Educating families about end-of-life care in advanced dementia:Acceptability of a Canadian family booklet to nurses from Canada, France, and Japan

Background: The families of people with late-stage dementia need to be informed about the course of the dementia and the comfort/ palliative care option. A booklet was written for that purpose and can be provided to family members by physicians and nurses. Methods: The acceptability of the booklet for nurses was tested in Canada (French and English version), France (French Canadian version) and Japan (translated and adapted version). Results: Overall, 188 nurses completed a survey questionnaire. The booklet was accepted best in Canada and less so in France and Japan. Despite regional variation, the majority of the nurses perceived the booklet as useful for families. The French and Japanese nurses also reported a greater need for palliative care education in advanced dementia. Conclusion: The booklet may help nurses educate families about end-of-life issues in dementia palliative care, but local adaptation of the booklet content and physician engagement are necessary.

An Extended Implementation of the Great Deluge Algorithm for Course Timetabling

Course Scheduling consists of assigning lecture events to a limited set of specific timeslots and rooms. The objective is to satisfy as many soft constraints as possible, while maintaining a feasible solution timetable. The most successful techniques to date require a compute-intensive examination of the solution neighbourhood to direct searches to an optimum solution. Although they may require fewer neighbourhood moves than more exhaustive techniques to gain comparable results, they can take considerably longer to achieve success. This paper introduces an extended version of the Great Deluge Algorithm for the Course Timetabling problem which, while avoiding the problem of getting trapped in local optima, uses simple Neighbourhood search heuristics to obtain solutions in a relatively short amount of time. The paper presents results based on a standard set of benchmark datasets, beating over half of the currently published best results with in some cases up to 60% of an improvement.