a b s t r a c tWhen there are safety representatives (SRs) at the workplace higher levels of preventive action have been observed. However, no study has analyzed workers' health and safety results when workers (do not) know they have SRs. Based on data from the VII Spanish Working Conditions Survey (2011), this paper explores differences in the intensity of self-reported preventive action among workers reporting to have SRs at their workplaces, workers reporting not having them, and workers unaware of SRs' existence. The sample included employees aged 16-65 years working at firms with 6 workers or more (n = 5562). A multinomial logistic regression was undertaken to study the association between the reported existence of SRs and levels of preventive action (high, intermediate and non-existent), comparing workers unaware of SRs' existence to those reporting to have SRs and those reporting no SRs. Models were adjusted by socio-demographic and employment-related features. It was found that workers reporting SRs' existence were protected by greater preventive action, both at the intermediate (aOR = 2.87, 95% CI 2.39-3.44) and high level (aOR = 10.26, 95% CI 7.27-14.50), and that there were no statistically significant differences between workers reporting not to have SRs and those unaware of SRs. Our results draw attention to a group of workers who might have SRs without being aware of it and remain less protected by preventive action. These workers would benefit from interventions aimed at making SRs known and available to all workers.

Interaction with workers is influenced by a more prevalent technical-legal view of the SRs' role and by unequal power relations between workers and management. Poor interaction with workers might lead to decreasing SRs' effectiveness.

Background Amenable mortality, or premature deaths that could be prevented with medical care, is a proven indicator for assessing healthcare quality when adapted to a country or region's specific healthcare context. This concept is currently used to evaluate the performance of national and international healthcare systems. However, the levels of efficacy and effectiveness determined using this indicator can vary greatly depending on the causes of death that are included. We introduce a new approach by identifying a subgroup of causes for which there are available treatments with a high level of efficacy. These causes should be considered sentinel events to help identify limitations in the effectiveness and quality of health provision. Methods We conducted an extensive literature review using a list of amenable causes of death compiled by Spanish researchers. We complemented this approach by assessing the time trends of amenable mortality in two high-income countries that have a similar quality of healthcare but very different systems of provision, namely, Spain and the United States. This enabled us to identify different levels of efficacy of medical interventions (high, medium, and low). We consulted a group of medical experts and combined this information to help make the final classification of sentinel amenable causes of death. Results Sentinel amenable mortality includes causes such as surgical conditions, thyroid diseases, and asthma. The remaining amenable causes of death either have a higher complexity in terms of the disease or need more effective medical interventions or preventative measures to guarantee early detection and adherence to treatment. These included cardiovascular diseases, diabetes, hypertension, all amenable cancers, and some infectious diseases such as pneumonia, influenza, and tuberculosis. Conclusions Sentinel amenable mortality could act as a good sentinel indicator to identify major deficiencies in healthcare quality and provision and detect inequalities across populations.