Social relationships and healthy ageing : epidemiological evidence for the development of a local intervention programme

Introduction
In view of the growing number of older people in our society and the related consequences for health and well-being, research focussing on healthy ageing is essential. Already, the associations between supportive social relationships and healthy ageing have been established. However, there is as yet no consensus about whether or not it is the structure of the social network, its function- ing or a combination that is most important for health, and in addition, about which aspects of structure and function are important.

Aim
The main objective of this thesis was to investigate aspects of the structure and functioning of social relationships and their influence on mental, physical and social health in older people. This was relevant to obtain scientific evidence for practice-based research to support local policy making on healthy ageing.

Methods
Different characteristics and functions of social relationships, such as frequency of contact, different sources of social network ties, satisfaction with relationships, positive and negative perceptions of social support and social engagement have been analysed in cross-sectional and prospective studies. Cross-sectional data are from six community health services in the eastern part of the Netherlands. The overall sample size constituted of 24,936 people aged 65 and over (response 79%). Prospective data are from the Doetinchem Cohort Study. The first examination round (1987-1991) comprised 12,448 men and women aged 20 to 59 years. The overall response rate was 62% for the baseline measurement and 79%, 75% and 78% for rounds 2, 3 and 4 respectively.

Results
Cross-sectional analyses showed that satisfaction with the social contacts was strongly related to physical (OR 2.36; 95% CI 2.11-2.64), mental (OR 4.65; 95% CI 4.20-5.15) and self-perceived health (OR 2.52; 95% CI 2.29-2.78). Longitudinal analyses underlined this finding by showing that unfavourable levels of social support were predictive for health-compromising behaviours and poor health over a 10-year period of follow-up, and for increased mortality risk over a 15- year period of follow-up (HR 1.57; 95% CI 1.03-2.39). Furthermore, neighbours were found to be an important source of the social network ties of older people in relation to physical (OR 1.87; 95% CI 1.68-2.07), mental (OR 1.53; 95% CI 1.39-1.69) and self-perceived health (OR 1.42; 95% CI 1.30-1.54).

Further exploration of the relationship between social support and loneliness using structural equation modelling identified that social support in everyday situations may serve as a good start- ing point for health promotion activities to prevent loneliness. To better target health promotion activities for healthy ageing, analyses were performed to group older people into subgroups with similar social engagement activity patterns. Five clusters were identified: 1) less socially engaged elderly; 2) less socially engaged caregivers; 3) socially engaged caregivers; 4) leisure-engaged elderly; and 5) productive-engaged elderly. Older people who were not engaged in any social activity other than the care for a sick person, were identified as a possible target group, given the relatively high share of unhealthy people among them. In this non-socially engaged target group, the prevalence of loneliness was 48%, compared to 30% in the socially engaged groups; poor self-perceived health: 41% compared to 14%; poor mental health: 25% compared to 9%; poor physical health: 27% compared to 2%.

Conclusion
Well-functioning social relationships were favourably associated with health. By integrating all results, the local data have strengthened the scientific evidence-base for local policy making and have contributed to the development of an evidence-based community intervention supporting social participation among older people.