Influence of the family cluster effect on psychosocial variables in families undergoing BRCA1/2 genetic testing for cancer susceptibility
Article first published online: 2 MAR 2011
Copyright © 2011 John Wiley & Sons, Ltd.
Volume 21, Issue 5, pages 515–523, May 2012
How to Cite
Lapointe, J., Abdous, B., Camden, S., Bouchard, K., Goldgar, D., Simard, J. and Dorval, M. (2012), Influence of the family cluster effect on psychosocial variables in families undergoing BRCA1/2 genetic testing for cancer susceptibility. Psycho-Oncology, 21: 515–523. doi: 10.1002/pon.1936
- Issue published online: 20 APR 2012
- Article first published online: 2 MAR 2011
- Manuscript Accepted: 20 JAN 2011
- Manuscript Revised: 18 JAN 2011
- Manuscript Received: 8 OCT 2010
- BRCA1/2 testing;
- cancer susceptibility;
- design effect;
- family clustering;
- psychosocial issues;
- statistical methods
Objective: This study evaluated the influence of the family cluster effect on behavioral and psychological variables among individuals undergoing BRCA1/2 genetic testing for cancer susceptibility.
Methods: French-Canadian women (n = 552) and men (n = 104) from 140 different families undergoing BRCA1/2 testing between years 1998 and 2004 completed a self-administered questionnaire on a wide range of behavioral and psychological variables. The impact of the family cluster effect on 29 variables was assessed using the intraclass correlation coefficient (ICC) as computed from multilevel random-effect models.
Results: ICC values were statistically significant for 22 of the 29 variables. Overall, the mean ICC value was 0.10. The ICC values for knowledge about hereditary breast and ovarian cancer and frequency of alcohol consumption were 0.25. The ICC value for the Impact of Event Scale, which is a commonly used measure of cancer-specific distress, was 0.14.
Conclusions: Our results suggest that the family cluster effect influences the majority of behavioral and psychosocial variables. When studying psychosocial aspects of genetic testing for cancer susceptibility, the family cluster effect should be routinely accounted for when determining sample size and statistical methods. Copyright © 2011 John Wiley & Sons, Ltd.