Currently, we have developed exponential random graph models (ERGMs) for incomplete networks. These models accommodate network data that have been obtained using cascade sampling or snowball sampling designs or where eligible respondents have declined participation in the study (NHGRI Protocol # 05-HG-N249; PI: Laura Koehly). The 80 interview transcripts from members of 13 families with known mutations for hereditary nonpolyposis colo-rectal cancer (HNPCC) have been coded and will be used to illustrate the statistical models that have been developed for partial network data. We will be using these models to investigate the role of the familial social structure in communications about genetic testing and counseling and disclosure of mutation status. Preliminary analyses indicate that dissemination patterns of hereditary risk information are not only related to biological relationships. Although family members tend to discuss family risk with their first degree relatives, they reach other family members through their social relationships as defined by cohesiveness, advice-seeking, and lack of conflict. The accuracy and robustness of the ERGMs for partial networks will be evaluated using complete network data that will be obtained from an ongoing study examining the role of the familial social context in disclosure of family risk information, screening behaviors, and adjustment to risk status (NHGRI Protocol # 95-HG-0165; PI: Don Hadley). We have begun recruitment of both at-risk and not at-risk family members from 51 families with identified HNPCC germline mutations and anticipate that we will be in the field collecting data until October, 2008. These individuals will be asked to provide information regarding their family relationships (e.g. cohesiveness, support, and conflict) as well as the communication of family risk information. Additionally, we are investigating the role of social support processes in women coping with their hereditary risk for breast-ovarian cancer (NCI Protocol # 01-C-0009; PI: Jennifer Loud). This research uses the Colored Eco-genetic Relationship Map (CEGRM) to assess the social support networks of study participants. Currently,200 participants have been recruited into the study. CEGRM assessments and psychosocial measurements are obtained at baseline and at three annual follow-ups. Our analyses suggest that cooperative support processes among sisters are both positively and negatively associated with well-being. Shared emotional support resources among sisters appears to facilitate positive adaptation, whereas large numbers of shared informational resources among sisters appears to be associated with increased somatization suggesting a contagion effect. Finally, we are investigating the dissemination process for complex disease risk information based on family health history and the development of family level strategies to address this risk (NHGRI Protocol # 07-HG-N140; PI: Laura Koehly). This research uses the CDCs Family Healthware to provide risk information based on participants family history and behavioral recommendations based on participants current health behaviors. We will use this software to provide feedback to participants from Mexican American households in the Houston, TX area and assess how this feedback motivates family communications about common, complex diseases and the development of cooperative strategies, such as encouragement to screen, to address this risk. At the time of this report, 97 participants had completed baseline assessments (32 households) and 17 had completed the 1-month follow-up assessment.
Showing the most recent 10 out of 29 publications
