Excessive Reassurance Seeking, Depression, and Interpersonal Rejection: A Meta-Analytic Review Lisa R. Starr and Joanne Davila State University of New York at Stony Brook Coyne’s (1976a, 1976b) interactional theory of depression predicts positive associations between excessive reassurance seeking (ERS) and both depression and interpersonal rejection. A growing body of research has supported the ERS model, but this work has yet to be systematically reviewed. A meta-analysis of 38 studies (N = 6,973) revealed an aggregate effect size (r) of .32 between ERS and concurrent depression. Moderator analyses showed effect sizes were significantly stronger for studies with self-report measures, compared with interviews, and for samples with higher percentages of women and were marginally stronger for studies with community samples, compared with clinical samples. A second meta-analysis of 16 studies yielded a weighted mean effect size of .14 between ERS and concurrent rejection, with studies assessing target-reported rejection showing stronger effect sizes than studies assessing informant-reported rejection and studies examining romantic relationships yielding marginally stronger effect sizes than studies examining nonromantic relationships. Prospective studies are qualitatively reviewed. Results support the ERS model (with several important caveats) but under- score the need for methodological diversity in future research. Keywords: excessive reassurance seeking, meta-analysis, depression, interpersonal rejection Interpersonal models have guided understanding of the etiology, course, consequences, and treatment of depression (Armsden & Greenberg, 1987; Gotlib & Hammen, 1992; Mufson et al., 2004). Coyne’s (1976a, 1976b) interactional theory of depression has emerged as one of the most influential frameworks for studying interpersonal aspects of depression. In this model, mildly de- pressed people attempt to assuage feelings of guilt and low self- worth by seeking reassurance from others. At first, others provide support, but the depressed person doubts its authenticity and con- tinues to seek reassurance until the other person grows annoyed and rejects them. The rejection exacerbates their symptoms as the cycle continues. Joiner and colleagues (Joiner, Alfano, & Metalsky, 1992; Joiner, Metalsky, Katz, & Beach, 1999) point to excessive reassurance seeking (ERS) as the active ingredient in this process. They define ERS as the relatively stable tendency to repeatedly request reas- surance from others that one is lovable and worthy, despite pre- vious attempts by others to provide such reassurance. ERS has been implicated as a risk factor for the development, maintenance, and worsening of depression (Joiner & Metalsky, 2001; Joiner, Metalsky et al., 1999), a force behind interpersonal rejection (Benazon, 2000; Joiner et al., 1992), and a mediator of depression contagion (Joiner, 1994). The past decade and a half has seen an explosion of research on ERS, investigating the role of related interpersonal constructs (Davila, 2001; Shaver, Schachner, & Mikulincer, 2005), neurological underpinnings (Minnix et al., 2004), and clinical implications (Stellrecht, Joiner, & Rudd, 2006). Although ERS and its association with depression and rejection have been examined in substantial work, there has yet to be a quantitative, systematic synthesis of the research that clarifies what is and is not known about ERS. In addition, numerous aspects of the model have not been sufficiently developed, tested, or refined. Our meta-analysis was designed to shed light on these issues. First, we quantitatively summarize current knowledge of ERS. Given the prominence of the ERS model within depression research, it is crucial to ensure that it fits with existing data to ensure that the conceptual premises guiding understanding of depression and its treatment are supported. As two other aspects of Coyne’s (1976a, 1976b) theory, depression contagion and the depression–rejection link, have previously been supported meta-analytically (Joiner & Katz, 1999; Segrin & Dillard, 1992), a meta-analytic review of the ERS literature would lend convergent support to this model. Sec- ond, we use meta-analysis to look for patterns in the data that cannot be easily deciphered by examining individual study reports. For example, the link between ERS and depression has been relatively consistent across studies, which on the surface suggests a well-established finding with little need of replication. However, most of these studies have had very similar samples and method- ology (an aspect of the ERS literature that has been appropriately Lisa R. Starr and Joanne Davila, Department of Psychology, State University of New York at Stony Brook. Portions of these results were presented at the 2007 Annual Meeting of the Association for Behavioral and Cognitive Therapies, Philadelphia, Pennsylvania. This research was supported by National Institute of Mental Health Grants F31MH082545 (awarded to Lisa Starr) and R01 MH063904-1A2 (awarded to Joanne Davila). The content is solely the responsibility of Lisa R. Starr and Joanne Davila and does not necessarily represent the official views of the National Institute of Mental Health or the National Institutes of Health. We thank Catherine B. Stroud and Jennie Park for their assistance with study identification and coding, as well as Anne Moyer for her statistical consultation and comments on drafts of this article. Correspondence concerning this article should be addressed to Lisa R. Starr, Department of Psychology, State University of New York at Stony Brook, Stony Brook, NY 11794-2500. E-mail: lisa.starr@stonybrook.edu Journal of Abnormal Psychology Copyright 2008 by the American Psychological Association 2008, Vol. 117, No. 4, 762–775 0021-843X/08/$12.00 DOI: 10.1037/a0013866 762