Systematic Review of Workplace Rehabilitation Interventions for Work- Related Low Back Pain

1
Systematic Review of Workplace Rehabilitation
Interventions for Work- Related Low Back Pain

• Dr Renee Williams, Professor Muriel Westmorland,
• Chia-Yu Lin, Gloria Schmuck Marg Creen
• School of Rehabilitation Science McMaster
  University Hamilton Ontario, Canada

2
Objectives

• To review literature on effectiveness of
  workplace-based rehabilitation interventions for
  injured workers with work-related low back pain
  (LBP) - focused on secondary prevention
  interventions
• To determine quality of articles using quality
  guidelines (MacDermid, 2004) levels of evidence
  (Centre for Evidence Based Medicine 2001)

3
Background Why LBP?

• LBP is major challenge to employers, employees
  service providers ( Tulder et al 1995 Wadell
  1987)
• Reviews have examined effectiveness of workplace
  interventions for LBP (van Tulder et al, 2000
  Heymans et al, 2004)
• Inconclusive - methodological problems (sample
  sizes, inadequate descriptions of interventions)
• Did not distinguish between clinical vs workplace
  
• primary vs secondary prevention (Tveito et al,
  200 Volinn, 1999)

4
Methods

• 2 academics, 1 experienced clinician, 1
  Disability Management consultant 1
  kinesiologist
• Inclusion Criteria
• Intervention was carried out at workplace
• Participants had work-related LBP
• Intervention involved secondary prevention
• Study design was prospective or cross-sectional
• Study published in English

5
Search Strategy

• MEDLINE, CINAHL (Cumulative Index to Nursing
  Allied Health Literature), EMBASE (Excerpta
  Medica Database) AMED (Allied Complementary
  Medicine) Jan 1982 to Apr 2005
• Search Terms
• back pain return to work disabled persons job
  accommodation rehabilitation occupational
  health services workplace

6
Search Strategy . . .

• 1,224 abstracts - 198 full text articles
  identified
• 54 articles randomly allocated to 2 pairs
  reviewers
• 15 articles/10 studies met eligibility criteria
• Reasons For Exclusion
• Intervention was not workplace based
• Participants did not sustain work-related LBP
• Article was case study not primary research
• Intervention was primary prevention not
  secondary prevention

7
Review by Team

• 2 pairs reviewers evaluated articles using
• Quality guidelines (MacDermid, 2004)
• 24 specific research design elements
• 2 best score (high quality) 1 acceptable
  (fair quality) 0 poor (low quality or unmet
  criteria)
• Total possible score 48
• Levels of evidence (Centre for Evidence Based
  Medicine, 2001) were assigned

8
Levels of Evidence (Centre for EBM, 2001)

• 1a Systematic review of RCT studies
• 1b individual RCT
• 2a Systematic review of cohort studies
• 2b Individual cohort study
• 3a Systematic review of case control studies
• 3b Individual case control study
• 4 Case series
• 5 Expert opinion

9
Results (15 articles)

• 4 RCTs
• Karjalainen et al, 2004 2003 Alexandre et al,
  2001 Loisel et al, 1997
• 7 Cohort Studies With Control Groups
• Anema et al, 2004 Durand Loisel, 2001 Cooper
  et al 1998 Grayzel et al, 1997 Yassi et al,
  1995 Cooper et al, 1996 Linton, 1991
• 4 Cohort Studies Without Control Groups
• Anema et al, 2003 Bakhtiar et al, 2002 Jellema
  et al 2002
• Cooper et al, 1997

10
Quality Ratings
11
Quality Ratings . . .
12
Categories

• Early RTW/modified work
• Clinical interventions with occupational
  interventions
• Ergonomics including exercises lumbar supports
• Exercise workplace visit
• Supervisor involvement for return to work

13
1. Early RTW/Modified Work

• 2 yr study in Canadian hospital (Yassi et al,
  1995 Cooper et al, 1998 1997 1996)
• 60 nurses with LBP (intervention group) 158
  nurses (control)
• Intervention assessment, treatment modified
  work
• Findings - rates of back injury lost-time
  injuries decreased by 23 (intervention)
  increased by 44 (control) decreases in pain
  disability

14
2. Clinical Interventions With Occupational
Interventions

• Loisel et al (1997) - 130 employees from 31
  workplaces
• Randomized to
• usual care
• clinical/rehab intervention
• occupational intervention
• combination of 2 3
• Findings - full intervention group returned to
  work 2.4 times faster than usual care

15
2. Clinical interventions With Occupational
Interventions . . .

• As part of Loisel study, Durand Loisel (2001)
  examined therapeutic return to work (TRW) with 28
  employees 99 (control)
• Findings - at 2 yr follow up, 93 in TRW working
  had less pain but control group was not
  randomly selected

16
3. Ergonomics Including Exercises Lumbar
Supports

• Anema et al (2003) evaluated participatory
  ergonomics (n 35)
• Findings satisfied with program but no control
  group small sample size
• Anema et al (2004) examined ergonomic supports
  for 1631 employees (Netherlands, Denmark, Sweden,
  Germany, Israel USA)
• Findings ergonomics was effective on rate of
  RTW

17
3. Ergonomics Including Exercises Lumbar
Supports . . .

• Bakhtiar et al, (2002) examined ergonomics
  exercises in 15 milk vendors 15 goldsmiths
• Findings - 13/15 14/15 recovered but
• description of intervention/results limited
• Alexandre et al, (2001) evaluated exercise
  education in 27 nursing aids 29 (control)
• Findings - reduction in pain but retrospective
  study

18
3. Ergonomics Including Exercises Lumbar
Supports

• Jellema et al, (2002) studied lumbar supports in
  59 home care workers
• Findings disability decreased
• but no control group psychometric properties
  of survey tool were not tested

19
4. Exercise Workplace Visit

• Karjalainen et al, (2004 2003) Randomized to
• group 1 (n 56) Mini intervention (light
  mobilization graded activity
• group 2 (n 51) Mini intervention worksite
  visit
• group 3 (n 57) Usual care
• Findings - average sick days group 1 19 group
  2 28 group 3 41
• At 2 yrs decreased pain sustained workplace
  visit (group 2) did not add to its effectiveness
  

20
4. Exercises Workplace Visit . . .

• Grayzel et al (1997) - compared in-house PT (n
  10) to outside PT (n 120)
• Findings - lost days reduced in intervention
  group was 28 versus 57 in controls but small
  sample size

21
5. Supervisor Involvement for RTW

• Linton (1991) examined education about back
  neck injuries - 22 supervisors vs 14 control
  supervisors
• Findings - Intervention group rated knowledge
  higher, contacted sick employees more often, felt
  they initiated more ergonomic improvements but
  sampling bias male supervisors

22
Summary

• 15 articles - 8 medium (32) to very high (48)
• Key Messages
• Clinical interventions with occupational
  interventions are effective in RTW reducing
  pain disability
• Early RTW/modified work is effective in reducing
  rates disability
• Exercises ergonomic interventions are important

23
Limitations

• Quality guidelines psychometric properties?
• Some outcome measures discussed in articles were
  unknown
• Selection bias of articles

24
Conclusions

• More quality studies on workplace rehabilitation
  interventions for LBP are necessary to promote
  early safe return to work to reduce number of
  lost-time injuries

25
Questions