Predictors of Successful Return to Work In a Cohort of Workers Undergoing Carpal Tunnel Release Surgery

1
Predictors of Successful Return to Work In a
Cohort of Workers Undergoing Carpal Tunnel
Release Surgery
Benjamin C. Amick, Ph.D.1 Rochelle H.
Habeck2 Janet Ossmann, Ph.D. 1 Holly Fossel
3 Jeffrey N. Katz, M.D., M.S. 3
1. University of Texas School of Public
Health 2. W.E. Upjohn Research Institute 3.
Brigham Womens Hospital
Paper Presented at the Workers Compensation
Research Group Meeting October 25-26, 2002,
Cambridge, MA
2
Research Collaborators

• Maine Medical Assessment Foundation
• Robert Keller, Ellen Schneiter
• Maine Health Information Center
• Cynthia Barrata, Alice Chapin, Nancy Mooney
• W.E. Upjohn Institute
• H. Allan Hunt
• Institute for Work and Health
• Harry Shannon

3
Labor Markets and Health Framework
Global Economy
Society
Social Context Labor market context
Social Actors Employers Labor Market
Intermediaries Communities Community- Based
Organization Trade Unions Business
Associations Government
Social Hierarchy
Illness Injury Producing Mechanisms (Labor
Market Experiences)
Health Status
Amick and Lavis, 2000
4
Background

• The fact of return to work may not be the most
  appropriate outcome
• Research focuses on a small conceptual set of
  predictors
• Multidimensional RTW research is confounded by
  tremendous medical treatment variability
• Most research does not consider timing of
  predictors
• No research has considered the role of the
  organization per se

5
Heuristic Model
Economic, Legal Environment
Demographic, Psychosocial Conditions And
Clinical Status
EARLY
LATE
Family Environment
Job and Organizational Conditions
6
We Hypothesize

• Clinical, worker and economic/legal factors will
  predict successful return to work at 2 months
  following carpal tunnel surgery
• The employment situation (job and organizational
  conditions) will be important at 6 and 12 months
  post surgery

7
Defining the Sample

• Sample 197 workers at baseline with
  CTS undergoing carpal tunnel release and 181
  completed at least 1 follow-up
• Recruited in MD offices throughout Maine
• Working at time developed CTS
• CTS documented with nerve conduction
• Homogeneous treatment

8
Defining the Sample

• Exclusions
• No follow-up information
• No information on 6 month outcome
• No baseline work role functioning
• 128 patients at 2 month follow-up and 122 at 6
  months

9
Defining Successful Return to Work
The ability worker to meet work demands given
their current physical and emotional health status
Poor success translates into lost productivity
for the business and increased job insecurity for
the worker
10
Defining Successful Return to Work

• 15 item work role functioning assessed at
  baseline, 2 and 6 months

• In general skewed towards better functioning

11
Defining Successful Return to Work

• Problem of not back at work
• Definitely not successful work role functioning
• But many reasons for a person to not be back at
  work
• Create three-level outcome
• 0, not yet back at work and not working because
  of health
• 1, back at work but not functioning well
• (lt90 score on WRF)
• 2, back at work and functioning well
• (gt 90 on WRF)
• Note gt 90 is a useful norm for healthy WRF

12
Statistical Approach

• Ordered logit regression in STATA 7.0 with
  predictor variables assessed at baseline for 2
  month outcomes and baseline and two months for
  six month outcomes
• Model building following Hosmer and Lemeshow,
  Applied Logistic Regression
• Non-Proportionality of odds assessed using the
  Brandt test
• Model fit described with McKelvey and Zavonias
  R2
• Baseline predictors always included when change
  variable assessed

13
Two Month Results
14
Six Month Results
15
To Summarize
Improved Self Efficacy
Supportive Organization
Pre-Surgical Work Role Functioning
EARLY 2 Mos.
LATE 6 Mos.
WC Claimant
Baseline Depression
16
Conclusions

• This work highlights the importance of
• Using multidimensional models in examining the
  return to work process
• Using multiple time points following injury or
  medical intervention
• Using successful work role functioning measure

17
Suggestive Interventions

• Results suggest interventions targeting improved
  individual and organizational agency
• Improving worker self-efficacy
• Self-efficacy is confidence the worker has in
  his or her ability to effectively manage pain and
  health in return to work process
• Developing highly supportive organizational
  policies and practices
• People-Oriented Culture
• Safety Climate (Active Safety Leadership, Safety
  Training and Safety Diligence)
• Ergonomics Policies and Practice
• Disability Management

18
Future Study

• Larger, more diverse samples needed to elaborate
  multidimensional model and timing of predictors.
• Should we begin interventions?
• Should we worry about job satisfaction or
  satisfaction with medical care?
• Multi-level studies where organizational context
  more accurately specified are needed to identify
  OPPs amenable to intervention

19
Thank Youwww.benamick.com
Research Support Provided By NIOSH
R01-O503523-01A1 The Robert Wood Johnson
Foundation Workers Compensation Research
Initiative 038151 The Arthritis
Foundation NIAMS AR36308