Summary The most common confined spaces in Québec workplaces are tanks, bins, manholes, sewers, pipes and vats. Workers enter these confined spaces for a variety of reasons, including to perform maintenance (e.g., repairs, inspection, cleaning and unclogging). However, confined spaces pose health and safety risks, for workers concerned. These risks may be atmospheric, chemical, biological, mechanical, physical (trips and falls, for example) or ergonomic. Our review of the serious/fatal accident investigation reports issued by the Commission des normes, de l’équité, de la santé et de la sécurité du travail (CNESST) for 1998 to 2017 showed an average 2.6 deaths per year (53 deaths) in confined spaces in Québec, accounting for about 4% of all fatal work accidents in the province during the period. Work in confined spaces is regulated in Québec by the Regulation respecting occupational health and safety and the Safety Code for the construction industry. Confined space risk reduction measures have traditionally taken the form of work permits, natural or mechanical ventilation, gas measurements and use of personal protective equipment (PPE). Although inherently safer design and collective protection measures are recommended in standards and may be more effective approaches, they are still marginal in the Québec workplace according to an earlier IRSST study (Chinniah et al., 2016). This study looks at measures at the top of the hierarchy of risk control for confined spaces, that is, measures to eliminate or reduce risks at the source and collective protection measures. In the interests of simplicity, the term “inherently safer design and collective protection” or “ISD-CP” is used in this project to refer to such measures. ISD-CP includes all measures introduced when designing or modifying (i.e., re-engineering or retrofitting) a confined space to reduce risks for all stakeholders (e.g., relocating a rescue/intervention site outside the confined space to eliminate the need for entry). As existing knowledge in the literature and in practice in the field is not structured, the goal of this study was to build a knowledge base on ISD-CP measures for confined space risk management. This knowledge base includes a breakdown of applicable risk reduction principles; information on the environment, circumstances and anticipated impact of risk reduction; and technical and organizational barriers that must be considered. A three-phase methodology was used to build the knowledge base. In Phase 1, fault tree analysis (FTA) was used to investigate ten fatal accidents in confined spaces. This made it possible to identify the primary design shortcomings responsible for the accidents and to demonstrate the potential for ISD-CP measures in these situations. In Phase 2, fifteen experts in confined space risk management (e.g., trainers, consultants, practicing preventionists and client-side designers) were consulted about introducing and using ISD-CP measures. From these consultations, barriers as well as factors favorable to the adoption of ISD-CP in the eyes of these experts were identified. Key levers proved to be designer consideration of the entire life cycle of the confined space, long-term investment planning by the user and a commitment to OHS in the user-designer relationship. In this phase, we were also able to propose a model for designers and users that structures ISD-CP measures for confined spaces according to the following principles: P1) completely eliminate the confined space; P2) declassify the confined space by reducing risks; P3) eliminate the need for entry regarding a specific task; P4) reduce the need for entry in general ; P5) improve access, intervention and rescue. Note that a risk reduction approach must always be based on a complete risk assessment and not just on regulatory compliance (e.g., definition of a confined space). During Phase 3 of the project, 19 case studies were conducted in industry to identify ISD-CP measures that are or could be used for real targeted interventions. The following sectors of activity are represented in the sample: water treatment, power distribution/telecommunication, pulp and paper, parts manufacturing, chemical treatment, transportation and industrial maintenance/civil engineering. To render usable the 112 solutions identified in connection with 30 confined spaces, a searchable relational database was developed. Each solution was associated in the knowledge base with a context (a confined space, a task), a problem (risks, problematic design features), a risk reduction mechanism (principles P1 to P5, a type of measure) and barriers to its implementation (technical, organizational). Presented in this report by type of confined space (tank, vat, access shaft/chamber, chimney, interior of industrial equipment), the solutions break down as follows: 6% are applications of principles P1 and P2; 54% are applications of principles P3 and P4; and 40% are applications of principle P5. Solutions applying principles P1 and P2 are limited, notably because modification of existing spaces (principles P3 to P5) is the primary approach in the situations faced by our research participants. Research targeting manufacturers of confined spaces might help grow the number of solutions applying principles P1 and P2. Lastly, this knowledge base and the fault tree analyses developed should be the subject of a knowledge translation activity that will provide users interactive access to all the data. Dissemination of the knowledge base will help raise awareness among designers, integrators and preventionists of the importance of reducing confined space risks and the opportunities do so with the help of ISD-CP measures.