Introduction 2002 1998 2006 1994 1994 1999 1994 2003 1997 1982 1994 1996 1997 1997 2002 2005 1994 1997 1997 2003 2003 2000 1996 2006 2005 As it is very likely that the results of international studies are not comparable to the Dutch situation, the aims of this study were: (a) to investigate the mean (and median) duration of absenteeism due to depressive symptoms in the Dutch working population by sector and company size, and (b) to investigate gender and age influences on the mean (median) duration of absenteeism due to depressive symptoms. 2000 1993 1996 2000) 1996 2000) 2006 1996 2000) 2000 Methods Data The prevention, supervision and medical examination of sickness absence is a task of the employer for which he can choose to engage an occupational health department. ArboNed (the second largest occupational health department in the Netherlands) employs an absenteeism registration system for their affiliated companies. We register sickness absence and its causes in about 15% of the total Dutch working population. In our population, commercial services are over-represented (57% vs. 41%) and non-commercial services (e.g. health care, civil servants) are under-represented (17% vs. 34%) as compared to the total working population in the Netherlands. All absence periods that started between April 2002 and November 2005, and were diagnosed as depression were selected from the registration system. Absence periods were encoded as depression by the occupational physician when symptoms of depressed mood (such as feelings of profound sadness or emptiness) and/or reduced interest in activities that used to be enjoyed lasted for at least two weeks in combination with 3 or more of the following symptoms: decreased energy, decreased motivation, appetite changes, disturbed sleep pattern, agitation or inhibition, guilt feelings, low sense of self-worth, impaired concentration, or self-destructive thoughts. The symptoms must be accompanied by evident suffering and adverse effects on personal as well as social functioning. T df P Absence duration 1 Absence duration was not corrected for part-time return to work, unless mentioned otherwise. When the number of absence days is corrected for part-time return to work, this is done by dividing the number of absence days by 1/reintegration percentage. For example, when the employee returns to work for 50% the number of absence days is divided by 2. 2002 Statistical methods 1958 If the main effect was significant, the Tukey post hoc test was performed in order to determine which groups differed from each other. Results 1 2 df P Table 1 Number and course of depressive episodes  Men Women Total n % n % n % Number of employees 4,522  5,018  9,540  Number of absences 4,735 100.0 5,175 100.0 9,910 100.0 Returned to work within a year 3,205 67.7 3,355 64.8 6,560 66.2 One year of incapacity for work 719 15.2 903 17.4 1,622 16.4 Left the employment during the illness 307 6.5 380 7.3 687 6.9 Not returned to work at end of observation period 504 10.6 537 10.4 1,041 10.5 1 Fig. 1 Survival function of the duration of absence due to depression in men and women by age  2 Table 2 Mean (95% CI) and median (95% CI) duration of depressive episodes by age, company size and sector in men and women  Number of absences Mean (95% CI) Median (95% CI) Men Women Men Women Men Women Age   <35 years 1,301 2,105 184 (177–191) 204 (198–209) 153 (143–163) 182 (171–193)   35–44 years 1,610 1,729 200 (194–206) 217 (211–223) 181 (170–192) 206 (193–219)   45–54 years 1,357 1,081 212 (205–219) 223 (215–230) 199 (184–214) 223 (205–241)   ≥55 years 456 248 205 (193–217) 222 (205–238) 182 (163–201) 207 (163–251) Company size   <75 employees 1,349 1,214 214 (207–221) 226 (218–233) 198 (182–214) 229 (210–248)   75–500 employees 1,174 1,225 200 (192–207) 210 (203–217) 181 (167–195) 201 (187–215)   500–5,000 employees 1,421 2,032 196 (189–203) 213 (207–218) 174 (163–185) 200 (188–212)   >5,000 employees 531 423 188 (177–198) 208 (195–220) 170 (157–183) 179 (145–213) Sector   Construction industry 338 39 206 (192–220) 206 (174–239) 182 (151–213) 189 (133–245)   Health care 196 1,218 212 (194–230) 214 (207–221) 202 (163–241) 200 (184–216)   Trade 524 406 205 (194–216) 212 (199–224) 183 (163–203) 200 (171–229)   Catering industry 176 345 197 (177–216) 203 (189–217) 168 (136–200) 174 (144–204)   Industry 1,077 341 189 (182–197) 205 (192–219) 163 (149–177) 192 (167–217)   Education and public sector 165 196 232 (213–251) 242 (224–260) 236 (199–273) 272 (223–321)   Transportation and communication 417 332 196 (183–208) 208 (194–221) 182 (159–205) 177 (151–203)   Commercial services 645 682 213 (203–223) 219 (209–229) 196 (176–216) 212 (187–237)   Other/unknown 1,197 1,616 194 (186–201) 212 (206–218) 171 (160–182) 201 (188–214) Total 4,735 5,175 200 (196–204) 213 (210–217) 179 (172–186) 201 (193–209) In all age categories, women with depressive symptoms were absent longer than men. Elderly employees had a longer duration of absence and a higher risk of reaching disability. Employees in companies with less than 75 employees were absent for a longer period than employees working in large sized companies. Employees in the educational and public sector had the longest absence duration, followed by employees in commercial services and health care. Men working in the industrial sector had the shortest duration of absence due to depressive symptoms. 3 T df P F P Table 3 Mean (95% CI) number of absence days corrected for partial work resumption of depressive episodes by age, company size and sector in men and women  Men Women Total Mean (95% CI) Mean (95% CI) Mean (95% CI) Age    <35 years 140 (134–146) 156 (151–160) 150 (146–153)   35–44 years 153 (148–158) 169 (164–174) 161 (158–165)   45–54 years 164 (158–170) 176 (169–183) 169 (165–174)   ≥55 years 163 (152–173) 178 (163–192) 168 (160–176) Company size   <75 employees 166 (160–172) 176 (169–182) 171 (166–175)   75–500 employees 154 (148–161) 166 (159–172) 160 (156–165)   500–5,000 employees 150 (144–156) 163 (158–168) 158 (154–161)   >5,000 Employees 134 (125–142) 157 (146–168) 144 (137–151) Sector   Construction industry 164 (152–176) 160 (132–189) 164 (152–175)   Health care 168 (152–185) 167 (161–174) 168 (162–173)   Trade 158 (149–168) 166 (155–177) 162 (154–169)   Catering industry 153 (137–170) 163 (151–176) 160 (150–170)   Industry 143 (137–150) 157 (145–169) 147 (141–152)   Education and public sector 181 (162–199) 196 (179–213) 189 (177–201)   Transportation and communication 141 (130–151) 151 (139–163) 145 (138–153)   Commercial services 162 (154–171) 164 (155–172) 163 (157–169)   Other/unknown 151 (145–158) 166 (161–172) 160 (156–164) Total 154 (150–157) 165 (162–169) 160 (158–162) F P F P Discussion 1997 2003 1998 2003 2006 2000 1996 2000) 1996 2000) 1996 2000) 2003 1993 1995 2000 1996 In accordance with our third hypothesis (H3), the periods of absence due to depressive symptoms were shorter in large-scale companies than in smaller ones. The reintegration percentage is higher in large companies, resulting in a lower number of absence days. Probably, large companies have more opportunities for part-time return to work. Moreover, large companies have structured protocols as to how to deal with long-term absence from work. In the transportation and communications sectors, part-time return to work was most often observed. In our sample, these are mostly large companies in the postal and telecommunications sector, offering more reintegration possibilities. Although blue-collar workers generally have higher sickness absence than white-collar workers, depressed employees in industrial settings had shorter absence durations. A possible explanation might be that employees in the industrial sector are more susceptible to stigmatization of having mental problems, and return to work earlier. Also, in blue-collar workers, depressive symptoms may more often be masked by or diagnosed as somatic diseases. 2000 Duration of absence due to depression compared to other absenteeism studies 1996 2000 2003 2006 1996 2000 2003 2003 2003 Duration of a depressive episode compared to the duration of absence 2002 2003 2004 1997 1994 2002 The duration of a depressive episode in The Netherlands is comparable to other countries, whereas the duration of absenteeism is longer. It could be hypothesized that Dutch employees return to work when they are completely recovered, while in other countries employees start working (part-time) during recovery. We think that differences in social legislation and benefits contribute to these differences. In most other European countries compensation schemes apply to work-related diseases (“risque professional”). In The Netherlands, however, the social insurance system does not take into account the cause of the disease (“risque social”). Moreover, social benefits could have contributed to the relatively long absence duration. Dutch employers are obliged to pay the employee on sick-leave at least 70% of his/her salary. In almost all cases the full salary is paid during the first year of sick-leave, which does not motivate employees to return to their work. Limitations of the study Our population is not a random sample from the total Dutch working population, because employers voluntarily engage our occupational health department. Therefore, results are not representative of the whole Dutch working population. However, the considerable sample size and the use of registered, rather than self-rated information on sickness absence, are strong points of this study. 2003 2006 1996 2003 2004 Clinical implications Because of the risk of chronicity, it is important that occupational physicians recognize depressive symptoms in an early stage of the absence episode It is recommended to develop and apply tools for recognizing employees at risk for chronic depression Special attention should be given to employees in educational and public services, commercial services and health care as they are at risk for longer absence duration in case of depressive symptoms. Study limitations Diagnosis of depressive symptoms in absent employees may be biased towards absences with a longer duration. Mildly depressed employees, who did not report ill or returned to work before consulting the occupational physician, were not included Depressive symptoms were not strictly assessed according to DSM criteria, limiting the comparability to other studies Our absence registration system does not account for comorbidity.