Background: Older people living in nursing homes are at a high risk of becoming malnourished, falling and developing pressure ulcers. In Sweden the national quality registry Senior Alert was developed to support prevention in these areas. Prevention according to Senior Alert follows a preventive care process of four steps, including risk assessment, analysis of the causes of risk, to determine and perform appropriate actions, and finally, to evaluate the care given.

Aim: The overall aims of this thesis were to investigate how the preventive care process in Senior Alert functions as a tool for preventive work among older persons living in nursing homes, and to investigate the results of risk assessments and actions.

Design: The thesis is based on three longitudinal quantitative studies (I, III, and IV) and one qualitative study (II). In Studies I and III, process- and patient results were compared among different groups of nursing home residents, with a follow-up time of 6 months. In Study IV, associations between the assessment instruments and the outcomes of weight loss, falls and pressure ulcers were investigated. The qualitative study (II) was based on focus group interviews with healthcare professionals and was analyzed using content analysis.

Results: The residents included in the registry during the later years (2013-2014) had a higher proportion of registered preventive actions in the three areas, and were followed up more frequently regarding weight and new assessments than residents included during the earlier years (2010-2012). Nevertheless, regardless of risk, only 30% were reassessed, and 44% of the residents at risk for malnutrition were followed up for body weight within 6 months. No difference in weight change was found between a group of residents included in Senior Alert and a second group receiving ‘care as usual’. Generally, the mobility variables in the risk assessment instruments had the strongest associations with the tested outcomes of weight loss, fall and pressure ulcers, albeit in different ways. Healthcare professionals described that Senior Alert stimulated better teamwork while at the same time they experienced the increased documentation and time constraints as aggravating circumstances. They also described a lack of reliability of the assessment instruments in that they overrated the risks compared to their own clinical judgement. Healthcare professional’s knowledge about the evaluation part of the process was low.

Conclusion: The evaluation and follow-up step of the preventive care process was not sufficiently applied. This was expressed by the participants in the focus groups and was also reflected in registry data by the varying time to follow-up and the poor event registration. As a consequence, the sample to measure outcomes within 6 months became small. Therefore, larger samples are needed to study longitudinal outcomes, if a fixed system-mandatory time point for follow-up is not implemented. A committed leadership is important to improve the preventive work and to stimulate follow-up of results.