Occupational Medicine 2014;64:56–63Advance Access publication 10 December 2013 doi:10.1093/occmed/kqt142

© The Author 2013. Published by Oxford University Press on behalf of the Society of Occupational Medicine. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Factors influencing return to work after illness in France

C. Pélissier1,2, L. Fontana1,2 and F. Chauvin2,3

1Department of Occupational Health, University Hospital, Saint Etienne 42055, France, 2Department of Occupational Health, University of Jean Monnet, Saint Etienne 42000, France, 3Lucien Neuwirth Institute of Cancer Research, Saint Priest en Jarez 42270, France.

Correspondence to: C. Pélissier, Department of Occupational Health, University Hospital, Saint Etienne 42055, France. Tel: +33 (0)477828325; fax: +33 (0)477828139; e-mail carole.pelissier@chu-st-etienne.fr

Background Few studies have been published about the factors influencing return to work after sickness absence.

Aims To identify medical and occupational factors influencing the type of fitness certificate given by occu-pational physicians before employees return to work after sickness absence.

Methods A cross-sectional study was undertaken over 3 months in several health services in France. Workers undergoing a medical examination before returning to work after a period of sickness absence of at least 3 weeks were included. Medical and occupational factors were collected using a questionnaire. The relationship between different factors and certification of fitness was assessed by univariate and multivariate analyses.

Results Among the 402 workers included, 64% were considered fit to return to work. Being older, strenuous work, prolonged sick leave and fear of returning to work appeared to be negative factors influencing the return to a previous job. In contrast, having an education level higher than secondary school, being satisfied at work, perception of very good health and benefitting from satisfactory professional relationships appeared to favour return to work. We developed a predictive score of not being fit to return to work after illness.

Conclusions Our study highlighted the relationship between medical and occupational factors with problems returning to work. The predictive score may be used by occupational physicians as a screening tool to identify those who are likely to have difficulties returning to work after illness, so that their working conditions can be modified to take this into consideration.

Key words Certificate of fitness; determining factors; medical examination before returning to work; return to work; score; sick leave.

Introduction

Return to work following illness has been reported to be dependent on medical and occupational factors [1–7]. While certain factors appear to be obstacles, others appear to be facilitators [1–9]. A delayed, partial or non-return to work has major consequences for both the indi-vidual and society [3,4]. However, to our knowledge, few studies have assessed the role of these factors and most of these studies have considered subjects with the same illness [1–4].

In France, occupational physicians are at the heart of the process of maintaining employees in employment after a period of sick leave. Under French law, employees

have to see an occupational physician before returning to work following a period of sickness absence. At the medi-cal examination before return to work, the physician delivers a certificate of fitness to return to the previously held job. This certificate takes into account the health status of the employee, i.e. possible physical or psycho-logical sequelae of the illness, as well as the job-related constraints. Thus, the occupational physician can certify employees fit to return to their previous employment, but with adjustments. The physician can also recommend a change in position in the company, so that it is better adapted to an employee’s health status. However, if no acceptable solution is possible within the company, the employee can be made redundant. Every year in France,

at Monash U

niversity on Decem

ber 5, 2014http://occm

ed.oxfordjournals.org/D

ownloaded from

C. PéLISSIER ET AL.: FACTORS INFLUENCING RETURN TO WORK 57

~700 000 people have restrictions on their fitness for their job because of health problems and between 20 000 and 40 000 people lose their job following illness [10].

For these reasons, we undertook a survey on a popu-lation of workers to identify medical and occupational factors, including socio-professional and psychosocial factors, influencing the type of certificate of fitness given by the occupational physician at the medical visit before their return to work.

Methods

A descriptive, cross-sectional, multicentre study was undertaken between March and June 2009 using a questionnaire. Our target population was composed of workers from the agricultural, industrial and ser-vices sectors who were followed up by the occupational health services (OHS) in St Etienne, France, and the surrounding region. Subjects presenting at the medical visit before returning to work after a period of sickness absence of at least 21 days were included; this visit has been obligatory under French labour laws since 2009. Those presenting at the medical examination follow-ing a workplace accident or professional illness were excluded from the study.

A questionnaire covering the medical and occupational factors, including socio-professional, and psychosocial factors likely to influence the type of certificate of fitness was developed for this study. The factors were chosen based on an analysis of published studies related to the return to work following illness: after cancer [3], after low back pain [2,4–8] and after myocardial infarction [9]. In addition, we also used factors from the questionnaires of previous French surveys, SUMER, SAMOTRACE and Inadéquation Santé-Travail [11–13].

Job satisfaction was assessed by asking the subject their degree of agreement with the following sentence from the SUMER survey: ‘on the whole, I’m satisfied with my job’ [11]. Every occupational physician who took part in the study received the study protocol and was required to propose the questionnaire to every employee satisfy-ing the inclusion criteria. After giving written informed consent, the subjects anonymously completed the first part of the questionnaire during the medical visit, and the occupational physician completed the second part.The questionnaire covered the following factors:

• Socio-professional factors: gender, age, number of children, highest degree, social security cover, socio-professional category, company size, type of work contract, actual number of hours worked per week, length of time in the company, employment sector and position held.

• Occupational risks (chemical, physical, psychoso-cial, biological, manual handling risks): physical difficulty of work, autonomy at work, certificate of

fitness, possible workplace adjustments and steps taken by the occupational physician following the certificate of fitness.

• Psychosocial factors: feelings about the idea of returning to work, level of job satisfaction, level of satisfaction about the quality of relationship with their employer and colleagues, contact with people in the company during sick leave and psychologi-cal support.

• Medical factors: duration of sick leave, self-assess-ment of health status, disabled worker status, disability status, recognition of occupational ill-ness, comorbidities, previous deficits, new deficits related to the illness, length of illness, means of informing the employee’s physician about the ill-ness, treatment and attendance to a medical visit prior to return to work. Visual analogue health scales were used to assess general health status (1: very poor health; 10: very good health).

• The number of subjects who refused to partici-pate and their reasons were anonymously col-lected by the occupational physicians. The data from the questionnaires were entered using SAS version 9.1 by one person who verified that the data entry was accurate for each questionnaire. As this was a cross-sectional, observational study without human experimentation, ethical commit-tee approval was not required for this study.

The International Classification of Diseases (ICD-10), International Standard Classification of Occupations (ISCO-88) and Anatomical Therapeutic Chemical (ATC) classification systems were used to code quali-tative data collected in the questionnaires [14–16]. The quantitative variables were converted into dummy vari-ables by transforming them into ordinal qualitative vari-ables. The variable ‘certificate of fitness for the previously held position’ was the explained variable. We distin-guished this variable from the other types of certificate, i.e. fit for the same job with restrictions, fit but requiring modification of duties and unfit. The socio-professional and medical characteristics of the population were ana-lysed with descriptive methods. We performed univari-ate analysis of the factors likely to influence the type of certificate of fitness using the chi-squared test (SAS 9.1), with the significance level set at P ≤ 0.05. Variables that were significantly associated with the certificate of fitness were introduced into a stepwise logistic regres-sion model. A variable ‘score’ was created to estimate the probability of not being fit to return to work, using the main predictive variables from the multivariate analyses. The score was externally validated with a second set of 103 employees included from June to August 2009 with the same inclusion criteria. The values for each variable in the score were verified for both the first and second set of employees.

at Monash U

niversity on Decem

ber 5, 2014http://occm

ed.oxfordjournals.org/D

ownloaded from

58 OCCUPATIONAL MEDICINE

Results

In total, 422 subjects, recruited by 30 occupational physicians, completed the questionnaire. We were able to analyse 402 questionnaires: two subjects refused to participate, and the duration of sickness absence was insufficient for 18 subjects. The population included 215 women and 187 men, with a mean age of 42.3 years (SE = 0.57).

The socio-professional categories found in our group were predominantly labourers (43%) (Table  1). Most of the subjects had not completed secondary school. Almost 90% of subjects had a permanent work contract. More than two-thirds of the subjects worked >35 h per week. Nearly half of the subjects worked in a company with at least 200 employees and more than two-thirds were exposed to risks related to manual handling.

The mean duration of sick leave was 17.7 weeks. The subjects’ mean rating of their health status was 6.7 on a visual Analogic Scale (VAS) ranging from 1 to 10. Two-thirds of subjects had had their health problems for <1 year. One-third of the subjects already had a health problem prior to their present illness, predominantly involving the musculoskeletal system. Overall, 90% of the subjects received treatment during their sick leave and 58% were still taking treatment for their illness at the visit.

Regarding the idea of returning to work, 37% (n = 149) of the subjects were enthusiastic, 33% (n = 134) fear-ful and 28% (n = 113) were indifferent. Eighty per cent (n  =  319) reported that they were satisfied with their work. More than two-thirds were happy with their rela-tionship with their employer (73%, n = 291) and 91% were satisfied with their relationship with their colleagues (n = 362).

During sick leave, 74% (n = 296) of the subjects main-tained contact with their colleagues, 38% (n = 153) with their employer, but only 12% (n = 50) with the occupa-tional physician. Overall, 35% (n = 143) of subjects had psychological support during their sick leave, with 23% (n = 105) being supported by their family, 20% (n = 83) by friends and 15% (n = 63) by a professional.

At the visit before returning to work, 65% (n = 257) of subjects were deemed to be ‘fit for their previous job’, 24% (n = 96) ‘fit, but requiring certain restrictions or modifications to their post’ and 11% (n = 45) were ‘unfit for work’. When modifications were required, two-thirds required modification of working hours, and almost one-third required physical modifications; almost one in five required modifications in their tasks.

In univariate analysis, both medical and occupational factors were associated with the certificate of fitness delivered by the occupational physician (Tables 2 and 3): older than 53 years of age, education level lower than the end of secondary school, permanent contract, previous health problems, new health problems, disabled worker

status, treatment, fear at the idea of returning to work and job dissatisfaction. Where the results of medical tests were forwarded to the occupational physician, this was negatively associated with return to the previously held post. However, working for a company with at least 200 employees, not performing physical tasks at work, having a recent health problem, being cured, having had short-term sickness absence were all positively associated with return to the previously held post. Health status assessed as good by the subject and having good social support were also positively associated with return to previous employment.

Multivariate analysis found four variables remained statistically significant (Table  4): (i) self-assessment of health status, (ii) duration of sickness absence, (iii) edu-cation level and (iv) results of medical tests forwarded to the occupational physician.

A tool was developed, using logistic regression, to calculate a score for the risk of not being ‘fit for their previously held job’. This tool was elaborated using a combination of the four variables that were statistically significant in the multivariate analysis (Table 5). Using this tool, the higher the score, the lower the risk of not being fit for their previously held job. The external vali-dation process confirmed the validity of the initial multi-variate model and the significant association of the four factors to predict being fit to return to a previous job.

Discussion

In our study, we identified factors, e.g. lack of job sat-isfaction, fear of returning to work and unsatisfactory relationship with the employer, education level lower than the end of secondary school and having a physically demanding post, that were associated with the risk of not being fit to return to the previous job. In the multivariate analysis, only the following four factors remained cor-related with difficulties of returning to a previous post: (i) self-perception of health, (ii) the duration of sickness absence, (iii) educational level and (iv) having results from additional tests related to their illness forwarded to the occupational physician.

To our knowledge, only a few studies, none in France, have reported the factors influencing the certification of fitness given by occupational physicians [9,17–19]. Our survey had the advantage of including workers with diverse health problems and coming from very different employment sectors. In addition, our participation rate was high (96%). However, we were not able to assess the association between the type of certification given and the health problems responsible for the sickness absence, as our study population was underpowered for this. In addi-tion, not all the eligible employees were invited to par-ticipate due to refusal from the occupational physicians. Also, certain physicians may not have proposed the study

at Monash U

niversity on Decem

ber 5, 2014http://occm

ed.oxfordjournals.org/D

ownloaded from

C. PéLISSIER ET AL.: FACTORS INFLUENCING RETURN TO WORK 59

systematically to all the eligible employees during the con-sultation, due to organizational constraints. The question-naire, which was based on questions validated in other studies (SAMOTRACE, SUMER), contained certain questions requiring a subjective response, which could have introduced information bias. Any potential bias was minimized by the adjustments in the multivariate analysis.

The results from our survey are coherent with those reported in the literature. Few studies have reported that lower education level was negatively associated with returning to a previously occupied position [20,21].

Our study highlighted the favourable role of good social support on the resumption of work. Social support

was found to be an important work-related prognostic factor associated with return to work for workers with low back pain, with cancer or with distal lower extremity fractures [7,18,22].

Our study highlighted the favourable role of job satisfaction on the return to work following sickness absence. The results from a meta-analysis of almost 500 studies showed a strong relation between job satisfac-tion and both mental and physical health [23]. Another study assessing the views of vocational rehabilitation professionals found that work motivation and positive attitude to returning to work of the subjects on sick-ness absence were associated with early return to work

Table 1. Socio-professional description of study population

Socio-professional variables n (%)

Socio-professional categories Labourers 170 (43) Employees 104 (26) Intermediate professionals 53 (13) Other 69 (18)Employment sectors Health, social work and education 123 (31) Industry 106 (26) Construction 50 (12) Other 123 (31)ISCO-88 classification 1 Directors, executive management and managers 8 (2) 2 Intellectual and scientific professions 18 (5) 3 Intermediate professions 80 (22) 4 Administrative employees 34 (9) 5 Service personnel for private individuals, retailers and shopkeepers 69 (19) 6 Farmers and qualified labourers in agriculture, forestry and fishing sectors 2 (1) 7 Skilled trades in industry and crafts 81 (22) 8 Machine operators and assemblers 28 (7) 9 Elementary professions 50 (13)Highest degree Level less than or equal to the end of secondary school 315 (79) Level higher than secondary school 83 (21)Work contract Permanent 359 (89) Other 42 (11)Actual number of hours per week <35 h 103 (26) ≥35 h 291 (74)Company size ≥200 employees 180 (45) 50–199 employees 92 (23) 10–49 employees 89 (22) <10 employees 41 (10)Professional risks Manual handling risks 274 (68) Physical risks 135 (34) Psychosocial risks 130 (32) Accident risks 83 (21) Chemical risks 79 (20) Biological risks 78 (19) None of these propositions 22 (6)

at Monash U

niversity on Decem

ber 5, 2014http://occm

ed.oxfordjournals.org/D

ownloaded from

60 OCCUPATIONAL MEDICINE

[24]. A vocational, multidisciplinary rehabilitation pro-gramme for patients on long-term sickness absence has been reported to improve return to work [25].

In our study, a good result from the self-assessment of health status had a positive influence on returning to work. A  study on work resumption following myocar-dial infarction reported that patients who believed their illness had serious consequences had more difficulty returning to work [26]. The results from other studies confirmed the role of illness perception and subjective complaints on the worker’s fear of returning to work after a long illness [27,28]. In another study, the roles of the worker’s health perception and expectation of recovery in relation to the resumption of work were highlighted [29]. In addition, this study reported the influence, to a lesser degree, of the professional function, co-worker support and workplace arrangement [29]. Based on a literature review of the prognostic factors for relapse, chronicity and non-resumption of work in patients suf-fering from common low back pain, poor general health was reported to be a negative prognostic factor [5]. In a study involving subjects with back pain in Quebec, a

number of modifiable determinants for work resump-tion in good health were identified, i.e. beliefs and fears about work, functional incapacities, feeling of personal effectiveness and certain features of the patient–physi-cian relationship [30].

Our study highlighted the relationship between med-ical and occupational factors and difficulty returning to work. Some of these factors, such as the workers’ self-assessment of health status, fear of resuming work and job satisfaction, appeared to be factors that could be modified to promote work resumption. Job satisfac-tion is dependent on a number of factors and, thus, appears to be the most difficult to change. However, the employee’s understanding of their illness and their fitness to return to work could be improved. Improving social support and maintaining contact with colleagues from the company during sick leave are factors that can be modified in order to reduce the fear of returning to work. In France, although employees are offered the possibility of meeting their occupational physician dur-ing sickness absence, with the principal aim of preparing their return to work, they rarely accept the opportunity.

Table 2. Univariate analyses of individual and medical factors significantly associated with fitness to return to a previously held job

Individual and medical factors Categories Certificate of fitness

Fit for the previously held job

Fit with restriction, workplace arrangement or unfit

n (%) n (%)

Age (years)** ≤32 76 (74) 26 (26)≥33 to <45 61 (62) 37 (38)≥45 to <53 66 (67) 33 (33)≥53 years 54 (54) 45 (46)

Previous health problems*** Yes 64 (49) 66 (51)No 193 (72) 75 (28)

New health problems*** Yes 29 (31) 64 (69)No 228 (75) 77 (25)

Prior recognition of disabled worker status*** Yes 10 (28) 26 (72)No 247 (68) 115 (32)

Treatment administered during sickness absence*** Yes 225 (63) 132 (37)No 32 (78) 9 (22)

Treatment ongoing after sickness absence*** Yes 127 (55) 103 (45)No 130 (77) 38 (23)

Duration of illness (years)*** <1 187 (74) 64 (25)>2 42 (49) 44 (51)1–2 26 (45) 31 (55)

Evolution of the condition*** Cured 100 (91) 10 (9)Stabilized 98 (54) 84 (46)In remission 14 (50) 14 (50)Other 34 (47) 26 (43)

Duration of sickness absence*** ≤8 weeks 152 (76) 48 (24)>8 weeks 105 (53) 93 (47)

Self-assessment of health status*** <8/10 121 (52) 111 (48)≥8/10 136 (82) 30 (18)

Results of medical examination forwarded to occupational physician***

Yes 55 (47) 61 (53)No 202 (72) 80 (28)

*P < 0.05; **P < 0.01; ***P < 0.001.

at Monash U

niversity on Decem

ber 5, 2014http://occm

ed.oxfordjournals.org/D

ownloaded from

C. PéLISSIER ET AL.: FACTORS INFLUENCING RETURN TO WORK 61

Other studies are necessary to improve our understand-ing of the relevant factors and, therefore, improve our ability to offer guidance to prevent socio-professional exclusion.

We found four factors that remained significantly associated with returning to a previously held post in the multivariate analysis including all the individual, medi-cal and professional variables. Using these factors, we developed a tool to predict the risk of not being fit to return to work after illness. This tool is simple to use and only requires knowing the self-assessment of the subject’s health status, the duration of sickness absence, the subject’s education level and if the results of medical

examinations were forwarded to an occupational physi-cian. This score may be used by occupational physicians as a screening tool to identify subjects at risk of having difficulties returning to work before they return to work, and thus enable the adjustment of their working con-ditions to ensure return to work. This score could also be used by GPs to identify subjects at risk of difficul-ties when returning to work in order to advise them to contact their occupational physician. Future studies are needed to assess the usefulness of the score for the early detection of employees requiring reorganization of their post or change of post in order to improve the prepara-tion of their return to work.

Table 3. Univariate analyses of occupational factors associated with fitness to return to a previously held job

Occupational factors Categories Certificate of fitness

Fit for the previously held job

Fit with restriction, workplace arrangement or unfit

n (%) n (%)

Highest education level*** Less than or equal to end of secondary school

136 (57) 103 (43)

Higher than secondary school 119 (77) 36 (23)Company size (number of employees)** ≥200 130 (73) 49 (27)

10–59 56 (64) 32 (36)50–199 51 (57) 39 (43)<10 20 (49) 21 (51)

Permanent contract* Yes 223 (63) 132 (37)No 34 (79) 9 (21)

Professional risks Physical* Yes 77 (58) 56 (42) No 180 (68) 85 (32)Biological* Yes 59 (76) 19 (24) No 198 (62) 122 (38)Manual handling* Yes 272 (61) 106 (39) No 126 (72) 35 (28)

Physical nature of work Physical exertion** Yes 159 (60) 106 (40) No 93 (73) 34 (27)

Feelings about returning to work*** Fearful 65 (49) 68 (51)Enthusiastic 104 (71) 42 (29)Indifferent 86 (76) 27 (24)

Job satisfaction** Yes 213 (67) 103 (33)No 40 (51) 38 (49)

Satisfaction about the quality of relationship with the employer*

Unsatisfied 21 (48) 23 (52)Rather unsatisfied 37 (60) 25 (40)Rather satisfied 147 (69) 66 (31)Very satisfied 47 (63) 27 (37)

Contact maintained during sickness absence

Colleagues** Yes 201 (69) 91 (31) No 56 (53) 50 (47)Occupational physician*** Yes 12 (24) 38 (76) No 245 (70) 103 (30)

*P < 0.05; **P < 0.01; ***P < 0.001.

at Monash U

niversity on Decem

ber 5, 2014http://occm

ed.oxfordjournals.org/D

ownloaded from

62 OCCUPATIONAL MEDICINE

Key points

• The results from this study show that both medi-cal and occupational factors play a role in the employees’ fitness to return to work after sickness absence.

• We developed a quick and easy-to-use predictive score to assess the risk of employees having dif-ficulty returning to work.

• This score could be used by occupational physi-cians to assess the risk of inappropriate return to work for a given employee allowing modification of working conditions on return to work.

Funding

Saint Etienne University Hospital.

Acknowledgements

The authors would like to thank all of the occupational phy-sicians and the subjects who agreed to participate in this

study and Margaret Haugh (MediCom Consult) for editorial assistance.

Conflicts of interest

None declared.

References

1. Berglind H, Gerner U. Motivation and return to work among the long-term sicklisted: an action theory perspect-ive. Disabil Rehabil 2002;24:719–726.

2. Krause N, Dasinger LK, Deegan LJ, Rudolph L, Brand RJ. Psychosocial job factors and return-to-work after compen-sated low back injury: a disability phase-specific analysis. Am J Ind Med 2001;40:374–392.

3. Spelten ER, Sprangers MA, Verbeek JH. Factors reported to influence the return to work of cancer survivors: a litera-ture review. Psychooncology 2002;11:124–131.

4. Fayad F, Lefevre-Colau MM, Poiraudeau S et  al. Chronicité, récidive et reprise du travail dans la lombalgie: facteurs communs de pronostic. Annal Réadapt et Méd Phys 2004;47:179–189.

Table 4. Multivariate analysis of medical and occupational prognostic factors for being fit to return to a previously held job

Fit for the previously held job

OR 95% CI

Self-assessment of health status by worker (visual analogue health scale)*** <8/10 1*** ≥8/10 4.11 2.48–6.80Duration of sickness absence** >8 weeks 1** ≤8 weeks 2.45 1.53–3.91Highest degree obtained** Less than to end of secondary school 1** More than or equal to end of secondary school 2.28 1.38–3.74Results of medical examination forwarded to occupational physician** Yes 1** No 2.14 1.31–3.55

CI, confidence interval; OR, odds ratio.*P < 0.05; **P < 0.01; ***P < 0.001.

Table 5. Scoring rules for the tool to assess the risk of not being fit for a previously held job

Subscore

Yes No

Q1. Self-assessment of health status by employee (VAS ≥ 8/10) 1.5 0Q2. Duration of sick leave: ≤8 weeks? 0.5 0Q3. Highest degree obtained: more than or equal to high school? 0.5 0Q4. Results of medical examinations forwarded to occupational

physician?0 0.5

Score Q1 + Q2 + Q3 + Q4 0–0.5 1–1.5 2–3Risk of not being fit for a previously job High Moderate Low

at Monash U

niversity on Decem

ber 5, 2014http://occm

ed.oxfordjournals.org/D

ownloaded from

C. PéLISSIER ET AL.: FACTORS INFLUENCING RETURN TO WORK 63

5. Steenstra IA, Verbeek JH, Heymans MW, Bongers PM. Prognostic factors for duration of sick leave in patients sick listed with acute low back pain: a systematic review of the literature. Occup Environ Med 2005;62:851–860.

6. Hoogendoorn WE, Bongers PM, de Vet HC, Ariëns GA, van Mechelen W, Bouter LM. High physical work load and low job satisfaction increase the risk of sickness absence due to low back pain: results of a prospective cohort study. Occup Environ Med 2002;59:323–328.

7. Heymans MW, de Vet HC, Knol DL, Bongers PM, Koes BW, van Mechelen W. Workers’ beliefs and expectations affect return to work over 12  months. J Occup Rehabil 2006;16:685–695.

8. Anema JR, Cuelenaere B, van der Beek AJ, Knol DL, de Vet HC, van Mechelen W. The effectiveness of ergonomic interventions on return-to-work after low back pain; a pro-spective two year cohort study in six countries on low back pain patients sicklisted for 3–4 months. Occup Environ Med 2004;61:289–294.

9. Varaillac P, Sellier P, Iliou MC, Corona P, Prunier L, Audouin P. Return to work following myocardial infarc-tion. Medical and socio-professional factors. Arch Mal Coeur Vaiss 1996;89:203–209.

10. CINERGIE. Agir contre l’exclusion professionnelle des personnes handicapées. http://handitrav.fr/cinergie/asso-ciation.html (27 November 2013, date last accessed).

11. Enquête Sumer 2003. http://travail-emploi.gouv.fr/etudes-recherches-statistiques-de,76/statistiques,78/con-ditions-de-travail-et-sante,80/les-enquetes-surveillance-medicale,1999/l-enquete-sumer-2003,15982.html (27 November 2013, date last accessed).

12. Cohidon C, Murcia M et al. Samotrace-volet « épidémi-ologie et entreprise », résultats intermédiaires à 1 an (3000 questionnaires), Régions Centres, Pays de la Loire et Poitou-Charentes (2007). Santé et travail 1–4. http://www.invs.sante.fr/publications/2007/samotrace/samotrace.pdf (27 November 2013, date last accessed).

13. Enquête inadéquation Sante-Travail. http://travail-emploi.gouv.fr/IMG/pdf/Inadequation_Sante_-_Travail_2005_-_depliant.pdf (27 November 2013, date last accessed).

14. CIM-10 Version 2008. http://apps.who.int/classifica-tions/icd10/browse/2008/fr (27 November 2013, date last accessed).

15. Anatomical therapeutic chemical (ATC). http://www.whocc.no/atc/structure_and_principles/ (27 November 2013, date last accessed).

16. CITP Classification Internationale Type de Professions. http:// www.ilo.org/public/french/bureau/stat/isco/ (27 November 2013, date last accessed)

17. Slebus FG, Kuijer PP, Willems JH, Sluiter JK, Frings-Dresen MH. Prognostic factors for work ability in

sicklisted employees with chronic diseases. Occup Environ Med 2007;64:814–819.

18. MacKenzie EJ, Morris JA Jr, Jurkovich GJ et al. Return to work following injury: the role of economic, social, and job-related factors. Am J Public Health 1998;88:1630–1637.

19. Di Lorenzo L, De Metrio R, De Francesco G, Pirris A, Gigante MR. Criteria and modalities of assessment of work fitness for jobs in food industry. Med Lav 1999;90:473–483.

20. Choi KS, Kim EJ, Lim JH et al. Job loss and reemployment after a cancer diagnosis in Koreans — a prospective cohort study. Psychooncology 2007;16:205–213.

21. Abbott J, Berry N. Return to work during the year follow-ing first myocardial infarction. Br J Clin Psychol 1991;30 (Pt 3):268–270.

22. Taskila T, Lindbohm ML, Martikainen R, Lehto US, Hakanen J, Hietanen P. Cancer survivors’ received and needed social support from their work place and the occupa-tional health services. Support Care Cancer 2006;14:427–435.

23. Faragher EB, Cass M, Cooper CL. The relationship between job satisfaction and health: a meta-analysis. Occup Environ Med 2005;62:105–112.

24. Dekkers-Sánchez PM, Wind H, Sluiter JK, Hw Frings-Dresen MH. What promotes sustained return to work of employees on long-term sick leave? Perspectives of voca-tional rehabilitation professionals. Scand J Work Environ Health 2011;37:481–493.

25. Braathen TN, Veiersted KB, Heggenes J. Improved work ability and return to work following vocational multidisci-plinary rehabilitation of subjects on long-term sick leave. J Rehabil Med 2007;39:493–499.

26. Petrie KJ, Weinman J, Sharpe N, Buckley J. Role of patients’ view of their illness in predicting return to work and func-tioning after myocardial infarction: longitudinal study. Br Med J 1996;312:1191–1194.

27. Oyeflaten I, Hysing M, Eriksen HR. Prognostic fac-tors associated with return to work following multi-disciplinary vocational rehabilitation. J Rehabil Med 2008;40:548–554.

28. Dekkers-Sanchez PM, Wind H, Sluiter JK, Frings-Dresen MH. What factors are most relevant to the assessment of work ability of employees on long-term sick leave? The physicians’ perspective. In Arch Occup Environ Health 2013;86:509–518.

29. Schultza IZ, Crook J, Meloche GR et al. Psychosocial fac-tors predictive of occupational low back disability: towards development of a return-to-work model. J Occup Rehabil 2003;12:233–256.

30. Dionne C, Bourbonnais R, Frémont P, Rossignol M, Stock S. Le pronostic occupationnel des travailleurs aux prises avec des affections vertébrales. Etudes et Recherches 2003;R-356:1–147.

at Monash U

niversity on Decem

ber 5, 2014http://occm

ed.oxfordjournals.org/D

ownloaded from