Investigating the Relationship between Stress and Self-Rated Health during the Financial Crisis and Recession in 2008: The Mediating Role of Job Satisfaction and Social Support in Spain

As can be seen in the manuscript, the methodology section has been modified and the information on methodology has been expanded (pp 21-24)

1.     Repeated cross-sectional study using Spanish Surveys from2006 to 2017, a total of 32.105 participants (47.4% women) aged 16 years and over (M = 42.3, SD = 10.7) answered a series of questions about work-related stress (PV), self-rated health (CV), job satisfaction and social support (mediator variables) through the National Health Survey (NHS) Prevalences of work-related stress, self-rated health, job satisfaction and social support were calculated standarized by aged. We performed Mediation/moderation analysis with Macro Process for SPSS to analyse the roll of social support and job satisfaction in the relationship between self-rated health and work-related stress among the Spanish working population.

2.     Thank you for your comment, the word "discussion" has been removed from the text because, as you pointed out, it was not correct.

1.     In response to your suggestion, this information has been better explained and expanded in the introduction section. Two bibliographical citations have also been added to support our argument. The bibliographical citations added to the text are shown below (and are included in the reference section in yellow) (pp 59-67)

Many authors have described a protective effect of employment on self-rated health. Studies carried out in Spain during the economic crisis …. However, Aguilar Palacio et al. in a study carried out with data provided by the National Health Surveys in Spain, showed that women, after entering the labour market, reported improvements in their self-perceived health during this period of economic recession.

2.     It is not a literal quote, but as suggested by the reviewer 3 we have restructured this section and this quote has been removed.

3.     The objective was indeed repeated and has been removed from the part you suggest.

4.     You are indeed right, and it has been amended.

5.     The methodology section is rewritten (pp 197-217).

We used a repeated cross-sectional study design. This study was based on information provided by the National Health Surveys (NHSs) in 2006, 2011 and 2017. The NHSs are representative …….. Seasonal effect was avoided by including autumnal months in the sample collection. The methodology applied allows comparability between surveys. Sample sizes ranged from 29,478 in 2006, 20,884 in 2011 and 22,903 in 2017.

In this study we selected from these health surveys the ‘working population’. The term ‘working population’ is used to refer to everyone aged 16 years and over who, …….hose still living under the care of their parents or legal guardians require parental permission until age 18 [40]. Therefore, in this study the participants were 32,105 members of the Spanish working population (47.4% women) aged 16 years and over (M = 42.3, SD = 10.7) during the study period (2006-2017).

6.     It was indeed necessary to make such a modification to the manuscript. Thank you for your suggestion. (pp 197-217).

We used a repeated cross-sectional study design. This study was based on information provided by the National Health Surveys (NHSs) in 2006, 2011 and 2017. The NHSs are representative ……. Seasonal effect was avoided by including autumnal months in the sample collection. The methodology applied allows comparability between surveys. Sample sizes ranged from 29,478 in 2006, 20,884 in 2011 and 22,903 in 2017.

In this study we selected from these health ……therefore, in this study the participants were 32,105 members of the Spanish working population (47.4% women) aged 16 years and over (M = 42.3, SD = 10.7) during the study period (2006-2017).

7.     As suggested we added clear information about practical implications (pp 493-506 and 512-514).

Human resource management practices …… According to the Job Demands and Resources model, increasing challenging demands would improve employee’s satisfaction, followed by providing them with useful resources to cope with these demands.

As the Social Cure Theory stated, multiple group membership could be considered a protective factor to cope with disease and affliction, based on recent empirical findings

First of all, we would like to begin by thanking you for your collaboration in the review of this manuscript. We believe that both their contributions and those of the other reviewers have made the text clearer and of higher quality. Thank you very much for your time and we hope that all your suggestions will be answered below.

1.     The manuscript has been revised and this has been clarified. Thank you for your appreciation and we think that after the revision the article has gained in quality. You are also right that as long as the explanation about the two mediators is not clear. For this reason, we have introduced the following phrases (pp 14-15).

·          or whether variables such as social support or job satisfaction attenuate the negative effects of stress on self-rated health

·         as well as the mediating role of social support and job satisfaction in the previous mentioned relationship,

2.             In response to your suggestion and to make it clearer in the manuscript, the specific question used in health surveys to ask about self-perceived health and how the variable work stress was created by making it a dichotomous variable has been added (in yellow). (pp 232-242).

Self-rated Health. This was measured through a question which aimed to measure respondents' perceptions of their general health status over the last twelve months, specifically, the question used in national health surveys is How would you say your health has been in the last twelve months?. The question offered 5 response options: very good, good, regular, poor and very poor health.

Work-related stress is measured in the NHSs through the following question: 'Overall, and bearing in mind the conditions under which you work, how stressful would you say your job is?' Response options range from 1 (Not at all Stressful) to 7 (Very Stressful). For this study, the variable work stress was recoded into a dichotomous variable in which all respondents who answered 1 were considered to have no work stress and the remaining responses (from 2 to 7) were grouped into the category of having work stress.

3.     As suggested, the procedures to recode the values are moved to the measurement section

1.     You were right. The correct thing to do is to explain in the methodology section all the analyses carried out, which is why the paragraph that referred to this has been removed from the results and placed in the corresponding section (methodology). (pp. 268-276)

To determine whether job satisfaction and perceived social support mediate the relationship between self-rated health and work-related stress, three mediation analyses were carried out using the Process macro for SPSS [46], one for each time point in the study period. Age (coded as 0 = < 40 years and 1 = > 40 years), sex (coded as 0 = men and 1 = women), education level (coded as 0 = non university and 1 = university) and profession (coded as 0 = unqualified jobs -social classes IV, V and VI- and 1 = qualified jobs -social classes I, II and III) were included as covariables. The bootstrapping technique with 10,000 subsamples was used to estimate the confidence interval (95%).

3.     The word has been corrected in the manuscript. Thank you very much for your review. (p. 279).

The analyses were carried out using IBM SPSS Statistics 24 ® and Stata 14®.

5.     Indeed, for some reason, the legend in figure 2 was not transcribed. We have added the legend (pp.352-356)

Figure 2 A, B and C: Results of the mediation analysis. Criterion variable: self-perceived health, explanatory variable: job stress and mediating variables: job satisfaction and perceived social support, where a, b, c, a', b', c' are the direct effects of medication. Indirect effects are represented by Bsatis (indirect effect of satisfaction) and Bapoyo (indirect effect of perceived social support).

6.     The quality of figure 2 has been improved.

7.     Following your indications, in the discussion the results section has been summarised as follows (pp. 393-401).

The results reveal how, during the recession, the working population in Spain decreased, particularly in terms of …… However, we also observed that perceived social support only had the same positive mediating effect in the years prior to the recession (2006).

8.     As suggested, we have expanded the discussion about the mediating variables' roles (pp. 447-461)

However, our results also show that only apply to men, younger workers, those with qualified jobs and those with a high education, but not to women……….This situation could place women, older workers and those with unqualified jobs or low education in an even greater resource loss spiral relative to men or (young) workers with qualified jobs and higher education and, therefore, be more vulnerable to additional losses (Dohrenwend, 1978).

Dear reviewer, thank you for your comments and for taking the time to review this manuscript. After modification we feel that the text is clearer and has gained in quality. We hope that we have responded to all your suggestions. You can review the changes made to the text point by point below.

1.     The title you propose seems to us to be appropriate and reflects very well the work presented. Thank you for your suggestion and the change has been made (pp 1-4).

Investigating the Relationship between Stress and Self-Rated Health during Financial Crisis and Recession in 2008: The Mediating Role of Job Satisfaction and Social Support in Spain

2.     As suggested, we have clarified the objective in the summary. However, we had to redo the last sentence of the background in order to further clarify the objective of the study (pp 14-15).

However, few studies focus on stress and self-rated health among the Spanish workforce or analyse which variables can act as a buffer against the negative effects of stress on self-perceived health. Aim: To analyse the mediator role of social support and job satisfaction in the relationship between work-related stress and self-rated health among the Spanish working population between 2006 and 2017.

3.     Indeed, the introduction was brief and needed more information to support and justify the need for this study. Following your suggestion, the appropriate modifications have been made and three bibliographical citations have been added to support the paragraphs introduced. The bibliographical citations are added within the references section (pp. 43-57).

Spain is one of the countries where the economic crisis has had the greatest impact. Although the crisis began in 2008, it was not until the first quarter of 2009 that its first consequences became apparent, and direct and indirect effects of its impact on health can be differentiated [9-10]:

A.The direct effects can be classified as: i) Influence on the social structure of the population: the economic crisis has influenced the welfare of the population by sharply decreasing employment, increasing poverty rates and the lack of social policies. ii) Influencing health systems to contain economic expenditure.

B.The indirect effects: loss of priority in health for both governments and the population itself as economic problems take precedence.

It is very difficult to monitor the health consequences of the various social, health and employability measures taken during the economic crisis of 2008. Monitoring this fact, the effects of the crisis on the health of different groups of the population, is important in order to assess whether the basic criteria of equity and efficiency with regard to citizens' health have been met.

The association between employment and health is well-known [11]. Many authors have described a protective effect of employment on self-rated health. Studies carried out in Spain during the economic crisis in this respect show divergent results depending on the methodology used. Some authors, such as Arroyo et al.[12] conclude that there are no statistically significant differences in the self-perceived health of the Spanish population before and during the economic crisis. However, Aguilar Palacio et al.[13] in a study carried out with data provided by the National Health Surveys in Spain, showed that women, after entering the labour market, reported improvements in their self-perceived health during this period of economic recession.

5.     We have restructured this section by adding the demand control model and the results of two studies pertinent to the objectives of the study. The bibliographical citations are added within the references section (pp. 89-95 and 104-109)

From the perspective of organizational psychology, there are many theoretical approaches to stress. Nevertheless, the Job Demands-Resource (JD-R) model ……resources (associated with psychological or physical strengths).

In this vein, several studies have demonstrated …… regardless of demand and control levels.

6.     As suggested, clarified this section relating stress, health, and social support, adding empirical research that supports previous relationships. The bibliographical citations are added within the references section (pp 140-154).

In fact, according to the JD- R model, social support could buffer the impact of job demands, considered as work- related stressors, on employee's health.

Thus, research on the relationships ….. Similarly, research suggests that high levels of social support (for medical staff) are positively related to self-efficacy and negatively to the degree of anxiety and stress

7.     The whole part of the methodology has been modified and the part of the selection of participants has also been clarified (pp. 209-217).

In this study we selected from these health surveys the ‘working population’. The term ‘working population’ ……. Therefore, in this study the participants were 32,105 members of the Spanish working population (47.4% women) aged 16 years and over (M = 42.3, SD = 10.7) during the study period (2006-2017).

8.     Along the same lines as above and in order to improve the understanding of this part of the manuscript, the entire methodology part has been modified and the part on the selection of participants has also been clarified. The part on instrument and procedure has been removed because the modification of the entire methodology did not give rise to this section. Finally, the methodology part is as follows (pp 198-217).

·          We used a repeated cross-sectional study design. This study was based on……. Sample sizes ranged from 29,478 in 2006, 20,884 in 2011 and 22,903 in 2017.

·          In this study we selected from these health surveys the ‘working population’. The term ‘working population’ ….. Therefore, in this study the participants were 32,105 members of the Spanish working population (47.4% women) aged 16 years and over (M = 42.3, SD = 10.7) during the study period (2006-2017).

(-)

(-)

7.     Thank you. As suggested we have renamed the section as Practical Implications.

8.      As suggested we added clear information about practical implications (pp 493-506 ,  512-514).

·          Human resource management practices …… According to the Job Demands and Resources model, increasing challenging demands would improve employee’s satisfaction, followed by providing them with useful resources to cope with these demands.

·          As the Social Cure Theory stated, multiple group membership could be considered a protective factor to cope with disease and affliction, based on recent empirical findings

·          In order to increase jobs and curb unemployment, the quality of jobs and thus the health of workers can be neglected at the political level. It is important to generate employment, but in healthy conditions, where we are able to articulate exposure to new occupational hazards such as current biological hazards, with increased socialisation of workers, stress reduction and satisfaction. Occupational health services must pay attention to these new circumstances and to the new challenge of trying to protect workers' health in an era where less is more.