A RETURN ON INVESTMENT IN EMPLOYEE ASSISTANCE PROGRAMMES : A THEORETICAL UNDERPINNING

INTRODUCTION Mental health conditions such as stress, depression, substance abuse and relational problems are major contributors to sickness, absenteeism and lower job performance (Nissly & Mennen, 2002). Steward, Ricci, Chee, Hahn and Morganstein (2003) indicate that employees with depressive disorders present with almost four times more healthrelated lost productivity than those without depressive symptoms. This amounts to 5.6 hours per week lost by employees with depressive disorders in comparison with 1.5 hours lost by employees who do not suffer from these disorders, costing the United States economy up to $44 billion per year. Steele (2002) maintains that workdays lost because of substance abuse alone cost the US economy $5 billion dollars a year. In South Africa the Bytes Technology Group conducted a study (Bytes, 2011) indicating that local companies lose between R12 to R20 billion a year to absenteeism. An article in Rapport Loopbane (2 July 2006) also indicated that the cost of absenteeism for the South African economy was estimated at R20 billion for that year. It is unlikely that statistics have changed significantly over the last five years and the country continues to lose revenue through employee absenteeism.


INTRODUCTION
Mental health conditions such as stress, depression, substance abuse and relational problems are major contributors to sickness, absenteeism and lower job performance (Nissly & Mennen, 2002).Steward, Ricci, Chee, Hahn and Morganstein (2003) indicate that employees with depressive disorders present with almost four times more healthrelated lost productivity than those without depressive symptoms.This amounts to 5.6 hours per week lost by employees with depressive disorders in comparison with 1.5 hours lost by employees who do not suffer from these disorders, costing the United States economy up to $44 billion per year.Steele (2002) maintains that workdays lost because of substance abuse alone cost the US economy $5 billion dollars a year.In South Africa the Bytes Technology Group conducted a study (Bytes, 2011) indicating that local companies lose between R12 to R20 billion a year to absenteeism.An article in Rapport Loopbane (2 July 2006) also indicated that the cost of absenteeism for the South African economy was estimated at R20 billion for that year.It is unlikely that statistics have changed significantly over the last five years and the country continues to lose revenue through employee absenteeism.

EMPLOYEE ASSISTANCE PROGRAMMES AND THEIR IMPACT ON PRODUCTIVITY
Absenteeism has many underlying causes and the health condition of the country's working population may be one of these.Employee Assistance Programmes (EAPs) are employer-sponsored programmes designed to alleviate a range of workplace problems using core technologies such as screening, assessment, brief intervention and referral to service providers, case management and follow-up services for mental health and substance-abuse problems (Jacobson & Attridge in Sweet & Casey, 2010).
These programmes operate at the individual employee level using aspects of behavioural science for the assessment and management of certain work-related and non-workrelated problems.EAPs facilitate the assessment of an individual's progress from the time they show problematic personal and work-related behaviour to when they have achieved full functioning in both these areas.Selvik, Stephenson, Plaza and Sugden (2004) conducted a study of the EAP at the US Department of Health and Human Services and recorded a pre-and post-test difference between specific performance measures.Productivity affected by emotional challenges improved by 73%, productivity affected by health problems improved by 66%, relationship challenges improved by 74%, and absenteeism and tardiness problems improved by 62%.It is thus possible to assume that behavioural science interventions on the individual employee level may have positive spinoffs for workplace performance.Attridge, Amaral, Bjornson, Goplerud, Herlihy and McPherson (2010) present an EAP business value model The workplace performance value category refers to the positive changes in employee performance as a result of the EAP.The benefit-cost category refers to the effect of the EAP on the employer-paid medical benefits used in areas of medical and pharmaceutical, mental health, disability and workers' compensation, regarded as direct costs to the employer.The organisational value category refers to savings owing to effective safety risk management, legal liability management and a positive improvement in organisational culture.This study focuses primarily on the workplace performance value category.

THE NEED TO PROVE VALUE THROUGH PROGRAMME EVALUATION
In an online article (December 20 II) the Chicago land Chamber of Commerce indicates that, according to a Business Insurance survey, EAPs in the US grew by 318% after 2002.Whereas in 2002 37.4 million lives had been covered, by 2011 the figure had grown to 156.9 million (http://www.chicagolandchambers.org).These statistics may actually be pointing to the real growth, representing as they do the growth of the top ten EAPs in the US.This does not include the growth associated with the smaller EAPs.
There has been a significant increase in company investment in employee wellbeing resulting from the EAP.Attridge (2010) indicates that over the last 30 years a number of studies had shown positive outcomes for EAPs in different areas, especially in the workplace performance and benefit-cost categories.Masi (1994) cited some of these studies in an article and refers to the Kennecott Copper Corporation, which estimated a 6:1 benefit-cost ratio resulting from employees' use of its EAP.Equitable Life Assurance found a $3 return for every $1 invested in its EAPs, when alcoholic employee absenteeism dropped from 8% to 4% after referral to, and treatment in, alcohol programmes.3M Company data suggested that 80% of the employees who used the EAP showed improved attendance, greater productivity, and enhanced family and community relations.The McDonnell Douglas study, often quoted as an example of a cost-effectiveness study, determined that the EAP treatment of alcoholism was more effective than others.Csiernik (2004) reflected on studies conducted during the 1990s showing different forms of benefit.In 1994 Bruhnsen, a unit in the Michigan Medical Centre, found that EAP clients had a higher employee retention rate but no difference in sick days from non-EAP users.They still recorded a cost savings of $65.341, reflecting significance in cost implications when positive change takes place in singular performance indicators.In 1998 the Chevron Corporation in the US found a significant return of 14: I on an investment, while fewer substance-abusing employees were dismissed and both supervisor-referred (mean 50%) and self-referred (mean 5%) clients were observed.Safety rates remained identical.A study (1996) of the Southern California Edison EAP showed that users had higher initial addiction treatment costs, but subsequently lower mental and physical health care costs.EAP users had 38% fewer medical costs over a 30-month post-treatment period.In 1999 the United States Postal Services reflected a benefit to cost ratio of 1.27: 1 for the first year after EAP use.By the fifth year, the ratio was 7.21: 1. Cost savings thus increased progressively after EAP use.
Despite evidence generated in such studies, Attridge (20 I0) maintains that the EAP industry persists in its perception that there is insufficient proof of its value.Seidi (20 I0) indicates that the necessity of evaluating EAPs is increasing, with purchasers insisting that providers prove the value of their product.Employee Assistance providers are thus under increasing pressure to ensure that their programmes have sound evaluation components able to measure the impact of their services.Seidi regards the development of specific metrics to measure the effectiveness of workplace counselling as being much more than just a sales pitch.He sees this meticulous business-focused analysis as the primary means by which the industry can demonstrate that an EAP is a professional, systematically-applied intervention that generates a return beyond its cost.The development of these analytical capabilities will save the industry from what he refers to as commoditisation.Seidi (2010) raised a concern about two forms of commoditisation.
The first refers to the risk of being treated as a commodity, assessed according to its price rather than the value it adds to the organizational bottom-line.The second refers to EAPs keeping programme costs down, with low utilisation figures that undermine their real impact.These forms of commoditisation are driven by both the purchaser of an EAP and its supplier, which works against the true value of the industry.The trading characteristics of the industry are important only up to the point where it generates sustainable revenue.The researchers concur with Seidi's view that what distinguishes this intervention strategy from others is the particular behavioural science knowledge that the EA specialists bring to the workplace.The development and refinement of its analytical capabilities is what will promote the development and refinement of the industry as a discipline in its own right.Evaluation studies are one such strategy.Like Seidi, Csiernik (2004) argued that an EAP should have in place a mechanism whereby the appropriateness, effectiveness and efficiency of service delivery could be evaluated.This study demonstrates that measurement is possible, but could nevertheless be greatly enhanced if the ability to evaluate were built into programme design.Sharar and Lennox (2009) maintain that measuring tools for detecting change over time and directly connected with the expected, agreed outcomes should be built into programme design.These tools should coordinate scientific methodology with EAP practice by focusing on personal issues that directly affect workplace productivity.They should be automated to Social WorklMaatskaplike Werk 2013:49(2) simplify and expedite survey methods for employees and all parties interested in the outcomes.

METHODOLOGY USED IN THIS STUDY
The researcher conducted a summative evaluation of two EAPs.Return on investment studies generally falls into the category of summative evaluation.Company A is a motor-vehicle production company and Company B a major financial institution.Both companies were using the same EAP service provider at the time of the evaluation.The EAP at Company A has been operating since 1995, that at Company B since 2000.
Company A employs a significant number of blue-collar workers, while company B employs a significantly higher number of white-collar workers.The researcher used a quantitative research strategy with a smaller qualitative component and used data triangulation in order to explore the value of both quantitative and qualitative data in programme evaluation (De Vos, Strydom, Fouche & Delport, 2011).The study focuses on EAP cases of employees who were referred to the programme via only the formal referral process, meaning that their line managers referred them because of identified performance challenges.Systematic sampling was followed, with the selection of every second person referred through the formal referral process for the demarcated period.
The sample was comprised of29 employees from Company A and 12 from Company B, all of whom were available during this specific referral period.Company B made less use of the formal referral procedure and five of the employees identified were not available for participation at the time.Twenty-four (24) line managers, 12 from each company, who were responsible for the referral of the employees, participated in the qualitative component of the study, engaging in semi-structured interviews.The human resource data for 20 of the respondents from Company A were available for the statistical data-analysis component of the study.Unfortunately, this step had to be omitted for the respondents from Company B, because of their low margin of permission.All the respondents and participants in the study gave their consent.Consent for access to HR data was obtained in addition to their initial participation in the programme.On-site counsellors at Company A and line managers from both companies played a supportive role in the distribution and completion of the questionnaires Respondents (employees) were consequently able to interact with people they knew well, rather than feeling exposed to a researcher with whom they were relatively unfamiliar.
A survey was conducted by means of questionnaires completed by the employee sample, which formed the quantitative part of the study.The survey focused on the following personal and performance indicators: personal-relational impact, work performance before and after participation, attendance, presentee ism, consideration of leaving the job, counselling and disciplinary action.The statistical data analysis focused on absenteeism and disciplinary action one year before programme participation and one year subsequent to that.The semi-structured interviews conducted with referral agents (line managers) as a qualitative data collection instrument focused on the support given to line managers for using the programme effectively.This covered their understanding of the referral process, their perception of the impact of employees' personal problems on productivity, the financial implications of productivity problems, the impact of employees' personal problems on teams, their perception of the programme's ability to effect change and their views on the consistency of this change.

RESULTS
The goal of the research was to evaluate EAPs in the South African context by exploring the return on investment in the programme by two corporate clients of a specific EAP service provider in the region.There was a specific focus on the counselling component of the programme involving the targeted employees using the programme via the formal referral process.
The objectives of the study were: • To investigate the impact of the EAP on organisations focusing on performance indicators such as absenteeism, on-the-job absenteeism, productivity and disciplinary action; • Using different data-collection tools, to provide guidelines for programme evaluation including ROI tools for determining the value of EAPs as well as identifying growth areas; • To investigate the consistency of data from different measuring instruments: selfreporting, qualitative impressions from referral agents and statistical data analysis.De Vos et al. (2011) warn that the results of all cost-benefit analyses should be considered carefully, as its value is time-bound and influenced by the estimates and assumptions on which it is based.This description of how these results should be viewed could to some extent provide an answer to the concern in Attridge (20 10) that, despite so many completed studies reflecting a positive return on investment, the value of the programme remains in question.At the same time, De Vos et al. (2011) maintain that these evaluation studies are necessary, as they create a foundation for policy decisions.They could therefore influence the direction ofEAP programme operations.
As far as this study is concerned, the authors consider that the results should be seen both for their possible ROI value, and because there are lessons to be learned about the important elements needed for programme evaluation studies.Aspects for consideration in this respect are recognition of the uniqueness of data sources and exploration of the value of the data they generate.The researchers strongly concur with Seidi's views on the need to build the discipline rather than commoditise the industry through consistent evaluation studies.
The data obtained in this study reflect a measure of change from one level of functioning to another.Change in relationships (personal and work-related) in following both the quantitative and qualitative approaches and change in identified performance indicators has been reported.
The following data set reflects the results from the self-reported questionnaires, a subsection of the quantitative component of the research.

='"
' " ::;: ~::I: ' "g.The self-reporting quantitative results show that personal problems had a negative impact on employee attendance, resulted in disciplinary and staff counselling processes for many of the respondents, and impacted significantly on employee work performance.The majority of respondents from both companies said their work performance was negatively affected by their personal problems.Opinions differed as to whether their performance necessarily included absenteeism patterns.Company A reported a less significant impact on attendance than did Company B. Respondents from Company B were generally of the opinion that their personal problems did not have an adverse effect on their relationships with their partners.However, they did report a general improvement in personal relationships after participating in the programme.One of the weaknesses of a study of this nature, one that uses employees who participated in the Social Work/Maatskaplike Werk 20l3:49(2) programme as an essential population, is that those who had left the company could not be included in the measurement.The self-reporting nature of this information is strengthened by the general consistency of viewpoints among respondents from the same company, qualitative feedback from line managers and data generated by the two different companies.
The statistical data for Company A show significant absenteeism before the EAP and a considerable decrease in the period following participation.The return on investment for this indicator was a saving of R14.60 for each Rl spent.One participant showed an increase in absenteeism, and another showed the same number of days absent.Nevertheless, from the previous 291 days for the 12-month period before the EAP was held, down to 116 days for the 12 months following the programme, there was an overall decrease of 174 days.During the 12-month period before participation in the EAP 14 of the 20 employees whose human resource data were available for analysis had disciplinary action taken against them.In the period following the EAP, this number was reduced to five.The ROI on this indicator showed a saving of R8.30 for each Rl spent.The absence of corresponding data from Company B strengthens the argument that data management significantly influences or even determines the feasibility of programme evaluation.If an organisation maintains its HR data on employee performance indicators, statistical data analysis can serve as a stable data source for ROI purposes.
The overall ROI calculation for the programme included the total annual amount paid by each company for their EAP, international benchmarks for productivity loss owing to troubled employees, the average income and the total number of employees using the programme during a specific period.The companies' overall reported improvement in performance for all programme users is also reflected in the data below.The ROI calculation below thus includes both the respondents in the study and the total number of employees who used the service during the period.Thus there are specifics for the participants and the overall performance of the programme.
Company A spends Rl m annually on their entire EAP programme.
• The company's success indicators for EAP participation indicated a 78% success rate.
• 65% of the respondents in this research who used the programme over that period indicated that their performance was affected by their personal problems.
• The average income is R6 000 per month, which amounts to R72 000 annually.
• International benchmarks indicate that organisations lose 28-35% of a troubled employee's income to lower productivity.
• Savings to company.
Company B spend R2 970 000 annually on their entire EAP programme.
• The company success indicators for EAP participation showed a 78% success rate.
• 72% of the respondents in this research who used the programme over that period indicated that their performance was affected by their personal problems.
• International benchmarks indicate that organisations lose 28-35% of a troubled employee's income to lower productivity.
The qualitative component was included in this study to obtain a more descriptive understanding of the impact of performance challenges over and above the numerical calculations.The participants selected for the qualitative component of the study were the line managers/supervisors responsible for referring the respondents to the EAP.As a result, their views on changes in attitude and work performance added objectivity to the views presented in the self-reporting component.
The qualitative results show consistency across the two companies in the line managers' assessment of the impact of employees' personal problems on work performance, team work and supervisory roles.They also highlighted the financial and reputational risk factors that may be the result of challenges to employee work performance.The line managers who participated in this study held generally positive views on the impact of the programme, and unsuccessful cases were regarded as individual instances of non-positive responses rather than as programme failures.The line managers also expressed the need for training that would empower them to use the programme more effectively, with a more comprehensive understanding of mental health issues.The quotations below reflect some of these views.

THE NEED FOR MORE COMPREHENSIVE AND CONSISTENT TRAINING FOR LINE MANAGERS
Company A: Participant 1: "We have been to a couple of in-house training on how to check for guys for attendance, look at absence on Mondays/Fridays, coming under the influence, signs to look outfor." consisting of three categories by means of which the programme's worth can be evaluated and expressed.These categories are like building blocks, with one component building on the other.The figure below illustrates this model.IOrganizational ~aluecategoryI ~-------.. •---•--u ~ace performance valuecategory Attridge et at.(2010): EAP Business Value Model FIGURE 2 REPORTED IMPROVEMENT AFTER EAP ATTENDANCE I 90% I 80% 70% 60% ' 50% 40% 30% 20%'" 10% 0%