38% of claims reported. Of these claims, 62.5% maintained employment 5 years post injury. The pattem suggests that environments posing less risk of injury have a lower claims rate than high risk environments but that retum-to-work outcomes are similar in both environments. As the impact of job demands was not examined, the pattem must be viewed cautiously. Nonetheless, an interesting future direction would be to investigate if the comparable retum-to-work outcomes holds true for the different types of industry and whether any particular factor(s) contributes to comparable retum-to-work outcomes from sectors that pose different injury risks. f/nmnizarion. Renaud (2002) empirically supported the contention that union workers operate in poorer working conditions compared to their non-union counterparts. Non-union workers presented with promotion and career opportunities, pleasant working envirorunents and greater locus of control over their productivity reported greater satisfaction in their jobs than union counterparts without such working conditions. The level of dissatisfaction in the job is due to working conditions and not union membership. Consistent Disability Management Strategies 58 with the findings of Habeck et al. (1991) it is more likely that such workplaces would be unionized as unions tend to organize in situations of poor working conditions and negative labour relations. Both these factors are conducive to high claims rates (Habeck et al., 1991). Accordingly, the expectation would be that unions are involved to ensure the presence of DM policies. The evidence for the contribution unionization makes to the presence of DM policies, however, is inconclusive. Akabas (1996) posits the move towards DM has a longstanding history with trade unions stating “unions cut their teeth on D M long before the concept was officially identified” (p.33). In contrast, Shoemaker (1989) reports that 64% of unions surveyed remained mute on the issue of DM in negotiated contracts. Research indicates that unions are successful in protecting the injured worker’s job but are reportedly less successful in maintaining the worker’s employment status (Butler et al., 2001). In the current study, of the 2 1 claims information provided on injured workers, union members maintained their employment five years post-injury in 43% of cases where unions were present in 29.2% of all companies surveyed. Unionized companies indicated they did not have specific clauses in the bargaining agreement related to DM in 42.8% instances. Language related to early retum-to-work was not included in the bargaining agreement for 62% of the unionized companies despite clear language and explicit policy being cited as one of the basic elements of successful DM (Westmorland & Williams, 2002). The pattem elicited from the data suggests that unionized workplaces continue to have higher claims rates (57.1%) than non-union establishments (42.9%) but continue to remain mute (64% from Shoemaker study, 62% in the current study) on the issue of specific disability management language in collective bargaining agreements. Disability Management Strategies 59 Compony Although the current study did not Gnd a significant difference between size of company and perceptions, policies and retum-to-work outcomes, existing research (Cunningham & James, 2000; Habeck, Hunt, 1998) does suggest that larger companies are more likely to be actively managing disability in the workplace; are better able to provide accommodations than medium to smaller sized companies; and that larger companies are more likely to have computerized record keeping of absenteeism and claim trends. The latter was certainly observed during the data collection stage of this study. In the smaller companies memory and paper files were relied upon for the initial identification of workers and the extraction of relevant claim information requested in the survey. Of the 15 companies that comprised Group 1 (payroll under $200,000) all relied on WCB and their insurance carrier for disability management support and return to work initiatives. Two companies indicated they had a designated contact person within the organization for working with WCB. In comparison, of the 21 companies comprising Group 4 (payroll over million dollars) 10 companies provided some form of disability management. Among the larger organizations one third of companies have specific DM language in their employment contracts and/or bargaining agreement. The pattem that larger companies are more active in managing disability continues to hold true in this study. In summary, the findings for objective (b) have found no differences between employer characteristics of (a) Industry type, (b) Unionization, (c) Hourly-paid workers, and (d) Company size that are more likely to contribute to (1) positive perceptions of DM, (2) the presence of DM policies, and (3) outcomes of the injured worker. Interesting pattems did emerge from the data. In particular, the pattems suggest that environments posing less risk of injury have a lower claims rate than high risk environments but that retum-to-work Disability Management Strategies 60 outcomes are similar in both environments; that unionized workplaces continue to have higher claims rates than non-union establishments but continue to remain mute on the issue of speciHc disability management language in collective bargaining agreements; and that larger companies are more active in disability management than smaller companies. 06/gchvg (cj Ejcaming thg rgZohonyAip dgmogrupAic individwaZ o/id rgrwrn fo wo/* OMTco/z/gj^. In an exploratory vein, this study looked at demographic factors that are supported by research to influence retum-to-work outcomes for the injured worker (refer to Worker Demographics, p20). To assess the significance between these factors and return to work outcomes, separate ANOVAs were conducted. The independent variables were Age, Gender, Tenure, Union membership, Education, Hourly pay-rate. Nature of injury and Location of injury. As the evidence (Butler et al., 1995; Shaw & Polatajko, 2002; Crook et al., 2 0 0 2 ) supports that these factors are implicated in return to work rates, it was expected that a difference would be found between these variables and return to work outcomes. Contrary to previous research the variables proved non-significant in this study (p<.01) suggesting that these factors do not contribute to outcomes for the injured worker. Although the findings are not significant, the descriptive data presented in Appendix E will be discussed further to describe some of the pattems that were found. Agg (W Tg/zwrg. The general consensus is that the older worker is associated with longer and less successful recovery periods and is considered to have less potential for maintaining employment. In support of the consensus, previous research (e.g. Blackwell et al., 2003; Butler et al., 1995) has found that age is a determinant in return to work outcomes. For example, of the 246 successful retum-to-work outcomes monitored by Blackwell, 211 Disability Management Strategies 61 were less than 50 years of age. While the prevalence of work disability increases with age (Berkowitz & Hill, 1989) and reports of disability are twice as likely for workers aged 45-54 than workers aged 25-34, the age factor is not a clear-cut explanation for the poorer return to work rates. Greenblum (1979) found that the older worker was less likely to be offered rehabilitative services. In examining a Vocational Rehabilitation Services Administration, Greenblum revealed that the effect of rehabilitation on employment was greater for older, middle-aged workers than for younger workers. Rehabilitated older workers maintained employment at a higher rate than non rehabilitated older workers whereas rehabilitation had less effect on employment rates of rehabilitated and non-rehabilitated younger workers. Baril et al. (2003) found a similar trend in their review of return to work outcomes among Quebec workers’ compensation claims. Return to work measures were more hkely targeted to workers within the age range 30-39 years. Although not conclusive, it appears that the older worker is generally disadvantaged in return to work potential, not by age, but by lack of access to DM strategies. The current study varies somewhat from this pattem. Sixty-two percent of injured workers in this study were 40 years or older and of these, only 15.4% did not maintain employment with their pre-injury employer. From all claims reported, the older worker returned to their regular duties in 38% of instances even though 85% of employers agreed the necessary accommodations were provided and the remaining 15% indicated they somewhat agreed that the necessary accommodations were provided. The employer’s willingness to accommodate the older worker would suggest that employers are predisposed to continue the employment of their own workers with whom they have established a work history. In the current study, the mean length of service for older employees was 15 years. Looking at all Disability Management Strategies 62 cases, the minimum duration of pre-injury employment was five years with 57% of cases employed for 13 or more years with the same employer. Furthermore, 85% of injured workers were in the middle to higher wage bracket; and the m^ority required company training or higher to perform their duties. The suggestion is that seniority may be indicative of a greater ability to adapt or compensate for limitation (Cater & Smith, 1999, Baril et al., 2002). It may also be reasonable to speculate that the return to work rate of the older worker reflects the work skill and knowledge possessed. Accommodating older workers may also be indicative of employers’ realization that these skills and knowledge cannot be so readily replaced anymore. Nature and Location o f injuries. The hypothesis that nature and location of injury would influence return to work outcomes was not supported. The inadequacy of the claims data did not detect a relationship that is generally confirmed by previous research. The expectation for the study was that low back sprains and strains would constitute the majority of injury and be consistent with WCB reports. While sprains and strains did constitute 6 6 .6 % of the nature of injuries, the predominant location of injury was upper limbs/shoulders (38.1%), followed by lower limbs (28.6%). Although the hndings were not as expected the low frequency of occurrence of back injury may be more reflective of the sample population. The focus of the study was on claimants who had successfully returned to work. Back iiijury is progressive and chronic and high risk for continued disability. Employers may have selectively omitted this type of claim. Interestingly, Baiil & Berthelette (2000) found that early retum-to-work measures were more likely applied in cases of injury to upper hmbs and shoulders. The speculation was that these iiquries were more easily accommodated by employers. This is the pattem that may have been tapped into in this study as severity of Disability Management Strategies 63 injury was not a factor that was investigated. The ease with which an employer can accommodate a disability is an area for consideration in future investigation of the determinants of outcomes for workers returning to the workforce. In summary, the findings for objective (c) examining the relationship between demographic factors of the individual retum-to-work outcomes were not significant. The pattems derived from the data suggest, however, that employers are predisposed to continue the employment of older workers with whom they have established a work history. The majority of cases reported in this study worked 10 or more years with the employer. The majority were in the higher wage and education brackets. It may, therefore, be reasonable to speculate that accommodating older workers is indicative of the employer’s realization that the skills and knowledge possessed by older worker cannot be readily replaced anymore. Retum-to-work outcomes may be dependent on the ease with which an employer can accommodate an injured worker rather than nature or location of injury. Disability Management Strategies 64 Conclusion This study examined factors at the organizational level that influence return to work outcomes for the injured worker. The objectives were (a) to examine the predictive value of DM perceptions, DM policies, and Retum-to-work outcomes for the injured worker; (b) to examine factors that influence DM perceptions, DM policies and Retum-to-work outcomes; and (c) to examine the relationship between demographic factors of the individual and Retum to work outcomes. The objectives of the study were accomplished. The findings of this study support that more positive perceptions towards DM are a good predictor for the presence of DM policies in the workplace. It was also demonstrated, albeit at a less stringent level, that DM policies are predictive of retum-to-work outcomes for injured workers. The study also supported that DM perceptions are not a singular predictor of retum-to-work outcomes. Overall, these results support the general body of literature that suggests effective DM relies on a work culture that combines a receptive and facilitative environment to assist the injured worker retum to the workplace (see for example Habeck et al., 1998; Shrey, 1998; Unger, 1999). While the remaining hypotheses were not supported, the data revealed some interesting pattems. For example, from the findings a pattem is revealed that suggests environments posing less risk of injury have a lower claims rate than higher risk environments but that retum-to-work outcomes are similar in both environments. A plausible explanation may be the employer's competitive need to improve WCB experience ratings and contain premium and disability related costs, thereby, leading to the implementation of DM strategies (see for example Dyck, 2003). Another pattem suggests that unionized workplaces continue to have higher claims rates than non-unionized Disability Management Strategies 65 workplaces but that unions continue to remain mute on the issue of specific disability management language in collective bargaining agreements (see Akabas, 1996; Butler et al., 2001; Shoemaker, 1989). Both these pattems would be interesting avenues for further study, particularly, the union's perceptions towards disability management strategies and the factors contributing to labour-management hesitancy to include specific DM language in bargaining agreements. The pattem derived from the data also suggests that employers are predisposed to continue the employment of older workers with whom they have established a work history. It would be interesting to explore whether this pattem is specific to the rural population used in this study or whether this pattem holds in more urban areas where the labour pool is larger. The most relevant limitation for this study was sample size. Sample size was based on detecting a medium effect. This proved to be an insufficient source of claims data to address the issue of outcomes for injured workers. The reported frequency of claims (A=21) was lower than expected and did not correspond to the survey response rate (N=12). This limitation draws particular attention to the need for collaboration with the WCB to supplement employer reports of claims data. While there is no reason to doubt the integrity of the information provided, some employers may have limited their co-operation and failed to report claims even though the claims met the inclusion criteria. A more effective approach to be considered in future investigations is the initial step of identifying eligible claims through WCB collaboration before approaching employers for their input. Moreover, WCB collaboration may avoid the potential for selection bias. In the current study, twenty-nine on-site visits were conducted. In these instances, the principle researcher had the opportunity of completing the survey questioimaire with the employer and Disability Management Strategies 66 witnessed the extraction of the required claim's data from the actual claim's file. While the employers were graciously co-operative in providing reliable and valid information on the claims, they may have been selective in the type of claims reviewed. Employers may have singled out the uncomplicated claims that were easily accommodated to portray the employer in a favourable light. Selectivity was similarly noted by Baril & Berthelette (2000) who found retum-to-work measures were more likely applied in cases that the employer could more readily accommodate. Furthermore, the inclusion criteria used in the current study {Minimum 6 weeks paid disability compensation by WCB) was intended to eliminate transient, short-term absences so 'genuine' medical conditions could be compared. Future studies should consider residual functional limitations following maximum medical recovery rather than severity of injury as a factor that impacts retum to work outcomes for the injured worker. To this end, structural equation modeling as originally intended for this study could be most aptly applied to address the issue of mediating factors that influence outcomes for the injured worker. This is an area worthy of more detailed research. To close on a positive note, the strength of the study lies in the potential for the survey instrument and it warrants further development. The questions relating to perceptions and policies loaded on clear dimensions and may be valuable in identifying incongruent attitudes towards disability management between different levels of an organization. Resistance to disability management is demonstrated by research to be problematic in the implementation and success of DM programmes. Such a tool could prove beneficial during the regular evaluation process of the DM programme highlighting programme deGciencies and/or educational needs. Actively addressing such issues within a workplace may maintain collaborative efforts between labour and management - a basic Disability Management Strategies 67 element for success. The availability of a tool that identifies areas of resistance to disability management strategies would have high utility for the employer. As a fumre direction this would merit further investigation. Disability Management Strategies 68 References Akabas, S. H. (1994). Workplace responsiveness: Key employer characteristics in support of job maintenance for people with mental illness. 7 7(3), 91-102. Akabas, S., Fine, M., & Yasser, R. (1982). Putting secondary prevention to the test: A study of an early intervention strategy with disabled workers. Jowmal q/"Pn7Ma/y Prevention, 2(3), 165-187. Akabas, S. H., & Gates, L. B. (1990). Organizational commitment: A key to successful implementation of disability management. A m erican Rehabilitation, 16(3), 9-14. Amick, B., Habeck, R.V., Hunt, A., Fossel, A.H., Chapin, R.B., Keller, H., & Katz, J.N. (2000). Measuring the impact of organizational behaviors on work disability prevention and management. Journal of Occupational Rehabilitation, 10(1), 21-38. Anema, J.R., Cuelenaere, B., van der Beck, A.J., Knol, D.L., de Vet, H.C.W., & van Mechelen, W. (2004). The effectiveness of ergonomic interventions on retum-towork after low back pain: A prospective tow year cohort study in six countries on low back pain patients sicklisted for 3-4 months. O ccupational and Environm ental M edicine, 61, 289-294. Aronoff, G. F., & Feldman, J. B. (2000). Preventing disability from chronic pain: a review and reappraisal. International R eview o f P sychiatry, 12, 157-169. Baldwin, M. L., Johnson, W. G., & Butler, R. J. (1996). The error of using retums-to-work to measure the outcome of health care. Amgncan JowmaZ a/^/Wwrr/faZ MerZZcmg, 29(6), 632-641. Baril, R., & Berthelette, D. (2(XX)). Components and organizational determinants of workplace interventions designed to facilitate early return to work (A-263). Montréal, Quebec: Institut de recherché en santé et en sécurité du travail du Québec. Baril, R., Berthelette, D., & Massicotte, P. (2003). Early return to work of injured workers: Multidimensional patterns of individual and organizational factors, ^a/gfy .5cZgncg, 47, 277-300. Disability Management Strategies 69 Beaudway, D. (1986). 3M: A disability management approach. Jow/TmZ q/^AppZW Cow/wgZZng, 77(3), 20-22. Blackwell, T., Leierer, S. J., Haupt, S., & Kampitsis, A. (2003). Predictors of vocational rehabilitation retum-to-work outcomes in workers' compensation. Z(eZiahZZZtaZZon CoM/wgZZngBwZZgtZn, ^6(2), 108-114. Bruyere, S. M., Erickson, W. A., & VanLooy, S. (2004). Comparative study of workplace policy and practices contributing to disabihty nondiscrimination. Z(gZia6ZZZmtZoM P.^cZioZogy, 49(1), 28-38. Bruyere, S., & Shrey, D. (1991). Disability management in industry: A joint labormanagement process. R ehabilitation Counseling Bulletin, 34{3), 227-243. Bulter, R. J., Johnson, W. G., & Baldwin, M. L. (1995). Managing work disability: Why first return to work is not a measure of success. Industrial an d L abor R elations Review, 48(3), 452-469. Butler, R. J., Baldwin, M. L. & Johnson, W. G. (2001). The effects of worker heterogeneity on duration dependence: Low-back claims in workers compensation. The R eview o f Econom ics and Statistics, 83(4), 708-716. Burkhauser, R. V., Butler, J. S., & Kim, Y. W. (1995). The importance of employer accommodation in the job duration of workers with disabilities: A hazard model approach. 2, 109-130. Canadian Centre for Occupational Health and Safety (1999). Onfano '.y 6o.yZc rraZnZng progrommg.' manwaZ. Retrieved February 14, 2003 from http://www.worksafesask.ca/files/ont wsib/certmanual/prelim.html#Table Canadian Human Rights Commission (1997). Human Rights and employment: Excerpts from the CHRC Annual Report. ITbrkZ^ Rgporf, 70(3), 6-8. Cater, B., & Smith, J.B. (1999). Inferring disability from post-injury employment duration. AppZZgd Economicf TLgtrgr.;, 6,747-751. Disability Management Strategies 70 Cohen, D., Parrinello, A., & Kelliber, T. (1990). Disability management on the Long Island Railroad. .9(1), 29-39. Cohen, J. (1992). A power primer. 112(1), 155-159. Crook, J., Milner, R., Schultz, I.Z., & Stringer, B. (2002). Determinants of occupational disability following a low back injury: A critical review of the literature. JowmaZ q/" O ccupational Rehabilitation, 72(4), 277-295. Cunningham, I., & James, J. (2000). Absence and return to work: Towards a research agenda. Personnel R eview , 29(1), 33-42. Dasinger, L., Krause, N., Thompson, P. J., Brand, R. J., & Rudolph, L. (2001). Doctor proactive communications, retum-to-work recommendation, and duration of disability after a workers' compensation low back injury. Journal o f O ccupational an d Environm ental M edicine, 43(6). Davis, D. H. (1983, March). Bridging the claims dollar gap through rehabilitation. Risk M anagement, 3-8. Durand M. J., & Loisel, P. (2001). Therapeutic return to work: Rehabilitation in the workplace. Work, 17, 57-63. Dyck, D. (2003). DZs^aZ)ZZZfy managemgnr.' TTieory, .yfrmegy and Zndwjtry procricg. (2"^ ed.). Markam, Ontario: Butterworths. Franche, R. L., & Krause, N. (2(X)2). Readiness for return to work following injury or illness: Conceptualizing the interpersonal impact of healthcare, workplace and insurance factors. JowmaZ q/'OccwporionaZ RgZia^ZZZtorZon, 72(4), 233-256. Frank, J., Sinclair, S., Hogg-Johnson, S., Shannon, H., Bombardier, C., Beaton, D., & Cole, D. (1998). Preventing disability from work-related low-back pain. Canadian MgdZcaZ A.y.yacZadon dawmaZ, 75(9(12), 1625-1632. Friesen, M., Yassi, A., & Cooper, J. (2001). Retum-to-work: The importance of human interactions and organizational structures. Wiork, 77(1), 11-22. Disability Management Strategies 71 Frymoyer, JW. & Nachemson, AL. (1991). Natural history of low back disorders. In JW. Frymoyer (Ed.). TTig japing.oWpractice. New York. Raven Press, 1537-50. Galvin, D. E. (1986). Health promotion, disability management and rehabilitation in the workplace. Rehabilitation Z,iteratwre, 47(9-10), 218-223. Galvin, D. E., & Schwartz, G. (1986). E/npioyer-boje^i ^i.rabiiity management ami rehabilitation initiative.^. ATA Institute: Washington, DC. Gilbride, D. (2000). Employers’ attitudes toward hiring persons with disabilities and vocational rehabilitation services. Journal o f Rehabilitation, 66(4), 17-23. Green, S. B., Salkind, N. J., & Akey, T. M. (2000). Using SPSS f o r w indow s: Analyzing and understanding data. (2”*^ed.). New Jersey: Prentice Hall. Greenwood, J. G. (1984). Intervention in work-related disability: The need for an integrated approach. Social Science & M edicine, 19(6), 595-601. Guinter, R., Eagels, S., Harringer, R., & Trusewych, T. (1995). AT&T Bell Lab’s ergonomic program aims to cure VDT workstation ills. O ccupational H ealth & Safety, 64(2), 3035. Habeck, R. V. (1999). Job retention through disability management. Rehabilitation CoMTiyellng Rwlletln, 42(4), 317-329. Habeck, R., Hunt, H. A., & VanTol, B. (1998). Workplace factors associated with preventing and managing work disability. Rehabilitation Cowna^ellng Rwlletln, 42(2), 98-143. Habeck, R., Leahy, M. J., Hunt, H. A, Chan, F., & Welch, E. M. (1991). Employer factors related to workers' compensation claims and disability management. Rehabilitation Cown.yellng Rwlletln, 34(3), 210-226. Disability Management Strategies 72 Habeck, R. V., Scully, S.M., VanTol, B., & Hunt, A.H. (1998). Successful employer strategies for preventing and managing disability. RgAabiZitation Cow/weZmg Bwlknn, 42(2), 144-162. Habeck, R. V., & Shrey, D. E. (1991). Interchange. Reliabifimtion Cowrwelmg BwZktin, 34(3), 178-182. Harder, H. & Voaklander, D.C (2003). Dwobilify TMWiage/Meni career mvarene.;s^ .$wrv^. Retrieved January 23, 2004 from http://www.iidcr.ca/VOL02 01 CAN/articles/ harder.shtml Hernandez, B. (2000). Employer attitudes toward workers with disabilities and their ADA employment rights: A literature review. Journal o f R ehabilitation 66(A), 4-17. Hester, E. J., & Decelles, P. G. (1990). The effect o f em ployer size on disability benefits and cost-containm ent practices. Topeka, KS: Menninger Foundation. Hirsch, B., MacPherson, D. A., & Dumond, T. M. (1997). Workers' compensation recipiency in union and nonunion workplaces. Industrial & L a b o r R elations Review, 50(2), 213236. Howe, B., & Johnson, D. (2000). R estraining E quality: Human R ights Com m issions in Canada. Toronto: University of Toronto Press. Human Resources Development of Canada (2002). Egwify Act. Retrieved February 20,2004 from http://info.load-otea.hrdc-drhc.gc.ca/workplace_equity/ review/response/backgrounder, shtml Infante-Rivard, C., & Lortie, M. (1996). Prognostic factors for return to work after a first compensated episode of back pain. DccwparioMoZ and EnvimnmcntaZ Mcd/cmc, J 3 ,488-494. Isemhagen, D. (2000). A model system: Integrated work injury prevention and disability management. 73, 87-94. Disability Management Strategies 73 Jarvikoski, A., & Tuunainen, K. (1978). The need for early rehabilitation among Finnish municipal employees. S'coWinavia/i TowmoZ MigcZZcing., 70(3), 115-120. Jarvikoski, A., & Lahelma, E. (1980). Early rgAohiZimnon or tAg workpZocg. (Monograph No. 6). New York: World Rehabilitation Fund. Jarvikoski, A., Lahelma, E. (1981). Early rehabihtation and its implementation at the work place. /ntgmarmnaZ JowmoZ q/^EgAaAzZZtarZon Eg.ygarcZz, 4(4), 519-530. Johnson, W. G., & Ondrich, J. (1990). The duration of post-injury absences from work. The R eview o f Econom ics and Statistics, 72, 578-586. Johnson, W.G., Butler, R. J., Baldwin, M. (1995). First spells of work absences among Ontario workers. In T. Thomason & R. Chaykowski, eds.. R esearch in Canadian W orkers’ com pensation. Kingston, Ontario: Industrial Relations Centre Press, Queens University. Karjalainen, K., Malmivaara, A., van Tulder, M., Roine, R., Jauhiainen, M., Hurri, H., & Koes, B. (2001). Multidisciplinary biopsychosocial rehabilitation for subacute low back pain in working-age adults. Spine, 26(3), 262-269. Kenny, D. T. (1995). Common themes, different perspectives: A systemic analysis of employer-employee experiences of occupational rehabilitation. Vocational RgZioAZZZatZo», 39(1), 54-78. Krause, N., Dasinger, L. K. & Neuhauser, F. (1998). Modified work and return to work: A review of the literature. JoumoZ q/'OccwpoZZonaZ EgZioAZZZmrZon, 8(2), 113-139. Lacerte, M., & Wright, G. R. (1992). Return to work determination. FAy.yZcaZ AfgdZcZng and EgAoAZZZrafZo». 8mrg q/^rAg Art EgvZgwa^, 6(2), 283-302. Loisel, P., Durand, M., Bertelette, D., Vézina, N., Baril, R., Gagnon, D., et al. (2001). Disability prevention: New paradigm for the management of occupational back pain. Disease Management Health Outcomes, 9(7), 351-360. Disability Management Strategies 74 Loisel, P., Gosselin, L., Durand, P., Lemaire, J., Poitras, S., & Abenhaim, L. (2001). Implementation of a participatory ergonomics program in the rehabilitation of workers suffering from subacute back pain. Applied Ergonomie, 32, 53-60. McDonald, T. & Harder, H. (2003). Older wor^era^ and d»ahd;ry management. Retrieved July 20, 2004 from http://www.iidcr.ca/VOL02 01 CAN/articles/ harder.shtml McMahon, B., Matkin, R. E., Growick, B., Mahaffey, D., & Gianforte, G. (1983). Recent Trends in private sector rehabilitation. Rehabilitation Counseling Bulletin, 27, 32-47. Mitchell, K. (2002). Best practices for creating an effective retum-to-work program. C om pensation & Benefits M anagement, 18(3), 34-37. Mobley, E. M., Linz, D. H., Shukla, R., Breslin, R. E. & Deng, C. (2000). Disability Case Management: An impact assessment in an automotive manufacturing organization. Journal o f O ccupational and Environm ental M edicine, 42(6), 597-602. National Institute of Disability Management & Research. (2003). Disability Management in the Workplace: A guide to establishing a joint workplace program (2“*^ed.). Port Albemi, BC: Author. Nickel, C., & Yangouyian, S. (1996), Disabled doesn’t mean deficient. O ccupational H ealth & 3q^ry, 63(4), 52-69. O’Brien, E. (2003). Employers’ benefits from workers’ health insurance. TTze Md^anA: Gwa/Terly, 87(1), 5-43. Pati, G. C. (1985). Economics of rehabilitation in the workplace, dawrmd Rg/iahddadan, 37(4), 22-30. Petty, D. M., & Fussell, E.M. (1997). Employer attitudes and satisfaction with supported employment. on Awdjm & OtAgr Deve/opmentaZ DMahd/dg.y, 72(1), 15-23. Disability Management Strategies 75 F*ransky, G., Shaw, W., & McLellan, R. (2001). Employer attitudes, training, and retum-towork outcomes: A pilot study. TgcAnoZogy, 7J(2), 131-138. Renaud, S. (2002). Rethinking the union membership/job satisfaction relationship: Some empirical evidence in Canada. /ntgmnhoMaZ JowmaZ 23(2),137-150. Salkever, D. S., Shinogle, J., & Pumshotaman, M. (2000). Employers' disability management activities: Descriptors and an exploratory test of the Enancial incentives hypothesis. JowmaZ q/^DccwporZonaZ RgZia^ZZZfatZon. 70(3), 199-214. Salkever, D. S., Shinogle, J. A., & Purushothaman, M. (2001). Employer disability management strategies and other predictors of injury claims rate and costs. Journal o f Safety Research, 32(2), 157-185. Schein, E. H. (1992). O rganizational Culture an d Leadership (2f^ ed.). San Francisco: Jossey-Bass. Schultz, I. Z., Crook, J. M., Berkowitz, J., Meloche, G. R., Milner, R., Zuberbier, O. A., & Meloche, W. (2002). Biopsychosocial multivariate predictive model of occupational low back disability. Spine, 27(23), 2720-2725. Schwartz, G. (1984). Disability costs; The impending crisis. B usiness an d Health, 1(6), 25-28. Schwartz, G. Ë., Watson, S. D., Galvin, D. E., & Lipoff, S. (1989). 77i€ (ZMO^iZZry managgmg/rt Washington, DC, Washington Business Group on Health/Institute for Rehabilitation and Disability Management. Scott, L., & Curtis, J. (2003). Making the connection: Looking to improve the health of your employees and the organization's bottom Une? Consider how disability management fits into the company's strategic plan. R gn^ tf Canada, 27(4), 7581. Shaw, W. S., Feuerstein, M., Miller, V. I., Lincoln, A. E., Berger, R. H., & Wood, P. M. (2000). Ergonomics and workplace accommodation to improve outcomes in a large workers' compensation system. Rehabilitation Engineering and Assistive Disability Management Strategies 76 Shaw, L., & Polat^ko, H. (2002). An application of the occupational competence model to organizing factors associated with return to work. TTig Conndmn VbwmaZ o/" Occwpotio/wZ TTigmpy, 69(3), 158-195. Shaw, W. S., Robertson, M. M., Pransky, G., & McLellan, R. K. (2003). Employee perspectives on the role of supervisors to prevent workplace disability after injuries. 73(3), 129-142. Shoemaker, R. J. (1989). Co/porate to garZy rgtwrn to wort policy. Unpublished doctoral dissertation, Western Michigan University, Michigan. Shoemaker, R. J., Robin, S. S., Robin, H. S. (1992, July/August/September). Reaction to disability through organization policy: Early return to work policy. Journal o f Rehabilitation, 18-24. Shrey, D. E. (1996). Disability management in industry: The new paradigm in injured worker rehabilitation. D isability an d Rehabilitation, 18(8), 408-414. Shrey, D. E. (1998). Effective worksite based disability management programs. In P. King (Ed.), Sourcebook o f occupational rehabilitation (pp. 389-409). New York: Plenum Press. Shrey, D. E., & Bangs, S.A. (1991). Returning social security beneficiaries to the work force: A proactive disability management model. R ehabilitation Counseling Bulletin, 34(3), 257-274. Shrey, D. E., & Olshesky, J. A. (1992). Disability management and industry-based work return transition programs. PAy.yZcaZ MedicZng omJ 6(2), 303-314. Silversides, A. (1998). Disability management efforts can reduce number of iryuries, improve bottom line. CanodZan MedZcaZ JoumaZ, 159(3), 268-270. Stock, S., Deguire, S., Baril, R., & Durand, M. (1999). DWacZg.y o W^ z c ZZZZo ZZn g rgmm to wZzZi /MWJCwZo.y^eZemZ dZ.yordgr.s. report on tZie gwebec gwaZZtatZve 3tiaZy Zn tZie EZectrZc anzZ EZectronZc Sector q/'W ortrea^ PZia.ye 7. Quebec: Institut de recherché en santé et en sécurité du travail du Québec. Disability Management Strategies 77 Steeves, L. & Smithies, R. (1998) Disability management in Canada: Rights and responsibilities. EmpZoygg /oMnwl. 23(1), 37-39. Tabachnick, BG., & FideU, LS. (2001) MA: Allyn & Bacon. mwZrivanarg (4^ ed). Boston, Tate, D. G., Habeck, R. V., & Galvin, D. E. (1986). Disability management: Origins, concepts, and principles of practice. Journal o f a p p lied rehabilitation counseling, 77(5), 5-17. Tate, D. G., Habeck, R. V., & Schwartz, G. (1986). Disability management: a comprehensive framework for prevention and rehabilitation in the workplace. R ehabilitation Literature, 47(9-10), 230-235. Thompson, D. (2001). In support of STD: with STD and LTD claims soaring, it is time to rethink disability management. Benefits C anada, 25(11), 23-27. Traver, N. (1990). Opening doors for the disabled. Time, 135(23), 54. Unger, D. (1999) Workplace Supports: A view from employers who have hired supported employees. Focus on Autism an d O ther D evelopm en tal D isabilities, 14(3), 167-176. Votel, T. (1993). Correct PPE helps alleviate persistent back injuries, compensation expenses. Hgolfh & 62(4), 65-69. Westmorland, M. & Buys, N. (2002). Disability management in a sample of Australian selfinsured companies. Dimhihty and Rg/mhiZimtmn, 24(14), 746-754. Westmorland, M. G., & Wilhams, R. (2002). Employers and policy makers can make a difference to the employment of persons with disabilities. Di.yahihiy and ReAa6ddadan, 24(15), 802-809. William, J. R. (1997). Methods toward a science of behaviour and experience (5^ ed). PaciEc Grove, CA: Brooks/Cole Publishing. Disability Management Strategies 78 Workers' Compensation Board of British Columbia (2004). Annual Statistics. Retrieved August 23,2004 from http://www.worksafebc.com/facts_and_figures/default.asp World Health Organization (1981). prevgn/ion omf rgAahi/iration.- Report r/ie WHO expert comm/ttee on prevention am/ re/io6i/itation. World Health Organization Technical Report Series 668. World Health Organization, Geneva. Wright, T.A. & Bonett, D.G. (2002). The moderating effects of employee tenure on the relation between organizational commitment and job performance: A meta-analysis. Journal o f A p p lied Psychology, 87(6), 1183-1190. Yelin, E., Henke, C., & Epstein, W. (1986). W ork disability among persons with musculoskeletal conditions. A rthritis and Rheumatism, 29(11), 1322-33. Zabrodski, R. (1998). Does facilitating an early return to work improve worker outcome after an occupational injury? A m erican A cadem y o f D isability E valuating Physicians, 7(3), 29-44. Disability Management Strategies 79 Appendix A Letter of Introduction Disability Management Strategies 80 C om pany Official C om pany Name S treet address Tow n Postal Code Dear Name, W ith the enclosed questionnaire we are asking fo r your participation. W e are gathering inform ation about the im pact of disability m anagem ent strategies in im proving the em ploym ent retention rate am ong w orkers who have incurred a w ork injury. W e have random ly selected com panies covered by W C B insurance in the Prince G eorge region and are contacting organizations like yours to furthe r understand the nature o f com pany policy with regard to disability m anagem ent and the injured worker. W hether or not your organization provides disability m anagem ent strategies, we w ould greatly appreciate your feedback. W hile participation in the survey is voluntary, there is no substitute fo r the know ledge and insights tha t you can provide. Y ou r participation in com pleting the attached questionnaire is valued and appreciated as understanding the im pact of com pany policy will be of benefit to com panies and injured workers. W e have provided an inform ation sheet covering all the details of the study. Should you have any questions about the questionnaire or research project, please feel free to contact us. W e will also m ake telephone contact with you over the com ing w eeks to discuss the project. For your convenience we have a dedicated phone line ( answ er your queries and fax line i for the return of your com pleted questionnaire. W e look forw ard to speaking with you directly. W e tha n k you in advance for your participation, your valuable time, and assistance. Y ours sincerely. Disability Management Strategies 81 Appendix B Sample Questions from Questionnaire Disability Management Strategies 82 Disability M anagement Policies Please read each of the following statem ents and indicate (by circling) the degree to which you agree. Disagree Som ew hat Disagree 2 Neutral 3 Som ew hat A gree 4 1. W e regularly educate and inform all em ployees about our early return-to-w ork options. 2. W e have specific clauses in the em ploym ent contract and/or bargaining agreem ent requiring disability managem ent. 3. W e have specific clauses in the em ploym ent contract and/or bargaining agreem ent requiring early return-to-w ork. 4. A ny of our workers can utilize our disability m anagem ent program m es to assist with their return-to-work. Agree 5 Perceptions of Disability Management Please read each of the following statem ents and indicate (by circling) the degree to which you agree. Disagree 1 Som ew hat Disagree 2 Neutral 3 S om ew hat A gree 4 1. An em ployee returning to duty with w orkplace restrictions lowers m orale within that w ork area. 2. An employee returning to work with workplace restrictions hinders the recovery process of that worker. 3. A ccom m odating w orkers with w orkplace restrictions is costly to the employer. 4. A ccom m odating w orkers with w orkplace restrictions is a waste of em ployer resources. A gree 5 Disability Management Strategies 83 Demographic Information Regarding Injured Worker 1. A g e:____________________________________ 2. Gender: M ____________ F ______________ 3. Year of Hire: 4. Union Member: Y e s_______ No 5. Education required for his/her job: a. No specific education required b. Company training c. Trade School d. College e. University Injured Worker Return-to-Work Information Please read each of the following statem ents and indicate (by circling) the degree to which you agree. Disagree Som ew hat Disagree 2 Neutral 3 S om ew hat Agree 4 Agree 5 1. T his w orker returned fully to his/her regular duties. 1 2 3 4 5 2. This w orker’s earnings were subsequently reduced upon returning-to-work (e.g., he/she was accom m odate in an lower-paying alternate position; went part-tim e etc). 1 2 3 4 5 1 2 3 4 5 3. This worker had subsequent relapses of sim ilar condition fo r sim ilar durations. Disability Management Strategies 84 Appendix C DM Policy Histogram Disability Management Strategies 85 DM Policy Histogram 12- 10- 0 8- § 1 e- 4— 2- M ean = 4 0 .8 5 S td. D ev. = 1 1 .2 6 7 N = 72 10 20 30 40 50 Sum of Policy 60 70 Disability Management Strategies 86 Appendix D DM Perceptions Histogram Disability Management Strategies 87 DM Perception Histogram Mean = 46.53 Std. Dev. = 8.9 N = 72 40 50 Sum of Perception Disability Management Strategies 88 Appendix E Descriptive Data Disability Management Strategies 89 Pull Pay Re­ lapse Company N Claims Pull Duty Manufacture Construction Repair Hospitality Retail Sales/Service Transportation Service Forestry Lumber Trucking Healthcare Education Consulting 5 2 1 2 1 2 2 8 3 1 1 1 3 2 1 1 1 1 1 1 1 1 1 1 1 1 2 2 1 1 Accom Reti 5 3 3 10 1 1 2 2 2 1 3 2 2 2 1 8 3 3 3 3 2 1 1 1 1 2 3 6 1 5 2 11 2 High Risk Forestry/Lumber/Manufacture/Construction Low Risk Consulting/Educ ati on/Retail/S ales/Service 1 Claims 10 2 2 2 2 1 2 % 47.6 38.0 Retain 6 % 60.0 62.5 Disability Management Strategies 90 Full Full Company Demographics Industry Type Manufacture/Constmction/Repair Retail/Hospitality/SaIes,sevice Forestry/Lumber/T rucking Health care/Education Consulting Sample Claims Duty N % N % N % N % 18 18 14 28.6 23.8 23.8 14.3 9.5 5 4 5 23.8 19.0 23.8 4 3 5 3 11 25.0 25.0 19.4 15.3 15.3 19.0 14.3 23.8 14.3 9.5 21 51 29.2 70.8 57.1 42.9 7 9 8 33.3 38.1 8 42.9 38.1 17 20 35 23.6 27.7 48.7 4 4 13 19.0 19.0 62.0 3 3 9 14.3 14.3 42.8 4 4 9 19.0 19.0 42.8 15 18 18 20.8 25.0 25.0 29.2 4 4 9.5 19.0 19.0 52.5 3 4 7 11 6 5 5 3 2 Pay 0 4.8 1 2 Relapse Accom. Retain N N % N % 28.6 23.8 19.0 14.3 9.5 5 5 23.8 23.8 9.5 9.5 8 57.1 38.0 9 5 3 4 13 14.3 19.0 62.0 2 % 9.5 4.8 2 1 0 9.5 4.8 2 1 6 5 4 3 2 2 2 0 Unionization Yes No Hourly-paid Hourly < 40% Hourly < 80% Hourly >81% Company Size Payroll under $200,000 Payroll under $599,000 Payroll under $999,000 Payroll over $1 million 21 12 2 11 1 4.7 14.3 19.4 33.3 9 2 2 4 9 9.5 9.5 19.0 42.8 4 19.0 9.5 2 1 1 4 4.8 4.8 19.0 4.8 9.5 1 2 0 3 12 2 4 4 14.3 10 25.0 30.8 11 9.5 19.4 19.4 47.6 42.9 23.8 4.8 9.5 52.5 1 11 2 4 2 6 9.5 19.0 9.5 28.6 Individual Demographics Age (Mean=43.95 yrs) Less than 40 yrs 40 yrs & greater 8 13 38.1 61.9 7 8 87.5 61.5 6 12 75.0 92.3 4 84.6 87.5 2 2 5 3 62.5 84.6 11 37.5 84.6 76.9 62.5 9 5 69.2 62.5 Gender Male Female 13 8 61.9 38.1 10 3 76.9 37.5 11 7 3 23.1 25.0 10 5 Disability Management Strategies 91 Individual Demographics Tenure (Mean=13 yrs) Less than 13 yrs 13 yrs or greater Education None Company Trade College University Sample N % Claims N % Full Duty N % Full Pay N % Relapse N % Accom, N % Retain N % 9 12 42.8 57.1 6 7 8 9 88.9 75.0 2 3 22.2 25.0 7 8 6 8 66.7 66.7 28.6 33.3 23.8 9.5 4.8 4 4 3 5 5 4 83.3 71.4 80.0 3 50.0 14.3 20.0 4 5 4 3 5 4 2 1 0 0 0 1 71.4 80.0 50.0 1 100 0 1 1 0 0 50.0 71.4 80.0 50.0 100 100 0 1 1 00 6 7 5 2 1 66.7 58.3 6 6 .6 1 57.1 60.0 50.0 1 1 0 0 1 1 77.8 66.7 6 6 .6 1 1 Pay-rate $8-11 $12-15 $16-19 $20-23 over $24 7 7 4 4.8 9.5 33.3 33.3 19.0 Nature of Injury Sprain Strain Fracture Other 7 7 3 4 33.3 33.3 14.3 19.0 Location of Injury Head/neck Lower back Trunk Upper limbs/Shoulders Lower limbs 1 2 1 3 3 8 6 4.8 14.3 14.3 38.1 28.6 1 1 3 5 3 5 3 3 2 1 0 0 50.0 42.8 71.4 75.0 2 5 4 6 4 4 71.4 42.8 6 5 3 3 1 0 0 50.0 0 2 1 2 100 85.7 57.1 0 6 1 0 0 0 2 3 2 7 5 3 85.7 71.4 3 100 1 75.0 1 1 50.0 42.8 14.3 42.8 33.3 25.0 0 0 66.7 66.7 62.5 66.7 1 1 00 66.7 87.5 83.3 1 1 2 1 2 4 6 6 1 2 100 57.1 85.7 50.0 85.7 85.7 33.3 50.0 2 3 5 5 5 5 2 4 5 2 3 1 0 33.3 33.3 25.0 16.7 1 1 66.7 100 62.5 83.3 2 1 5 5 1 0 0 50.0 71.4 71.4 50.0 57.1 71.4 66.7 75.0 100 66.7 33.3 62.5 83.3 Disability Management Strategies 92 Perception Mean SD Frequency 1 2 3 4 5 Q7. 2.94 1.51 Q8. 3.49 1.35 Q9. 3.65 1.28 QIO. 3.58 1.22 Q ll. 4.10 1.02 Q12. 4.06 0.85 10 4 5 6 2 0 0 16 17 6 1 0 8 1 2 2 1 17 25 18 42 9 18 15 2 0 2 0 8 2 2 23 19 22 13 35 Q5. 3.18 1.53 Q6. 2.57 1.56 Q7. 3.00 1.52 Q8. 2.88 1.66 Q9. 3.68 1.42 QIO. 3.78 1.50 Q ll. 3.56 1.54 16 9 14 12 21 28 19 7 19 9 18 25 7 13 12 4 15 4 5 22 26 Qi 3.96 1.24 Q2. 4.08 1.24 Q3. 2.76 1.32 Q4. 3.94 1.33 Q5. 2.26 1.37 Q6. 4.29 3 12 25 14 10 5 9 3 11 13 37 29 18 10 7 36 5 4 10 14 39 Q l. 2.76 1.52 Q2. 2.50 1.52 Q3. 2.25 1.32 Q4. 3.78 1.44 24 7 17 10 14 31 3 22 3 13 33 4 25 4 6 8 14 11 8 1 .1 1 1 Q13. 3.26 1.14 Q14. 3.40 1.11 4 4 51 4 2 8 38 12 15 Q12. 3.07 1.55 Q13. 3.22 1.48 Q14. 3.85 1.30 20 7 9 20 16 15 9 7 3 15 16 31 2 1 23 27 0 Policy Mean SD Frequency 1 2 3 4 5 11 1 14 13 33 10 14 5 15 6 2 1 11 5 6 6 24 26 16 34 11 19 18