A sustainable working life in the car manufacturing industry: The role of psychosocial factors, gender and occupation.

AIMS: In order to add to the existing knowledge about factors associated with retirement timing, in the car industry, it is useful to consider the psychosocial working conditions prior to retirement. This case-control study aimed to investigate relationships between psychosocial job factors and extended work after the age of 62 years among workers in the car industry in Sweden.
METHODS: A study invitation with a survey was sent to workers in one of Sweden's largest car manufacturing company, who were employed 2005-2015 and either retired at the age 55-62 years or working at 63 years or older. Psychosocial variables such as job demand-control (JDC) and effort-reward imbalance (ERI) were recorded through the survey. Multiple logistic regression models were used to investigate associations between psychosocial variables and retirement in 572 cases that had continued to work ≥ 63 years, and 771 controls who had retired at 62 or earlier.
RESULTS: No associations were found between JDC-variables and retirement in the total sample or gender stratified analyses, but high demands-low control (high strain) was related to retirement before the age of 63 years in blue-collar workers. In contrast, high strain was related to continuing to work after 62 years for white-collar men and, high ERI was associated with extended work for the total sample of white-collar workers, and white-collar men, however these effects became non-significant in fully adjusted models.
CONCLUSIONS: The relationships between psychosocial factors and extended work after 62 years were inconsistent, with high strain being related to retiring earlier for blue-collar workers.

Introduction

The ageing European population calls for the implementation of policies that promote healthy ageing and a sustainable working life [1]. Indeed, governments have implemented policies aiming to encourage later retirement, including raising the age for state pensions and legislation against age and disability discrimination [2]. However, there has been some concern regarding polices to extend working life and closure of early exit pathways. Certain groups within the population, with very demanding jobs, may still need the option of early retirement [3]. Continuing to work at an older age is a complex issue, and there is limited knowledge of push and pull determinants. A few identified ‘push factors’ include chronic diseases, physical demands and poor working conditions and ‘pull factors’ constitutes one’s spouse not working, care-taking of relatives and leisure time expectations [4,5]. Norms about working and retiring, economic incentives, attitudes at the workplace, work satisfaction, social relationships at work and home are also important factors for an extended working life [6].

The car manufacturing industry is one of the largest industries worldwide, and is viewed as one of the most important and strategic industries in the manufacturing sector. Although a substantial amount of research have been conducted within the car industry [7] there is a lack of knowledge of factors influencing the timing of retirement in this sector. The car industry provides a large variety of jobs, both manual jobs at the manufacturing sites and office work involving administration, design development and marketing. The role of workplace factors in timing of retirement most likely differs between blue- and white-collar workers, since performing different jobs and, thus, face different working conditions [8]. There is evidence that white-collar workers with higher education tend to work to a later age compared to blue-collar workers [9]. A Swedish study of retirement intentions found that white-collar workers, in high demand -high control jobs, would consider continue to work after the statutory age for old age retirement, whereas most blue-collar workers wanted to retire early [10]. Another study in workers aged 45+ [11] found that job resources (job control and social support) were positively related to work enjoyment and negatively related to early retirement intention. These relationships were found in both blue- and white-collar workers, but were stronger in blue-collar workers. There were also associations between job demands and early retirement intention. It is further plausible that gender plays a role regarding factors determining retirement timing. A previous study found that the strength of a number of push and pull factors relating to retirement decisions were different for men and women. Demanding jobs was a push factor for women and socially rewarding jobs was a pull factor for men [12]. In Sweden, women retire on average one year earlier than men (women = 63 years; men = 64 years) [13]. Women tend to have lower level positions and the working conditions for men and women may differ even in the same profession and organisation. In addition, women spend more time taking care of home and family compared to men [14].

There is extensive evidence that psychosocial factors at work influence physical and mental health [15,16]. Current knowledge about stressing psychosocial work environment mainly relies on two theoretical models, the demand–control–support model (JDC-S model) [17], and the effort–reward imbalance (ERI) model [18]. Indeed, psychosocial factors have been related to retirement timing, and several studies have found associations between low levels of job control and early retirement [19]. Similarly, job strain (combination high demands-low control) have been associated with intentions to leave work early [20]. Effort-reward imbalance has also been associated with intentions to early retirement [21]. A longitudinal study conducted in an English population aged 50+, found that high job demands were associated with preferences for early retirement, and high decision authority was associated with preferences for later retirement [5]. Another study [22] illustrated that stressful jobs decreased the risk of early retirement. One proposed explanation was that stressful jobs were seen as more meaningful, stimulating and satisfying. However, it was also found that men with low levels of job control retired earlier compared to men with high levels of control.

Aims

In order to add to the existing knowledge about factors associated with retirement timing, in the car industry, it is useful to consider the psychosocial working conditions prior to retirement.

This case-control study therefore aims to investigate relationships between psychosocial job factors, using the JDC- and ERI-models, and extended work after the age of 62 years among workers in the car industry in Sweden. The age of 62 years as criteria was selected because in this particular company, due to beneficial retirement schemes, there is no particular economic gain by working after 62 years of age. Consequently, early retirement could indicate that other factors, such as the work environment, could be associated with the retirement decision.

Methods

Study population

The present study is a case-control study, conducted as part of “AgeCap—a center for aging and health”. Potential study subjects constituted all individuals employed at Volvo Cars at sites located in the West county of Sweden during 2005–2015 and either retired at the age 55–62 or working ≥63 years during the observation years. Access to personnel who fulfilled inclusion criteria was available to us through Volvo’s staff registers. In total, 3025 persons fulfilling inclusion criteria and with a valid postal address, were invited to the study. Out of those, 1871 men and women returned filled-in questionnaires, yielding a response rate of 61.9%. Some subjects completely missed fill-in psychosocial questions, and were thus, deleted. As job demand-control and effort-reward imbalance and some of the covariates, were analysed using sum scores, subjects with <50% missing items received imputed values, described under statistical analysis. Subjects lacking >50% filled-in items per each variable were excluded. Since the amount of subjects with >50% missing items varied between JDC and ERI, calculations for these psychosocial dimension were carried out in separate subgroups. The final sample used for JDC analyses constituted of 1320 subjects and in the ERI analyses 1305 subjects. This study has been approved by the Regional Ethical Review, Gothenburg, Sweden (dnr 371–15). The consent was written.

Through the register we also obtained information on employment status, i.e. if being currently active at work or retired, year of retirement and year or birth from which age at retirement could be calculated. This information was used to categorize staff into cases or controls. The cases were employees who during the observation period had either retired ≥63 years, or who were currently working and aged ≥63 years. Controls constituted all study base subjects who retired at 55–62 years. We defined cases as subjects continuing to work ≥63 years, rather than those retiring early, in order to emphasize work dimensions that benefit continued work. Individuals who had been laid off were not included in the study.

Questionnaire

All potential participants were sent a questionnaire by post. The questionnaires recorded demographics, occupational history and position, shift work, physically and psychosocial work conditions, retirement circumstances, previous diseases, stress at home and life events. The same variables were measured for cases and controls, but the wording slightly differed since cases could be either retired or still working, while all controls were retired. The cases were asked to consider their current working environment, or the working environment at their last workplace at Volvo. The controls were asked to consider the working environment at their last workplace at Volvo.

Psychosocial variables

Job demand-control was measured with the Swedish Demand-Control-Support Questionnaire (DCSQ) [23]. For the purposes of analysis, all demand and control items were both positively inverted, i.e. a high scores equated either high demands or high control, and then tallied separately. Median scores for demand and control were 12 and 20, respectively. Both variables were dichotomized into high or low by the median values of the distributions, and combined into: high strain (high demand-low control), active (high demand-high control), passive (low demand-low control) and low strain (low demand-high control). In the regression analyses low strain was used as reference [24].

Reward was assessed using the Effort-Reward Imbalance at Questionnaire—short version (ERI-S) [25,26]. Effort and reward items were positively inverted and then summed up. According to standard procedure, a ratio value was created (Σeffort/(Σreward*0.4286)) and then dichotomized by the established cut-off 1.0, where a score above the cut-off represent reciprocity between work efforts and received rewards. ERI-ratio scores <1.0 were used as reference value.

Statistical analysis

The statistical software package SAS version 9.2 for Windows (SAS Institute; Cary; NC) was used for all regression analyses. Multiple logistic regression models were used to investigate associations between psychosocial variables and extended work ≥ 63 years of age. Several available covariates could be important in the analyses of associations between psychosocial work conditions and retirement, but since our cohort size is fairly small and some analyses are stratified by gender or blue-/white-collar workers, we wish to slim the models by only including covariates relevant for our analyses. We therefore conducted, stepwise purposeful selection as proposed by Hosmer & Lemeshow [27] for associations between our psychosocial variables of interest and retirement at 62 years of age (yes/no). Correlation coefficient values > 0.4 was considered as co-linearity and cut-off for inclusion was Likelihood ratio 0.25. The potential covariates considered were: civil status, age of partner in relation to the participant, country of birth, retirement during a year of recession, being offered a beneficial retirement deal, if retirement was the persons own decision, amount of years since retirement, previous diseases, work ability at the time of retirement, need for recovery after work, life events, stress at home, frequency of heavy physical demands, quality of leadership, leadership position and social support at work.

According to established procedure the selection process began with checking for co-linearity. Retirement during economic recession, being offered a beneficial pension deal and amount of years retired, displayed co-linearity. Since retirement during recession year displayed strongest effect in relation to retirement, this variable was chosen for the next step. We then carried out univariate analyses of each potential covariate and retirement as outcome. All variables meeting the cut-off in the univariate analyses constituted the full model. These variables and main psychosocial measures were then entered together in a logistic regression analyses. Variables with Likelihood ratio >0.25 were excluded. The excluded variables were then added back one at a time and reinserted to the model if meeting inclusion criteria. The remaining variables constituted the reduced model. Since main effects in the reduced models were only reduced with 4.4%, i.e. less than criteria of 15%, compared to the full model, the reduced model was kept.

After concluding the selection process the following confounders remained: age difference to partner, frequency of heavy physical demands, leadership quality, having a leadership position, retired during an economic recession, whether it was the person’s own decision to retire and life events.

Age difference to partner was captured by one item in the survey about spouse/partners age with was compared to the study participants age, and analysed as a categorical variable: younger, the same age or older. Frequency of heavy physical job demands was measured with an instrument developed by the Swedish work Environment Authority and constituted seven items inquiring about physical work tasks and a response scale (1–6): 1 = “No not at all” to 6 = “Almost all the time”, sample item: “Does your job require that you a certain amount of the time work purely physically?”. Leadership quality was measured with standard items from the Copenhagen Psychosocial Questionnaire (COPSOQ) defining “leader” as the participant’s immediate superior [28]. Having a leadership position (yes/no) was meaured by one item “What is/was your latest occupational position at Volvo Cars”. The text repsons was then coded as leadership position = “yes” if having any of the following positions; CEO, manager, director, executive, leader or foreman. Retirement during recession years was defined as years when Volvo due to economic downturns were forced to lay off large number of employees. These years where 2006, 2009 and 2013. The variable was dichotomized as retired during recession year (yes/no). Whether it was the person’s own decision to retire was captured with one item created for this study: “Did you chose yourself to retire or did you feel pressured (by for example health, employer etc.): Have either retired or still working; if retired the response scale ranged response scale (1–5): 1 = “It was entirely my own decision to retire” to 5 = “I could not chose at all, I was forced to retire”. Life events defined as having a severely ill close relative, measured with one item from a scale developed by Welin et al [29], “Have any of your closest relatives been severely ill or had a serious accident, and has this affected your decision to retire?”, with a categorical response scale “no”, “yes, and it affected my wish to retire” “yes, and it affected my wish to continue working.

Three models were calculated; first a crude model, a second model was adjusted for work variables frequency of heavy physical demands, leadership quality and having a leadership position and retirement during crisis years and if it was the person’s own decision to retire. In the third model off-work factors were added: age difference to partner and life events. JDC and ERI were analysed separately. All missing items for demand, control, effort and reward were imputed accordingly: subjects with ≥50% missing items per variable were excluded. For subjects with <50% missing items per a variable, mean scores of the remaining items in each variable, were imputed on individual level.

Results

In total, analyses were based on 1343 subjects (85.4% men), who had filled-in any psychosocial questions, and then further divided into subsamples for analyses using JDC (n = 1312) or ERI dimensions (n = 1307). In the final sample 42.6% were cases and 57.4% were controls, and 25% of women were cases versus 75% controls (Table 1).

Table 1

Cohort characteristics according to demographics and psychosocial variables.

	All		Cases		Controls
All, N (%)	1343		572	(42.6)	771	(57.4)
-Men	1140		521	(45.7)	619	(54.3)
-Women	196		49	(25.0)	147	(75.0)
Blue-collar, N (%)	561		236	(42.1)	325	(57.9)
-Men	458		215	(46.7)	243	(53.3)
-Women	95		19	(19.8)	77	(80.2)
White-collar, N (%)	782		336	(43.0)	446	(57.0)
-Men	680		306	(45.0)	374	(55.0)
-Women	100		30	(30.0)	70	(70.0)
Heavy physical demands, mean (sd)	2.0	(1.1)	2.0	(1.0)	2.0	1.0
Leadership quality, mean (sd)	12.7	(3.5)	12.7	(3.4)	12.6	3.5
Having a leadership position (yes), N (%)	321	(26.5)	126	(23.8)	195	(28.6)
Retired in recession years (yes), N (%)	626	(46.6)	115	(20.1)	511	(66.3)
Own decision to retire, mean (sd)	2.7	(1.3)
Age difference to partner
• Younger	695	(69.2)	321	(72.0)	374	(67.0)
• Same age	106	(10.6)	51	(11.4)	55	(9.9)
• Older	203	(20.2)	74	(16.6)	129	(23.1)
Life event affecting retirement (yes), N (%)	174	(13.3)	81	(14.4)	93	(12.4)
Participants experiencing high strain, N (%)*
All	496	(37.8)	215	(38.0)	281	(37.7)
-Men	4108	(36.3)	196	(37.8)	212	(34.9)
-Women	88	(47.1)	19	(39.6)	69	(49.6)
Blue-collar	186	(34.5)	61	(24.5)	98	(31.3)
-Men	140	(31.0)	61	(28.5)	79	(33.2)
-Women	46	(52.9)	6	(33.3)	40	(58.0)
White-collar	235	(30.4)	116	(34.7)	119	(27.1)
-Men	203	(30.2)	105	(34.5)	98	(26.6)
-Women	32	(32.0)	11	(36.7)	21	(30.0)
Participants with an ERI-ratio>1.0, N (%)**
All	503	(38.5)	230	(40.9)	274	(36.5)
-Men	414	(36.9)	210	(40.9)	204	(33.6)
-Women	89	(47.9)	20	(40.7)	69	(50.0)
Blue-collar	208	(39.0)	88	(38.3)	120	(39.5)
-Men	168	(37.5)	81	(38.2)	87	(36.9)
-Women	40	(46.5)	7	(38.9)	33	(48.5)
White-collar	295	(38.2)	142	(42.8)	153	(34.7)
-Men	246	(36.5)	129	(42.7)	117	(31.5)
-Women	49	(49.0)	13	(43.3)	36	(51.4)

* based on n = 1312;

** based on n = 1307

Multiple logistic regression analyses between psychosocial variables in total-sample analyses (Table 2), revealed mostly small and non-significant effects. A few calculations illustrated surprising results, such as augmented odds ratios for remaining at work if experiencing imbalance between effort and reward. However, these results became non-significant in fully adjusted models.

Table 2

Regression analyses of psychosocial conditions between cases with extended work ≥63 years and controls who retired ≤62 years.

		Crude		Model 2		Model 3
		ratio	95% CI	ratio	95% CI	ratio	95% CI
			p-value		p-value		p-value
All n = 1312	High strain	1	0.7–1.4	1.1	0.8–1.7	0.9	0.5–1.5
			0.9		0.5		0.7
	Active	0.9	0.6–1.3	0.9	0.6–1.4	0.9	0.5–1.4
			0.5		0.8		0.6
	Passive	1.1	0.8–1.7	1.2	0.8–1.8	0.9	0.5–1.5
			0.4		0.3		0.7
	Low strain	1	Reference	1	Reference	1	Reference
n = 1307	ERI ≥ 1.0	1.2	0.96–1.5	1.4	1.0–1.8	1.2	0.8–1.7
			0.1		0.02		0.7
Men n = 1125	High strain	1.2	0.8–1.7	0.9	0.5–1.4	1	0.6–1.7
			0.3		0.6		0.95
	Active	0.9	0.6–1.4	0.8	0.5–1.4	0.9	0.5–1.6
			0.7		0.5		0.8
	Passive	1.3	0-9-1.7	0.9	0.5–1.4	0.9	0.5–1.9
			0.2		0.6		0.8
	Low strain	1	Reference	1	Reference	1	Reference
n = 1121	ERI ≥ 1.0	1.4	1.1–1.8	1.4	1.0–2.0	1.3	0.9–2.0
			0.01		0.05		0.2
Women n = 187	High strain	0.5	0.2–1.4	1.2	0.3–6.0	0.9	0.1–10.5
			0.2		0.8		0.4
	Active	0.5	0.1–1.8	0.8	0.1–4.6	0.4	0.03–4.1
			0.3		0.8		0.4
	Passive	0.8	0.3–2.3 0.6	1.8	0.4–8.6	0.7	0.01–7.8
					0.5		0.8
	Low strain	1	Reference	1	Reference	1	Reference
n = 186	ERI ≥ 1.0	0.7	0.4–1.4	0.7	0.2–2.3	1.1	0.2–5.0
			0.3		0.6		0.9

Model2: adjusted for frequency of heavy physical demands, leadership quality and having a leadership position and retirement during crisis years and if it was the person’s own decision to retire

Model3: model 2 + agedifference to partner and life event

Analyses conducted among blue-collar workers (Table 3), showed that high strain jobs meant notable lower odds ratios for continued work ≥63 years, in both unadjusted and fully adjusted models (OR 0.4; 95% CI 0.1–0.96. Active work was also associated to lowered OR of continued work, but results were non-significant. When stratifying by gender similar, but non-significant results were found among men. Few blue-collar women resulted in wide confidence intervals in fully adjusted analyses. Results for effort-reward imbalance mostly were small and non-significant, except for analyses in women which displayed considerably lowered odds ratios, but due to few women, confidence interval were wide.

Table 3

Regression analyses of psychosocial conditions between cases with extended work ≥63 years and controls who retired ≤62 years in blue-collar workers.

		Crude		Model 2		Model 3
		ratio	95% CI	ratio	95% CI	ratio	95% CI
			p-value		p-value		p-value
All	High strain	0.5	0.3–0.9	0.4	0.3–0.9	0.4	0.1–0.96
n = 539			0.01		0.03		0.04
	Active	0.6	0.4–1.1	0.7	0.4–1.2	0.7	0.3–1.7
			0.08		0.2		0.4
	Passive	1	0.6–1.7	1.2	0.7–2.2	1.3	0.4–3.7
			0.99		0.5		0.6
	Low strain	1	Reference	1	Reference	1	Reference
n = 534	ERI ≥ 1.0	0.95	0.7–1.4	1	0.6–1.9	0.8	0.4–1.7
			0.8		0.9		0.3
Men n = 452	High strain	0.7	0.4–1.2	0.7	0.4–1.3	0.5	0.2–1.6
			0.2		0.3		0.3
	Active	0.7	0.4–1.2	0.7	0.4–1.3	0.8	0.3–1.9
			0.2		0.3		0.6
	Passive	1.2	0.7–2.2	1.5	0.8–3.0	1.9	0.6–5.9
			0.5		0.2		0.3
	Low strain	1	Reference	1	Reference	1	Reference
n = 448	ERI ≥ 1.0	1.1	0.7–1.6	1.3	0.7–2.4	1	0.5–2.3
			0.8		0.5		0.97
Women	High strain	0.2	0.01–2.8	N/A	<0.001–0.4	N/A	0
N = 87			0.08		0.01
	Active	0.3	0.03–2.9	0.1	0.02–1.8	N/A
			0.3		0.1
	Passive	0.4	0.044–3.8	0.1	0.03–1.8	N/A
			0.4		0.08
	Low strain	1	Reference	1	Reference	1	Reference
n = 86	ERI ≥ 1.0	0.7	0.2–1.9	N/A		0.3	0.04–2.1
			0.5				0.2

Model2: adjusted for social support, retirement during economic crisis years

Model3: model 2 + civil status, previous diseases, life events, stress at home

Regression analyses in white collar-workers (Table 4) displayed several unexpected results as high strain was related to considerably increased odds for continued work in older ages for total sample and for men, but the effects did not persist in fully adjusted models. In total sample and male white-collar workers high imbalance between efforts and just reward at work meant increased odds for working ≥ 63 years of age, but these effects became non-significant when adjusting for all potential covariates. Among white-collar women both high strain and ERI illustrated increased odds of continued work but confidence intervals were wide and non-significant.

Table 4

Regression analyses of psychosocial conditions between cases with extended work ≥63 years and controls who retired ≤62 years in white-collar workers.

		Crude		Model 2*		Model 3**
		ratio	95% CI	ratio	95% CI	ratio	95% CI
			p-value		p-value		p-value
All	High strain	1.6	1.0–2.3	1.3	0.8–2.2	1.4	0.8–2.6
			0.04		0.3		0.2
	Active	1.1	0.7–1.6	1	0.6–1.7	1.1	0.6–1.9
			0.8		0.9		0.8
	Passive	1.2	0.8–1.9	1	0.6–1.8	1.1	0.6–2.0
			0.4		0.9		0.9
	Low strain	1	Reference	1	Reference	1	Reference
n = 773	ERI ≥ 1.0	1.4	1.1–1.9	1.4	0.9–2.1	1.3	0.8–2.1
			0		0.1		0.2
Men n = 673	High strain	1.7	1.1–2.6	1.4	0.8–2.3	1.4	0.8–2.7
			0.02		0.3		0.2
	Active	1.2	0.8–1.8	1.1	0.8–2.4	1.2	0.6–2.2
			0.5		0.8		0.6
	Passive	1.3	0.8–2.2	1	0.6–1.9	1.1	0.6–2.1
			0.3		0.9		0.8
	Low strain	1	Reference	1	Reference	1	Reference
n = 673	ERI ≥ 1.0	1.6	1.2–2.2	1.5	0.97–2.3	1.4	0.9–2.3
			0.003		0.06		0.2
Women	High strain	0.9	0.3–2.9	1.8	0.3–9.3	1.4	0.1–14.2
			0.9		0.5		0.8
	Active	0.5	0.1–1.8	0.5	0.1–4.2	0.4	0.03–4.4
			0.3		0.6		0.4
	Passive	0.7	0.2–2.4	1.8	0.3–10.0	0.9	0.1–9.5
			0.5		0.5		0.9
	Low strain	1	Reference	1	Reference	1	Reference
n = 100	ERI ≥ 1.0	0.7	0.3–1.7	1.7	0.4–7.5	2.3	0.3–15.5
			0.5		0.4		0.4

Model2: adjusted for social support, retirement during economic crisis years

Model3: model 2 + civil status, previous diseases, life events, stress at home

Discussion

This case-control study found inconsistent results regarding relationships between psychosocial factors, measured by the JDC- and ERI-model, and extended work ≥ 63 years. In total sample and gender stratified analyses no JDC-variables were significantly related to retirement in models included all potential covariates. When stratifying by blue- or white-collar work it was found that, for blue-collar workers high strain was related to retirement at 62 or earlier in all models. When stratifying by gender similar, but non-significant results were found for the blue-collar men. Surprisingly, for white-collar workers high strain was related to notably increased odds ratios for continuing to work after 62 years, but these results did not persist in fully adjusted models. When analysing white-collar men and women separately it was found that high strain was related to continuing to work after 62 years in men. Similar contradictive results were found for imbalance between effort and reward, where high ERI was associated with increased odds for continuing to work for all white-collar workers, and for white-collar men, but these results were non-significant in the fully adjusted models.

Previous research have found that job strain is associated with intentions to leave work early [20], although gender and sector specific differences have been reported [11,12]. Gender stratified analyses in the current study did not support that high strain was related to earlier retirement for women. However, when stratifying by blue- or white-collar sector, high strain was related to retirement before 63 years for blue-collar workers. Our findings are similar to the results from a previous study [11], that found a relationship between job demands and early retirement intention, although in both blue- and white-collar workers. Blue-collar workers in the car manufacturing industry have a different working situation compared to white-collar workers with more physically demanding and strenuous working conditions [8]. Moreover, blue-collar workers generally tend to retire at an earlier age compared to white-collar workers [9]. It is therefore plausible that the process that drive retirement intention work differently for the two groups. Some demands may be more important for blue-collar workers and others may be more relevant to white-collar workers when predicting retirement intention [11].

A major limitation of this study is the risk of recall bias, as some of the participants retired up to ten years prior to completing the questionnaire. Indeed, the difficulty in assessing past exposures and the risk of recall bias is a well-known problem in case-control studies [30], where the estmiate of exposure is usually based on past records or recall of past events by participants. The accuracy of recall in interviews has been studied for a number of exposures with varying results [31]. It has been suggested that bias or imprecision in estimated exposure can be difficult to avoid in case-control studies, and therefore it is important to consider if these errors are likely to differ between cases and controls [31]. In order to assess the magnitude of this problem we performed sensitivity analyses and investigated differences between the psychosocial variables (job demand, job control and the ERI-ratio) depending on the time lag between the survey and retirement. The time since filling in the survey was categorized as currently working, or retired since 1 years, 2–4 years or more than 5 years and analysed using two-way ANOVA, with a p-value of 0.05 as significance level. Analyses illustrated that there were no statistically significant differences in the examined psychosocial variables, between those currently working or leaving work recently or more than 5 years ago, regardless of total-sample analyses or when stratifying by cases or controls. Nevertheless, recall bias can be a threat to the validity of the study. Only subjective measures of working conditions were used in the study as the main focus was the psychosocial working environment. In the current study, the JDC model was used to assess demands and this model was developed when studying blue-collar working conditions [16] and is therefore suitable to assess the job demands of this group. However, it may not be the best model to capture some of the demands faced by white-collar workers such as cognitive demands. Moreover, there was a small sample of women, thus the results relating to gender comparisons need to be treated with caution due to the limited statistical power. Further limitations included the fact that health status just before retirement, and individual differences (for example coping) were not measured. These factors could have an influence on the decision to retire.

This study can have practical implications because it contains data from the largest car manufacturer in Sweden, stratifies for blue- and white-collar work and gender, and includes important confounders such as life events. The results indicate that demanding psychosocial working conditions are related to earlier retirement for blue-collar workers. Hence, it is important to promote sustainable working conditions for these groups in order to retain older workers. For example, it may be useful to examine the job design to investigate if it is possible to decrease demands and provide more job autonomy for blue-collar workers. However, this case-control study investigates associations and not causation, and it is important to note that prospective studies will be required to determine what job factors that have a causal role in the timing of retirement.

Conclusions

In conclusion, this study found that for blue-collar workers high strain was related to retirement before 63 years.

(DOCX)

Click here for additional data file.

18 Aug 2019

PONE-D-19-15489

A sustainable working life in the car manufacturing industry: The role of psychosocial factors, gender and occupation

PLOS ONE

Dear Dr Gyllensten,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Below, you will see that three experts in the field have reviewed your paper. Overall, two of them stated it has a certain potential, but more work is needed before considering its publication in PLoS ONE. As -in my view- most of the appended comments and issues are amendable, I would like to give the authors the chance to address and respond all the comments provided by our reviewers. Although all the comments should be properly addressed, please pay special attention to those related to the methods, data analysis, scope and limitations of the study,

We would appreciate receiving your revised manuscript by Oct 02 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Sergio A. Useche, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

1. Please include a copy of any survey questions or questionnaire used in the study that have not been published previously, in both the original language and English, as Supporting Information. Please ensure that you have provided citations for any scales published previously.

2. We note that you have included in the following statement in your Acknowledgements "Volo Cars [sic]". We feel that the involvement of Volvo Cars should be included in the Competing Interests statement. Please state what role Volvo had in  conducting the study and decision to publish. Please include an updated Competing Interests statement in your Cover Letter and we will update the submission form on your behalf.

3. Thank you for including your financial disclosure statement; "The study was partly funded by AgeCap - a centre for aging and health at Gothenburg University. The funder did not have any involvement in the study, in the writing of the manuscript or the decision to submit the paper for publication.  "

Please provide an amended statement that declares *all* the funding or sources of support (whether external or internal to your organization) received during this study, as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now.  Please also include the statement “There was no additional external funding received for this study.” in your updated Funding Statement.

Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

4. Thank you for including your ethics statement: "The study was approved by the Ethical Review Board in Gothenburg in Sweden, dnr 371-15. The consent was written."

Please amend your current ethics statement to include the full name of the ethics committee/institutional review board(s) that approved your specific study.

Once you have amended this/these statement(s) in the Methods section of the manuscript, please add the same text to the “Ethics Statement” field of the submission form (via “Edit Submission”).

For additional information about PLOS ONE ethical requirements for human subjects research, please refer to http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research.

5. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

6. We note you have included a table to which you do not refer in the text of your manuscript. Please ensure that you refer to Tables 1 & 4 in your text; if accepted, production will need this reference to link the reader to the Table.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

Reviewer #3: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: No

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This study uses state of the art measures of stress at work – job strain (JDC) and effort-reward imbalance (ERI) – to evaluate the influence of stress at work on early retirement (age 55-62) vs continuing to work age >63. The findings of high strain being related to early retirement in blue collar workers but both high strain and high ERI being related to continued work in white collar men are interesting and of potential importance with respect to reducing early retirement in blue collar workers (assuming that’s considered a desirable goal). There are some concerns that need the authors’ attention.

1. Rather than, or in addition to, covarying several variables, there are some – especially previous diseases, social support at work, and stress at home – that, rather than being confounders, could be moderators (which can be documented by testing their interactions with job strain and/or EFI) of the influence of JDC and/or ERI on retirement time. If blue collar workers with previous disease, for example, are more likely to retire early if they are in high strain jobs, they could be targeted for interventions that might help them cope better and remain at work. If those with high support at work are less likely to retire early even if in high strain jobs, interventions to increase support at work could help workers in high strain jobs continue to work longer.

2. It would also be good to test the interaction of blue x white collar x JDC/ERI, to clearly document that effects of JDC and ERI on time of retirement are moderated by blue/white collar status.

3. There is no mention in the text that I can find of Table 1, which could include whether Cases and Controls differ significantly on any of the characteristics – e.g., 25% of women being cases vs 75% controls.

4. While the ratios and 95% CIs do document effects that are statistically significant, it would be helpful if they also included p-values for these effects.

5. The authors do note that a major limitation of this study is the risk of recall bias – e.g., those blue collar workers who retired before age 63 may be more likely to recall their job as having high strain than those who work past age 63. The authors need to note that before conclusions can be drawn regarding causality, the effect of high strain to increase incidence of early retirement in blue collar workers and of high strain and ERI delaying retirement in white collar workers, it will be essential to do a prospective study of current workers to determine whether high strain predicts early retirement in blue collar workers and high strain and ERI predict later retirement in which collar men before any conclusions can be drawn (and interventions developed and evaluated) regarding causality. If the JDC and ERI assessments were done while any of the workers in the current study were still working, it could be possible to see if JDC or ERI levels predict retirement ages in them.

Reviewer #2: 1. Is the manuscript technically sound, and do the data support the conclusions?

The most important limitation of this study is that psychosocial working conditions is measured after the decision to retire or not, and often many years later. This means that one may primarily measure the respondent’s justification for his/her decision and not how he/she felt about the working conditions in the time leading up to the decision. There is a lot of psychological research on cognitive dissonance and similar phenomena that give rise to considerable skepticism about the validity of such data.

Another limitation, although one shared with most studies in this tradition, is that only subjective measures of working conditions are used. This should, however, be clearly acknowledged. The research cannot address the issue of whether the working conditions in an objective sense has any effects or not.

Even beyond this, it should be made clear that findings are only descriptive, above all in the Aims section. Causal language should be avoided throughout the paper.

2. Has the statistical analysis been performed appropriately and rigorously?

A better justification of the selection of covariates is needed. Why are these particular covariates included? What about potential confounders not included?

I find several of the covariates problematic. What are, for instance, the implications of including health related variables that are likely to be endogenous to working conditions?

On the other hand it is notable that there are no measures of working and employment conditions apart from JDC and ERI. This means that the effects of these specific variables cannot be singled out. One illustration of this problem is provided by the descriptive results on distinguishing between management and other job positions (line 325ff.). Indeed, it seems strange to me that that this variable is not included as a covariate. Another obvious confounder (probably not measured) is the wage level.

Statistical power may be problematic (particularly in analyses of women), and should be addressed.

3. Have the authors made all data underlying the findings in their manuscript fully available?

No further comments.

4. Is the manuscript presented in an intelligible fashion and written in standard English?

No further comments.

Reviewer #3: This manuscript describes a study examining the association between psychosocial work characteristics and the propensity to retire early versus continue to work past the age of 62. The results are thought-provoking, indicating associations in different directions for blue collar male and white collar male car manufacturing workers.

However, the research design is somewhat problematic (and a little unclear). All employees who worked at a particular car company during the years 2005-2015, and either retired during this time (after age 55) or continued working past the age of 62 were included in the sample. The first potentially problematic issue is that people who were 55 years or older and who were laid off during this time were included in the sample (I think). Since the study is examining employees’ choices about retirement, it seems more appropriate to exclude the employees who did not have a choice.

The second problem is the timing of the data collection. Some of the participants may have retired 10 years prior to the survey data collection. Much literature suggests that recall bias can be a very real threat to validity--- that a person’s current state can greatly influence his or her report of previous exposures. While the authors attempt to dispel the seriousness of this potential threat, I find myself unconvinced.

Thirdly, the authors are not very clear as to the time period that participants were asked to consider when answering the psychosocial work characteristics questions. Were they asked to consider the work conditions being experienced at the time of their retirement? And what about those still working? Were they asked to report on the conditions at the time when they could have retired? The stress at home question seems to have asked about the last 5 years which could have been substantially post-retirement for some. To sum, a retrospective study design lends itself to several potential threats to validity.

In terms of the sample, the number of women is quite small. When the stratified regressions are conducted, there are very few in each of the categories of the psychosocial work characteristics variables. The study might be better served by only including men, and thus not making gender comparisons.

While the authors briefly review the associations between psychosocial working conditions and employee health, they do not examine the potential relationships between retiring at a later age and employee health. What if retiring at a later age is good for employers but bad for employee health? This topic needs to be addressed in the introduction.

Lastly, the discussion section does not clearly explain how this study moves the field forward. What do we know after reading this study that we didn’t know before? This study seems to be another flawed study that adds to the inconsistencies in the previous research. The same possible mechanisms/explanations for the effects were discernable from previous research.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

5 Nov 2019

Note: Page numbers refer to the pages on the revised manuscript with track changes.

Editors comments:

Comment 1. A document has been included with supporting information including survey questions that have not been published previously.

In the methods section on page 7 citations for scales published previously have been included.

Comment 2. Updated competing interest statement. Thank you for updating the submission form on our behalf.

Volvo Cars provided contact details, age, details of level of employment and time of employment, of all employees that had worked/ and or retired at Volvo between the years of 2005-2015 that were 55 years or older. Volvo did not have any further involvement in conducting the study or in the decision to publish the study.

Comment 3. The funding statement has been updated.

The study was funded by AgeCap – a centre for aging and health at Gothenburg University and by the Swedish Research Council for Health, Working life and Wellfare. The funders did not have any involvement in the study, the writing of the manuscript or the decision to submit the paper for publication. There was no additional external funding received for this study.

Comment 4. The full name of the ethics committee has been included in the methods section on page 6, and in the submission form.

Comment 5. Data available on request.

a) Data cannot be shared publicly, but, anonymised data is available with an approval from an ethical review board. For data requests, contact: Department of occupational and environmental medicine, Gothenburg University, Box 414, 405 30, Gothenburg, Sweden.

Comment 6. Table 1 and 4 is now referred to in the text in the results section on page 10 and 14.

Reviewer 1

Comment 1. We agree that mentioned analyses may add more information to the decision of retiring. Exploratory analyses were conducted testing interaction effects between the main psychosocial variables (job demand-control / effort-reward imbalance) and the suggested variables (previous diseases, social support at work). The interactive effects were mostly small and all were non-significant. We conclude that these variables are not moderators, and will not add the results to our paper.

Comment 2. Exploratory analyses showed interactive effects between high strain and blue-/white-collar worker. However, the paper already includes analyses stratified by blue- white-collar worker, and illustrates effects of psychosocial variables may differ depending on being blue- or white-collar workers.

Comment 3. Table 1 is now mentioned in the text on page 10.

Comment 4. P-values are included in table 2-4 in the manuscript on pages 12, 13, 14. Some calculation errors were noted and corrected. Our results remain largely similar

Comment 5. Conclusions regarding causality has been removed on the following pages.

In the abstract on page 2.

In the aims on page 5.

In the implications on page 21.

Reviewer 2

Comment 1. In limitations on page 20 a sentence has been added that acknowledges that only subjective measures of working conditions were used in the study.

Casual language has been removed in the following sections:

In the abstract on page 2.

In the aims on page 5.

In the implications on page 21.

Comment 2. We evaluated several confounders we identified as important, based on previous literature and meetings with Volvo staff (including mangers, white-collar and blue-collar workers). The following variables were considered: civil status, age of partner in relation to the participant, country of birth, retirement during a year of recession, being offered a beneficial retirement deal, if retirement was the persons own decision, previous diseases, work ability at the time of retirement, social support at work, need for recovery after work at the time of retirement and stress at home at the time of retirement. Since working with a fairly small cohort, we performed stepwise purposeful selection according to Lemeshow & Hosmer (2000) reference 26, to slim our models, which resulted in the confounders presented in the paper.

We had access to other work condition variables (e.g. noise, physical demands), but our main aim with this paper was to focus on psychosocial variables. Also, this paper examines both white- and blue-collar Volvo workers. Noise and physical demands likely have little effect among white-collar workers. We plan to examine effects from physical work factors among blue-collar workers in future studies.

As suggested we examined other psychosocial variables that might be important such as managerial/executive position and overtime per week. Both variables had little effect and did not full-fill the criteria according to Lemeshow & Hosmer (2000).

We did not have access to wage level, which is a limitation, but Volvo workers at this time, where provided with a beneficial pension deal, so that pension would be similar regarding if retiring at 62 or statutory Swedish pension age 65. Wage may therefore be of less importance in this cohort, than other similar studies.

Comment 3. The limited statistical power realting to the the small sample of women is addressed in the limitations on page 20.

Reviewer 3

Comment 1. The reviewer highlighted that people over 55 years that were laid off appeared to be included in the sample. Individuals who had been laid off were not included in the sample. Thank you for highlighting that this was unclear. A paragraph clarifying this has been included in the study population section on page 6.

Comment 2. We agree that a major limitation of this study is the risk of recall bias. In the manuscript we have revised the text relating to limitations on page 20 to further acknowledge this problem.

Comment 3. The reviewer highlighted that the time period that the participants were asked to consider their psychosocial environment was unclear. The cases were asked to consider their current working environment, or the working environment at their last workplace at Volvo. The controls were asked to consider the working environment at their last workplace at Volvo. Information about this has been added in the section describing the questionnaire on page 7.

Comment 4. We agree that sample of women is quite small and the problem with power has been acknowledged in the section on limitations on page 20. However, despite the small sample we still think there is a value with including women in the analysis.

Comment 5. A recognition of the fact that late retirement can be problematic for certain employees have been added in the beginning of the introduction on page 3.

Comment 6. There is currently an active debate, especially regarding job demand-control, that these models that were developed several decades ago, and due to a changing labor market, may have different meaning compared to when they were created. During the time of the forming of the models, the typical high strain work situation was found in blue-collar workers e.g. in line production. Many, however, argue that due to technology, several white-collar workers in modern labor market have increasing demands linked to less boundaries and longer work hours. The understanding how to interpret and measure psychosocial work conditions needs to be updated, but to do so, there is a need to first establish limitations in standard methods.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

4 Dec 2019

PONE-D-19-15489R1

A sustainable working life in the car manufacturing industry: The role of psychosocial factors, gender and occupation

PLOS ONE

Dear Dr Gyllensten,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The paper has been reviewed for a second time by two acknowledged experts in the topic addressed in your manuscript. Although some improvements are evident, it is true (especially considering the feedback received from Reviewer # 2) that more precision, extension and efforts are needed in regard to the amendments they asked for during their first review.

Actually, and considering the aforementioned, one of the reviewers have advised the rejection of the paper. However, I believe the manuscript has potential, and the topic is worth of investigation; thus, I would like to encourage the authors to perform an adequate revision of it, involving all the comments received in both review phases, with more level of detail and accuracy in regard to what is expected from reviewers. Please carefully follow the journal guidelines and cover all the pertinent queries.

We would appreciate receiving your revised manuscript by Jan 18 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Sergio A. Useche, Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have done a good job of responding to my concerns in this revision. That said, there remain some concerns that need to be addressed.

1. The exploratory analyses they performed in response to my Comment 1 showed no significant interactions between job variables and other psychosocial variables (previous diseases, social support at work, etc.) and hence support the conclusion that effects of job characteristics on retirement time were not moderated by these other variables. Rather than NOT adding these results to the paper, I believe it would be better to report these non-significant interactions in the paper.

2. Similarly, I would include in the revised ms the results of the exploratory analyses that showed significant interactions between high strain and blue/white collar status – thereby strengthening the case that blue/white collar status moderates the influence of job factors on retirement time.

3. The changes in text they have made to reduce conclusions regarding implications of their findings for causality of early retirement are good. In addition, I believe it would be good to indicate at the end of the Discussion that prospective studies will be required to document further that job factors are playing a causal role in retirement time.

4. A minor point: They need to insert “p-value” beneath “95% CI” at the top of the columns in Table 4.

Reviewer #2: Unfortunately, I do not find much improvement in this version. Although a few sentences are changed, the whole justification for the paper set up in the Introduction is the need for knowledge about causes (a few examples, 52-53, 62-64, 101-102). Im the paper is not intended to address this, then what is the contribution? I do not find any clear statement of that.

With regard to recall bias, problems of confounding, etc. there is no concrete and serious discussion of this might affect the results of the analyses.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

27 Feb 2020

Reviewer #1

1 & 2. We agree to these well-thought of and insightful comments, but after thorough discussion in our project group we have concluded the following. It’s correct that several of the chosen covariates could possible also be moderators. We do, however, believe that there are many studies in this field with a similar design examining psychosocial variables and using models with covariates that could be both moderators or covariates. Even though it may add interesting information, is not common to calculate or display results on interaction effects. We still carried out exploratory analyses evaluating interaction, but none were significant. In the previous review we had other models, but as suggested from reviewer 2 we included more work related covariates and interaction effects between for example psychosocial variables and socioeconomic status are now not significant anymore. We agree to your suggestion that it may benefit the study in some ways, but we have decided not to expand our results with interaction analyses, since most studies do not and since interaction effects were non-significant it may not contribute that much.

3. Thank you for this helpful suggestion. We have added a sentence about the fact that prospective studies are needed in order to find out more about casual factors. Page 19.

4. Thank you, p-value has been added to Table 4. Page 14.

Reviewer #2

Contribution of the paper

We acknowledge that this study cannot discuss causes of retirement. However, we believe that this study adds to existing knowledge regarding the associations between psychosocial factors and the timing of retirement. There is little research investigating these factors in the car manufacturing industry. We have revised the aim in order to emphasize that we are investigating the relationships between psychosocial work factors and retirement, and thereby aim to add to existing knowledge. Page 5.

Problems with recall bias

We agree that this can be a serious problem, and have added further reasoning regarding this issue in case-control studies in the discussion. Page 17-18.

Problems with confounding

We agree that our models needed to be revised and more covariates should be considered. Since modelling and used covariate is one of the main issues in the reviewer comments we have expanded the methodology section in which we explain the methodology for inclusion of covariates: stepwise purposeful selection as proposed by Hosmer & Lemeshow [1] (page 8). The rationale for using this method is that theoretically a broad variety of available covariates could be important in analyses in psychosocial work environment and retirement. However, since our cohort size is fairly small, and include analyses that are stratified by gender or blue-/white-collar workers, we wish to slim the models by only including covariates relevant for our analyses. Meaning, if a variable does not fulfill the criteria in the established selection process, they have little relevance for the model and evaluated effects.

The following potential confounders, which previous have been identified in the literature or by practical experience as important in the retirement process, were tested with stepwise purposeful selection as proposed by Hosmer & Lemeshow [1]:

civil status, age of partner in relation to the participant, country of birth, retirement during a year of recession, being offered a beneficial retirement deal, if retirement was the persons own decision, amount of years since retirement, previous diseases, work ability at the time of retirement, need for recovery after work, life events, stress at home, physical job index, quality of leadership, managerial position and social support at work.

Per suggestion we especially added work related covariates such as frequency of physically heavy job tasks [2], leadership quality [3] and leadership position in the company (yes/no). The reviewer suggest salary as an important co-variate, unfortunately we do not have access to this variable. It has been suggested that for those with high status jobs, the salary per se is not determining for retirement, but rather social status and identity tied to the position. For blue-collar workers physical health and actual retirement benefits, is more important. If health if failing, salary may have less important. Furthermore, this cohort is slightly unique as the employees, due to a very generous deal with Volvo Cars, could retire at 62 years, and receive similar pension as if they would have retired at 65 years, which was the statutory retirement age during our observation period. Thus, salary hopefully, have less importance for in this particular cohort. The reviewer also suggested we remove health as a covariate given its endogenous property in relation to work conditions. We did include it in our purposeful stepwise confounding testing, and as it now, including all new work related variables, the variable previous diseases, did not fulfil the inclusion criteria, it is no longer included in our models.

1. Lemeshow S, Hosmer D: Applied Logistic Regression (Wiley Series in Probability and Statistics. Hoboken: Wiley-Interscience; 2 Sub edition; 2000.

2. Arbetsmiljöverket: Arbetsmiljön 2013. In: Arbetsmiljöstatistisk Rapport 2014:3. vol. 2014:3; 2014.

3. Kristensen TS, Hannerz H, Høgh A, Borg V: The Copenhagen Psychosocial Questionnaire-a tool for the assessment and improvement of the psychosocial work environment. Scandinavian journal of work, environment & health 2005:438-449.

Submitted filename: response to reviewers.docx

Click here for additional data file.

5 Mar 2020

PONE-D-19-15489R2

A sustainable working life in the car manufacturing industry: The role of psychosocial factors, gender and occupation

PLOS ONE

Dear Dr Gyllensten,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Thanks for the work done in regard to the previous set of revisions requested by our reviewers. Below, you will find an additional set of comments (most of them considerably minor, but still important) raised by the Reviewer #1. Please address all them with all the possible rigor for submitting the paper to a final round of reviews, and the potential acceptance of the paper in case the changes and rationales were accepted.

We would appreciate receiving your revised manuscript by Apr 19 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Sergio A. Useche, Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have responded appropriately to my concerns 3 and 4. I have ongoing concerns with their argument for not reporting interactions.

Concern 1. Reporting nonsignificant interactions between job variables and JDS and ERI – the authors did assess these interactions and found that “none were significant.” This indicates that none of these job variables moderated effects of JDS and ERI on retirement, and I see no reason not to report this absence.

Concern 2. In response to my prior review they indicated they had done exploratory analyses that did show significant interactions between high strain and blue/white collar status. It is important to report this interaction, because it justifies doing separate analyses of JDS and ERI associations with retirement time in blue- and white-collar workers.

In the text regarding the regression analyses shown in Table 4 for white-collar workers, they neglect to report that, in addition to high strain being associated with extended work in all and male workers, high ERI is also associated with extended work in all and male subjects -- ratios 1.4 (1.1-1.9) and 1.7 (1.2-2.2) respectively. They need to mention this in text.in Abstract, Results and Discussion.

They note that associations of high strain jobs with lower odds ratio of continued work >63 years are significant in both unadjusted and fully adjusted models in blue-collar workers. In their report of high strain jobs being associated with increased odds of continuing to work in all and male white-collar workers, they say (line 281) this effect “persisted in fully adjusted models and in men.” Inspection of Table 4 shows that this (and ERI’s) association was not significant in analyses with Models 2 and 3 adjusted, which is noted in next sentence (lines 283-284). They need to correct line 281.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

19 Mar 2020

Reviewer #1

Concern 1 & 2We deeply regret that we expressed ourselves unclear, but in the previously response letter we reported that the interaction effects between psychosocial variables and socioeconomic status (division: blue-/white-collar workers) were NOT significant anymore when entering the additional confounders requested in the previous revision by reviewer #2. So therefore, we prefer not to report these in the article. Also, if reporting on interactive effects for: psychosocial variables*other job variables; psychosocial variables * socioeconomic, then there is little justification to why we should not report interaction between main psychosocial variables and all covariates, and this would shift the focus of the study somewhat. As our study design focuses on psychosocial variables and relationships to continued work, we are not sure if this add much value, especially as the interactions mentioned are non-significant. Regarding adding the interaction because this is a justification for doing separate analyses of blue- and white-collar workers, we would argue that it is common practice to do separate analyses for these two groups. Moreover, it in the case of Volvo, it is clear that they have very different working environments.

We do, nevertheless, appreciate the reviewer’s valuable suggestions regarding how to investigate the complexity in a retirement process as multiple variables have indeed a joint effect on an executed retirement or not. We are also very interested in such joint effects on psychosocial variables and socioeconomic markers, as similar topics have arisen in other of our research projects. We therefore plan for our next study to look at retirement in clusters of persons with specific set of several physical and psychosocial work conditions and socioeconomic markers (more than just blue-/white-collar worker, which may be too simplified). However, since the design used in the current paper is in accordance with general practice in analyses with job demand-control and effort reward imbalance (two of the authors main research focus is these psychosocial job variables) we felt it important to start with a first study using a standard design, and then build on those findings to make a more complex study in our next paper.

Regarding the comment on Table 4, that we neglected to report that high ERI is associated with extended work in all, and males - we did report this in the Results, but agree that the text was unclear. We have revised the text on page 14 to make it more clear.

We have also added this information in Abstract – p 2, and in the Discussion – p 16.

Regarding the mistake in line 281, we have corrected this mistake both in the Results – line 281, and in the Discussion - page 16. Once again, we are thankful for the thorough review.

Submitted filename: response to reviewers.docx

Click here for additional data file.

7 Apr 2020

PONE-D-19-15489R3

A sustainable working life in the car manufacturing industry: The role of psychosocial factors, gender and occupation

PLOS ONE

Dear Dr Gyllensten,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

After assessing this revised version of the manuscript, the Reviewer has expressed their satisfaction with most of the improvements and clarifications done in your last round of revisions. Nevertheless, an additional issue (that is important) has been raised and requires your attention. Please see the comments below.

Also, I noticed that the raw data supporting the study results is not already available in your submission. In this regard, please consider that PLOS Data policy is quite strict in this regard ("PLOS journals require authors to make all data necessary to replicate their study’s findings publicly available without restriction at the time of publication. When specific legal or ethical restrictions prohibit public sharing of a data set, authors must indicate how others may obtain access to the data"), and it must be fully available in order to accept the paper for publication.

We would appreciate receiving your revised manuscript by May 22 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Sergio A. Useche, Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: They do note in the Results and early in the Discussion that the finding of high strain (and ERI) association with more likelihood to continue work becomes nonsignificant in the fully adjusted models. This casts doubt on the validity of this association -- suggesting that the extensive discussion of it (lines 328-349, P. 17) should be omitted.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

9 Apr 2020

Comment from editor: I noticed that the raw data supporting the study results is not already available in your submission. In this regard, please consider that PLOS Data policy is quite strict in this regard ("PLOS journals require authors to make all data necessary to replicate their study’s findings publicly available without restriction at the time of publication. When specific legal or ethical restrictions prohibit public sharing of a data set, authors must indicate how others may obtain access to the data"), and it must be fully available in order to accept the paper for publication.

Response: In our response to reviewers submitted in October 2019, we included the following text. The statement includes information regarding how the data can be obtained. We hope that this is sufficient information regarding availability .

“Data cannot be shared publicly because information about health and symptoms are regarded as sensitive information, and when sharing such data there has to be an approval from a Swedish Ethical committee (according to Swedish law). However, anonymised data is available with an approval from an ethical review board. For data requests, contact: Kristina Gyllensten, Department of occupational and environmental medicine, Gothenburg University, Box 414, 405 30, Gothenburg, Sweden. The name of the data set is ‘The Volvo Work Ability Study’.

Reviewer #1: They do note in the Results and early in the Discussion that the finding of high strain (and ERI) association with more likelihood to continue work becomes nonsignificant in the fully adjusted models. This casts doubt on the validity of this association -- suggesting that the extensive discussion of it (lines 328-349, P. 17) should be omitted.

Response: Thank you for pointing this out. Lines 328-349 has now been deleted.

Submitted filename: Response to reviewers.docx

Click here for additional data file.

28 Apr 2020

A sustainable working life in the car manufacturing industry: The role of psychosocial factors, gender and occupation

PONE-D-19-15489R4

Dear Dr. Gyllensten,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Sergio A. Useche, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

1 May 2020

PONE-D-19-15489R4

A sustainable working life in the car manufacturing industry: The role of psychosocial factors, gender and occupation

Dear Dr. Gyllensten:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Sergio A. Useche

Academic Editor

PLOS ONE