A randomized controlled experiment was conducted to investigate the effects of expressive writing and fact-control writing about experiences with ovarian cancer on emotional well-being and quality of life one month after writing in women diagnosed with ovarian cancer. Moderation effects of emotional approach coping and social constraints were predicted based on a matching hypothesis for the fit between the demands of writing and individual differences (Niles et al, 2014) and the role of social constraints on disclosure (Lepore & Revenson, 2007). Cancer-related avoidance and cancer-related intrusive thoughts were examined as mediators on the relationships between social constraints and emotional well-being at follow up and social constraints and quality of life at follow up. Results showed that participants in the expressive writing condition reported increased emotional well-being at follow up, controlling for baseline levels of emotional well-being, but there were no differences in quality of life at follow up between the two writing conditions, controlling for baseline quality of life. There were no differences in reported average negative affect post writing sessions between the two writing conditions. In a model predicting emotional well-being at follow up, expressive writing had a positive effect, social constraints had a negative effect, and emotional approach coping had no effect. The hypothesized moderation effects between emotional approach coping, social constraints and writing condition in predicting emotional well-being at follow up were not detected and support for the matching hypothesis proposed by Niles et al (2014) was not found. In a marginally significant model, social constraints had a significant effect and interacted with writing condition to predict quality of life at follow up, such that participants with high baseline levels of social constraints benefitted

more from the fact control condition, whereas participants with low baseline levels of social constraints benefitted more from the expressive writing condition. Finally, cancer-related intrusive thoughts mediated the effect of social constraints on emotional well-being at follow up but not on quality of life at follow up. Cancer-related avoidance was not found to mediate the effect of social constraints on either emotional well-being at follow up or quality of life at follow up.