Advisor

Nancy J. Chapman

Date of Award

1-1-1985

Document Type

Dissertation

Degree Name

Doctor of Philosophy (Ph.D.) in Urban Studies

Department

Urban Studies and Planning

Physical Description

4, xiv, 277 leaves: ill. 28 cm.

Subjects

Health and environmental sciences, Social interaction, Mortality, Health

DOI

10.15760/etd.451

Abstract

Several recent longitudinal studies of large community populations have revealed that people with more extensive network resources live longer. However, it is not known whether this occurs because social ties prevent disease or retard its progression once it occurs. The purpose of this research was to: (1) determine the relationship between social network indicators and mortality in an urban sample; (2) extend that knowledge by addressing the relationship between networks and disease incidence and disease progression; (3) delineate which specific network sectors were the strongest predictors of the health related outcomes. This was uniquely possible because measures of the three dependent variables were available within the same data set at the Kaiser Permanente Center for Health Research. The research design was longitudinal, based on survey data. The conceptual framework posited that social support delivered via social networks modifies disease states. The setting was the Northwest Region, Kaiser Permanente Health Care Plan, an HMO serving the Portland/Vancouver SMSA. The sample included 2603 adults who participated in a 1970 household interview survey. Their health service utilization data from 1967-73 has been computerized and linked with the survey information. As of 1982, 376 have died. To measure the independent variables, four summary social network indexes (scope, size, frequency of contact, and interaction) were prepared according to a network model based upon the survey questions available, network theory, and prior research. Indexes representing nine relationship domains were constructed. Control variables included age, sex, SES, health status indicators, and health behaviors. Multiple logistic regression was used to assess hypothesis 1 and ordinary regression was used to assess hypotheses 2 and 3. Each of the four summary network measures was a statistically significant predictor of 12 year mortality. Network scope was the strongest predictor. Marital, family, and kin relationships were not predictive of death. Extended ties of close friends, other friends, work associates, and social leisure activities were significant predictors. No relationship was found between network scope, disease incidence, or disease progression, so it is still unclear how social connections act to decrease mortality.

Description

Portland State University. School of Urban and Public Affairs.

Persistent Identifier

http://archives.pdx.edu/ds/psu/4503

Share

COinS