Transcript Examining Outcomes of EAP and Work/Life Product Integration

An Examination of Clinical
Outcomes from EAP and
Work/Life Product
Integration
Melissa Back Tamburo, PhD, LCSW-C
Chesapeake Chapter, EAPA
September 1, 2011
EAP Background
“Worksite based programs
designed to assist: a) work
organizations in addressing
productivity issues, and b)
employee clients in
identifying and resolving
personal concerns including,
but not limited to, health,
marital, family, financial,
alcohol, drug, legal,
emotional, stress or other
personal issues that may
affect job performance”
(EAPA, 2011)
Work/Life Background
W/L programs are “actions taken by
employers and employees to help
the workforce effectively handle the
growing pressure and
responsibilities of both work and
personal lives, to live and work up
to their full potential, and to achieve
both life balance and increased
productivity” (Boston College
Center on Work and Family, 1999)
Integration Background

Customers of EAP and W/L services are driving
program integration

Cost advantages to integration (Peck, 2001; Roberts, 2000;
Stein, 2002; Swihart & Thompson, 2002; Turner & Davis, 2000;
Willaman, 2001)

“One Stop Shopping” for employees reduces confusion
about where to go for help (Stein, 2002; Turner & Davis, 2000;
Willaman, 2001)

Professional associations (EAPA, EASNA &
AWLP) have studied integration in three part
study (Herlihy, Attridge & McCormick, 2003; Herlihy, Attridge & Turner,
2002)
Purpose of Study


To introduce clinical outcomes in the discussion
of impact of integration of EAP and Work/Life
(W/L) services
To examine whether there are differences
between program models in treatment
effectiveness
Political Economy Theory

Examination of
interrelation between
organization and an
economy system (Wamsley
& Zald, 1973)

Model to examine
various pressures
(market fluctuations,
labor shortages,
competition) influence
on decision making
Analysis of Literature

Policy

Case Studies

Surveys

Trade Literature
Design & Data Sources

Secondary data analysis using information from
EAP Case Closing Forms from a large
Behavioral Health Company’s (BHC) regional
office between April, 2002 and June, 2003
(N=5,792)

Quasi-experimental 2 group design

Proxy pretest design (Trochim, 2000)
Stand Alone Model
W/L Client
EAP Client
W/L
Intake
EAP
Intake
W/L
Program
EAP
Program
W/L
Data
EAP
Data
Partnership Model
W/L Case
W/L Client
EAP
Intake
EAP Client
W/L Case
with Clinical
Features
EAP Case
W/L
Data
EAP
Data
Sample
N=5,792

High tech company (N=3,976)


Stand Alone EAP
Communication conglomeration (N=1816)

Partnership Model EAP & W/L
Measures
Independent Variables
Independent Variable: Program Model


Stand Alone: up to 8 sessions of face-to-face EAP
counseling, CISM, member and management
training
Partnership: mean number of sessions of EAP
counseling (between 2 operating units), CISM,
member and management training; partnership with
W/L vendor (dependent care issues, academic
concerns, life management consultation and
materials and financial consultation)
Measures
Dependent Variables

Level of functioning scale (LOF)

Single-item questions measuring


5-point Likert type scale with categories of







Overall functioning
Excellent
Above average
Good
Below average
Poor
Proxy pretest measure
Post treatment measure
Measures
Dependent Variables


Global Assessment of Functioning Scale (GAF)
One of the most widely used measures of impairment
and functioning in clinical and research settings (Basco,
Krebaou & Rush, 1997)



Single scoring scale for evaluating psychological, social
& occupational functioning with rating on 0 to 100
point scale
Excellent inter-rater reliabilities (Startup, Jackson & Bendix, 2002)
Excellent reliability (ICC >.74) (Hilsenroth et. al, 2000)
Measures
Covariates

Pre-test scores

Number of visits
Data Analysis

Examine characteristic differences in sample

Examine differences in proportions for penetration,
traditional W/L presenting problems

ANCOVA for Overall LOF & GAF (separate)



DV = posttest score
IV = program model
CVs = pretest score & number of visits
Data Analysis
Specific Presenting Problems

Select cases with problem





Alcohol/drug
Mental health
Relationship
Traditional W/L
ANCOVA for Overall LOF & GAF (separate)



DV = posttest score
IV = program model
CVs = pretest score & number of visits
Results!
Differences in Sample
Race
Race/Ethnicity
Caucasian
African American
Hispanic
Stand Alone
N
%
3308
84.6
286
7.3
141
3.6
Partnership
N
%
1232
69.3
366
20.6
120
6.8
Asian/Pacific
Islander
Native
American/Alaskan
88
2.3
28
1.6
4
.1
6
.3
Other
Missing Data
82
67
2.1
1.7
25
39
1.4
2.1
Differences in Sample
Referral Type
Test of proportions
statistically significant
(z=420.00, p<.01)
with Partnership
model having larger
proportion of
Mandatory referrals
Model
N
%
Stand
Alone
13/
3,976
.3%
Partnership
43/
1,816
2.4%
Research Question 1
Do utilization rates increase with
program integration?
Test of proportions
Model
N
%
statistically significant
(z=24.04, p<.01), with Stand Alone 3,976/214,757 1.85%
the Partnership
model having larger
proportion of
employees using the
program
Partnership 1816/52,447 3.46%
Research Question 2

Is there a significant
difference between
program models for
the number of cases
presenting with
traditional W/L
issues?

YES, but opposite of
hypothesis

Partnership Model = 12%
Stand Alone Model = 10%


Test of proportions significant (z=2.177,
p<.05)
Research Question 3
Do scores for Overall Level of Functioning differ
between program models?
Stand Alone Overall LOF
scores slightly higher
than Partnership scores
after adjusting for
pretest scores &
number of visits
(F=90.414, p<.01)
Model
Adj.
Mean
Stand
Alone
2.83 1.13
Partnership
2.55
SD
1.26
Research Question 4
Do Global Assessment of Functioning scores
differ between program models?
2550 cases with complete
GSF pre & post scores
Model
Adj. SD
Mean
 NO significant
difference between
program models on
Stand Alone 69.36 11.15
GAF scores after
adjusting for pretest
scores & number of
Partnership 68.62 10.53
visits
F=3.397, p=.065

Research Questions
Are there significant differences in outcomes
between program models for those clients
presenting with:
5. alcohol/drug involvement?
6. relationship issues?
7. mental health issues?
8. traditional W/L issues?
RQ5-8: Specific Presenting Problems
See Handout
Presenting Problem
LOF Scale
GAF
Alcohol/Drug
Not significant
Not significant
Relationship
Significant
SA slightly higher
scores
Significant
SA slightly higher
scores
Not significant
Significant
SA slightly higher
scores
Not significant
Mental Health
Traditional W/L
Not significant
Discussion


Statistical versus Practical differences
Non-significant findings
GAF
 Alcohol/drug presenting problem


Sample size and results
Increased size, smaller effects found significant
 Increased power, decreased false retention of null
(Type II error)

Discussion

Sample differences
Race
 Mandatory referrals
 Utilization


Call flow

Role of intake assessment
Discussion

Increased marketing efforts of Partnership
model

EAP role as gatekeeper

“turf issues”
Implications for Theory

Need organizational analysis to fully apply
political economy theory
Proprietary/confidential information
 Role of union in support of research


ROI

Cycle of innovation
Implications for Practice

Supports literature that integration increases
accessibility and visibility of EAP

Supports proposal that EAP maintain lead role
due to training and clinical credentials of staff
(King, 2002)
Implications for Policy

Continued struggle for identity for EAP


Domain expertise in clinical assessment
Findings mixed on specific focus for EAP &
W/L



Clinical differences not practically significant in all areas except
alcohol/drug
More Traditional Work/Life cases in Partnership model
EAP clinical skills create distinction between
other potential collaborators
Future Research







Data with demographic information
Include W/L data
Measures
Organizational analysis
Include more levels of integration & integration
partners
Explore relationships between race and access
Explore assessment process
Limitations

Use of administrative data

Lack of organizational information

Outcome measures limitations
Strengths

Fills gap in literature, adding clinical outcomes
to impact of EAP and W/L integration

Data from leading provider of EAP services in
the country

Technical support and access to industry experts
from BHC
Conclusions

Partnership Model had higher penetration rates

Outcomes did not substantially improve with
access to increased resources; the SA model had
slightly better outcomes than Partnership model

EAP retain clinical lead role

Vast research trajectory
6 years later…..

Data integration & benefits to programming

Apples & Oranges, make fruit salad!

Wellness…new player, new door to enter

Domain expertise holds true
Resources

Boston College Center for Work & Family


http://www.bc.edu/centers/cwf.html
Global Assessment of Functioning Scale

http://en.wikipedia.org/wiki/Global_Assessment_o
f_Functioning
Question & Answer