This study set out to evaluate a small scale model of care for people with dementia that aimed to support residents within a daily household routine (termed household care). Stemming from a pluralistic evaluation, this thesis examines issues identified as important to residents' experience, namely the ability of the care houses to provide a home for life, the process and effectiveness of recreating a homely environment and the implications of group-living. Three case studies were investigated with multiple methods, including semi-structured interviews with staff, relatives and representatives of managing agencies, structured observation and assessment of dependency. Respondents believed care housing was superior to its alternatives and attributed perceived improvements in residents' well-being to the nature of staff support and the 'homely' setting. Agency representatives were preoccupied with their relationships with each other and how to sustain and expand this model of care. Staff focused on the nature and conditions of the work. Kin were keen that residents should settle and stay in the care houses. Residents' support needs at least matched entry criteria but over half had to move out to hospital because of physical illness and behavioural problems. This study suggests that it was very difficult for residents to accept the houses as 'home'. Residents' and their relatives' participation was, in practice, limited but one house was particularly successful in implementing household care; reasons are suggested for this. Residents' interactions with each other seemed to be increased by household care but their relationships were influenced by a number of factors, including the presence of dementia. The latter exacerbated the tensions of group-living. The challenge for care houses was to sustain care as dementia progressed and extend good quality care to those with more substantial behavioural and physical support needs.

This study involved 58 persons with dementia living in three rural Canadian long-term care (LTC) facilities. In an attempt to find the relationship between these person's possible pain and levels of agitation, data on five proxy indicators of pain were collected and correlated with scores from the Pittsburgh Agitation Scale (PAS). Results indicated that three of the resident pain measures were significantly correlated with PAS scores. In particular, the palliative consultant pain ratings and the DS-DAT were strongly correlated with total PAS scores, and the five PAS sub-factors. Importantly, the PAS sub-factor of resistance to care was strongly correlated with three of the pain variables. Major study implications include the need for increased use of palliative pain consultants in LTC, and the need for nursing staff to realize that when demented residents resist care, it may be a potential clue that the resident is experiencing untreated pain. / xv, 116 leaves ; 29 cm.