BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

Beskæftigelsesudvalget 2019-20
BEU Alm.del Bilag 307
Offentligt

Journal Pre-proof

Working time characteristics and long-term sickness absence: a large

Ann Dyreborg Larsen , Annina Ropponen , Johnni Hansen ,

Ase Marie Hansen , Henrik A. Kolstad , Aki Koskinen ,

¨ ¨

Mikko I. Harma , Anne Helene Garde

PII:

DOI:

Reference:

To appear in:

Received date:

Revised date:

Accepted date:

S0020-7489(20)30123-1

https://doi.org/10.1016/j.ijnurstu.2020.103639

NS 103639

International Journal of Nursing Studies

15 November 2019

1 May 2020

2 May 2020

Please cite this article as: Ann Dyreborg Larsen , Annina Ropponen , Johnni Hansen ,

¨ ¨

Ase Marie Hansen , Henrik A. Kolstad , Aki Koskinen , Mikko I. Harma , Anne Helene Garde ,

Working time characteristics and long-term sickness absence: a large register-based

study of Danish and Finnish nurses,

International Journal of Nursing Studies

(2020), doi:

https://doi.org/10.1016/j.ijnurstu.2020.103639

This is a PDF ﬁle of an article that has undergone enhancements after acceptance, such as the addition

of a cover page and metadata, and formatting for readability, but it is not yet the deﬁnitive version of

record. This version will undergo additional copyediting, typesetting and review before it is published

in its ﬁnal form, but we are providing this version to give early visibility of the article. Please note that,

during the production process, errors may be discovered which could affect the content, and all legal

disclaimers that apply to the journal pertain.

2020 Published by Elsevier Ltd.

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

Working time characteristics and long-term sickness absence: a large register-based study of Danish and

Finnish nurses

Ann Dyreborg Larsen

, Annina Ropponen

, Johnni Hansen

, Åse Marie Hansen

a,d

, Henrik A. Kolstad

, Aki

Koskinen

, Mikko I. Härmä

, Anne Helene Garde

a,d

The National Research Center for the Working Environment, Lersoe Parkalle 105, DK-2100 Copenhagen,

Denmark

Finnish Institute of Occupation Health, P.O.Box 40, FI-00032 Työterveyslaitos, Finland. @MikkoHaermae,

@AnninaRop

Danish Cancer Society Research Center, Strandboulevarden 49, DK-2100 Copenhagen, Denmark

Department of Public Health, Copenhagen University, Gothersgade 160, DK-1123 Copenhagen, Denmark

Department of Occupational Medicine, Danish Ramazzini Centre, Palle Juul-Jensens Boulevard 99, Aarhus

University Hospital, DK-8200 Aarhus, Denmark

*Corresponding author: Dr. Ann Dyreborg Larsen, The National Research Centre for the Working Environment,

Lersoe Parkalle 105, DK-2100 Copenhagen. Email: [email protected], phone: +45 39 16 54 77, fax: +45 39 16 52

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

Abstract

Background:

Working time regimes in Denmark and Finland share many similarities such as nursing personnel

working in highly irregular shift systems. Yet, there are also differences for example in policy on when and how

the employer are compensated for sickness absence.

Objective:

We aimed to investigate the association between different working hour characteristics and long-

term sickness absence and whether these associations differed within various age groups in two large datasets

of nursing personnel from Denmark and Finland.

Design:

Based on objective payroll data we used Poisson regression models to calculate incidence rate ratios

with 95 % confidence intervals to prospectively assess the risk of long-term sickness absence in relation to

annual working hour characteristics. The analyses were adjusted for age, sex, short-term sickness absence, and

weekly working hours.

Setting(s):

Danish and Finnish nursing personnel.

Participants:

31 729 Danish and 6970 Finnish nursing personnel with

. Whole-Time

Equivalent, registered

in the database 1 year, 18-67 years of age with less than 30 days sickness absence in baseline year 2008.

Methods:

Working

hour characteristics were assessed for 2008: time of day; day; evening; night. Duration of

shift; long shifts (9-12 hours); very long shifts (12-24 hours); quick returns (< 11 hours between two shifts); long

weeks (> 40 hours/week); very long weeks (> 48 hours/week); and

o se uti e ight shifts

5 night shifts).

Long-term sickness absence was assessed as first incidence of 30 or more consecutive days off in 2009-2015.

Results:

The Danish data showed having evening work or five or more consecutive night shifts were associated

with higher risk of long-term sickness absence. When excluding pregnant women, night work was also

associated to higher risk of sickness absence. When stratifying on age groups, we observed a lower risk of

sickness absence in the youngest age groups and a higher risk among the oldest. The Finnish results showed a

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

higher risk of sickness absence when working nights, longs shifts, quick returns, and long work weeks. When

stratifying on age groups, the results showed similar tendencies as the Danish.

Conclusions:

The results show that the scheduling of working hours is likely to affect the risk of long-term

sickness absence and that the risk differs in different age groups. No consistent picture was found for the

results from Denmark and Finland. Differences may be due to contextual differences thus comparison of risk of

sickness absence in relation to working hours between countries should be performed with caution.

Tweetable abstract:

A recent study from Denmark and Finland shows higher risk for long sickness absence

among nurses with five or more consecutive night shifts

What is already known about the topic?



Previous studies indicate an association between working time arrangements and sickness absence but

evidence is not well-established as some studies also report no associations or decreased risk of

sickness absence.



With many studies based on self-reported working time and sickness absence there is a call for studies

of high quality with objective, detailed assessment of shift work exposure for example by use of payroll

data.

What this paper adds



Results show that the scheduling of working hours is likely to affect risk of long-term sickness absence

and that the risk is higher among older nurses.



There were differences in the association of working hours and sickness absences between Denmark

and Finland which may be due to differences in compensation policy.

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

Keywords:

cohort study; health personnel; prospective study; shift work; shift work schedule; sick days; sick

leave

Background

Objective

Shiftwork including nights and long working hours are common in a society with 24 hours’ availability. Based on

numbers from the European Working Condition Survey (2015), shift and night work were estimated to affect

nearly 20 % of the working population where long weekly working hours could affect up to 40 % (1). Shift and

night work have been associated with cardiovascular diseases, diabetes, injuries and cancer (2-6) whereas long

working hours have been associated with cardiovascular diseases and depression and sleep disturbances (7-9).

In turn, some evidence indicates that shift and night work and long working hours are associated with both

short- and long-term consequences of the chronic diseases such as sickness absence (10-12).

Increasing numbers of days with sickness absence from work due to illness is an increasing problem in most

European countries. In Denmark and Finland, numbers of days with sickness absence per year have increased

from 6.7 (Denmark) and 8.4 (Finland) in 2000 to 9.0 (Denmark) and 9.3 (Finland) in 2016 (13). Sickness absence

and long-term sickness absence in particular have large social consequences in terms of lost productivity and

economical compensations (14) as well as individual consequences in terms of affected health, social exclusion,

financial difficulties and exit from the workforce (15-17). Further, previous studies have shown long-term

sickness absence to be a predictor of early retirement and disability pension (18, 19). Denmark and Finland

share to a large extent similar working life structures, working hour regulations and health care system.

However, previous studies including payroll data from Denmark and Finland revealed differences in working

hour characteristics and distribution of type of work schedules (20, 21). Both Denmark and Finland are covered

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

by the Nordic welfare model including a national social security scheme for sickness absence but there are

differences e.g. in relation to compensation, disability pensions and when an employer can discharge an

employee.

Some studies find that work time schedules are related to sickness absence (12, 22, 23), but the associations

between work time schedule, including shift and night work, long shifts (i.e. long working hours) and long

weekly working hours, are not well-established as other studies have found no associations or decreased risk of

sickness absence (24-26). In the review by Merkus

et al,

the authors considered evidence in regards to sickness

absence to be inconclusive with respect to rotating shifts, shift work including nights, fixed night work, and for

8-hour and 12-hour shifts. Call for studies of high quality with detailed assessment of shift work exposure are

thus needed (27). Studies using objective working hour data has increased in number during recent years, but

they are still scarce in studies on sickness absence (12, 22, 28).

Therefore, based on objective high quality data

a aila le fro

e ployers’ pay-roll

based registers with

detailed information on working hour characteristics, the present study aims to investigate the association

between timing and length of work shifts, short time between shifts (quick returns), number of consecutive

night shifts, and weekly working hours and the risk of long-term sickness absence (long-term sickness absence,

30 consecutive days) among female nursing personnel in the public health care sector in Denmark and

Finland. Further, we want to test whether the associations would differ in various age groups. As the literature

suggest an association between several working hour characteristics and sickness absence, we hypothesize that

time of day, duration of shifts and work weeks along with work patterns and number of consecutive night shifts

are associated with risk of long-term sickness absence. Further, we hypothesize that the risk increase with the

intensity of the shifts: the more shifts per year, the higher risk and with age as age can reflect longer exposure

time.

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

More specifically, the research questions were:

Are working hour characteristics in shift work a predictor for long-term sickness absence? Is time of day

(evening shift, night shift), duration of shift (long shift

hours a d <

hours

, very

lo g shift

hours

and

< 24 hours)) or work patterns (quick returns (< 11 hours between two shifts), long weeks (calendar weeks with

> 40 hours/week), very long weeks (calendar weeks with > 48 hours/week); consecutive night shifts (periods of

fi e

night shifts or more in a row) associated with increased risk of long-term sickness absence? Do the

associations differ between age groups?

Methods

In the current study, we analyze payroll data from Denmark and Finland for the associations between working

hour characteristics and risk of long-term sickness absence and further, to test if these associations differ in

different age groups.

Data

Data sources comprise data from two large databases: the Danish Working Hour Database (29) and the

Working Hours in the Finnish Public Sector Study database (30). Both databases include detailed information

on working time and absence from work based on daily information from payroll data.

The Danish Working Hour Database is a nationwide database with payroll data from the five administrative

Danish regions from 2007-2015 with a total of 265 702 unique participants. The regional employees include all

public hospital employees. For each participant, the database includes data on starting and ending time for all

shifts, age, sex, profession and absence from work (sickness absence, holiday, parental leave, etc.).

The Working Hours in the Finnish Public Sector Study data used in this study includes payroll data from five

hospital districts and workers from one social health care department of one town (The Finnish Public Sector

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

Study hospital cohort). Data on working time was retrieved from the historical records of the shift scheduling

programme: Titania® (CGI Finland) for 2008-2015 (30). A total of 40 351 unique participants were included.

Study population

To ensure homogeneous study populations for comparison between countries, the datasets were harmonized

and we restricted to nurses and assistant nurses only. Further, we restricted to participants with at least a 50 %

time position to exclude those having part-time jobs due to ill health. Participants were included if they had

been registered in the database at least one year to ensure that all participants contributed equally to the

exposure assessment and were between 18 and 67 years of age. To minimize the risk of recurrent sickness

absence, we excluded all participants who had 30 days or more sickness absence in their first year in the cohort

(2008). The flowcharts for the Danish and the Finnish data is shown in figure 1. Although varying in size, the

two cohorts share many similarities shown as percentage of excluded participants except when it comes to the

number of participants with less than 50 % positions which is more common in the Danish data than in the

Finnish data.

Figure 1 Flowchart Danish and Finnish Data

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

Working hours

Working hours were coded similarly both in Denmark and Finland based on the joint definitions (20, 30).:

Exposure to different working hour characteristics were calculated as 12-month average values based on 2008

data. In the Danish data, when comparing 2008 values of working hour characteristics with the mean values of

2008-2012, estimates generally lies within 5 percentage points why there is no indication that 2008 differs in

regards to the distribution of shifts compared to other years. Similarly for the Finnish data, as comparison

between years showed working time variables within individuals to be relatively stable in the years 2008-2013

(30)

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

Shifts were classified according to

time of day

including dayshifts (defined as shifts starting after 06:00 hours

and ending before < 21:00), evening shifts (at least 3 hours

et ee

a d<

a d ight

shifts (at

least 3 hours between 23:00–06:00 (both included)). The definitions of day, evening and night shifts were not

mutually exclusive but prioritized with night work as the highest priority, then evening and lowest day work.

Shifts were also classified according to duration with

long shifts

defi ed as shifts lasti g

hours) and

very

long shifts

hours a d <

hours

and < 24 hours). Further, we also looked

quick returns

(defined as less

than 11 hours between two shifts),

long weekly working hours

(calendar weeks with > 40 hours/week);

very

long weekly working hours

(calendar weeks with > 48 hours/week) and

consecutive night shifts

(periods of five

or more consecutive night shifts).Exposure information was summed by individuals and grouped into

categories:

day shift

(0, 1-100, 101-200 and >200 shifts/person/year);

evening shift, night shift, long shift, very

long shift, long weeks,

very

long weeks,

and quick

returns

(0, 1-12, 13-50, and >50 per person/year); and

consecutive night shifts (number of periods with five or more)

(0, 1-12 and >12 per person/year).

Outcome

Long-term sickness absence was defined as 30 consecutive days of sickness absence, and data was obtained

from the Danish Working Hour Database and the Working Hours in the Finnish Public Sector Study database in

the years 2009-2015. The analyses were restricted to the first incidence of long-term sickness absence.

Covariates

The included covariates were based on pay-roll data available in the Danish Working Hour Database and the

Working Hours in the Finnish Public Sector Study database: Age ( 30 years, >30 and 40 years, > 40 and 50

years, > 50 years of age), sex (male/female), short-term (<30 consequent days) sickness absence in 2008

(yes/no) (due to the Danish legislation where the municipality and job centers take actions after 30 days) were

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

treated as categorical variables, and mean weekly working hours in 2008 which were treated as a continuous

variable.

Statistical methods

We performed the statistical analyses for Denmark and Finland separately, and used Poisson regression (zero-

inflated) models

–

fulfilling the requirements for this type of model - to calculate the incidence rate ratio (IRR)

with 95% confidence intervals (CI) to prospectively assess the risk of long-term sickness absence in relation to

working hour characteristics. Annual working hour characteristics were calculated for 2008 and analyses ran to

first occurrence of long-term sickness absence in 2009-2015 like previous studies (20, 30).

Three statistical models were tested: Model 1: Working hour characteristics in 2008 and long-term sickness

absence in 2009-2015, adjusted for age and sex. Model 2: Model 1 with further adjustment for short-term

sickness absence in 2008, and model 3: Model 2 with further adjustment for weekly working hours (not for the

analysis of long and very long weekly working hours). The stepwise adjustment did not alter the results

significantly, and we therefore only present data from model 3 in the tables. Full tables are presented in the

supplementary digital content (A and B). Further, we stratified on age groups (>30, 31-40, 41-49, <50 years of

age).

We included secondary analyses where we excluded pregnant women (only applicable for Danish data since

this information was not available in the Finnish data). We hypothesized that some long-term sickness absence

might be related to pregnancy. If ever a registration of parental leave, the woman was identified as having

been pregnant and excluded. And further, in a sensitivity analysis we adjusted for organizational unit (only

applicable for the Finnish data) as this might reflect different work tasks and activity levels. However, the risk

estimates were similar in direction and magnitude as in the model 3, hence we chose not to present these

results (data not shown).

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

All analyses were conducted in SAS v. 9.4 (Danish data) and Stata 15.1 MP (Finnish data).

Ethical approval

The databases were approved for research use by the following local data protection agencies: The Danish Data

Protection Agency (2015-57-0074) and the ethics committee of the Hospital District of Helsinki and Uusimaa

(HUS 1210/2016). The need for individual written consent by participants is deemed unnecessary according to

national regulations in Denmark (Databeskyttelsesloven, nr 502 af 23/05/2018) and in Finland when approved

by the ethics committee of the Hospital District of Helsinki and Uusimaa (HUS 1210/2016).

Results

The descriptive socio-demographic characteristics of the Danish and the Finnish study populations are

presented in table 1. The cohorts consisted primarily of women (88-93%) with a mean age of 42 and 47 years,

respectively. With respect to the Danish data, the distribution of age in the smallest group were those under

30 years of age and in the largest group those between 40-50 years of age (19.6 and 31.7 %, respectively).

Similarly, for the Finnish data, the distribution of age in the smallest group were those under 30 years and in

the largest groups those above 50 years of age (6.6 and 43.7 %, respectively). Both cohorts had short-term

sickness absence of 7-8 days in 2008 (mean: 6.7 days (Denmark), 8.0 days (Finland)). Weekly working hours

were on average 29 hours in the Danish data and 32 hours in the Finnish data. Table 1 further presents the

distributions of the working hour characteristics showing similarities in distributions between the two study

populations in most of the exposure categories but quick returns were more common in Finland and

consecutive night shifts were most common in Denmark.

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

Table 1 Socio-demographic and work schedules variables.

Denmark

Mean

100

88.4

11.6

42.3

6227

8463

10051

6988

19.6

26.7

31.7

22.0

6.7

5.6

8.0

29.0

6.1

32.0

6.3

10.6

463

1734

1726

3047

6.6

24.8

24.7

43.7

Finland

Mean

100

93.0

7.0

46.9

11.2

Total population

Sex

Women

Men

Age in 2008

<= 30 years

>30 & <=40 years

> 40 & <=50 years

> 50 years

Short-term sickness absence (sum of

spells with less than 30 consequent

days) in 2008

Weekly working hours in 2008

Working hour characteristics

Day (least 3h 06:00-and <21:00)

0 shift/person/year

1-100

101-200

>200

Evening (least 3h 18:00 and < 02:00)

0 shift/person/year

1-12

13-50

>50

Night (least 3h 23:00 and 06:00)

0 shift/person/year

1-12

13-50

>50

- < 12 h)

0 shift/person/year

1-12

13-50

>50

Very Long shifts

h - < 24 h)

0 shift/person/year

1-12

13-50

>50

Quick returns (<11 h between shifts)

0 shift/person/year

1-12

13-50

>50

Long weeks (>40h/week)

0 weeks/person/year

1-12

13-50

>50

Very long weeks (>48h/week)

0 weeks/person/year

Long shifts

31729

28034

3695

6970

6485

485

986

20752

7632

2359

9587

8363

8905

4874

15972

8217

5258

2282

3.1

65.4

24.1

7.4

30.2

26.9

28.1

15.4

50.3

25.9

16.6

7.2

155

1982

4062

771

2205

2274

2473

1762

2728

2330

150

2.2

28.4

58.3

11.1

31.6

32.6

35.5

0.3

25.2

39.0

33.4

2.0

7233

16009

6615

1872

18704

11022

1867

136

12062

16150

3382

135

8999

19480

3250

19024

22.8

50.5

20.8

5.9

59.0

34.7

5.9

0.4

38.0

50.9

10.7

0.4

28.4

61.4

10.2

60.0

683

3251

2883

153

3071

3397

438

331

1073

4375

1191

868

3788

2314

3103

9.8

46.6

41.4

2.2

44.0

48.7

6.3

0.9

4.7

15.3

62.8

17.0

12.5

54.3

33.2

44.5

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

1-12

13-50

>50

Consecutive night shift (

0 person/year

1-12

>12

12181

524

29448

1784

497

38.4

1.7

92.8

5.6

1.6

3688

179

6585

340

52.9

2.6

93.8

5.6

0.6

Table 2 presents the results from the Poisson regression analyses of the Danish Working Hour Database and

the Working Hours in the Finnish Public Sector Study database on working hour characteristics and incidence

rate ratio (IRR) for long-term sickness absence. In both databases, day shifts were associated with lower risk of

long-term sickness absence, only statistically significant in the Danish data though. In the Danish Working Hour

Database, working more than 50 evening shifts per year were associated with higher risk of long-term sickness

absence. Working less than 50 night shifts per year seems to be associated with lower risk on long-term

sickness absence. Working very long shifts, quick returns, long weeks and very long weeks were all associated

with lower risk of long-term sickness absence, whereas 12 or more spells a year of consecutive night shifts of

five or more shifts were associated with higher risk of long-term sickness absence. In the Working Hours in the

Finnish Public Sector Study database more than 50 night shifts per year were associated with a higher risk of

long-term sickness absence which also counted for long shifts, quick returns and long weeks.

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

Table 2 Poisson regression analyses with Incidence rate ratios for the association of working hour characteristics and 1

incidence of

long-term sickness absence adjusted for age, sex, previous short-term sickness absence and weekly working hours in Danish and

Finnish data. *not adjusted for weekly working hours. Bold marking represent statistically significant values at a 0.05 level

1st incidence of long-term sickness absence

Denmark

Finland

Incide

Person

95%

Person

Incidence

nce

Cases

Years at

Confidence

Years at risk Rate Ratio

Rate

risk

Iimits

Ratio

Day

least h

<21:00)

0 shift/person/year 304

4720

Ref

308

ref

1-100 6021

110166

0.65

0.50-0.85

308

9706

0.70

0.39- 1.24

101-200 2114

41427

0.52

0.40-0.69

947

24898

0.92

0.52- 1.64

>200 532

13286

0.39

0.29-0.53

127

4224

0.75

0.41- 1.37

Evening (least 3h

18:00 and < 02:00)

0 shift/person/year

2684

51117

Ref

457

12299

Ref

1-12 2237

45455

0.96

0.86-1.06

460

12474

1.05

0.92- 1.20

13-50 2493

48058

1.00

0.90-1.09

475

14264

0.94

0.83- 1.07

>50 1557

24970

1.18

1.06-1.34

0.43

0.11- 1.71

Night (least 3h 23:00

and 06:00)

0 shift/person/year

4661

83744

Ref

449

11844

ref

1-12 2219

44998

0.87

0.79-0.95

464

13095

1.03

0.90- 1.17

13-50 1409

29031

0.89

0.81-0.98

433

13250

1.02

0.89- 1.16

>50 682

11826

1.11

0.97-1.27

947

1.41

1.05- 1.90

Long shifts

- < 12

0 shift/person/year

2012

38132

Ref

2619

ref

1-12 4315

85691

0.99

0.90-1.09

662

18530

1.04

0.84- 1-29

13-50 1835

35654

0.89

0.80-1.00

588

17065

1.06

0.86- 1.32

>50 509

10122

0.88

0.75-1.04

922

1.42

1.00- 2.01

Very Long shifts

- < 24 h)

0 shift/person/year

5398

98527

Ref

623

16974

ref

1-12 3008

59971

0.91

0.84-0.98

695

19707

1.02

0.92- 1.14

13-50 520

10419

0.92

0.80-1.06

2009

0.93

0.71- 1.22

>50 45

683

1.44

0.90-2.31

446

1.33

0.81- 2.19

Quick returns (<11 h

between shifts)

0 shift/person/year 3415

63258

Ref

886

ref

1-12 4609

87267

0.96

0.89-1.04

116

3492

1.38

0.87- 2.21

13-50 916

18334

0.88

0.78-0.99

924

25250

1.62

1.04- 2.53

>50 31

740

0.65

0.37-1.13

333

9504

1.65

1.04- 2.60

Long

weeks*(>40h/week)

0 shift/person/year 2532

47120

Ref

103

3497

ref

1-12 5575

104555

0.94

0.86-1.03

855

22013

1.40

1.13- 1.74

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

13-50

>50

Very long weeks*

(>48h/week)

0 shift/person/year

1-12

13-50

>50

Consecutive night shift (

0 person/year

1-12

>12

864

17925

0.76

0.67-0.86

436

13626

1.24

0.98- 1.58

5419

3395

157

8281

524

166

100569

66290

2741

157783

9460

166

0.91

1.03

1.11

1.88

Ref

0.84-0.98

0.79-1.34

Ref

0.96-1.29

1.43-2.48

543

837

1330

15315

23439

383

37753

2110

272

1.08

1.12

1.14

1.08

ref

0.97- 1.21

0.66- 1.92

ref

0.90-1.43

0.59-1.95

When stratifying by age groups (table 3), the Danish data showed a tendency of lower risk of long-term

sickness absence in younger age groups (>30, 31-40 years of age) and a higher risk of long-term sickness

absence in older age groups (41-50, <50 years of age) for all working hour characteristics except from day work,

which still showed that the higher number of day shifts, the lower risk of long-term sickness absence, whereas

consecutive night shifts showed increased risk of long-term sickness absence when 30 years of age. In the

Finnish data, we saw tendencies, however few results were statistically significant (table 4).

Table 3 Danish Working Hour Database: Poisson regression analyses with Incidence Rate Ratios for the association of working hour

characteristics and 1 incidence of long-term sickness absence (long-term sickness absence) in age categories in a model adjusted for

sex. Bold marking represent statistically significant values at a 0.05 level

1st incidence of long-term sickness absence

> 30- <=40

> 40 - <=50 years of age

years of age

(n =10051 )

(n =8463 )

Incidence

95%

Incidence

95%

Rate

Confidence

Rate

Confidence

Ratios

Iimits

Ratios

Iimits

< =30 years of age

(n =6227)

Incidence

Rate

Ratios

Day (least 3h

06:00 <21:00)

0 shift/person/year

1-100

101-200

>200

Evening (least 3h

18:00 and < 02:00)

0 shift/person/year

95% Confidence

Iimits

>50 years of age

(n =6988)

Incidence

Rate

Ratios

95%

Confidence

Iimits

1.19

0.81

0.45

Ref

0.75-1.89

0.49-1.32

0.17-1.20

0.86

0.74

0.50

Ref

0.58-1.27

0.49-1.11

0.31-0.83

0.84

0.76

0.47

ref

0.62-1.14

0.55-1.03

0.33-0.66

0.82

0.66

0.43

Ref

0.60-1.11

0.48-0.90

0.30-0.61

Ref

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

1-12

13-50

>50

Night (least 3h

23:00 and 06:00)

0 shift/person/year

1-12

13-50

>50

Long shifts

12 h)

0 shift/person/year

1-12

13-50

>50

Very Long shifts

12 h - < 24 h)

0 shift/person/year

1-12

13-50

>50

Quick returns (<11

h between shifts)

0 shift/person/year

1-12

13-50

>50

Long weeks

(>40h/week)

0 shift/person/year

1-12

13-50

>50

Very long weeks

(>48h/week)

0 shift/person/year

1-12

13-50

>50

Consecutive night

shift (

0 person/year

1-12

>12

0.92

0.85

0.93

0.74-1.15

0.69-1.06

0.69-1.25

0.86

0.94

1.09

0.74-1.00

0.81-1.10

0.89-1.32

1.12

1.24

1.53

0.98-1.28

1.09-1.41

1.31-1.78

0.95

1.11

1.27

0.80-1.14

0.94-1.32

1.07-1.51

0.85

0.71

0.65

Ref

0.71-1.02

0.58-0.86

0.46-0.92

0.84

0.86

0.88

Ref

0.73-0.96

0.74-1.01

0.70-1.11

0.94

0.98

1.21

ref

0.83-1.14

0.84-0.96

0.99-1.48

0.92

1.22

1.41

ref

0.79-1.08

0.98-1.51

1.11-1.80

0.99

0.88

0.77

Ref

0.81-1.20

0.70-1.12

0.53-1.10

0.94

0.83

0.86

Ref

0.81-1.08

0.69-0.99

0.66-1.14

1.12

1.03

0.98

Ref

0.99-1.27

0.89-1.20

0.78-1.23

1.14

1.10

1.01

ref

0.97-1.34

0.91-1.33

0.77-1.34

0.94

0.76

0.65

Ref

0.81-1.10

0.56-1.03

0.05-7.76

0.87

0.77

0.91

Ref

0.77-0.99

0.61-0.98

0.42-1.97

0.87

1.06

1.25

Ref

0.79-0.97

0.85-1.31

0.63-2.51

0.96

1.10

2.52

ref

0.83-1.10

0.80-1.52

0.91-6.99

0.91

0.73

1.93

Ref

0.77-1.08

0.56-0.96

0.40-9.31

0.94

0.88

0.63

Ref

0.83-1.07

0.71-1.08

0.25-1.60

1.07

1.00

0.65

Ref

0.96-1.19

0.84-1.18

0.31-1.36

1.12

0.95

0.56

Ref

0.97-1.28

0.77-1.18

0.24-1.30

0.95

0.61

Ref

0.79-1.14

0.46-0.82

0.94

0.71

Ref

0.82-1.06

0.56-0.90

1.07

0.96

Ref

0.95-1.19

0.81-1.15

1.09

0.98

Ref

0.94-1.26

0.79-1.22

0.78

0.43

Ref

0.67-0.90

0.17-1.04

0.86

0.85

Ref

0.77-0.97

0.49-1.47

1.06

1.18

Ref

0.95-1.18

0.81-1.71

1.17

1.25

Ref

1.03-1.34

0.82-1.89

0.86

0.54

Ref

0.63-1.18

0.17-1.65

1.01

1.62

Ref

0.78-1.30

0.88-2.97

1.19

1.20

Ref

0.96-1.49

0.81-1.77

1.37

1.58

1.02-1.85

1.10-2.29

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

Table 4 The Working Hours in the Finnish Public Sector Study: Poisson regression analyses with Incidence Rate Ratios for the

association of working hour characteristics and 1 incidence of long-term sickness absence (long-term sickness absence) in age

categories in a model adjusting for sex. Bold marking represent statistically significant values at a 0.05 level

1st incidence of long-term sickness absence

> 40 -

≤50

years of age

> 30-

≤40

years of

(n = 1726)

age (n = 1734)

Incide

nce

Rate

Ratio

95%

Confidenc

e limit

Incide

nce

Rate

Ratio

95%

Confidence

limit

≤30

years of age

(n = 463)

Incid

ence

95%

Rate

Confidence

Rati

limits

Day (least 3h 06:00

<21:00)

0 shift/person/year

1-100

101-200

>200

Evening (least 3h

18:00 and < 02:00)

0 shift/person/year

1-12

13-50

>50

Night (least 3h 23:00

and 06:00)

0 shift/person/year

1-12

13-50

>50

Long shifts

- < 12

0 shift/person/year

1-12

13-50

>50

Very Long shifts

- < 24 h)

0 shift/person/year

1-12

13-50

>50

Quick returns (<11 h

between shifts)

0 shift/person/year

1-12

13-50

>50

Long weeks

(>40h/week)

0 shift/person/year

1-12

13-50

>50

Very long weeks

>50 years of age

(n =3047)

Incidec

e Rate

Ratio

95%

Confidence

limit

0.34

0.52

0.30

ref

0.04, 2.56

0.07- 4.03

0.02- 4.82

0.82

0.89

0.69

Ref

0.36-1.87

0.39- 2.00

0.28- 1.71

0.72

0.94

0.70

ref

0.26, 1.97

0.35- 2.54

0.24- 2.00

1.45

2.08

1.55

Ref

0.20- 10.37

0.29- 14.79

0.22- 11.16

2.33

1.93

ref

0.68- 7.98

0.53- 7.07

1.13

0.96

Ref

0.85- 1.50

0.71- 1.28

1.35

1.14

ref

1.03- 1.77

0.86- 1.51

0.89

0.88

0.63

Ref

0.74- 1.07

0.74- 1.04

0.16- 2.54

0.40

0.74

2.58

ref

0.16- 1.05

0.29- 1.88

0.53- 12.42

1.01

1.00

1.36

Red

0.75- 1.36

0.74- 1.34

0.68- 2.72

1.07

1.06

1.39

ref

0.81- 1.41

0.80- 1.40

0.68- 2.87

1.05

0.97

1.40

Ref

0.88- 1.25

0.80- 1.16

0.97- 2.06

0.35

0.57

2.47

ref

0.70- 1.58

0.20- 1.61

0.48- 12.76

1.05

0.94

0.99

Ref

0.70- 1.58

0.62- 1.43

0.41- 2.40

1.06

1.17

1.29

ref

0.67- 1.67

0.74- 1.85

0.57- 2.92

1.17

1.18

1.74

Ref

0.85- 1.60

0.86- 1.62

1.10- 2.75

1.22

1.37

ref

0.58- 2.61

0.17- 10.67

1.03

1.17

1.41

Ref

0.81- 1.30

0.73- 1.87

0.74- 2.68

1.05

0.86

1.40

ref

0.84- 1.32

0.51- 1.45

0.45- 4.42

1.02

0.78

0.87

Ref

0.88- 1.18

0.51- 1.18

0.28- 2.70

2.42

2.34

3.00

ref

0.31- 19.11

0.31- 17.69

0.35- 25.81

1.61

1.75

1.62

Ref

0.72- 3.61

0.83- 3.73

0.75- 3.51

1.13

1.24

1.20

ref

0.51- 2.49

0.61- 2.51

0.58- 2.49

1.67

2.34

2.39

Ref

0.66- 4.22

0.97- 5.65

0.98- 5.81

1.64

1.28

ref

0.56- 4.79

0.37- 4.38

1.68

1.25

Ref

1.11- 2.54

0.80- 1.94

1.28

1.01

ref

0.87- 1.87

0.67- 1.54

1.52

1.41

Ref

1.12- 2.08

1.02- 1.94

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

(>48h/week)

0 shift/person/year

1-12

13-50

>50

Consecutive night shift

(

0 person/year

1-12

>12

2.01

ref

0.90- 4.50

1.24

2.59

Ref

0.97- 1.58

1.05- 6.38

1.14

1.26

ref

0.91- 1.43

0.47- 3.42

1.01

0.80

Ref

0.87- 1.17

0.33- 1.93

0.79

4.91

ref

0.19-3.31

0.67-36.17

1.51

1.72

Ref

1.04-2.18

0.55-5.36

1.13

0.73

ref

0.73-1.74

0.18-2.94

1.02

1.15

0.68-1.53

0.47-2.77

When excluding pregnant women in the analyses on data from the Danish Working Hour Database, for most

exposure categories the results were similar to those presented in table 2. However, an increased risk of long-

term sickness absence was found when working more than 50 evening shifts (IRR: 1.26, 95 % CI: 1.12-1.42) and

more than 50 night shifts (IRR: 1.20, 95 % CI: 1.03-1.40). Working few night shifts was, however, associated

with lower risk of long-term sickness absence (IRR: 0.89, 95 % CI: 0.81-0.99). Similarly, long weeks were

associated with lower risk of long-term sickness absence (IRR: 0.84, 95 % CI: 0.72-0.97). Five or more

consecutive night shifts was associated with an increased risk of long-term sickness absence when pregnant

women were excluded (IRR: 1.91, 95 % CI: 1.43-2.57) (table placed in the supplementary digital material C). It

should be noted that the mean age of the study population increased from 42.3 to 45.4 years when excluding

women who were pregnant in 2009-2015.

Discussion

This study comprising two large scale, population-based databases from Denmark and Finland with objective

daily working hour data in the nursing sector including over 38 000 employees provided a unique possibility to

investigate the role of working hour characteristics as predictors of long-term sickness absence. Main analyses

of the Danish data showed that participants working evening shifts and participants with five or more

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

consecutive night shifts had higher risk of long-term sickness absence whereas participants with day shifts, few

night shifts per year, very long shifts and quick returns had lower risk of long-term sickness absence. Exclusion

of all pregnant women analyses (in the Danish data) showed a higher risk of long-term sickness absence among

evening and night workers along with participants working five or more consecutive night shifts. When

stratifying on age groups, there was a tendency of lower risk of long-term sickness absence in the youngest age

groups and higher risk of long-term sickness absence in the oldest age groups. Hence, the association seen

between evening, night work and consecutive night work and long-term sickness absence showed increased

risk only in the oldest of the age groups.

The results based on the Finnish data showed a higher risk of long-term sickness absence when working nights,

longs shifts, quick returns, long work weeks and very long work weeks. When stratifying on age groups, the

results showed, however, a similar tendency as the results from the Danish data, i.e. higher risk of long-term

sickness absence in older age groups. However, due to low power in age stratified analyses of Finnish data,

these results should be interpreted with caution.

The analyses of the Danish data showed no association between night work and risk of long-term sickness

absence, but when excluding women who were pregnant in 2008-2015 or stratifying on age, results showed an

association between night work and long-term sickness absence indicating an effect modification of pregnancy

and/or age. This might be due to age since exclusion of ever pregnant women, the remaining cohort is on

average older (mean age 42.3 vs. 45.4 years). This corresponds to the results in the Finnish data where

participants on average are older than the Danish participants.

Previous studies are not fully conclusive in regard to night work and sickness absence. A Finnish study has

shown shift work - but not night work - to be associated with increased risk of long-term sickness absence (31).

An American payroll data study on policemen showed a higher rate of sick leave among night shift workers

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

compared to day workers (23). In a Dutch study, night work was not associated with increased risk of sickness

absence whereas shift work was associated with lower risk of sickness absence (26). The meta-analyses by

Merkus et al found the evidence in regards to night work and long-term sickness absence to be inconclusive

but suggest that the association of night work and long-term sickness absence is dependent on the intensity of

the night work (27). This is supported in both the Danish and the Finnish data when stratified on age, where

associations between night work and long-term sickness absence are only found among those working more

than 50 night shifts per year. There are many ways to schedule night work and this may affect how night work

is associated with long-term sickness absence.

In the Danish data, we found a higher risk of long-term sickness absence among workers with 50 evening shifts

or more per year. When stratifying on age, we saw that this association was only observed among those aged

40+. Effects of evening work have been observed in other Danish studies of female workers’ caring of elderly

(32). However, in the Finnish data, evening work was not associated with increased risk of long-term sickness

absence. The difference between countries could be a result of different distribution of evening workers in

Denmark and Finland.

When excluding pregnant women, the lower risk of long-term sickness absence when working very long shifts

or long or very long weekly working hours found in the main analyses of the Danish data, only remained

statistically significant for long weekly working hours underlining the need for attention to sickness absence

related to pregnancy. In the Finnish data, it was not possible to exclude pregnant women due to lack of

information.

Working long shifts or long weeks were found to be associated with long-term sickness absence in the Finnish

data. This is in line with the finding from an earlier study where the authors conclude in regards to both short-

term and long-term sickness absence

that

orking long shifts on hospital wards is associated with a higher risk

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

of sickness absence for registered nurses and health care assistants (12). Results from the Danish data did not

support this.

Quick returns (less than 11 hours between two shifts) were in the Danish data associated with lower risk of

long-term sickness absence, and in the Finnish data associated with higher risk. In the Danish data, however,

when excluding pregnant women as in the Finnish data, we found a higher risk of long-term sickness absence.

The distribution of quick returns is very different in the two countries as seen in table 1 which to some extent

might explain the different results found in the main analyses. Earlier studies have indicated that age might

affect the association between quick returns and higher fatigue (both during work and free days) and

difficulties to fall asleep (33). Further, a study pointed out that changes in quick returns are associated with

changes in perceived work-life conflict (21). Quick returns could reflect higher work demands, but could also be

preferred if leading to more consecutive days off.

With respect to the Danish data, five or more consecutive night shifts were associated with increased risk of

long-term sickness absence in all age groups and when pregnant women were excluded. The results were not

supported by the Finnish data, but as table 1 indicates, only few of the Finnish workers were in the high

exposure group (12 or more periods per year). Five or more consecutive night shifts could reflect the work

schedules of permanent night workers. Previous studies have found permanent night workers to report poorer

health and more absenteeism (34) along with difficulties of falling asleep and fatigue (35). However, the studies

also found higher job satisfaction with colleagues (35).

In regards to the differences in risk estimates when excluding pregnant women, previous studies have shown

pregnancy to be related to higher risk of sick leave (36, 37) in particular in relation to night shifts (38). It could

be discussed if the exclusion of pregnant women

re eals a truer

effect of night work on long-term sickness

absence or if the results simply just reflect another age distribution. In the study we excluded pregnant

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

women, who are typically younger, and we were therefore left with an older study population. When

stratifying the analyses on age, we found in both the Danish and the Finnish data that the higher age, the

higher risk of long-term sickness absence in several of the working hour characteristics. Older age is a known

risk factor for sickness absence (39, 40) but studies of night and shift work and cardiovascular disease and

cancer have also linked years of exposure to higher risk of morbidity and mortality (6, 41-43). As we have no

information on years of shift work, we cannot rule out the possibility that higher risk in the older age groups

also could reflect more years of exposure.

The differences seen in the Danish and the Finnish results could, however, also be due to differences in

legislation and rules in regards to sickness absence benefit (44). The report by the Nordic Social Statistical

Committee underlines several differences in rules governing payment of cash assistance to employees in the

event of sickness. One example is the length of the employer period which in Denmark are 30 days and in

Finland 10 working days (44). In the current study, we have defined long-term sickness absence as 30

consecutive days or more which might collide with the Danish reimbursement period of 30 days which local

authorities are involved in after this date. There is therefore the risk that to avoid the involvement of

authorities, a person returns to work before being fully recovered. Further, in Denmark, employers are

allowed to discharge an employee after 120 days of absence due to illness in one year (45) which also could

affect the number of persons with long-term sickness absence.

Strengths and limitations

The analyses for this study are based on large, representative and registerbased payroll data of high quality,

which is a major strength compared to earlier studies based on self-reported or cross-sectional data (12, 32) .

Furthermore, we utilized prospective study design with eight years of follow-up and very precise information

on both long-term sickness absence and working hours from registries without recall bias, attrition or selection

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

based on exposure (30). As we had access to two national samples, Danish Working Hour Database and the

Working Hours in the Finnish Public Sector Study database, we utilized strict protocol for harmonization of

samples to unify our definitions and handling of data to make comparable analyses. The distribution of nursing

personnel across the week is rather similar in the two countries (20) which justifies that we can use the same

definitions of working hour characteristics across the data from the two countries. Furthermore, an evident

strength is the possibility to analyze irregular shift systems. Hence, this study adds to the existing knowledge

based on other types of epidemiological studies which have mainly utilized self-reported working hours or shift

types in shift work (46, 47).

We estimated the working hour characteristics based on one year (2008), as this was manageable for both

cohorts and provided us follow-up time to identify enough cases for analysis of first incidence of long-term

sickness absence. Due to the varying nature of irregular working hours in the public health care sector and no

major legislative changes within the 2008-2015, i.e. time of this study, we do expect that any annual effects on

the results would be minimal. However, further studies would be warranted to test the findings over several

years of working hour data and even with a longer follow-up time. This study has been limited to nine main

working hour characteristics for the study, i.e. day, evening, and night shifts, long or very long shifts, long or

very long weeks, quick returns, and consecutive night shifts. Assessment of shift combinations or whether long

shifts took place during the day, evening or night has been excluded although relevant in future together with

specifying sickness absence according to diagnosis. Lastly, the utilization of payroll data only limited us from

adjustment for marital status, children or other life situation as our data do not include this. Future studies

should account for this. Further, our data was based on mainly female workers which might limit the

generalization.

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

Finally, as stated in the method section analyses showed working hours characteristics to be relative stable

over the years in both the Danish and the Finnish data. However, it does not exclude that on an individual level

people may shift especially from night work to day work during the years due to ageing and that this could

cause bias due to misclassification of exposure.

Conclusions

Our results based on register-based, and detailed working hour data of characteristics of shift work add to the

rare studies of associations with long-term sickness absence. The results show that the scheduling of working

hours is associated with the risk of long-term sickness absence. In the Danish data we found a higher risk when

working evening shifts and five or more consecutive night shifts. In addition, night shifts and weekly working

hours of more than 48 hours were associated with risk of long-term sickness absence among non-pregnant

women and in the age group above 50 years. In the Finnish data, we found a higher risk of long-term sickness

absence among those working nights, long shifts, quick returns or long weeks. Both cohorts showed the highest

risk of long-term sickness absence among those of highest age. The differences in result between the two

cohorts may be due to contextual differences, e.g. in the legislations and rules related to sickness absence

benefits. Comparison of risk of long-term sickness absence in relation to working hours between countries

should thus be performed with caution.

Acknowledgement

None

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

Contributorship statement

ADL, AR, MH and AHG were responsible for the conception and design of the study, JH, ÅMH, HAK, AK, MH and

AHG for acquisition of data, and ADL and AR for analyses. All authors were responsible for interpreting data,

ADL and AR drafted the article, and all the authors revised it critically for important intellectual content. All the

authors gave their final approval of the version to be submitted.

Conflict of interest

None

Funding sources

The project is part of the WOW project funded by NordForsk, Nordic Program on Health and Welfare [grant

number 74809]. MH, AK and AR are also supported by EU grant no 826266 (CO-ADAPT), and ADL and AHG are

also supported by the Danish Working Environment Research Fund [grant number 23-2012-09/20120220951].

Funding organizations have no involvement in study design, the collection of data, analysis and interpretation

of data, the writing of the report, or in the decision to submit the article for publication.

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

References

Eurofound. Sixth European Working Conditions Survey 2015.

https://wwweurofoundeuropaeu/surveys/data-visualisation/sixth-european-working-conditions-survey-2015

[Internet]. 2015.

Puttonen S, Harma M, Hublin C. Shift work and cardiovascular disease - pathways from circadian stress

to morbidity. Scand J Work Environ Health. 2010;36(2):96-108.

Gan Y, Yang C, Tong X, Sun H, Cong Y, Yin X, et al. Shift work and diabetes mellitus: a meta-analysis of

observational studies. Occupational and environmental medicine. 2015;72(1):72-8.

Larsen AD, Hannerz H, Moller SV, Dyreborg J, Bonde JP, Hansen J, et al. Night work, long work weeks,

and risk of accidental injuries. A register-based study. Scand J Work Environ Health. 2017.

Nielsen HB, Larsen AD, Dyreborg J, Hansen AM, Pompeii LA, Conway SH, et al. Risk of injury after

evening and night work - findings from the Danish Working Hour Database. Scandinavian journal of work,

environment & health. 2018;44(4):385-93.

Hansen J. Night Shift Work and Risk of Breast Cancer. Curr Environ Health Rep. 2017:10-0155.

Virtanen M, Kivimaki M. Long Working Hours and Risk of Cardiovascular Disease. Current cardiology

reports. 2018;20(11):123.

Virtanen M, Jokela M, Madsen IE, Magnusson Hanson LL, Lallukka T, Nyberg ST, et al. Long working

hours and depressive symptoms: systematic review and meta-analysis of published studies and unpublished

individual participant data. Scandinavian journal of work, environment & health. 2018;44(3):239-50.

Virtanen M, Ferrie JE, Gimeno D, Vahtera J, Elovainio M, Singh-Manoux A, et al. Long working hours

and sleep disturbances: the Whitehall II prospective cohort study. Sleep. 2009;32(6):737-45.

10.

Karkkainen S, Ropponen A, Narusyte J, Mather L, Akerstedt T, Silventoinen K, et al. Night work as a risk

factor of future disability pension due to musculoskeletal diagnoses: a prospective cohort study of Swedish

twins. European journal of public health. 2017;27(4):659-64.

11.

Ropponen A, Narusyte J, Mather L, Mittendorfer-Rutz E, Akerstedt T, Svedberg P. Night work as a risk

factor for future cause-specific disability pension: A prospective twin cohort study in Sweden. Chronobiology

international. 2018;35(2):249-60.

12.

Dall'Ora C, Ball J, Redfern O, Recio-Saucedo A, Maruotti A, Meredith P, et al. Are long nursing shifts on

hospital wards associated with sickness absence? A longitudinal retrospective observational study. Journal of

nursing management. 2018.

13.

OECD.stat. OECD.stat Health status. Absence from work due to illness 2019 [cited 2019. Available from:

https://stats.oecd.org/index.aspx?queryid=30123.

14.

Stromberg C, Aboagye E, Hagberg J, Bergstrom G, Lohela-Karlsson M. Estimating the Effect and

Economic Impact of Absenteeism, Presenteeism, and Work Environment-Related Problems on Reductions in

Productivity from a Managerial Perspective. Value in health : the journal of the International Society for

Pharmacoeconomics and Outcomes Research. 2017;20(8):1058-64.

15.

Sieurin L, Josephson M, Vingard E. Positive and negative consequences of sick leave for the individual,

with special focus on part-time sick leave. Scand J Public Health. 2009;37(1):50-6.

16.

Vingard E, Alexanderson K, Norlund A. Swedish Council on Technology Assessment in Health Care

(SBU). Chapter 9. Consequences of being on sick leave. Scandinavian journal of public health Supplement.

2004;63:207-15.

17.

Bryngelson A. Long-term sickness absence and social exclusion. Scandinavian journal of public health.

2009;37(8):839-45.

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

18.

Salonen L, Blomgren J, Laaksonen M, Niemela M. Sickness absence as a predictor of disability

retirement in different occupational classes: a register-based study of a working-age cohort in Finland in 2007-

2014. BMJ open. 2018;8(5):e020491.

19.

Kivimaki M, Forma P, Wikstrom J, Halmeenmaki T, Pentti J, Elovainio M, et al. Sickness absence as a risk

marker of future disability pension: the 10-town study. J Epidemiol Community Health. 2004;58(8):710-1.

20.

Garde AH, Harris A, Vedaa O, Bjorvatn B, Hansen J, Hansen AM, et al. Working hour characteristics and

schedules among nurses in three Nordic countries - a comparative study using payroll data. BMC nursing.

2019;18:12.

21.

Karhula K, Koskinen A, Ojajarvi A, Ropponen A, Puttonen S, Kivimaki M, et al. Are changes in objective

working hour characteristics associated with changes in work-life conflict among hospital employees working

shifts? A 7-year follow-up. Occupational and environmental medicine. 2018;75(6):407-11.

22.

Ropponen A, Koskinen A, Puttonen S, Harma M. Exposure to working-hour characteristics and short

sickness absence in hospital workers: A case-crossover study using objective data. International journal of

nursing studies. 2019;91:14-21.

23.

Fekedulegn D, Burchfiel CM, Hartley TA, Andrew ME, Charles LE, Tinney-Zara CA, et al. Shiftwork and

sickness absence among police officers: the BCOPS study. Chronobiol Int. 2013;30(7):930-41.

24.

Bernstrom VH. Long working hours and sickness absence-a fixed effects design. BMC public health.

2018;18(1):578.

25.

Laaksonen M, Pitkaniemi J, Rahkonen O, Lahelma E. Work arrangements, physical working conditions,

and psychosocial working conditions as risk factors for sickness absence: Bayesian analysis of prospective data.

Annals of epidemiology. 2010;20(5):332-8.

26.

van Drongelen A, Boot CR, Hlobil H, van der Beek AJ, Smid T. Cumulative exposure to shift work and

sickness absence: associations in a five-year historic cohort. BMC public health. 2017;17(1):67.

27.

Merkus SL, van DA, Holte KA, Labriola M, Lund T, van MW, et al. The association between shift work

and sick leave: a systematic review. Occup Environ Med. 2012;69(10):701-12.

28.

Vedaa O, Pallesen S, Erevik EK, Svensen E, Waage S, Bjorvatn B, et al. Long working hours are inversely

related to sick leave in the following 3 months: a 4-year registry study. International archives of occupational

and environmental health. 2019;92(4):457-66.

29.

Garde AH, Hansen J, Kolstad HA, Larsen AD, Pedersen J, Petersen JD, et al. Payroll data based

description of working hours in the Danish regions. Chronobiology international. 2018;35(6):795-800.

30.

Harma M, Ropponen A, Hakola T, Koskinen A, Vanttola P, Puttonen S, et al. Developing register-based

measures for assessment of working time patterns for epidemiologic studies. Scand J Work Environ Health.

2015;41(3):268-79.

31.

Airaksinen J, Jokela M, Virtanen M, Oksanen T, Koskenvuo M, Pentti J, et al. Prediction of long-term

absence due to sickness in employees: development and validation of a multifactorial risk score in two cohort

studies. Scandinavian journal of work, environment & health. 2018;44(3):274-82.

32.

Tuchsen F, Christensen KB, Nabe-Nielsen K, Lund T. Does evening work predict sickness absence among

female carers of the elderly? Scandinavian journal of work, environment & health. 2008;34(6):483-6.

33.

Harma M, Karhula K, Ropponen A, Puttonen S, Koskinen A, Ojajarvi A, et al. Association of changes in

work shifts and shift intensity with change in fatigue and disturbed sleep: a within-subject study. Scandinavian

journal of work, environment & health. 2018;44(4):394-402.

34.

Burch JB, Tom J, Zhai Y, Criswell L, Leo E, Ogoussan K. Shiftwork impacts and adaptation among health

care workers. Occupational medicine (Oxford, England). 2009;59(3):159-66.

35.

Karhula K, Hakola T, Koskinen A, Ojajarvi A, Kivimaki M, Harma M. Permanent night workers sleep and

psychosocial factors in hospital work. A comparison to day and shift work. Chronobiology international.

2018;35(6):785-94.

BEU, Alm.del - 2019-20 - Bilag 307: Orientering om resultater fra NFA-artikel om risikoen for langtidssygefravær afhængigt af arbejdets organisering tidsmæssigt, fra beskæftigelsesministeren

฀฀฀฀฀฀฀฀฀

36.

Kaerlev L, Jacobsen LB, Olsen J, Bonde JP. Long-term sick leave and its risk factors during pregnancy

among Danish hospital employees. Scandinavian journal of public health. 2004;32(2):111-7.

37.

Melsom AM. Long-term sickness absence during pregnancy and the gender balance of workplaces.

Scandinavian journal of public health. 2014;42(7):627-34.

38.

Hammer PEC, Garde AH, Begtrup LM, Flachs EM, Hansen J, Hansen AM, et al. Night work and sick leave

during pregnancy: a national register-based within-worker cohort study. Occupational and environmental

medicine. 2019;76(3):163-8.

39.

Boschman JS, Noor A, Sluiter JK, Hagberg M. The mediating role of recovery opportunities on future

sickness absence from a gender- and age-sensitive perspective. PloS one. 2017;12(7):e0179657.

40.

Ervasti J, Joensuu M, Pentti J, Oksanen T, Ahola K, Vahtera J, et al. Prognostic factors for return to work

after depression-related work disability: A systematic review and meta-analysis. Journal of psychiatric research.

2017;95:28-36.

41.

Torquati L, Mielke GI, Brown WJ, Kolbe-Alexander T. Shift work and the risk of cardiovascular disease. A

systematic review and meta-analysis including dose-response relationship. Scandinavian journal of work,

environment & health. 2018;44(3):229-38.

42.

Vetter C, Devore EE, Wegrzyn LR, Massa J, Speizer FE, Kawachi I, et al. Association Between Rotating

Night Shift Work and Risk of Coronary Heart Disease Among Women. JAMA. 2016;315(16):1726-34.

43.

Jorgensen JT, Karlsen S, Stayner L, Andersen J, Andersen ZJ. Shift work and overall and cause-specific

mortality in the Danish nurse cohort. Scand J Work Environ Health. 2017;43(2):117-26.

44.

Nordic Social Statistical Committee. Social Protection in the Nordic Countries - scope, expenditure and

financing. 2017.

45.

Nordic Social Statistical Committee. Sickness absence in the Nordic Countries. 2015.

46.

Catano VM, Bissonnette AB. Examining paid sickness absence by shift workers. Occup Med (Lond).

2014;64(4):287-93.

47.

Akerstedt T, Kecklund G. What work schedule characteristics constitute a problem to the individual? A

representative study of Swedish shift workers. Applied ergonomics. 2017;59(Pt A):320-5.