Is There an Estimation Bias in Occupational Health and Safety Surveys? The Mode of Administration and Informants as a Source of Error

Abstract

Information quality deficiencies have been detected in occupational safety and health surveys in Europe, which typically gather self-reported data responded by employers or their representatives. For instance, their low response rates and informant profiles make estimations on establishments with safety representatives (SRs) unreliable. We tested the mode of administration and informants as sources of error regarding establishments with SRs in Catalonia, Spain. Two sources of information were compared: the Second Catalan Survey of Working Conditions 2011 (IICSWC)—with a methodology similar to surveys conducted at the state and European level—and the Progam on Prevention of Risks Management in Companies (PPRMC)—in which the labor authority collected data using a documentary verification in another sample of establishments. Percentage of establishments with SRs was estimated using the data from the PPRMC and also the differences in percentage between sources and informant profiles (with 95 percent confidence interval). Results show that the IICSWC overestimates the percentage of establishments with SRs.

Keywords

estimation bias data quality employee participation information systems occupational health survey research

In order to assess the state of occupational safety and health in companies, surveys on occupational safety and health management (Departament d’Empresa i Ocupació [DEMO] 2012a; European Agency for Safety and Health at Work 2009a; European Foundation for the Improvement of Living and Working Conditions (EUROFOUND) 2009; Instituto Nacional de Seguridad e Higiene en el Trabajo (INSHT) 2009) and those on working conditions (DEMO 2012b; EUROFOUND 2010; INSHT 2012) are the most valuable sources of information. These surveys allow us to obtain indicators to guide and define not only public policies but also actions carried out by the agents involved in this issue (García and Gil 1996). Additionally, in social sciences such as sociology, industrial relations, and public health, these survey data have been used extensively for research purposes. A number of studies relying upon these surveys can be found, for instance, with the goal to analyze the situation, distribution and evolution of working and employment conditions, the extent of worker participation, or health-related impacts (Jones et al. 2013; Ollé-Espluga et al. 2015; Van Aerden et al. 2015; Smith et al. 2008).

However, occupational safety and health management surveys have certain deficiencies concerning information quality, owing to biases in the data gathered (GREDS-EMCONET 2013; Lucy and Sinclair 2012; Petrakos, Kleideri, and Ieromnimon 2011; Walters et al. 2012). The origin of these biases may be related to the profile of the informant (Peersman et al. 2014; Violán et al. 2013), the manner in which the information is collected (European Agency for Safety and Health at Work 2010; GREDS-EMCONET 2013; Lucy and Sinclair 2012; Martínez-Sánchez et al. 2013; Petrakos et al. 2011; Walters et al. 2012), and, connected to this, the response rate (Marta-Pedroso, Freitas, and Domingos 2007).

The estimate of establishments with safety representatives (SRs) stands as a prime example of the need to examine data quality in occupational safety and health management surveys. Worker participation in occupational health, especially that of the SR, has been associated with an improvement in health at work (Menéndez, Benach, and Vogel 2008; Walters 2006). This improvement has been shown in terms of direct impacts, such as the reduction in injuries due to accidents at work (Reilly, Paci, and Holl 1995) and in the prevalence of work-related illnesses (Mygind et al. 2005), but above all in terms of indirect impacts, such as the improvement in prevention policies in companies (Coutrot 2009). In the European Union, Framework Directive 89/391/CEE (European Council Directive No. 89/391/EEC 1989) recognizes the workers’ right to be consulted and participate in matters of health and safety at work. This Directive was transposed into the Spanish Law 31/1995 on Prevention of Risks at Work (1995), which regulates the number of SRs to be designated according to the number of workers in the company or workplace. Specifically, the designation of SRs is stipulated in companies or establishments with more than five employees (Real Decreto Legislativo 1995); establishments with fewer than six employees belonging to companies with more employees may have SRs. The constitution of a Health and Safety Committee is also stipulated in establishments with more than 49 employees.

Despite the relevance of worker representative participation in preserving health and safety at work, this is a relatively disregarded topic; moreover, the validity of existing data raises concern. For instance, data on establishments with SRs show a high degree of variability even though similar data collection methods are used. According to European sources (European Agency for Safety and Health at Work 2009b), 70 percent of establishments in Spain with more than nine employees were estimated to have SRs, whereas the Spanish survey for the same year (INSHT 2009) estimated that 43.4 percent of establishments with more than five employees had SRs, ranging from 27.9 percent in establishments with 6 to 9 employees to 100 percent in establishments with over 500 employees. In Catalonia, the 2011 Catalan survey showed that there were SRs in 47.8 percent of the companies with more than five employees based at a single workplace (GREDS-EMCONET 2013).

Reliability concerns arise regarding different aspects. On one hand, the aforementioned surveys present very low response rates: 8.7 percent in the case of Europe (Petrakos et al. 2011) and 21.2 percent in the case of Catalonia. On the other hand, a possible bias was detected in one recent European survey related to the fact that access to the workers’ representatives, as informants, was conditioned by the consent of the business owner (65 percent of the cases) (Riedmann and European Foundation for the Improvement of Living and Working Conditions 2010). These biases may lead to an overestimation of the companies with SRs (Lucy and Sinclair 2012; Walters et al. 2012).

Both at international and national levels, few studies have evaluated the quality of the surveys on occupational safety and health management and those available are of a qualitative nature (Lucy and Sinclair 2012; Petrakos et al. 2011). Driven by the importance of gathering good quality, reliable data (García and Gil 1996; Peruga et al. 2000), the Catalan administration via the Sub-Directorate General of Health and Safety at Work (SGHSW) carried out the Program on Prevention of Risks Management in Companies (PPRMC) in 2011 along with the development of the Second Catalan Survey of Working Conditions (IICSWC) (DEMO 2012a). In both cases, data were gathered via the same questionnaire (DEMO 2012a), but two different collection methods were used. In the IICSWC, the data were gathered in the manner typical of these types of surveys, self-reported by employers or their representatives, whereas the PPRMC was oriented to minimizing biases, placing special emphasis on the response rate, the method of data collection and the profile of the informant. For this reason, the data from the PPRMC may be considered to be the best available.

Given the importance of SRs for the health of workers, the variability in the estimates available on SRs and the scarcity of studies that give a quantitative evaluation of the quality of the information collected in occupational safety and health management surveys, this study aims to estimate, from the data of the PPRMC, the percentage of establishments in Catalonia with SRs as well as to evaluate the possible differences in estimates according to the source of the information (PPRMC and IICSWC) and the profile of the informant.

Data and Methods

Design and Reference Population

Cross-sectional study

The reference population consisted of those establishments in Catalonia with one or more workers affiliated with social security on March 30, 2011 (N = 226,355) (Tesorería General de la Seguridad Social and Ministerio de Empleo 2011).

Sources of Information

The data were obtained from the PPRMC and the IICSWC, each source with its own sample of establishments (421 and 1,675, respectively), which made up the units of the study. Both samples were obtained from a design stratified according to provinces, economic activity, and size of the social security contribution account. The sampling error when estimating prevalences for each sample was calculated taking the stratified design into account and using software SAS 9.2 (PROC SURVEY), for a confidence interval (CI) of 95 percent and under the assumption of maximum indetermination (p = q = 0.5). With an infinite population and an assumption of simple random sampling, the sampling errors are for proportions estimations, 4.78 percent in the PPRMC and 2.39 percent in the IICSWC. As a result of its larger sample size, IICSWC has narrower CIs than the ones from PPRMC.

The period of data collection was from October 1 to December 15, 2011, for the PPRMC, and from June 14 to August 2, 2011, for the IICSWC, with a response rate (American Association for Public Opinion Research 2011) of 81.2 percent in the PPRMC and 21.2 percent in the IICSWC. In previous publications, detailed information can be found regarding the methodology used in the sampling design and the fieldwork of the IICSWC (DEMO 2012a).

Instruments and Methods of Data Collection

The same questionnaire was applied to the PPRMC and the IICSWC (DEMO 2012a). However, while in the case of the IICSWC, the questionnaire was self-completed by the employers or their representatives who transmitted the information online, by telephone, or on paper (86.7 percent, 9.5 percent, and 3.7 percent of the establishments, respectively); for the PPRMC, data were collected by the labor authority. That is to say, technicians of the SGHSW carried out a documentary revision of the information in the establishments, in the framework of interviews with the persons responsible for occupational safety and health in the companies and in the presence of the SRs, where they existed. This verification involved checking certificates or documents to prove or disprove the information on management of risk prevention provided by the companies.

Study Variables

The dependent variable was “establishment with SR/s” (Yes, No), considering “Yes” to mean having at least one SR. Independent variables were “source of information” (PPRMC and IICSWC), “profile of the informant” (SGHSW technician, general manager, person responsible for human resources, person responsible for occupational safety and health, person responsible for other matters, administrative staff, and others, without special responsibility), and “size of establishment” (1–5 employees, 6–9 employees, 10–49 employees, 50–249 employees, 250–499 employees, and >499 employees). The size of the establishment was considered as a stratification variable in estimating the percentage according to the information source.

For all of the variables, the response categories does not know and does not answer were considered as lost values, except in the case of establishment size, for which a deterministic imputation was made based on the size associated with the social security contribution account code (64 companies from the IICSWC). The PPRMC had no missing values in reference to the variables included in the analysis. As for the IICSWC, missing values regarding the variable establishment with SRs were 78 (4.66 percent) and 184 (11 percent) with regard to the variable profile of the informant.

Statistical Analysis

A descriptive analysis was made of the independent variables for both samples (PPRMC and IICSWC). Then, an estimate was made of the percentage of establishments with SRs and the differences in the percentages between the two sources of information, using the PPRMC as a reference, with the respective 95 percent CI, globally (for all establishments and for those with more than five employees) and by establishment size. Finally, the percentage of establishments with SRs was estimated as were the differences in percentages, according to the different profiles of informants, using the technicians of the SGHSW (PPRMC) as a reference. These estimates were obtained from the samples, weighted in accordance with the stratification variables of the sampling design, in order to guarantee representativeness. Standard errors were accounted for the sample design in each case.

Results

In Table 1, the samples from the two sources of information are described in terms of the study variables. When compared to the distribution of establishments in Catalonia, the sample of the PPRMC is more similar to the reference population than the IICSWC. The distribution of the study units by establishment size presents differences between the sources: the stratum of the PPRMC with the most units is that which corresponds to establishments with 1 to 5 employees, while in the case of the IICSWC, the stratum of 10 to 49 employees stands out. As regards the profile of the informant of the IICSWC, the highest percentage corresponds to persons responsible for occupational safety and health.

Table 1.

Description of the Samples from the Two Information Sources (PPRMC and IICSWC) in Terms of the Study Variables.

Variable Name	Categories	PPRMC, N (%)	IICSWC, N (%)	Reference Population, N (%)
Establishment size
	1–5 Employees	284 (67.46)	253 (15.1)	178,345 (78.79)
	6–9 Employees	45 (10.69)	105 (6.27)	18,267 (8.07)
	10–49 Employees	59 (14.01)	626 (37.37)	24,469 (10.81)
	50–249 Employees	19 (4.51)	513 (30.63)	4,504 (1.99)
	250–499 Employees	9 (2.14)	103 (6.15)	453 (0.20)
	>499 Employees	5 (1.19)	75 (4.48)	317 (0.14)
Source of information
	PPRMC	421	Category not applicable	Category not applicable
	IICSWC	Category not applicable	1,675	Category not applicable
Profile of the informant
	SGHSW^a technician	421	Category not applicable	Category not applicable
	General manager	Category not applicable	341 (20.36)	Category not applicable
	Person responsible for human resources	Category not applicable	288 (17.19)	Category not applicable
	Person responsible for OSH^b	Category not applicable	508 (30.33)	Category not applicable
	Person responsible for other matters	Category not applicable	24 (1.43)	Category not applicable
	Administrative staff	Category not applicable	128 (7.64)	Category not applicable
	Others, without special responsibility	Category not applicable	202 (12.06)	Category not applicable

Source. Program on prevention of risks management in companies (PPRMC) and Second Catalan Survey of Working Conditions (IICSWC). Sub-Directorate General for Health and Safety at Work (2011), source of data are PPRMC and IICSWC.

^aSGHSW = Sub-Directorate General for Health and Safety at Work.

^bOSH = occupational safety and health.

According to the PPRMC, there are SRs in 6.6 percent (with a 95 percent CI: 4.6–9.4) of all establishments in Catalonia and 21.6 percent (15.1–29.7) if only the establishments with five or more employees are considered. This percentage tends to increase as establishment size increases: from 0.4 percent (0.1–2.5) in establishments with 1 to 5 employees to 79.3 percent (30–97.2) in establishments with more than 499 employees (Table 2).

Table 2.

Estimates of the Percentages and the Differences in the Percentages of Establishments with Safety Representatives—Globally and Stratified by Size of Establishment—by Information Source.

Establishment Size	PPRMC (a)	IICSWC (b)	Comparison (b − a), Percentage Difference
Establishment Size	% (95% CI)	% (95% CI)	% (95% CI)
1–5 Employees	0.36 (0.05 to 2.54)	10.04 (6.52 to 15.17)	9.68 (7.55 to 11.81)
6–9 Employees	6.80 (2.21 to 19.1)	42.04 (25.08 to 61.11)	35.23 (26.14 to 44.32)
10–49 Employees	26.34 (16.57 to 39.15)	47.77 (35.7 to 60.12)	21.44 (8.83 to 34.04)
50–249 Employees	40.65 (19.96 to 65.29)	87.47 (81.2 to 91.86)	46.82 (21.75 to 71.89)
250–499 Employees	100.00^a	92.71 (80.03 to 97.58)	−7.29 (−33.61 to 19.03)
>499 Employees	79.33 (29.97 to 97.18)	93.26 (81.47 to 97.76)	13.94 (−122.81 to 150.68)
Total >5 employees^b	21.56 (15.15 to 29.73)	49.82 (39.4 to 60.2)	28.30 (20.08 to 36.44)
Total	6.60 (4.58 to 9.42)	27.53 (21.90 to 33.96)	20.93 (17.66 to 24.19)

Source. Program on prevention of risks management in companies (PPRMC) and Second Catalan Survey of Working Conditions (IICSWC). Sub-Directorate General for Health and Safety at Work (2011).

^aNo variability in responses.

^bThe total has also been calculated for establishments with more than five employees, as they are the ones for which the regulations stipulate the designation of safety representatives.

Comparing the data from the IICSWC with those from the PPRMC, the differences in percentages of establishments with SRs are 20.9 percent overall (17.7–24.2), 28.3 percent (20.1–36.4) for establishments with more than 5 employees, and rising to 46.8 percent (21.7–71.9) for those with 50 to 249 employees (Table 2).

When data gathered from the different informant profiles of the IICSWC are compared to that collected by SGHSW technicians, greater differences in percentages are observed when informants are persons responsible for occupational safety and health, 44 percent (36.9–51.3), and persons responsible for human resources, 33.2 percent (25.2-41.1) (Table 3).

Table 3.

Estimates of the Percentages and the Difference in the Percentages of Establishments with Safety Representatives, According to the Profile of the Informant.

Source and Profile of the Informant	% (95%CI)	Percentage Difference,^a % (95%CI)
Program on prevention of risks management in companies (PPRMC)
SGHSW^b technician	6.60 (4.58 to 9.42)	Reference category
Second Catalan Survey of Working Conditions (IICSWC)
General manager	16.05 (10.18 to 24.38)	9.45 (5.63–13.28)
Person responsible for human resources	39.75 (22.04 to 60.62)	33.15 (25.24–41.06)
Person responsible for OSH^c	50.70 (32.37 to 68.84)	44.10 (36.90–51.29)
Person responsible for other matters	25.49 (5.41 to 67.16)	18.89 (3.90–33.87)
Administrative staff	27.09 (13.29 to 47.39)	20.49 (13.9–27.08)
Others, without special responsibility	21.21 (10.15 to 39.06)	14.61 (6.73–22.49)

Source. Program on prevention of risks management in companies (PPRMC) and Second Catalan Survey of Working Conditions (IICSWC). Sub-Directorate General for Health and Safety at Work (2011).

^aDifference in the percentage of the estimate corresponding to each informant profile of the IICSWC with respect to the estimate corresponding to the SGHSW technicians in the PPRMC.

^bSGHSW = Sub-Directorate General for Health and Safety at Work.

^cOSH = occupational safety and health.

Discussion

Taking the case of establishments with SRs, we provide a quantitative evaluation of the information quality of occupational safety and health management in companies, specifically of the biases in the estimates. We compared the percentage of establishments with SRs in Catalonia, collected according to the PPRMC, to figures obtained from the IICSWC (DEMO 2012a); where the former collected data using a method focused on minimizing biases (GREDS-EMCONET 2013; Lucy and Sinclair 2012; Petrakos et al. 2011; Walters et al. 2012) and the latter used a different method, one commonly used at the state (INSHT 2009) and European levels (European Agency for Safety and Health at Work 2009a; EUROFOUND 2009).

According to the PPRMC, there are SRs in 21.6 percent of establishments with more than five employees in Catalonia (establishments in which regulations stipulate the designation of one or more SRs). The IICSWC yields much higher estimates of establishments with representatives, for the whole set of establishments in Catalonia and for those with more than five employees.

In this study, we did focus not only on the mode of administration as a source of measurement error but also on the respondent (Alwin 1991). We observed that the percentage of establishments reported to have one or more SRs varies not only according to the method of data collection (PPRMC or IICSWC) but also according to whom the informant is. With respect to the information collected from SGHSW technicians, overestimations are higher when informants are the persons responsible for occupational safety and health. This is highly significant bearing in mind that they are the recommended informants in some surveys (Lucy and Sinclair 2012), given their training and duties (Real Decreto 1997).

The results of the PPRMC differ from other surveys on occupational safety and health management using employers or their representatives as informants. For example, the percentages of establishments with more than five employees which have SRs gathered in the IICSWC and in the Spanish survey (INSHT 2009) are both roughly double the estimate of the PPRMC (49.8 percent and 43.4 percent vs. 21.6 percent). Moreover, the European survey (European Agency for Safety and Health at Work 2009b) estimates that 70 percent of establishments with more than nine employees have SRs in Spain, well above the figures obtained by both the IICSWC and the PPRMC (56.6 percent and 30.4 percent, respectively). It would appear then that the overestimate observed in the IICSWC would also occur in the different occupational safety and health management surveys conducted at the state and European levels.

This generalized overestimating of establishments with SRs might be partly related to the low rates of response of the surveys. Estimates increase because companies participating in the surveys tend to be those most committed to occupational safety and health (Lucy and Sinclair 2012; Walters et al. 2012) or those that are least reticent in replying. In the IICSWC, 4.7 percent of the companies did not respond to the question on the existence of SRs, meaning that the percentage of establishments without SRs may be even higher. At the European level, several studies point out the need to improve response rates and to obtain information on the companies that do not respond, in order to correct biases associated with the low rates (Lucy and Sinclair 2012; Marta-Pedroso et al. 2007; Petrakos et al. 2011).

In line with other studies (European Agency for Safety and Health at Work 2010; Lucy and Sinclair 2012; Peersman et al. 2014), the results obtained point to the fact that the method of data collection and the profile of the informant are potential sources of bias. However, some considerations must be taken into account in order to interpret the results obtained and, specifically, to evaluate possible explanations for the differences observed between the two sources.

Firstly, the collection of data in the two samples (IICSWC and PPRMC) was not carried out simultaneously (third and fourth quarter of 2011). This fact might have led to a change in the number of establishments with one or more SRs, mainly related to the calling of union elections. However, due to their proximity in time, it cannot be imagined that the quantity of union elections being called would have been a possible explanation of the differences observed between the estimates of each source of information. In the case of the IICSWC, despite the data being gathered in the third quarter, the month of August was avoided in order to facilitate access to the companies.

Secondly, a bias may also occur as a result of using the establishment as a unit of data collection and analysis (Petrakos et al. 2011)—since the regulation stipulates the designation of SRs according to the number of employees in the establishment or company—as well as the fact that the units of the sample (social security contribution account code) and of analysis (establishment) and the categories of the variable size of establishment in the sampling and analysis phases were not the same. For this reason, we explored the distribution of the samples according to the different categories of stratification and of analysis. It was detected that the IICSWC underrepresented small establishments or overrepresented medium and large establishments with respect to the population, whereas the distribution of the PPRMC corresponded to that of the population, which lends more external validity to the results obtained by the PPRMC. Even so, the results reported according to size of the establishment were not affected by the use of different sampling strata, thanks to the use of a stratified analysis.

Concerning the methodology of the data comparison, it has not been possible to carry out a specific analysis of the validity of the information from the IICSWC (Riegelman and Hirsch 1992; Viladrich and Doval 2007) because the establishments participating in the two samples were different. Therefore, the decision to use the difference in the percentages and CIs between the two sources of information (Gardner and Altman 1988; International Committee of Medical Journal Editors 2014) was made.

Information provided by the IICSWC leads to a generalized overestimation of the percentage of establishments in Catalonia with SRs, which might be explained by the informant profiles and the method of data collection employed in the survey. It is necessary to carry out more specific studies to determine the most appropriate profile of informant as well as to make changes in the way data are collected in an attempt to improve, for example, the response rates (Lucy and Sinclair 2012; Marta-Pedroso et al. 2007; Peruga et al. 2000; Petrakos et al. 2011; Walters et al. 2012). It would also be advisable to explore the existence of biases in the rest of the information on occupational safety and health management compiled via surveys of this type. Hence, the study suggests there is a need for the labor authority and the institutions responsible, not only in Catalonia but also at the state and European level, to focus their efforts on creating more complete and better quality information systems (Walters et al. 2012)—specifically in surveys—upon which to base more efficient actions and policies in the area of occupational health. Also, the results show the low level of representative participation in occupational safety and health in establishments in Catalonia, highlighting the need to establish interventions that encourage worker participation in the companies, given its proven relationship with the improvement in prevention of risks management and, indirectly, that of workers’ occupational health.

Footnotes

Acknowledgments

The authors wish to thank Pere Jódar (Universitat Pompeu Fabra), Carlota de Martí (DYM), Josep María Losilla and Albert Navarro (Autonomous University of Barcelona), Iris Delgado (Universidad del Desarrollo, Chile), and Lucía Artazcoz (Public Health Agency of Barcelona and Universitat Pompeu Fabra) for their valuable contributions that have helped to improve the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Alwin

Duane F.

1991. “Research on Survey Quality.” Sociological Methods & Research 20:3–29.

American Association for Public Opinion Research. 2011. Standard Definitions: Final Dispositions of Case Codes and Outcome Rates for Surveys, Revised 2011. 7th ed. Ann Arbor, MI: American Association for Public Opinion Research.

Coutrot

Thomas

. 2009. “Le Rôle Des Comités D’hygiène, de Sécurité et Des Conditions de Travail En France: Une Analyse Empirique.” Travail et Emploi 117:25–38.

DEMO (Departament d’Empresa i Ocupació). 2012a. Segona Enquesta Catalana de Condicions de Treball—Empreses. Barcelona, Spain: Direcció General de Relacions Laborals i Qualitat en el Treball, DEMO, Generalitat de Catalunya.

DEMO (Departament d’Empresa i Ocupació). 2012b. Segona Enquesta Catalana de Condicions de Treball - Treballadors. Barcelona, Spain: Direcció General de Relacions Laborals i Qualitat en el Treball, DEMO, Generalitat de Catalunya.

European Agency for Safety and Health at Work. 2009a. “Enterprise Survey-information for Respondents.” Retrieved December 9, 2014 (https://osha.europa.eu/en/riskobservatory/enterprise-survey).

European Agency for Safety and Health at Work. 2009b. “ESENER Survey Results 2009/Does Your Establishment Have a Health and Safety Representative?” Retrieved December 17, 2014 (https://osha.europa.eu/sub/esener/en/front-page/217).

European Agency for Safety and Health at Work. 2010. European Survey of Enterprises on New and Emerging Risks. Managing Safety and Health at Work. Bilbao, Spain: European Agency for Safety and Health at Work. Retrieved November 29, 2012 (http://osha.europa.eu/en/publications/reports/esener1_osh_management).

European Council Directive No. 89/391/EEC. 1989. Council Directive 89/391/EEC of 12 June 1989 on the Introduction of Measures to Encourage Improvements in the Safety and Health of Workers at Work.

10.

EUROFOUND (European Foundation for the Improvement of Living and Working Conditions). 2009. “European Company Survey (ECS) 2009.” Retrieved December 9, 2014 (http://eurofound.europa.eu/surveys/ecs/2009/european-company-survey-2009).

11.

EUROFOUND (European Foundation for the Improvement of Living and Working Conditions). 2010. “Fifth European Working Conditions Survey—2010.” Retrieved December 9, 2014 (http://eurofound.europa.eu/surveys/ewcs/2010/european-working-conditions-survey-2010).

12.

García

Montserrat

Gil

Enrique

. 1996. “Los Sistemas de Información Y La Vigilancia En Salud Laboral.” Revista Española de Salud Pública 70:393–407.

13.

Gardner

Martin J.

Altman

Douglas G.

. 1988. “Estimating with Confidence.” British Medical Journal 296:1210–11.

14.

GREDS-EMCONET. 2013. Qualitat de l’ocupació, participació i salut laboral a Catalunya. Resultats de la II Enquesta Catalana de Condicions de Treball. Barcelona, Spain: Direcció General de Relacions Laborals i Qualitat en el Treball. Departament d’Empresa i Ocupació. Generalitat de Catalunya. Retrieved October 2, 2016 (http://treball.gencat.cat/web/.content/03_-_centre_de_documentacio/documents/01_-_publicacions/06_-_seguretat_i_salut_laboral/ii_ecct/arxius/ii_ectc_qualitat_ocupacio_participacio.pdf).

15.

INSHT. 2009. Encuesta Nacional de Gestión de La Seguridad Y Salud de Las Empresas 2009. Retrieved October 4, 2014 (www.insht.es/Observatorio/Contenidos/InformesPropios/Desarrollados/Ficheros/Informe_%20ENGE%202009.pdf).

16.

INSHT. 2012. VII Encuesta Nacional De Condiciones De Trabajo 2011. Retrieved October 2, 2016 (http://www.oect.es/InshtWeb/Contenidos/Documentacion/FICHAS%20DE%20PUBLICACIONES/EN%20CATALOGO/OBSERVATORIO/Informe%20(VII%20ENCT).pdf).

17.

International Committee of Medical Journal Editors. 2014. “Recommendations for the Conduct, Reporting, Editing and Publication of Scholarly Work in Medical Journals.” Retrieved December 7, 2014 (www.icmje.org/icmje-recommendations.pdf).

18.

Law 31/1995 on Prevention of Risks at Work. 1995. Ley de Prevención de Riesgos Laborales. LEY 31/1995 (8 de Noviembre 1995).

19.

Jones

Melanie K.

Latreille

Paul L.

Sloane

Peter J.

Staneva

Anita V.

. 2013. “Work-related Health Risks in Europe: Are Older Workers More Vulnerable?” Social Science & Medicine 88:18–29.

20.

Lucy

Daniel

Sinclair

Alice

. 2012. Ex-post Evaluation of the European Survey of Enterprises on New and Emerging Risks. Brighton, UK: The Institute of Employment Studies. Retrieved May 14, 2013 (osha.europa.eu/en/esener-enterprise-survey/ex-post-evaluation-of-esener.pdf).

21.

Marta-Pedroso

Cristina

Freitas

Helena

Domingos

Tiago

. 2007. “Testing for the Survey Mode Effect on Contingent Valuation Data Quality: A Case Study of Web Based versus in-person Interviews.” Ecological Economics 62:388–98.

22.

Martínez-Sánchez Jose

Marcela

Gallus

Silvano

Sureda

Xisca

Vecchia

Carlo La

Fernández

Esteve

. 2013. “Variabilidad En El Consumo de Cigarrillos Según La Fuente de Información En España (1993-2009).” Gaceta Sanitaria 27:61–63.

23.

Menéndez

María

Benach

Joan

Vogel

Laurent

. 2008. “El Impacto de Los Delegados de Prevención En La Salud Laboral: El Proyecto EPSARE.” Archivos de Prevención de Riesgos Laborales 11:5–7.

24.

Mygind

Karen

Borg

Vilhelm

Flyvholm

Mari-Ann

Sell

Lea

Jepsen

Karen Frydendall

. 2005. “A Study of the Implementation Process of an Intervention to Prevent Work-related Skin Problems in Wet-work Occupations.” International Archives of Occupational and Environmental Health 79:66–74.

25.

Ollé-Espluga

Laia

Vergara-Duarte

Montse

Belvis

Francesc

Menéndez-Fuster

María

Jódar

Pere

Benach

Joan

. 2015. “What Is the Impact on Occupational Health and Safety When Workers Know They Have Safety Representatives?” Safety Science 74:55–58.

26.

Peersman

Wim

Pasteels

Inge

Cambier

Dirk

Maeseneer

Jan De

Willems

Sara

. 2014. “Validity of Self-reported Utilization of Physician Services: A Population Study.” The European Journal of Public Health 24:91–97.

27.

Peruga

Armando

María León

Esther

Child

Raquel

Cruz

Aurelio

Hernández

Manuel

Arredondo

Anabella

Hernández

Carlos

Cuchí

Paloma

Zacarías

Fernando

. 2000. “Análisis de la participación en encuestas de cinco países: relevancia para la investigación en salud pública.” Revista Panamericana De Salud Pública = Pan American Journal of Public Health 7:249–54.

28.

Petrakos

Michalis

Kleideri

Maria

Ieromnimon

Antonia

. 2011. Encuesta Europea de Empresas Sobre Riesgos Nuevos Y Emergentes. Dublin: European Foundation for the Improvement of Living and Working Conditions. Retrieved October 4, 2014 (http://www.eurofound.europa.eu/pubdocs/2011/23/en/1/EF1123EN.pdf).

29.

Real Decreto. 1997. Real Decreto 39/1997, de 17 de Enero, Por El Cual Se Aprueba El Reglamento de Los Servicios de Prevención.

30.

Real Decreto Legislativo. 1995. Real Decreto Legislativo 1/1995, de 24 de Marzo, Por El Que Se Aprueba El Texto Refundido de La Ley Del Estatuto de Los Trabajadores.

31.

Reilly

Barry

Paci

Pierella

Holl

Peter

. 1995. “Unions, Safety Committees and Workplace Injuries.” British Journal of Industrial Relations 33:275–88.

32.

Riedmann, Arnold and European Foundation for the Improvement of Living and Working Conditions. 2010. European Company Survey 2009-Overview. Luxembourg: Office for Official Publications of the European Communities.

33.

Riegelman

Richard K.

Hirsch

Robert P.

. 1992. Cómo Estudiar Un Estudio Y Probar Una Prueba: Lectura Crítica de La Literatura Médica. OPS. Retrieved December 9, 2014 (http://iris.paho.org/xmlui/handle/123456789/3309).

34.

Smith

Mark

Burchell

Brendan

Fagan

Colette

O’Brien

Catherine

. 2008. “Job Quality in Europe.” Industrial Relations Journal 39:586–603.

35.

Tesorería General de la Seguridad Social and Ministerio de Empleo. 2011. “Cuentas de Cotización Del Régimen General, Excluidos Los Sistemas Especiales Agrario Y Empleados Del Hogar Y El Régimen Especial de La Minería Del Carbón.” Estadísticas/Mercado de trabajo/Empresas inscritas en la Seguridad Social. Retrieved December 10, 2014 (http://www.empleo.gob.es/series/).

36.

Van Aerden

Karen

Moors

Guy

Levecque

Katia

Vanroelen

Christophe

. 2015. “The Relationship between Employment Quality and Work-related Well-being in the European Labor Force.” Journal of Vocational Behavior 86:66–76.

37.

Viladrich

M. C.

Doval

. 2007. Medición: Fiabilidad Y Validez. Bellaterra, Spain: Laboratori d’Estadística Aplicada i de Modelització (Univeristat Autònoma de Barcelona).

38.

Violán

Concepción

Foguet-Boreu

Quintí

Hermosilla-Pérez

Eduardo

Valderas

Jose M.

Bolíbar

Bonaventura

Fàbregas-Escurriola

Mireia

Brugulat-Guiteras

Pilar

Ángel Muñoz-Pérez

Miguel

. 2013. “Comparison of the Information Provided by Electronic Health Records Data and a Population Health Survey to Estimate Prevalence of Selected Health Conditions and Multimorbidity.” BMC Public Health 13:251.

39.

Walters

David

. 2006. “One Step Forward, Two Steps Back: Worker Representation and Health and Safety in the United Kingdom.” International journal of health services 36:87–111.

40.

Walters

David

Wadsworth

Emma

Marsh

Katie

Davies

Rhys

Lloyd-Williams

Huw

. 2012. Worker Representation and Consultation on Health and Safety. An Analysis of the Findings of the European Survey of Enterprises on New and Emerging Risks (ESENER). Luxembourg: European Agency for Safety and Health at Work. Retrieved November 29, 2011 (http://osha.europa.eu/en/publications/reports/esener_workers-involvement).