Due to the rapid increase in the number of people suffering from dementia worldwide, more residential care is greatly needed. Traditionally, residential dementia care has been carried out with a medical- and nursing-based care approach, which has been criticized for focusing mainly on the care aspects. Experts argue that dementia care should also pay attention to creating the environment that offers a better balance between living, well-being, and care. In an effort for better balance, small-scale living facilities for older persons have been developed. However, limited research has been done to identify their impact on quality of life of residents.
The purpose of this study was to examine benefits of small-scale living for residents with dementia, compared to traditional long-term care in The Netherlands and Belgium. The primary outcome was quality of life, divided into nine different items.
This quasi-experimental study had three measurement points over 1 year of the study period: baseline, 6 months, and 12 months. Participants included 179 residents (51 Dutch and 47 Belgians in small-scale living and 47 Dutch and 30 Belgians in traditional living facilities) with dementia, aged 65 years or older from five long-term care settings. Nurses and nursing assistants completed the questionnaires. Three measurement instruments were employed in this study: (1) instruments to measure the primary outcome quality of life, (2) secondary outcomes being quality of life determinants, and (3) control variables. The primary outcome quality of life was measured by Qualidem (a quality of life instrument for people with dementia) and secondary outcomes included behavioral characteristics (behavioral problems as measured by the Neuropsychiatric Inventory-Nursing Home version) and depression (as measured by the Cornell Scale for Depression in Dementia), behavioral interventions (use of physical restraints), and social interaction that was measured with Revised Index of Social Engagement. Control variables were basic personal characteristics including cognitive impairment measured by the Standardized Mini Mental State Examination and activities of daily living measured by the Barthel Index. The data formed a hierarchically nested or multilevel data structure: Observations over time were nested within persons, which were then nested within settings. Accordingly, the data were analyzed using hierarchical linear modeling techniques. The primary analyses were three-level models, and each quality of life subscale was analyzed separately.
In the Dutch sample, residents in small-scale settings had higher mean scores on social relations, positive affect, and having something to do than residents in traditional settings. Moreover, mean scores on caregiver relation and negative affect remained stable over time among residents in small-scale settings, but decreased in traditional settings. These differences could not be explained by differences in behavioral characteristics, behavioral interventions, or social interaction. In the Belgian sample, fewer differences were found between traditional and small-scale settings. Nevertheless, residents in small-scale settings were reported to experience less negative affect than those in traditional settings, which could be explained by differences in depression. Over time, however, residents reported feeling more at home in traditional settings, whereas no such increase was found for small-scale settings. Moreover, the mean quality-of-life scores on restless behavior, having something to do, and social relations decreased in small-scale settings, but remained stable in traditional settings.
Proxy reporting by caregivers was identified to have the disadvantage of filtering subjective measures through another person’s point of view. However, observation by caregivers is known to be the best, most reliable, and valid alternative method. Due to ethical and practical considerations, the residents in the study could not be randomly assigned to conditions. The study was therefore conducted as a quasi-experimental design in actual real-life care settings, with differences between individual participants, long-term care settings, and countries. Although the study controlled for relevant background variables, there is always a possibility that institutions may have differed on other variables as well.