Content » Vol 40, Issue 10

Review article

Rehabilitation and work ability: A systematic literature review

Jaana Kuoppala, MD, PhD1 and Anne Lamminpää, MD, PhD2

From the 1Siinto, Sievi and 2State Treasury, Division of Insurance, Helsinki, Finland

OBJECTIVE: To evaluate the effects of rehabilitation on sickness absenteeism, return to work and disability pensions among persons of working age.

Method: Original articles published during 1970–2005 indexed in Medline and PsycINFO databases were studied systematically. The main search terms were rehabilitation, sick leave and disability pension. Out of 576 references, 41 potentially eligible publications were retrieved; other sources producing 21 articles. Forty-five studies were included in the analysis.

RESULTS: There is moderate evidence that return-to-work programmes decrease long sick leaves (risk ratio (RR) 0.46, range 0.25–1.10) and multimodal rehabilitation decreases the risk of disability pension (RR 0.64, range 0.52–1.14), counselling, exercise, multimodal medical rehabilitation or return-to-work programmes having no effect on return to work. Based on mainly weak evidence, early rehabilitation seems to reduce both absenteeism and disability pension.

CONCLUSION: Any type of rehabilitation may have an effect at an early stage of decreased work ability, being ineffective later on if applied as the only mode of rehabilitation. Where chronic disability is already present, multimodal medical rehabilitation needs to be combined with vocational rehabilitation in order to reduce absenteeism and disability pensions. It is essential that the workplace is integrated into rehabilitation.

Key words: rehabilitation, work ability, sick leave, disability pension.

J Rehabil Med 2008; 40: 796–804

Correspondence address: Anne Lamminpää, State Treasury, Division of Insurance, PO Box 10, Sörnäisten rantatie 13, FI-00054 State Treasury, Helsinki, Finland. E-mail: anne.lamminpaa@fimnet.fi

INTRODUCTION

Rehabilitation can be defined as measures required for coping with functional consequences of a disease, defect or trauma. The aim of rehabilitation is to improve work ability and functional capacity. Rehabilitation can be divided into medical, vocational or social rehabilitation. Medical rehabilitation aims at developing the functional and psychological abilities of the individual and, if necessary, his or her compensatory mechanisms, to enable him or her to attain self-dependence and lead an active life (1). Vocational rehabilitation aims, for example, at promoting employment opportunities for disabled persons in the open labour market (2). If a disease or a defect due to trauma affects functional capacity, the need for rehabilitation should be assessed. Rehabilitation can focus on health, functional or work ability or employment.

The need for rehabilitation and its importance in society have strengthened during the last decades. Forty-five percent of the Finnish population of working age and 40% of those still working have some kind of chronic disease or traumatic defect, and the subjective need for rehabilitation is significant (3). Fifteen percent of men and 22% of women had a need for vocational rehabilitation, whereas 21% of men and 24% of women had a need for some other type of rehabilitation (3).

The number of those retired in 2006 receiving old age pension (n = 27,733) was the same as that of those retired on disability pensions (n = 27,215) (4). The main disease categories were mental disorders (33%) and musculoskeletal diseases (32%), the most common diagnoses being depression (F32) and intervertebral disc disorder (M51). On the other hand, absenteeism lasting less than one year is most often due to musculoskeletal diseases (34%) and less frequently to mental disorders (25%) (5).

The challenge of rehabilitation in our society is to maintain employees’ work ability at a level sufficient to continue working in spite of diseases or disabilities. The age profile in Finland (6), rapid changes in work-life and stress at work, as well as long-lasting unemployment set demands for rehabilitation.

This study was a part of a larger project in which the scientific evidence on the associations between psychosocial factors at work and rehabilitation, job well-being and work ability was evaluated. This meta-analysis focuses on the effects of medical and vocational rehabilitation as well as early rehabilitation on sick leave and disability pension among people of working age.

METHODS

Literature search

Two literature databases were searched from 1970 to 2005: Medline in June–July 2005 and PsycINFO in November 2005. The search terms were rehabilitation, sick leave, disability pension, trials and cohort studies. The aim was to find all relevant original studies published in international journals. Meta-analyses and systematic reviews on rehabilitation were also examined to ensure that no important studies were missed.

A study was included in the analysis if it was original and the study population was of working age. The studies that were conducted in other than a true working environment, such as in classes or courses or among students, were excluded. In addition, those studies that did not provide information about study design and results in sufficient detail were excluded. Dissertations were excluded for practical reasons; international dissertations are generally difficult to obtain.

The abstracts were scrutinized, and articles that could potentially be included were acquired. Search terms, search strategy, the selection and use of publications were documented systematically. Detailed information was collected from each study included in the analysis.

Definitions

For the purposes of this study, rehabilitative measures that were targeted on employees with subjective symptoms concerning health but no diagnosed disease or disorder were classified as “early rehabilitation”, compared with “rehabilitation” that was aimed at employees with chronic diseases. Early rehabilitation in this sense has not been used in the international literature but is potentially an important concept in the context of work ability and job well-being, especially if prevention is thought to be more cost-effective than rehabilitation.

Evaluation of strength of evidence

Six factors affected the strength of evidence: study design, quality of studies, quality of results, applicability of results, number of studies and homogeneity of studies (Table I). Each study was assessed for the 4 first mentioned properties. The study quality was based on the study population (e.g. the population of a certain area, or all consecutive patients in a clinic), and the definition and measurement of predictor and outcome. The quality of results, on the other hand, was based on the sample size, control group, number of drop-outs and those missing from analyses, randomization, treatment allocation, follow-up time, and whether potential confounding factors were accounted for. The applicability of results was affected by study country, setting (e.g. population, work environment, healthcare), sex distribution, mean age and coverage (i.e. response rate, how many from the eligible base population participated in the study). The homogeneity of studies was assessed comparing the following 11 factors between studies: study country, setting, sex distribution, mean age, the measurement of predictor and outcome, follow-up time, the risk or distribution of outcome in the control group, effect measure (risk difference, risk ratio (RR), odds ratio, difference in means, correlation coefficient), effect (benefit, no effect, harm) and the continuity of the effects between studies.

Studies were required to meet the predefined criteria at each level of strength of evidence. The evaluation was hierarchical, i.e. only those studies were taken into consideration that fulfilled the criteria for the best possible level. The cut-off points for each criterion were based on the current practice, specialist opinion or common sense.Statistical analyses

Table I. Criteria for strength of evidence on intervention

Strength of evidence

Study design

Minimum study quality

Minimum results quality

Minimum results applicability

Number of studies

Homogeneity Index*

Good

RCT, CRCT, RCCT

Good

Good

Moderate

3

6/8 (2/3)

Moderate

RCT, CRCT, RCCT

Moderate

Moderate

Moderate

2

5/8 (2/3)

CT, CO, NCC

Good

Good

Weak

RCT, CRCT, RCCT

Weak

Weak

Weak

1

4/8 (1/2)

CT, CO, NCC

Moderate

Moderate

CC

Good

Good

Very weak

CS

Weak

Weak

Weak

1

3/8 (1/2)

*First fraction expresses the number of factors (e.g. population, outcome, follow-up time) that need to be homogenous between studies. The latter fraction expresses the number of studies that need to be similar, in order for the factor in question to be considered homogenous.

RCT: randomized controlled trial; CRCT: cluster-randomized controlled trial; RCCT: randomized controlled crossover trial; CT: clinical trial; CO: cohort study; NCC: nested case-control study; CC: case-control study; CS: cross-sectional study.

The rate difference per 1000 person-years and RR were considered the most optimal effect measures. The medians and ranges of the effect sizes are reported. A summary statistic for risk ratios were calculated using the inverse variance method. If there were no cases in either intervention or control group, 0.5 was added to each cell of the 2×2 table in order to calculate rate difference and RR.

RESULTS

Literature search

Out of 576 references obtained from Medline and PsycINFO, 41 potentially eligible publications were retrieved. Twenty-four studies were included in the analysis (7–30). Seventeen studies were excluded due to insufficient data (31–40), ineligible outcome (41–44) or study population (45–47). Eighteen eligible studies were found through manual search of the reference lists of the relevant reviews and the acquired original studies (48–65) and 3 publications through other searches (on capacity and work ability) during the whole project (66–68). Additionally, 10 studies would have been included if they had contained sufficient information (69–78). In all, 45 studies were included in the meta-analysis.

Description of studies

The characteristics of studies are shown in Tables II–III. There were 24 randomized controlled trials, 1 cluster-randomized trial, 3 clinical trials and 17 cohort studies. Seven studies were from Finland, 11 from Sweden, 5 from Norway, 7 from the Netherlands, 2 from Germany, 1 from France, 1 from the UK, 2 from Canada and 9 from the USA. Eleven studies were population-based, 11 were performed in occupational settings, 22 in healthcare settings and one in an insurance setting. Most of the studies had mixed populations concerning gender, although the distribution might not have been even, especially in the studies performed in work environments; 36 studies had both men and women in their study populations, 5 studies focused on males and one on females, and 3 studies did not report the gender distribution. Seven studies did not have any kind of internal control group but, instead, reported values before and after intervention. The age distribution was reported heterogeneously, but most studies, if not all, seemed to have the whole working age covered.

Table IIA. Study characteristics: trials

Study, year

Type of rehabilitation

Outcome

Start of recruitment

Follow-up,

years

Treated

n/N

Controls

n/N

Crude RD per 1000 person-years

Crude

RR

Randomized controlled trials

Bengtsson 1983 (48)

Exercise

Sick leave > 6 months

1973

0.5

16/46

16/42

–66

0.91

Alaranta et al. 1986 (7)

Multimodal

Disability pension

nr

1

8/106

7/106

9

1.14

Burgess et al. 1987 (50)

Return to work

Sick leave > 6 months

nr

1.1

19/77

17/76

21

1.10

Dennis et al. 1988 (51)

Counselling

Return to work

1983

0.5

92/99

88/102

133

1.08

Greenwood et al. 1990 (55)

Counselling

Sick leave > 6 months

1985

1.5

nr/121

nr/163

12

1.06

Kellett et al. 1991 (57)

Exercisec

Sick leave > 1 month

nr

1.5

nr/37

nr/48

–60

0.63†

Pilote et al. 1992 (62)

Counselling

Return to work

1987

0.5

82/94

87/91

–167

0.91

Alaranta et al. 1994 (8)

Multimodal

Disability pension

1988

1

4/152

7/141

–23

0.53

Berglund et al. 1994 (9)

Multimodal

Return to work

nr

1

80/90

76/91

54

1.06

Bjorndal 1994 (10)

Counselling

Return to work

1990

2.4

64/122

852/1636

2

1.01

Engblom et al. 1994 (52)

Multimodal

Disability pension

1986

1

25/66

29/59

–113

0.77

Froelicher et al. 1994 (54)

Multimodal

Return to work

1977

1

51/52

56/62

78

1.09

Lindh et al. 1997 (11)

Multimodal

Return to work

nr

1

nr/151

nr/134

–19

0.97

Loisel et al. 1997 (58)

Multimodal

Return to work

1991

1

nr/31

nr/26

nr

1.59†

Torstensen et al. 1998 (14)

Exercise

Return to work

nr

1

87/136

40/70

68

1.12

Fanello et al. 1999 (53)

Educationc

Physical well-being

1995

2

15/87

25/70

–92

0.48

Hofman-Bang et al. 1999 (56)

Multimodal

Return to work

1993

1

34/46

28/41

56

1.08

Hazard et al. 2000 (15)

Education

Sick leave

1996

0.5

14/217

12/202

12

1.10

Molde Hagen et al. 2000 (16)

Exercise

Return to work

nr

1

162/237

124/220

119

1.21†

Jensen et al. 2001 (17)

Multimodal

Disability pension

1995

1.5

19/117

15/48

–100

0.52

Marhold et al. 2001 (18)

Return to work

Sick leave > 6 months

nr

0.5

nr/36

nr/36

–545

0.72†

Verbeek et al. 2002 (20)

Return to work

Sick leave > 6 months

nr

1

nr/61

nr/59

–169

0.30

Nystuen & Hagen 2003 (23)

Psychological

Sick leave > 6 months

2001

1.25

nr/113

nr/100

67

1.14

de Boer et al. 2004 (27)

Vocationale

Disability pension

1997

2

9/53

13/47

–53

0.61

Cluster-randomized controlled trials

van der Klink et al. 2003 (24)

Return to work

Sick leave > 6 months

1995

1

nr/109

nr/83

–87

0.25

Clinical trials

Perkiö-Mäkelä & Riihimäki 1997 (61)

Ergonomicsc

Physical well-being

1989

0.04

nr/31

nr/33

–13

1.00

Arnetz et al. 2003 (21)

Return to work

Sick leave > 6 months

nr

1

0/65

65/73

–898‡

0.01‡

Landy et al. 2003 (22)

Educationc

Physical well-being

2002

0.25

136/164a

119/164b

415*

1.14*

Note: Each study might have reported several outcomes.

*No control group: the comparison between the values in the beginning and the end of the study.

†Adjusted value.

‡In order to calculate RD and RR, 0.5 added to each cell of 2×2 table.

n/N: number of cases in the group; RD: rate difference; RR: risk ratio; a: at the end of the study; b: at the beginning of the study; c: classified as early rehabilitation in this review; nr: not reported.

Table IIB. Study characteristics, quality and strength: trials

Study

Country

Setting

Control

Mean age* years [range]

Sex

Study quality

Results quality

Results applicability

Study strength

Randomized controlled trials

Bengtsson (48)

USA

Healthcare

Yes

56

FM

Good

Moderate

Moderate

Moderate

Alaranta et al. (7)

Finland

Healthcare

Yes

40

FM

Good

Moderate

Moderate

Moderate

Burgess et al. (50)

USA

Healthcare

Yes

51

FM

Good

Moderate

Moderate

Moderate

Dennis et al. (51)

USA

Healthcare

Yes

[nr–60]

M

Good

Moderate

Moderate

Moderate

Greenwood et al. (55)

USA

Work

Yes

39

FM

Good

Moderate

Moderate

Moderate

Kellett et al. (57)

Sweden

Work

Yes

42

FM

Good

Moderate

Moderate

Moderate

Pilote et al.62)

USA

Healthcare

Yes

51

FM

Good

Moderate

Moderate

Moderate

Alaranta et al. (8)

Finland

Healthcare

Yes

[30–47]

FM

Good

Moderate

Moderate

Moderate

Berglund et al.(9)

Sweden

Healthcare

Yes

[nr–65]

FM

Good

Moderate

Moderate

Moderate

Bjorndal et al. (10)

Norway

Population

Yes

44

FM

Good

Moderate

Good

Moderate

Engblom et al. (52)

Finland

Healthcare

Yes

52

M

Good

Moderate

Moderate

Moderate

Froelicher et al. (54)

USA

Healthcare

Yes

56

FM

Good

Moderate

Moderate

Moderate

Lindh et al. (11)

Sweden

Population

Yes

[20–55]

FM

Good

Good

Good

Strong

Loisel et al. (58)

Canada

Work

Yes

41

FM

Good

Poor

Moderate

Weak

Torstensen et al. (14)

Norway

Population

Yes

[20–65]

FM

Good

Moderate

Good

Moderate

Fanello et al. (53)

France

Work

Yes

38

nr

Good

Moderate

Moderate

Moderate

Hofman-Bang et al. (56)

Sweden

Healthcare

Yes

53

FM

Good

Moderate

Moderate

Moderate

Hazard et al. (15)

Canada

Population

Yes

38

FM

Good

Moderate

Moderate

Moderate

Molde Hagen et al. (16)

Norway

Population

Yes

[18–60]

FM

Good

Moderate

Good

Moderate

Jensen et al. (17)

Sweden

Population

Yes

[18–60]

FM

Good

Moderate

Good

Moderate

Marhold et al. (18)

Sweden

Population

Yes

[25–60]

F

Good

Moderate

Moderate

Moderate

Verbeek et al. (20)

Netherlands

Work

Yes

nr

FM

Good

Moderate

Good

Moderate

Nystuen & Hagen (23)

Norway

Population

Yes

40

FM

Good

Poor

Good

Weak

de Boer et al. (27)

Netherlands

Work

Yes

[50–nr]

FM

Good

Moderate

Moderate

Moderate

Cluster-randomized controlled trials

van der Klink et al. (24)

Netherlands

Work

Yes

nr

FM

Good

Moderate

Good

Moderate

Clinical trials

Perkiö-Mäkelä & Riihimäki (61)

Finland

Healthcare

Yes

37

M

Poor

Moderate

Moderate

Very weak

Arnetz et al. (21)

Sweden

Population

Yes

42

FM

Good

Poor

Good

Very weak

Landy et al. (22)

USA

Work

No

nr

FM

Moderate

Poor

Moderate

Very weak

*Mean age at start of study.

nr: not reported; F: female; M: male.

The quality and strength of studies are shown in Tables IIIA–B. Study quality was good in most of the studies, yet the results quality could be considered good in only 2 studies and poor in 12. The applicability of studies was good to moderate in all but one study. Thus, study strength was strong in one study, moderate in 23, weak in 11 and very weak in 10.

Table IIIA. Study characteristics: cohort studies

Study

Type of rehabilitation

Outcome

Start of recruitment

Follow-up, years

Treated n/N

Controls n/N

Crude RD per 1000 person-years

Crude RR

Boulay et al. 1982 (49)

Exercise

Return to work

1978

1

51/59

51/62

42

1.05

Rauscha et al. 1988 (67)

Multimodal

Return to work

1975

5

205/285

33/56

26

1.22

Perk et al. 1990 (60)

Multimodal & vocational

Return to work

1980

1

22/37

41/64

–46

0.93

Straaton et al. 1992 (64)

Multimodal

Return to work

1985

1.64

105/137

140/319

227

1.85†

Malcolm et al. 1993 (59)

Administrative

Disability pension

1990

1

154/604

125/329

–125

0.67

Mellin et al. 1993 (66)

Multimodal

Return to work

1988

1

108/193a

101/194b

39*

1.07*

Schmidt et al. 1995 (63)

Vocational

Return to work

1984

5.5

108/184

35/179

23

1.65†

van Doorn 1995 (65)

Job satisfaction

Sick leave > 6 months

1990

1

0/73

11/15

–766‡

0.01‡

Grahn et al. 1998 (12)

Multimodal & vocational

Disability pension

1994

0.6

1/115

4/107

–48

0.23

Jensen & Bodin 1998 (13)

Multimodal & vocational

Sick leave > 1 month

nr

1.5

nr/67

nr/28

33

1.07†

Beutel et al. 1999 (68)

Vocational

Return to work

1995

0.5

46/57

137/241

477

1.42

Arokoski et al. 2002 (19)

Multimodal & vocationale

Job well-being

nr

1.5

nr/265a

nr/265b

84*

1.47*†

Verbeek et al. 2003 (25)

Vocational

Return to work

nr

1

18/34

48/64

412

2.00†

Bauer & Odijk 2004 (26)

Return to work

Sick leave

1997

2

nr/501a

nr/501b

–11*

0.79*

Goine et al. 2004 (28)

Return to work

Disability pension

1989

5

42/1952a

71/2445b

–1*

0.79*†

Holopainen et al. 2004 (29)

Multimodal & vocationalc

Sick leave > 1 week

nr

5

nr/20a

nr/20b

–72*

0.07*

Nieuwenhuijsen et al. 2004 (30)

Return to work

Return to work

2001

1

nr/19

nr/66

nr

0.80†

Note: Each study might have reported several outcomes.

*No control group: the comparison between the values in the beginning and the end of the study.

†Adjusted value.

‡In order to calculate RD and RR, 0.5 added to each cell of 2×2 table.

n/N: number of cases in the group; RD: rate difference; RR: risk ratio; a: at the end of the study; b: at the beginning of the study; c: classified as early rehabilitation in this review; nr: not reported.

Table IIIB. Study characteristics, quality and strength: cohort studies

Study

Country

Setting

Control

Mean age* years [range]

Sex

Study quality

Results quality

Results applicability

Study strength

Boulay et al. (49)

USA

Healthcare

Yes

49

M

Moderate

Moderate

Moderate

Weak

Rauscha et al. (67)

Germany

Healthcare

Yes

51

FM

Moderate

Moderate

Moderate

Weak

Perk et al. (60)

Sweden

Population

Yes

[42–71]

FM

Good

Moderate

Moderate

Weak

Straaton et al. (64)

USA

Healthcare

Yes

[18–69]

FM

Good

Moderate

Moderate

Weak

Malcolm et al. (59)

UK

Work

No

nr

nr

Good

Poor

Poor

Very weak

Mellin et al. (66)

Finland

Healthcare

No

43

FM

Good

Poor

Moderate

Very weak

Schmidt et al. (63)

Netherlands

Healthcare

Yes

40

FM

Good

Good

Moderate

Moderate

van Doorn (65)

Netherlands

Insurance

Yes

nr

nr

Good

Poor

Moderate

Weak

Grahn et al. (12)

Sweden

Population

Yes

44

FM

Good

Moderate

Good

Weak

Jensen & Bodin (13)

Sweden

Healthcare

Yes

41

FM

Good

Moderate

Moderate

Weak

Beutel et al. (68)

Germany

Healthcare

Yes

40

FM

Good

Moderate

Moderate

Weak

Arokoski et al. (19)

Finland

Healthcare

No

43

FM

Moderate

Poor

Moderate

Very weak

Verbeek et al. (25)

Netherlands

Healthcare

Yes

42

FM

Moderate

Moderate

Moderate

Weak

Bauer & Odijk (26)

Norway

Work

No

40

FM

Good

Poor

Good

Very weak

Goine et al. (28)

Sweden

Work

No

36

FM

Good

Poor

Good

Very weak

Holopainen et al. (29)

Finland

Healthcare

No

37

M

Moderate

Poor

Moderate

Very weak

Nieuwenhuijsen et al. (30)

Netherlands

Healthcare

Yes

44

FM

Poor

Poor

Moderate

Very weak

*Mean age at start of study.

nr: not reported; F: female; M: male.

Effects of early rehabilitation

Evidence on early rehabilitation is scanty and the strength of it is weak at best (Table IV). Nevertheless, exercise seems to decrease sickness absences (RR 0.63, range not applicable (N/A), only one study available), and multimodal medical combined with vocational rehabilitation seems to increase both job (RR 1.47, range N/A) and physical well-being (RR 1.68, range 1.60–1.76) and decrease sick leaves (RR 0.24, range 0.07–0.85). Vocational rehabilitation may decrease the risk of disability pension (RR 0.61, range N/A). On the other hand, there is no evidence that education or ergonomics alone would be beneficial.

Table IV. Early rehabilitation and work ability

Type of rehabilitation Outcome

Strength of evidence

k/K

Nk

RR

RD per 1000 years, median [range]

References*

Median [range]

Mean (95% CI)

Education

Physical well-being

Weak

1/2

157

0.48!

0.48 (0.17–1.37)

–92!

53 [22]

Exercise

Physical well-being

Weak

1/1

85

1.07

1.07 (0.36–3.17)

14

57

Sick leave > 1 week

Weak

1/1

85

0.63

0.63 (0.17–2.35)

–60

57

Multimodal & vocational

Job well-being

Very weak

1/1

265

1.47

1.47 (0.73–2.95)

84

19

Physical well-being

Very weak

2/2

285

1.68 [1.60;1.76]

1.63 (0.87–3.06)

67 [37;98]

19, 29

Sick leave > 1 week

Very weak

2/2

285

0.24 [0.07;0.85]

0.70 (0.36–1.34)

–55 [–72;–38]

19, 29

Ergonomics

Physical well-being

Very weak

1/1

99

1.00

1.00 (0.32–3.16)

–13

61

Vocational

Sick leave > 1 month

Weak

1/1

71

0.91

0.91 (0.38–2.19)

–29

27

Disability pension

Weak

1/1

100

0.61

0.61 (0.18–2.07)

–53

27

*Studies with weaker strength listed in brackets.

!The result is contrary to expectations.

RR: risk ratio; RD: rate difference; k/K: number of studies providing best evidence out of all eligible studies.; Nk: total number of participants in the studies providing best evidence; 95% CI: 95% confidence interval.

Effects of rehabilitation

The strength of evidence on rehabilitation is mainly weak, yet there is moderate evidence that multimodal rehabilitation decreases the risk of disability pension (RR 0.64, range 0.52–1.14) and that return-to-work programmes decrease sick leaves lasting longer than 6 months (RR 0.46, range 0.25–1.10), but that counselling, exercise, multimodal medical rehabilitation or return-to-work programmes do not have an effect on return to work at one year (Table V). On the other hand, vocational rehabilitation and multimodal medical combined with vocational rehabilitation seem to increase return to work (RR 1.53, range 1.42–2.00; RR 1.50, range 0.93–2.41, respectively). Education, exercise or psychological rehabilitation alone do not seem to have any effect on sick leaves. However, all the rehabilitation modalities for which there was any evidence; administration (RR 0.67, range N/A), psychological (RR 0.46, range N/A), multimodal with (RR 0.23, range N/A) or without vocational (RR 0.64, range 0.52–1.14), and return-to-work programmes (RR 0.79, range N/A) seemed to decrease the risk of disability pension.

Table V. Rehabilitation and work ability

Type of rehabilitation Outcome

Strength of evidence

k/K

Nk

RR

RD per 1000 years,

median [range]

References*

Median [range]

Mean (95% CI)

Administrative

Return to work

Very weak

2/2

1262

1.30 [1.15;1.46]

1.25 (0.93–1.69)

164 [121;207]

59, 65

Disability pension

Very weak

1/1

933

0.67

0.67 (0.36–1.24)

–125

59

Counselling

Return to work

Moderate

3/3

2144

1.01 [0.91;1.08]

0.99 (0.76–1.30)

2 [–167;133]

10, 51, 62

Sick leave > 6 months

Weak

1/1

284

1.06

1.06 (0.46–2.42)

12

55

Education

Return to work

Weak

1/1

212

1.03

1.03 (0.74–1.44)

28

7

Sick leave

Weak

1/1

419

1.10

1.10 (0.33–3.69)

12

15

Sick leave > 6 months

Weak

1/1

212

0.93

0.93 (0.30–2.89)

–10

7

Psychological

Sick leave

Weak

1/1

214

0.95

0.95 (0.42–2.12)

–22

17

Sick leave > 6 months

Weak

1/1

213

1.14

1.14 (0.60–2.16)

67

23

Disability pension

Weak

1/1

214

0.46

0.46 (0.13–1.60)

–113

17

Exercise

Return to work

Moderate

4/5

928

1.09 [1.04;1.21]

1.10 (0.84–1.44)

90 [33;119]

14, 16, 48, 54 [49]

Sick leave > 6 months

Weak

1/1

88

0.91

0.91 (0.32–2.58)

–66

48

Multimodal

Return to work

Moderate

5/9

855

1.08 [0.97;1.50]

1.09 (0.85–1.39)

56 [–19;188]

9, 11, 52, 54, 56 [58, 64, 66, 67]

Sick leave

Weak

1/1

214

0.89

0.89 (0.42–1.88)

–44

17

Sick leave > 6 months

Weak

1/1

125

0.83

0.83 (0.39–1.75)

–116

52

Disability pension

Moderate

4/4

795

0.64 [0.52;1.14]

0.69 (0.40–1.22)

–62 [–113;9]

7, 8, 17, 52

Multimodal & vocational

Return to work

Weak

2/2

205

1.50 [0.93;2.41]

0.93 (0.42–2.06)

–46

58, 60

Sick leave > 1 month

Weak

2/2

303

1.00 [0.94;1.07]

0.99 (0.62–1.57)

–20 [–74;33]

12, 13

Disability pension

Weak

1/1

222

0.23

0.23 (0.03–1.83)

–48

12

Vocational

Return to work

Weak

4/4

1215

1.53 [1.42;2.00]

1.50 (1.09–2.05)

281 [23;477]

25, 63, 64, 68

Ergonomics

Return to work

Weak

1/1

104

1.12

nr

.

58

Return to work

Return to work

Moderate†

3/4

465

1.00 [1.00;1.08]

1.01 (0.85–1.20)

1 [1;70]

20, 24, 50 [30]

Sick leave

Very weak

1/1

501

0.79

0.79 (0.33–1.91)

–11

26

Sick leave > 6 months

Moderate

4/5

537

0.46 [0.25;1.10]

0.64 (0.39–1.05)

–128 [–545;21]

18, 20, 24, 50 [20]

Disability pension

Very weak

1/1

1952

0.79

0.79 (0.33–1.89)

–1

28

*Studies with weaker strength listed in brackets.

†Strength borrowed from lower-quality studies.

RR: risk ratio; RD: rate difference; nr: not reported; k/K: number of studies providing best evidence out of all eligible studies; Nk: total number of participants in the studies providing best evidence; 95% CI: 95% confidence interval.

DISCUSSION

It is plausible that rehabilitation methods such as education, counselling, exercise, medical therapy and ergonomics might improve an employee’s work ability at an early stage of a disease even though at any later stage they became ineffective if applied as the only mode of rehabilitation. There is not enough evidence either to support or reject this hypothesis. However, it does seem that multimodal medical rehabilitation should be combined with vocational rehabilitation if the aim is to increase employees’ return to work. It is evident that the workplace should be involved in the rehabilitation process; medical rehabilitation may be fruitless if the way of working and the circumstances at work do not also change.

Authors of other meta-analyses concerning the effect of rehabilitation on absenteeism and return to work have drawn similar conclusions about clinical implications (79–82). Some evidence suggests that a graded activity programme improves absenteeism outcomes in patients with subacute low back pain (79). Asthma self-management, involving self-monitoring coupled with regular medical reviews and a written action plan, improves health outcomes (80). In patients with chronic low back pain, multidisciplinary approaches including a psychological component compared with other active control conditions have a positive long-term effect on return to work (81), and multimodal and multidisciplinary rehabilitation programmes that are in some way work-related seem to reduce the number of sick days (82).

The endpoints of interest in rehabilitation often differ depending on the point of view; healthcare, occupational health, health economy or insurance are interested in the effect on sick leaves, return to work and disability pensions, whereas the employee is more concerned about his or her well-being, health and quality of life. We see rehabilitation as a means to help employees to remain at work in spite of chronic symptoms or disease, which implies that it would be important to evaluate both kinds of endpoint at the same time. Yet, we chose to restrict to the evaluation to sick leaves, return to work and disability pensions, which were more readily defined in the studies available.

There are numerous publications on rehabilitation but surprisingly little scientifically convincing evidence. Studies concerning rehabilitation have been performed mainly in Northern Europe and the USA and results from the southern countries of Europe or America are virtually absent. Rehabilitation is a form of health technology and, as with any type of intervention, its effect should be assessed using randomized controlled trial designs. Blinding of patients and care-givers is seldom possible, yet outcome can always be assessed in a blinded fashion. Sickness absences and disability pension due to disability are not subjective endpoints as such, but, at least in the Finnish healthcare and social system, the patient’s history in rehabilitation affects these endpoints. Thus, the effect of rehabilitation is more or less confounded due to this fact alone. The International Classification of Functioning, Disability and Health (ICF) as a common reference framework for functioning may contribute to improved outcome research in rehabilitation (83).

One of the cornerstones of the evaluation of evidence is the fact that the literature should be searched extensively. It is unlikely that the 2 literature databases we used in this review cover all of the studies ever performed on these topics. However, the hierarchical nature of the method we used compensates for this weakness; it is essential that all the studies providing strongest evidence have been included. We believe that we have found at least the major part of those studies. On the other hand, weak evidence does not become stronger by adding more poor quality studies. We also have to bear in mind that the quality of a study can be high and yet the strength of evidence it provides can be weak.

According to our findings, early rehabilitation may reduce both absenteeism and disability pension. If an employee has a chronic disability that decreases his or her work ability, multimodal medical rehabilitation needs to be combined with vocational rehabilitation for the best outcome. It is essential that the workplace is integrated into rehabilitation. It is possible effectively to improve an employee’s work ability by rehabilitation. We emphasize the importance of early rehabilitation, even though a great deal more research is needed to clarify the true potential of different types of rehabilitation.

REFERENCES

1. WHO Expert Committee on Medical Rehabilitation. Geneva: WHO; 1968.

2. ILO Vocational Rehabilitation and Employment (Disabled Persons) Convention (No. 159) and Recommendation (No. 168): United Nations Convention on the Rights of Persons with Disabilities. Geneva: International Labour Office; 2008.

3. Health and functional capacity in Finland: baseline results of the Health 2000 Health examination Survey. Helsinki: National Public Health Institute; 2004.

4. Statistical yearbook of pensioners in Finland 2006. Helsinki: Finnish Centre for Pensions; 2007.

5. Kansaneläkelaitos. Kelan tilastollinen vuosikirja 2006 [The Social Insurance Institute of Finland: Health insurance statistics 2006]. Helsinki: The Social Insurance Institute of Finland, Statistical Branch; 2007 (in Finish).

6. Statistical yearbook of Finland 2007. Helsinki: Statistics Finland; 2007.

7. Alaranta H, Hurme M, Einola S, Kallio V, Knuts LR, Törmä T. Rehabilitation after surgery for lumbar disc herniation: results of a randomized clinical trial. Int J Rehabil Res 1986; 9: 247–257.

8. Alaranta H, Rytokoski U, Rissanen A, Talo S, Ronnemaa T, Puukka P, et al. Intensive physical and psychosocial training program for patients with chronic low back pain: a controlled clinical trial. Spine 1994; 19: 1339–1349.

9. Berglund G, Bolund C, Gustafsson UL, Sjoden PO. One-year follow-up of the ‘Starting Again’ group rehabilitation programme for cancer patients. Eur J Cancer 1994; 30A: 1744–1751.

10. Bjorndal A. Oppfølgning av langtidssykmeldte: en cohortstudie fra Moss [Follow-up of persons on long-term sick-leave: a cohort study in the city of Moss]. Tidsskr Nor Laegeforen 1994; 114: 2857–2862 (in Norwegian).

11. Lindh M, Lurie M, Sanne H. A randomized prospective study of vocational outcome in rehabilitation of patients with non-specific musculoskeletal pain: a multidisciplinary approach to patients identified after 90 days of sick-leave. Scand J Rehabil Med 1997; 29: 103–112.

12. Grahn B, Ekdahl C, Borgquist L. Effects of a multidisciplinary rehabilitation programme on health-related quality of life in patients with prolonged musculoskeletal disorders: a 6-month follow-up of a prospective controlled study. Disabil Rehabil 1998; 20: 285–297.

13. Jensen IB, Bodin L. Multimodal cognitive-behavioural treatment for workers with chronic spinal pain: a matched cohort study with an 18-month follow-up. Pain 1998; 76: 35–44.

14. Torstensen TA, Ljunggren AE, Meen HD, Odland E, Mowinckel P, Geijerstam S. Efficiency and costs of medical exercise therapy, conventional physiotherapy, and self-exercise in patients with chronic low back pain: a pragmatic, randomized, single-blinded, controlled trial with 1-year follow-up. Spine 1998; 23: 2616–2624.

15. Hazard RG, Reid S, Haugh LD, McFarlane G. A controlled trial of an educational pamphlet to prevent disability after occupational low back injury. Spine 2000; 25: 1419–1423.

16. Molde Hagen E, Eriksen HR, Ursin H. Does early intervention with a light mobilization program reduce long-term sick leave for low back pain? Spine 2000; 25: 1973–1976.

17. Jensen IB, Bergström G, Ljungquist T, Nygren ÅL. A randomized controlled component analysis of a behavioral medicine rehabilitation program for chronic spinal pain: are the effects dependent on gender? Pain 2001; 91: 65–78.

18. Marhold C, Linton SJ, Melin L. A cognitive-behavioral return-to-work program: effects on pain patients with a history of long-term versus short-term sick leave. Pain 2001; 91: 155–163.

19. Arokoski JP, Juntunen M, Luikku J. Use of health-care services, work absenteeism, leisure-time physical activity, musculoskeletal symptoms and physical performance after vocationally oriented medical rehabilitation-description of the courses and a one-and-a-half-year follow-up study with farmers, loggers, police officers and hairdressers. Int J Rehabil Res 2002; 25: 119–131.

20. Verbeek JH, van der Weide WE, van Dijk FJ. Early occupational health management of patients with back pain: a randomized controlled trial. Spine 2002; 27: 1844–1851.

21. Arnetz BB, Sjogren B, Rydehn B, Meisel R. Early workplace intervention for employees with musculoskeletal-related absenteeism: a prospective controlled intervention study. J Occup Environ Med 2003; 45: 499–506.

22. Landy S, Brookfield RB, McGinnis J, Hawkes MN. Headache in the workplace: an evaluation of the impact of educational programming on employee disability. Headache and Pain 2003; 14: 156–160.

23. Nystuen P, Hagen KB. Feasibility and effectiveness of offering a solution-focused follow-up to employees with psychological problems or muscle skeletal pain: a randomised controlled trial. BMC Public Health 2003; 3:19.

24. van der Klink JJ, Blonk RW, Schene AH, van Dijk FJ. Reducing long term sickness absence by an activating intervention in adjustment disorders: a cluster randomised controlled design. Occup Environ Med 2003; 60: 429–437.

25. Verbeek J, Spelten E, Kammeijer M, Sprangers M. Return to work of cancer survivors: a prospective cohort study into the quality of rehabilitation by occupational physicians. Occup Environ Med 2003; 60: 352–357.

26. Bauer MN, Odijk J. Nærværsarbeid i Statoil Forpleining [Return-to-work strategies in Statoil Forpleining]. Tidsskr Nor Laegeforen 2004; 124: 2630–2632 (in Norwegian).

27. de Boer AG, van Beek JC, Durinck J, Verbeek JH, van Dijk FJ. An occupational health intervention programme for workers at risk for early retirement: a randomised controlled trial. Occup Environ Med 2004; 61: 924–929.

28. Goine H, Knutsson A, Marklund S, Karlsson B. Sickness absence and early retirement at two workplaces: effects of organisational intervention in Sweden. Soc Sci Med 2004; 58: 99–108.

29. Holopainen K, Nevala N, Kuronen P, Arokoski JP. Effects of vocationally oriented medical rehabilitation for aircraft maintenance personnel: a preliminary study of long-term effects with 5-year follow-up. J Occup Rehabil 2004; 14: 233–242.

30. Nieuwenhuijsen K, Verbeek JH, de Boer AG, Blonk RW, van Dijk FJ. Supervisory behaviour as a predictor of return to work in employees absent from work due to mental health problems. Occup Environ Med 2004; 61: 817–823.

31. Grayzel EF, Finegan AM, Ponchak RE. The value of in-house physical therapy. J Occup Environ Med 1997; 39: 344–346.

32. Fleten N, Johnsen R, Ostrem BS. Sykmeldte tror tiltak på arbeidsplassen kan redusere sykefravær [Sick-listed persons think job adjustments might reduce sick-leave]. Tidsskr Nor Laegeforen 1999; 119: 3730–3734 (in Norwegian).

33. Pal TM, de Monchy JG, Groothoff JW, Post D. Follow up investigation of workers in synthetic fibre plants with humidifier disease and work related asthma. Occup Environ Med 1999; 56: 403–410.

34. Tate RB, Yassi A, Cooper J. Predictors of time loss after back injury in nurses. Spine 1999; 24: 1930–1935.

35. Linz DH, Ford LF, Nightingale MJ, Shannon PL, Davin JS, Bradford CO, et al. Care management of work injuries: results of a 1-year pilot outcome assurance program. J Occup Environ Med 2001; 43: 959–968.

36. Sullivan MJ, Stanish WD. Psychologically based occupational rehabilitation: the Pain-Disability Prevention Program. Clin J Pain 2003; 19: 97–104.

37. Linton SJ, Börsma K, Jansson M, Svärd L. The effects of cognitive-behavioral and physical therapy preventive interventions on pain-related sick leave: a randomized controlled trial. Clin J Pain 2005; 21: 109–119.

38. Medin J, Bendtsen P, Ekberg K. Health promotion and rehabilitation: a case study. Disabil Rehabil 2003; 25: 908–915.

39. Nathell L. Effects on sick leave of an inpatient rehabilitation programme for asthmatics in a randomized trial. Scand J Public Health 2005; 33: 57–64.

40. Silvennoinen A, Honkalampi K, Raty H. Kuntoutuminen masennusta sairastavilla kuntoutustutkimusasiakkailla [The rehabilitation of depressed patients]. Psykologia 2004; 39: 284–290 (in Finish).

41. Glamser FD, DeJong GF. The efficacy of preretirement preparation programs for industrial workers. J Gerontol 1975; 30: 595–600.

42. Heikkilä H, Heikkilä E, Eisemann M. Predictive factors for the outcome of a multidisciplinary pain rehabilitation programme on sick-leave and life satisfaction in patients with whiplash trauma and other myofascial pain: a follow-up study. Clin Rehabil 1998; 12: 487–496.

43. Granger B. The role of psychiatric rehabilitation practitioners in assisting people in understanding how to best assert their ADA rights and arrange job accommodations. Psych Rehabil J 2000; 23: 215–223.

44. Haugstvedt KT, Graff-Iversen S, Haugli L. Kan sykmeldte endre sjelvforståelse slik at arbedsevnen øker? [Could persons on sick leave learn to think differently about themselves and increase their participation in the workplace?]. Tidsskr Nor Laegeforen 2004; 124: 2885–2887 (in Norwegian).

45. Borchgrevink GE, Kaasa A, McDonagh D, Stiles TC, Haraldseth O, Lereim I. Acute treatment of whiplash neck sprain injuries: a randomized trial of treatment during the first 14 days after a car accident. Spine 1998; 23: 25–31.

46. Klanghed U, Svensson T, Alexanderson K. Positive encounters with rehabilitation professionals reported by persons with experience of sickness absence. Work 2004; 22: 247–254.

47. Oka M, Otsuka K, Yokoyama N, Mintz J, Hoshino K, Niwa Sl, et al. An evaluation of a hybrid occupational therapy and supported employment program in Japan for persons with schizophrenia. Am J Occupat Ther 2004; 58: 466–475.

48. Bengtsson K. Rehabilitation after myocardial infarction: a controlled study. Scand J Rehabil Med 1983; 15: 1–9.

49. Boulay FM, David PP, Bourassa MG. Strategies for improving the work status of patients after coronary artery bypass surgery. Circulation 1982; 66: III43–III49.

50. Burgess AW, Lerner DJ, D’Agostino RB, Vokonas PS, Hartman CR, Gaccione P. A randomized control trial of cardiac rehabilitation. Social Sci Med 1987; 24: 359–370.

51. Dennis C, Houston-Miller N, Schwartz RG, Ahn DK, Kraemer HC, Gossard D, et al. Early return to work after uncomplicated myocardial infarction: results of a randomized trial. JAMA 1988; 260: 214–220.

52. Engblom E, Hamalainen H, Ronnemaa T, Vanttinen E, Kallio V, Knuts LR. Cardiac rehabilitation and return to work after coronary artery bypass surgery. Qual Life Res 1994; 3: 207–213.

53. Fanello S, Frampas-Chotard V, Roquelaure Y, Jousset N, Delbos V, Jarny J, et al. Evaluation of an educational low back pain prevention program for hospital employees. Rev Rhum Engl Ed 1999; 66: 711–716.

54. Froelicher ES, Kee LL, Newton KM, Lindskog B, Livingston M. Return to work, sexual activity, and other activities after acute myocardial infarction. Heart Lung 1994; 23: 423–435.

55. Greenwood JG, Wolf HJ, Pearson RJ, Woon CL, Posey P, Main CF. Early intervention in low back disability among coal miners in West Virginia: negative findings. J Occup Med 1990; 32: 1047–1052.

56. Hofman-Bang C, Lisspers J, Nordlander R, Nygren A, Sundin O, Ohman A, et al. Two-year results of a controlled study of residential rehabilitation for patients treated with percutaneous transluminal coronary angioplasty: a randomized study of a multifactorial programme. Eur Heart J 1999; 20: 1465–1474.

57. Kellett KM, Kellett DA, Nordholm LA. Effects of an exercise program on sick leave due to back pain. Phys Ther 1991; 71: 283–291.

58. Loisel P, Abenhaim L, Durand P, Esdaile JM, Suissa S, Gosselin L, et al. A population-based, randomized clinical trial on back pain management. Spine 1997; 22: 2911–2918.

59. Malcolm RM, Harrison J, Forster H. Effects of changing the pattern of referrals in a local authority. Occup Med 1993; 43: 211–215.

60. Perk J, Hedback B, Engvall J. Effects of cardiac rehabilitation after coronary artery bypass grafting on readmissions, return to work, and physical fitness: a case-control study. Scand J Soc Med 1990; 18: 45–51.

61. Perkiö-Mäkelä M, Riihimäki H. Intervention on seat adjustment among drivers of forest tractors. Int J Ind Ergon 1997; 19: 231–237.

62. Pilote L, Thomas RJ, Dennis C, Goins P, Houston-Miller N, Kraemer H, et al. Return to work after uncomplicated myocardial infarction: a trial of practice guidelines in the community. Ann Intern Med 1992; 117: 383–389.

63. Schmidt SH, Oort-Marburger D, Meijman TF. Employment after rehabilitation for musculoskeletal impairments: the impact of vocational rehabilitation and working on a trial basis. Arch Phys Med Rehabil 1995; 76: 950–954.

64. Straaton KV, Harvey M, Maisiak R. Factors associated with successful vocational rehabilitation in persons with arthritis. Arthritis Rheum 1992; 35: 503–510.

65. van Doorn JW. Low back disability among self-employed dentists, veterinarians, physicians and physical therapists in The Netherlands: a retrospective study over a 13-year period (N = 1,119) and an early intervention program with 1-year follow-up (N = 134). Acta Orthop Scand Suppl 1995; 263: 1–64.

66. Mellin G, Harkapaa K, Vanharanta H, Hupli M, Heinonen R, Jarvikoski A. Outcome of a multimodal treatment including intensive physical training of patients with chronic low back pain. Spine 1993; 18: 825–829.

67. Rauscha F, Muller C, Kiss H, Mlczoch J, Schuster J, Weber H, et al. Arbeitswiederaufnahme nach Herzinfarkt. Wien Klin Wochenschr 1988; 100: 605–610.

68. Beutel M, Dommer T, Kayser E, Bleichner F, Vorndran A, Schluter K. Arbeit und berufliche Integration psychosomatisch Kranker – Nutzen und Indikation der beruflichen Belastungserprobung. Psychother Psychosom Med Psychol 1999; 49: 368–374.

69. Sheppeard H, Bulgen D, Ward DJ. Rheumatoid arthritis: returning patients to work. Rheumatol Rehabil 1981; 20: 160–163.

70. Hurri H. The Swedish back school in chronic low back pain: Part I: Benefits. Scand J Rehabil Med 1989; 21: 33–40.

71. Linton SJ, Bradley LA, Jensen I, Spangfort E, Sundell L. The secondary prevention of low back pain: a controlled study with follow-up. Pain 1989; 36: 197–207.

72. Hellsing AL, Linton SJ, Kälvemark M. A prospective study of patients with acute back and neck pain in Sweden. Phys Ther 1994; 74: 116–124.

73. Ryan WE, Krishan MK, Swanson CE. A prospective study evaluating early rehabilitation in preventing back pain chronicity in mine workers. Spine 1995; 20: 489–491.

74. Straaton KV, Maisiak R, Wrigley JM, Fine PR. Musculoskeletal disability, employment, and rehabilitation. J Rheumatol 1995; 22: 505–513.

75. Petrie KJ, Weinman J, Sharpe N, Buckley J. Role of patients’ view of their illness in predicting return to work and functioning after myocardial infarction: longitudinal study. BMJ 1996; 312: 1191–1194.

76. Reinhold D, Sohr C. Frührehabilitation nach Herzinfarkt – Ergebnisse und Schluβfolgerungen. Z Gesamte Inn Med 1988; 43: 698–700.

77. Karjalainen K, Malmivaara A, Mutanen P, Roine R, Hurri H, Pohjolainen T. Mini-intervention for subacute low back pain: two-year follow-up and modifiers of effectiveness. Spine 2004; 29: 1069–1076.

78. Stikeleather J. An older worker’s decision to “push or protect self” following a work-related injury. Work 2004; 22: 139–144.

79. Hayden JA, van Tulder MW, Malmivaara A, Koes BW. Exercise therapy for treatment of non-specific low back pain. Cochrane Database Syst Rev 2005; (3): CD000335.

80. Gibson PG, Powell H, Coughlan J, Wilson AJ, Abramson M, Haywood P, et al. Self-management education and regular practitioner review for adults with asthma. Cochrane Database Syst Rev 2003; (1): CD001117.

81. Hoffman BM, Papas RK, Chatkoff DK, Kerns RD. Meta-analysis of psychological interventions for chronic low back pain. Health Psychol 2007; 26: 1–9.

82. Schonstein E, Kenny DT, Keating J, Koes BW. Work conditioning, work hardening and functional restoration for workers with back and neck pain. Cochrane Database Syst Rev 2003; (1): CD001822.

83. Cieza A, Geyh S, Chatterji S, Kostanjsek N, Ustun B, Stucki G. ICF linking rules: an update based on lessons learned. J Rehabil Med 2005; 37: 212–218.

Comments

Do you want to comment on this paper? The comments will show up here and if appropriate the comments will also separately be forwarded to the authors. You need to login/create an account to comment on articles. Click here to login/create an account.
Advertisement