Abstract

Background Low back pain (LBP) is among the most frequent causes of sickness absence in Norway, and it is thought that it could be reduced by 30–50% if present day knowledge was implemented in the workplace. Evidence-based interventions in occupational settings to prevent sickness absence are still lacking.

Aim To evaluate whether peer support would be able to modify general beliefs about LBP, pain experiences, health care utilization and sickness absence due to back pain.

Methods In addition to a media campaign in two Norwegian counties in 2002–05, aiming at improving beliefs about LBP in the general public, the ‘Active Back’ project trained a peer adviser in six participating workplaces. The task of this peer adviser was to provide information aimed at reducing fear of the pain, supportive advice and arrange for modifications of workloads, etc., for a limited period of time.

Results The prevalence of back pain remained constant throughout the study period, but self-reported intensity of LBP decreased at the end. There was a small decline in use of health care professionals and significant improvements in beliefs, in line with the messages of the campaign. Total sickness absence decreased by 27% and the LBP-related sickness absence by 49%.

Conclusion The combination of peer support and modified workload seemed to have additional effects to the general media campaign, and resulted in decline in sickness absence and improvements in beliefs about back pain.

Introduction

Low back pain (LBP) is among the most frequent causes of sickness absence in Norway [1]. Waddell and Burton [2] claim that sickness absence due to back pain may be reduced by 30–50% if present day knowledge was implemented in the workplace. The main message in this new knowledge, as stated in evidence-based guidelines [1–3], is to stay active and continue normal social and work activity, even if the back hurts. This reduces sickness absence due to back pain, and it also enhances recovery [2–4].

However, previous attempts to reduce absence due to LBP in occupational settings do not show any consistent positive effects, or any remarkable effect size. ‘Back schools’, lumbar supports, ergonomic interventions and risk factor modification have not yet demonstrated their effectiveness [3–6]. There is consensus that exercise is beneficial, but the effect size appears to be moderate [1,5,7]. So far we have not reached the goals set by Waddell and Burton. This may be due to a communication gap. The knowledge is new, and implementation hinges on dissemination as well as on acceptance of new treatment methods, by health care providers and by the general public.

The basis for the new interventions is the notion that misconceptions and negative beliefs about back pain constitute major reasons for sickness absence due to LBP. Beliefs about back pain precede sickness behaviour [8]. Low recovery expectations and fears that work may increase pain or cause harm are risk factors for chronic work disability [9,10]. The notion that the work itself produces pain is strong, and probably a major factor in fear of returning to work. While workplace studies have demonstrated that physical loading in an occupational setting may induce back pain [11], this does not necessarily imply any causal factor. Identical twin studies have shown very little impact on heavy physical loading over a lifetime on degenerative changes in the spine [12]. In recent years, the injury model for the origin of LBP has been questioned, and there is strong evidence that traditional biomedical education based on an injury model does not reduce future LBP and work loss [13].

In this paper, we examine a new attempt to bridge the communication gap between what we know and what is accepted by the employees with back problems. In two Norwegian counties, where there was a public information campaign on new ways to treat back pain, we also trained lay contacts to function as peer support. These lay contacts were ordinary employees, with personal experience with back pain. They were given the task of communicating new ways of treating back pain that should reduce the fear of pain, and were also given the task of modifying the work situation wherever possible. The peer support, therefore, had several aspects, emotional support and instrumental support in addition to information. Peers may actually be better in finding practical solutions than external ‘experts’. The aim of this study was to evaluate whether these lay contacts would be able to modify general beliefs about LBP, pain experiences, health care utilization and sickness absence due to back pain.

Methods

The ‘Active Back’ project took place in Vestfold and Aust-Agder counties in the southern part of Norway in 2002–05. The main target for the project was the general public, who were exposed to a media campaign involving a number of various media. In addition, the project also included specific activities for the health care providers and social security officers. The media campaign ran continuously with a specific website during the period, and at four times a particular pamphlet with advice for managing LBP and information about the project was distributed to all households in the two project counties. In addition, commercials were broadcasted on local TV, radio and cinemas. The health care providers were invited to one meeting in each county and provided with written material and posters to hang up at waiting rooms. The project is described in detail elsewhere (E.L. Werner, C. Ihlebæk, E. Lærum et al., submitted).

The project included a special intervention in six different workplaces with a total of 3500 employees (three from each intervention county) (Table 1). In addition to continuous contact with the general management in each company, this intervention consisted of

  • (i) Training and supervising one or several employees as peer advisers for back pain at each workplace,

  • (ii) Educational meetings with all employees and

  • (iii) Provision of written materials such as pamphlets and posters.

Table 1.

Demographics and response rates

No. Company branches No. of employees Response rate (%) Proportion of men (%) Mean age (SD) No. of years employed (SD) 
  2002 2004 2005 2002 2004 2005 2002 2004 2005 2002 2004 2005 2002 2004 2005 
Chemical industry; sales and production 731 759 726 65 40 41 74 67 71 44 (10.4) 45 (9.6) 46 (9.5) 15 (10.3) 17 (10.2) 17 (10.8) 
Electronics; sales and production 191 172 198 45 67 59 58 49 46 39 (9.4) 39 (10.3) 39 (9.4) 6 (3.8) 7 (4.5) 8 (5.0) 
Transportation/accommodation services 413 879 830 68 36 48 52 60 52 40 (10.8) 39 (12.0) 37 (11.6) 9 (6.8) 10 (7.1) 8 (7.0) 
Insurance/service to business 258 267 169 39 37 59 45 44 40 44 (10.1) 44 (9.3) 47 (8.5) 14 (9.6) 13 (9.6) 14 (10.0) 
Mechanical industry 282 237 150 45 22 40 93 85 95 43 (10.7) 44 (10.6) 46 (10.9) 13 (9.0) 17 (11.2) 19 (10.8) 
Community hospital 1343 1388 1381 67 61 38 19 21 19 43 (10.2) 44 (10.4) 44 (10.1) 11 (8.8) 12 (9.0) 13 (9.3) 
 Total 3218 3702 3372 61 43 45 45 43 44 43 (10.5) 43 (10.7) 42 (10.9) 12 (9.2) 12 (9.2) 13 (9.6) 
No. Company branches No. of employees Response rate (%) Proportion of men (%) Mean age (SD) No. of years employed (SD) 
  2002 2004 2005 2002 2004 2005 2002 2004 2005 2002 2004 2005 2002 2004 2005 
Chemical industry; sales and production 731 759 726 65 40 41 74 67 71 44 (10.4) 45 (9.6) 46 (9.5) 15 (10.3) 17 (10.2) 17 (10.8) 
Electronics; sales and production 191 172 198 45 67 59 58 49 46 39 (9.4) 39 (10.3) 39 (9.4) 6 (3.8) 7 (4.5) 8 (5.0) 
Transportation/accommodation services 413 879 830 68 36 48 52 60 52 40 (10.8) 39 (12.0) 37 (11.6) 9 (6.8) 10 (7.1) 8 (7.0) 
Insurance/service to business 258 267 169 39 37 59 45 44 40 44 (10.1) 44 (9.3) 47 (8.5) 14 (9.6) 13 (9.6) 14 (10.0) 
Mechanical industry 282 237 150 45 22 40 93 85 95 43 (10.7) 44 (10.6) 46 (10.9) 13 (9.0) 17 (11.2) 19 (10.8) 
Community hospital 1343 1388 1381 67 61 38 19 21 19 43 (10.2) 44 (10.4) 44 (10.1) 11 (8.8) 12 (9.0) 13 (9.3) 
 Total 3218 3702 3372 61 43 45 45 43 44 43 (10.5) 43 (10.7) 42 (10.9) 12 (9.2) 12 (9.2) 13 (9.6) 

The peer advisers were selected at each workplace as a non-medical person with considerable knowledge of the organization and the employees, such as a foreman, a union leader or a personnel officer. Whenever one of the employees had LBP, he or she was encouraged to present him or herself to the peer adviser before visiting a doctor or other health care provider. The peer adviser would then provide information aimed at reducing fear of the pain, supportive advice and arrange for modifications of workloads, etc., for a limited period of time. The main goal for the peer adviser was to encourage the employee with LBP to stay at work during the pain episode.

The peer advisers underwent specific training about LBP. They were taught about ‘red flags’ and ‘yellow flags’ and that they should not provide any kind of treatment, but only encourage the employee to stay active and at work while recovery would occur by itself. Any signs of red flags or insecurity about the pain would be referred to a medical doctor. Most often the employee himself would recognize the LBP from previous episodes. All peer advisers from all the cooperating workplaces were gathered together several times during the project to discuss the experiences and the project.

The overall message in the intervention was that the back is strong and LBP is not necessarily a sign of injury, but most likely caused by natural changes in the spine. Physical activities are not responsible for the pain and cannot cause any injury to the spine. Recovery from LBP takes its time and no treatment has been able to shorten this. Even though, it is often beneficial to alter workload in order to have a tolerable working day while one is waiting for spontaneous recovery.

All employees were given a questionnaire before (in April 2002), during (April 2004) and at the conclusion (June 2005) of the project. The questionnaires were delivered by the project manager personally to the companies where the personnel office or peer advisers provided the distribution and collection to and from each department. The questionnaire used in the survey was designed specifically for this study, and consisted of three parts. In addition to personal background data (age, gender, education, occupational status), we asked about personal experiences with LBP, behaviour at the last episode of acute LBP and nine statements concerning general attitudes towards LBP. The statements (some true and some false) were based partly on seven myths about LBP of Deyo [14] and reflected the messages of the campaign. The responses were recorded on a five-point Likert scale (1 = totally disagree, 2 = disagree, 3 = neither disagree nor agree, 4 = agree, 5 = totally agree). The prevalence of the two value categories in line with the messages of the campaign was calculated for 2002, 2004 and 2005. The statements are listed in Table 2.

Table 2.

Per cent of all responders responding in agreement with the messages of the campaign

Statements  2002 2004 2005 P-value 
Back pain recovers best by itself Agree/totally agree 13 17 20 <0.001 
Disc herniation requires surgery Disagree/totally disagree 15 21 29 <0.001 
Modern X-rays will usually find the cause of the back pain Disagree/totally disagree 27 37 45 <0.001 
LBP requires rest and tranquillity until recovery Disagree/totally disagree 42 57 58 <0.001 
Most often, back pain recovers by itself after some weeks Agree/totally agree 31 39 45 <0.001 
Everyone with back pain should have a spine X-ray Disagree/totally disagree 36 48 57 <0.001 
Bed rest is the mainstay of therapy Disagree/totally disagree 62 75 70 <0.001 
One recovers faster from back pain if one continues at work or returns as soon as possible Agree/totally agree 18 33 40 <0.001 
My perceptions about how to deal with back pain has changed the last year Agree/totally agree 14 17 17 0.015 
Statements  2002 2004 2005 P-value 
Back pain recovers best by itself Agree/totally agree 13 17 20 <0.001 
Disc herniation requires surgery Disagree/totally disagree 15 21 29 <0.001 
Modern X-rays will usually find the cause of the back pain Disagree/totally disagree 27 37 45 <0.001 
LBP requires rest and tranquillity until recovery Disagree/totally disagree 42 57 58 <0.001 
Most often, back pain recovers by itself after some weeks Agree/totally agree 31 39 45 <0.001 
Everyone with back pain should have a spine X-ray Disagree/totally disagree 36 48 57 <0.001 
Bed rest is the mainstay of therapy Disagree/totally disagree 62 75 70 <0.001 
One recovers faster from back pain if one continues at work or returns as soon as possible Agree/totally agree 18 33 40 <0.001 
My perceptions about how to deal with back pain has changed the last year Agree/totally agree 14 17 17 0.015 

Sickness absence was calculated as days of sick leave as a percentage of the potential maximum days at work per year in Table 3. The data on disability were obtained from the Norwegian National Insurance Administration (NIA) and reflected sickness absence due to all reasons and LBP, respectively. The diagnostic codes L02 (Back symptom/complaint), L03 (Low back symptom/complaint), L84 (Back syndrome without radiate pain) and L86 (Back syndrome with radiating pain) in the International Classification of Primary Care [15] were used to identify the disability data in the NIA.

Table 3.

Work absence in intervention companies and Norway in per cent of all possible days at work, as reported on sick leave certificate by doctor

 2001 2003 2005 Reduction in per cent 
Total work absence in intervention companies 5.77 5.74 4.24 −27% 
Work absence related to LBP in intervention companies 0.69 0.6 0.35 −49% 
Total work absence in Norway 6.6 7.33 5.75 −13% 
Work absence related to LBP in Norway 0.91 0.97 0.66 −28% 
 2001 2003 2005 Reduction in per cent 
Total work absence in intervention companies 5.77 5.74 4.24 −27% 
Work absence related to LBP in intervention companies 0.69 0.6 0.35 −49% 
Total work absence in Norway 6.6 7.33 5.75 −13% 
Work absence related to LBP in Norway 0.91 0.97 0.66 −28% 

In 2002, the Norwegian authorities introduced a particular cooperation concept between all work organizations with the specific aim to reduce sickness absence and disability [16]. This concept increased the number of days each individual could be absent from work without a doctor's certificate from 3 to 8 days. In this study, two of the six companies joined this concept already in 2002, while three more did so in 2003. Our results are thus based on sickness absence longer than 3 days in 2001, and longer than 8 days for five of the six companies in 2005, while the 2003 results are a mixture of the two.

SPSS 14.0 for Windows was used for statistical analyses. Only respondents with personal experiences of LBP were included in the study. Files of responses from each of the three questionnaires were merged into one file and the responses were recoded so that the highest value was in line with the project messages.

Results

The number of employees in the six firms increased by 15% during the first 2 years of the project, but then decreased so that the overall increase of employees from 2002 until 2005 was 5%. The response rate decreased from 61% in 2002 to 43% in 2004 and 45% in 2005.

In 2002, 71% of the respondents reported personal experience of LBP, while this percentage was 68% in 2004 and 67% in 2005 (P < 0.01). The point prevalence of LBP was also constant at the three measure points, when 13% reported to have LBP in 2002, 15% in 2004 and 13% in 2005 (P < 0.001). At the other end of the scale, 21% of the respondents reported it was >1 year since the last episode of LBP in 2002, 20% in both 2004 and 2005. For the intensity at their last episode of LBP, 34% reported minor problems in 2002 and 2004, 36% in 2005, whereas 20% reported severe back pain in 2002, 16% in 2004 and 13% in 2005 (P < 0.01).

The respondents were asked whether they visited any health professional at their last episode of LBP. In 2002, 34% reported not to have visited anyone, while this proportion was 37% in 2004 and 34% in 2005 (P = 0.26). The number who visited a doctor was 27, 22 and 18% (in 2002, 2004 and 2005, respectively) (P < 0.001). A chiropractor was visited by 18% in 2002 and 17% in 2004 and 2005 (P = 0.73), and a physiotherapist by 16, 13 and 12% (P < 0.05). The respondents were allowed to report contact with more than one provider, which may explain a stable proportion who visited anyone, while the number of visits to each of the professional groups seemed to decrease.

The employees were also asked about other self-coping strategies at their last episode of LBP, whether they increased or reduced any activities like exercises and training, leisure activities and use of medications. We were unable to find any significant change of such activities during the intervention period.

For all statements regarding beliefs about LBP, there was a significant improvement in line with the messages of the project (Table 2). However, only 17% of the respondents believed themselves that they had changed their perceptions towards LBP during the last year in 2004 and 2005, as compared with 14% in 2002 (P < 0.05).

There was a general decrease by 13% in sickness absence in Norway from 2001, which was the last whole year before the Active Back project took place, to 2005, which was the last year of the project (Table 3). In this period, the LBP-related sickness absence was reduced by 28% in the country. These numbers were nearly doubled in the project companies, as the total sickness absence was reduced by 27% and the LBP-related sickness absence by 49%.

Discussion

The reassurance by a fellow worker that back pain is not dangerous, that maintaining as much of normal social and occupational life as possible, and advice as how to manage the work, appears to have advantageous effects. Our study shows a significant improvement in beliefs about LBP among the employees with personal experiences of LBP in the six companies included in the study. There was a small decrease in the number of employees who visited a medical doctor or physiotherapist at their last episode of LBP. Self-coping strategies seemed unchanged during the project. The prevalence of LBP seemed stable during the intervention, but significantly fewer employees reported severe intensity of LBP in 2005 compared to 2002. The intervention took place at a time of general decrease in total sickness absence in Norway. However, reduction in total sickness absence and LBP-related sickness absence in these six companies was twice the reduction for the country as a whole, and substantially greater than what was achieved in the two project counties in general in the Active Back campaign (E.L. Werner, C. Ihlebæk, E. Lærum et al., submitted).

The results of this study should be interpreted with caution because of the absence of a control group. The intervention itself had relatively low costs. However, to run this as a proper, randomized control study far exceeds the funds available to us. We also do not know the exact number of injured workers who actually sought advice from the peer advisers, which prevents us measuring their direct impact. Several activities took place at the workplaces simultaneously, i.e. meetings, pamphlets and posters in addition to the media campaign in the same areas. The direct influences of each of these activities are not possible to measure in this study. Also the low response rate could be a limitation of the study. The demographics and rate of personal experiences of LBP imply though that the responders are representative of the employees in these firms. Although all respondents were anonymous, they were aware that the questionnaires were a part of the scientific evaluation of the project, which may have influenced the responses in order to be more positive than they really would be in a state of LBP.

Our intervention was based on contemporary evidence-based guidelines [1]. The main message is to stay active and avoid catastrophic thinking. It hurts, but LBP happens regardless of physical activities. However, there are still resistant factors among health providers and among the general public. Even if guidelines [1] state that the injury model is not useful in the explanation of common LBP, most patients are still told by their health care provider to be careful. Clinicians that have not left the injury model as the medical basis for their management of LBP patients may actually contribute to a prolonging LBP episode.

Negative beliefs about LBP have been found to be strong predictors of LBP disability [17]. However, although a media campaign in the area improved beliefs about LBP in the general public, this did not change health behaviour (E. L. Werner, C. Ihlebæk, E. Lærum et al., submitted). The peer support may represent a valuable tool to change beliefs as well as behaviour. Peer support (the buddy system) has proven efficient in some cases where change in health-related behaviours is the goal. Peer support has been found to be efficient for behaviours ranging from breastfeeding initiation [18] to prevention of migraine attacks [19]. As a remedy for smoking cessation, the results are inconsistent [20].

The emphasis on peer support to close the communication gap has avoidance of health professionals as a deliberate ‘side effect’. The idea of LBP as a medical condition due to some sort of injury may be counterproductive. The individual becomes a patient, and all parties concentrate on finding an organic lesion that may not exist, or there is no reason to treat. The communication gap may be easier to close if the information is given by a peer, granted that this peer has the necessary knowledge. The peer has another advantage to the health professional, he or she knows what the work situation really is, and may find better practical solutions.

An intriguing aspect of all these models was the combination of reassurance and delivery of modified work. Could the results have been achieved by modifications of work alone? The evidence is conflicting. In a review from 1998, the workers who were offered modified work returned to work twice as often as others [21]. However, a randomized trial from 2006 could not confirm this [22], and an ‘active sick leave’ program that enabled the employees to return to a modified work during their sickness absence could neither measure any economic benefits from this [23]. It is along with social sciences literature that transition from a change of attitudes into behavioural change is dependent on practical interventions in addition to the re-education [24,25].

Funding

Partly supported by the the Norwegian Directorate for Health and Social Affairs, and partly by the owner of the project, the Hospital of Rehabilitation, Stavern.

Conflicts of interest

None declared.

The authors thank Holger Ursin for valuable supervision and comments and Research Technician Nina Konglevoll for technical assistance—both at the Research Unit, Unifob Health, Bergen. We also thank Øyvind Sørbrøden for supervising the management of the project.

References

1.
The Norwegian Back Pain Network—The Communication Unit
Acute Low Back Pain—Multidisciplinary, Clinical Guidelines
 , 
2002
Oslo, Norway
Nasjonalt Ryggnettverk
 
[in Norwegian]
2.
Waddell
G
Burton
KA
Concepts of rehabilitation for the management of low back pain
Best Pract Res Clin Rheumatol
 , 
2005
, vol. 
19
 (pg. 
655
-
670
)
3.
Airaksinen
O
Brox
JI
Cedraschi
C
, et al.  . 
European guidelines for the management of chronic nonspecific low back pain
Eur Spine J
 , 
2006
, vol. 
15
 
Suppl. 2
(pg. 
S192
-
S300
)
4.
Hagen
EM
Eriksen
HR
Ursin
H
Does early intervention with a light mobilization program reduce long-term sick leave for low back pain?
Spine
 , 
2000
, vol. 
25
 (pg. 
1973
-
1976
)
5.
Tveito
TH
Hysing
M
Eriksen
HR
Low back pain interventions at the workplace: a systematic literature review
Occup Med (Lond)
 , 
2004
, vol. 
54
 (pg. 
3
-
13
)
6.
Verbeek
JH
van der Weide
WE
van Dijk
FJ
Early occupational health management of patients with back pain
Spine
 , 
2002
, vol. 
27
 (pg. 
1844
-
1851
)
7.
Linton
SJ
van Tulder
MW
Preventive interventions for back and neck pain problems: what is the evidence?
Spine
 , 
2001
, vol. 
26
 (pg. 
778
-
787
)
8.
Woby
SR
Watson
PJ
Roach
NK
, et al.  . 
Are changes in fear-avoidance beliefs, catastrophizing, and appraisals of control, predictive of changes in chronic low back pain and disability?
Eur J Pain
 , 
2004
, vol. 
29
 (pg. 
1818
-
1822
)
9.
Turner
JA
Franklin
G
Fulton-Kehoe
D
, et al.  . 
Worker recovery expectations and fear-avoidance predict work disability in a population-based workers' compensation back pain sample
Spine
 , 
2006
, vol. 
31
 (pg. 
682
-
689
)
10.
Haldorsen
EMH
Kronholm
K
Skouen
JS
Ursin
H
Predictors for outcome of a multimodal cognitive behavioural treatment program for low back patients—a 12 months follow-up study
Eur J Pain
 , 
1998
, vol. 
4
 (pg. 
293
-
307
)
11.
Waters
TR
Dick
RB
Davis-Barkley
J
Krieg
EF
A cross-sectional study of risk factors for musculoskeletal symptoms in the workplace using data from the General Social Survey (GSS)
J Occup Environ Med
 , 
2007
, vol. 
49
 (pg. 
172
-
184
)
12.
Battie
MC
Videman
T
Gibbons
LE
Fisher
LD
Manninen
H
Gill
K
Volvo Award in clinical sciences. Determinants of lumbar disc degeneration. A study relating lifetime exposures and magnetic resonance imaging findings in identical twins
Spine
 , 
1995
, vol. 
20
 (pg. 
2601
-
2612
)
13.
Waddell
G
Burton
AK
Occupational health guidelines for the management of low back pain at work: evidence review
Occup Med (Lond)
 , 
2001
, vol. 
51
 (pg. 
124
-
135
)
14.
Deyo
R
Low-back pain
Sci Am
 , 
1998
, vol. 
279
 (pg. 
29
-
33
)
15.
World Organization of Family Doctors
ICPC-2 International Classification of Primary Care
 , 
1998
2nd edn
Oxford
Oxford University Press
 
Prepared by the International Classification Committee of WONCA
16.
Nergaard
K
Agreement reached on sickness absence and an inclusive working life
2001
 
17.
Schultz
IZ
Crook
JM
Berkowitz
J
, et al.  . 
Biopsychosocial multivariate predictive model of occupational low back disability
Spine
 , 
2002
, vol. 
27
 (pg. 
2720
-
2725
)
18.
Hoddinott
P
Chalmers
M
Pill
R
One-to-one or group-based peer support for breastfeeding? Women's perceptions of a breastfeeding peer coaching intervention
Birth
 , 
2006
, vol. 
33
 (pg. 
139
-
146
)
19.
Merelle
SY
Sorbi
MJ
Passchier
J
The preliminary effectiveness of migraine lay trainers in a home-based behavioural management training
Patient Educ Couns
 , 
2006
, vol. 
61
 (pg. 
307
-
311
)
20.
Park
EW
Schultz
JK
Tudiver
F
Campbell
T
Becker
L
Enhancing partner support to improve smoking cessation
Cochrane Database Syst Rev
 , 
2004
, vol. 
3
  
CD002928
21.
Krause
N
Dasinger
LK
Neuhauser
F
Modified work and return to work: a review of the literature
J Occup Rehabil
 , 
1998
, vol. 
8
 (pg. 
113
-
139
)
22.
Seenstra
IA
Anema
JR
Bongers
PM
de Vet
HCW
Knol
DL
van Mechelen
W
The effectiveness of graded activity for low back pain in occupational healthcare
Occup Environ Med
 , 
2006
, vol. 
63
 (pg. 
718
-
725
)
23.
Scheel
I
Hagen
KB
Herrin
J
Carling
C
Oxman
AD
Blind faith? The effects of promoting active sick leave for back pain patients. A cluster-randomized trial
Spine
 , 
2002
, vol. 
27
 (pg. 
2734
-
2740
)
24.
Ajzen
I
Fishbein
M
Attitude-behaviour relations: a theoretical analysis and review of empirical research
Psychol Bull
 , 
1977
, vol. 
84
 (pg. 
888
-
918
)
25.
Eagly
AH
Chaiken
S
The Psychology of Attitudes
 , 
1993
San Diego, CA
Harcourt Brace Jovanovich