Abstract
In a context of demographic ageing and the increasing prevalence of major neurocognitive disorders, and in response to recommendations from health authorities, nursing homes in France are required to rethink their care practices. To this end, the Montessori method adapted for older adults is increasingly being advocated. As a promising alternative to traditional practices, it encourages a focus on preserved abilities, autonomy, and the meaningfulness of daily activities. However, few studies have empirically evaluated its effects and the conditions for its implementation within care settings.The primary aim of this thesis is to document the effects of the Montessori method, both as a non-pharmacological intervention, through a reading activity, and as a comprehensive person-centred care approach within nursing homes. Particular attention is paid to the rigorous assessment of its implementation, through the development of a specifically designed evaluation tool. This work also seeks to deepen understanding of the mechanisms that support the adoption of such an approach, notably by examining the role of care staff’s attitudes towards residents and the influence of the working environment.A first study (Study A) was conducted with 15 residents from a Memory Care Unit (MCU), to compare the effects of Montessori group reading activities with those of traditional group reading sessions. This single-case experimental design with multiple baseline study highlighted the benefits of the Montessori reading activities, particularly in terms of increased verbal interactions, higher engagement levels, and greater expressions of pleasure among participants. A second study (Study B) focused on the development of an evaluation grid to assess the degree of Montessori method implementation at the scale of an entire nursing home. Administered across 29 establishments, it revealed considerable heterogeneity in practices between institutions, both before and after the training. The tool proved sensitive to the effects of training in the Montessori approach. The third study (Study C), involving 262 residents from MCU, aimed to evaluate the effects of Montessori training on residents six months after the method’s implementation. The results showed a significant reduction in responsive behaviours and their impact on care staff. Furthermore, variations between establishments suggested a possible seasonal effect on these behaviours. Certain aspects of implementation, assessed using the previously developed grid, were found to correlate with observed behavioural improvements. Although residents’ overall independence declined between the two measurement points, 69% showed either stability or an improvement six months after staff training.Finally, the fourth study (Study D) involved the validation of the French version of the Personhood in Dementia Questionnaire, confirming a unidimensional structure. Mediation analyses revealed that decision-making flexibility plays a key role in translating individual attitudes of care staff into collective, person-centred practices. These findings underscore the importance of the organisational environment in the successful implementation of the Montessori approach.