The Effect of Social Network Types On the Health Care Utilization In the Korean Older Adults

Purpose: Social network can serve as a meaningful construct in predicting both physical and psychosocial outcomes of older adults (Beckman & Karachi, 2000; Dunkley, Roberts & Hag, 2001; Ell, 1984; Litwin & Landau, 2000). In this research, we focus older adults' health care utilization by their social network characteristics. By investigating how social network promotes or inhibits health service use, individuals most at risk could be identified and interventions can be designed to meet the needs of those at risk. Compared to the previous research on older adults' health care use focusing on one or two constructs of such as social support or informal support, our research is different in two respects: First, it employed a social network approach based on both structural and functional dimension of social ties. Second, it focused on a possible heterogeneity within older adults group in the service use pattern by gender and age level (the young old, old-old, the oldest)

Method: In order to identify an independent role of network in health service utilization, we examined the impact of social network as an enabling factor on older adults' health service utilization based on Anderson and Newman's health behavior model. We analyzed the 2008 data from the Korean Longitudinal Study of Aging(Kolas), focusing on respondents aged 65 years and older. The analytic sample was N= 4251. Social network measured by marital status, frequency of social activities, frequency of attendances at religious service, contact frequency with friends, children, and financial support from children. Health service utilization was measured with five services including outpatient care, dental care, prescription drug use, oriental medicine care, and hospitalization. The impact of predisposing, needs, and social network on health service use were analyzed with logistic regression.

Results: The general findings show that stronger ties embedded in the social network are associated with more health care utilization, which is consistent with previous social network research in other countries (Renner mark& Hag berg, 1999; Lit win, 1997) except that marital status is not significantly related with any of the health service use. On closer examination, the impact of social network is found to vary according to type of health service use. Compared to other service use, outpatient use, for example, higher level of social activities(Odds ratio =1.0169, p<.001) and higher level of financial support from the children(Odds ratio=1.019, p<.001) are strongly associated with social network than other service use. Service utilization varies by gender: men are less likely to use outpatient care(Odds ratio=.7724, p<.01) and oriental medicine care(Odds ratio=.5544, p<0.001). By age group, compared to the old-old, the oldest are less likely to use all health services except dental care.

Implication: Older adults with similar health problems may need different health and social services if they are in a different network. Similarly, older adults with the similar network characteristics may have similar types of problems. The findings on the different pattern of health service use by network type, gender, and age group can provide important information in designing needs assessment and program developments.