Page 1 of 17http://spider.apa.org/ftdocs/abn/1999/november/abn1084558.html 9/15/2000Social Support and the Course of Bipolar Disorder Sheri L. JohnsonDepartment of Psychology University of Miami Carol A. WinettDepartment of Psychology University of Miami Björn MeyerDepartment of Psychology University of Miami William J. GreenhouseDepartment of Psychology University of Miami Ivan MillerDepartment of Psychiatry and Behavior Brown University ABSTRACTThe current study prospectively examined the impact of social support on symptom severity and recovery from episodes in bipolar disorder, both as a direct influence and as a buffer of life events. Fifty-nine individuals with Bipolar I disorder were followed longitudinally with monthly symptom severity interviews. Social support was measured by the Interpersonal Support Evaluation List and the Interview Schedule for Social Interaction, and life events were assessed using the Life Events and Difficulties Schedule. Individuals with low social support took longer to recover from episodes and were more symptomatic across a 6-month follow-up. Results suggest a polarity-specific effect, in that social support influences depression but not mania. Discussion focuses on theoretical implications of a series of polarity-specific findings within the field. We gratefully acknowledge the contributions of Dena Rosenberg, Randy Fingerhut, Ruben Perczek, and Julie Dykstra for their help in rating and interviewing. This study is supported by grants from the National Alliance for Research on Schizophrenia and Depression and from National Institute of Mental Health Grant R29-MH55950. Correspondence may be addressed to Sheri L. Johnson, Department of Psychology, University of Miami, Coral Gables, Florida, 33124—0721. Electronic mail may be sent to [email protected] Received: August 5, 1998 Revised: February 20, 1999 Accepted: March 2, 1999 Bipolar disorder is severe, with devastating consequences for affected individuals and for society. Of individuals hospitalized for mania, 30% remain unemployed for 6 months ( Dion, Tohen, Anthony, &Waternaux, 1988 ) and 23% for 1 year ( Harrow, Goldberg, Grossman, & Meltzer, 1990 ). Across physical and psychiatric disorders, bipolar disorder is ranked as the sixth leading cause of disability ( Murray & Lopez, 1996 ). In 1991, costs for bipolar disorder among adult Americans totaled $45 billion ( Wyatt & Henter, 1995 ). The personal costs of this disorder are emphasized by the alarming Journal of Abnormal Psychology © 1999 by the American Psychological Association November 1999 Vol. 108, No. 4, 558-566 For personal use only--not for distribution.Page 2 of 17http://spider.apa.org/ftdocs/abn/1999/november/abn1084558.html 9/15/2000finding that as many as 19% of bipolar individuals die from suicide ( Isometsa, 1993 ). Beginning with Cade's discovery of the mood-stabilizing effects of lithium in 1949, remarkable advances have been made in pharmacotherapy of bipolar disorder. Double-blind randomized trials demonstrate the efficacy of lithium in reducing the severity and frequency of episodes ( Sachs, Lafer,Truman, Noeth, & Thibault, 1994 ). Newer medications, such as valproate and carbamazepine, have shown promise as well ( Ketter et al., 1998 ; McElroy, Keck, Pope, & Hudson, 1992 ). The efficacy of pharmacotherapy has contributed to a biological zeitgeist, which has been bolstered by twin studies demonstrating an extremely high concordance for bipolar disorder ( Bertelsen, Harvald, & Hauge,1977 ). For many years, it was assumed that bipolar disorder was understood and controlled using these biological approaches. Despite the gains from pharmacological approaches, many individuals still experience poor outcomes. Even with blood serum levels of lithium between 0.8 and 1.0, 32% of bipolar individuals relapsed over a 3-year period ( Keller et al., 1992 ). Treatments are less effective for bipolar depression than mania ( Hlastala et al., 1997 ). Within naturalistic studies, noncompliance contributes to even higher relapse rates: In one study, only 4% of patients who were prescribed lithium monotherapy sustained remission for 1 year ( Sachs et al., 1994 ). Beyond the poor outcomes, biological models are limited in predicting course of disorder. Several factors, such as seasonality, previous history, and family history predict course within some studies; however, these effects tend to explain a relatively small proportion of the variance in outcome (cf. Coryell, Endicott, & Keller,1992 ; Silverstone, Romans, Hunt, & McPherson, 1995 ). In summary, biological indexes do not fully account for the timing and severity of episodes. Noting that these gaps exist, researchers have recognized the need for multifactorial models of treatment and etiology ( Prien & Potter, 1990 ). Recent research has demonstrated the importance of psychosocial treatments as adjuncts to medication ( Miklowitz, 1996 ). In tandem with shifts in treatment approaches, psychosocial predictors of the course of the disorder have been examined more carefully. Life events ( Hammen, Ellicott, & Gitlin, 1992 ; Johnson & Roberts, 1995 ) and personality ( Swendsen, Hammen, Heller, & Gitlin, 1995 ) predict course. Most relevant to social relationships, expressed emotion appears highly predictive of relapse ( Miklowitz, Goldstein, & Nuechterlein,1987 ; Miklowitz, Simoneau, Sachs-Ericsson, Warner, & Suddath, 1996 ; O'Connell, Mayo, Flatow,Cuthbertson, & O'Brien, 1991 ; Priebe, Wildgrube, & Muller-Oerlinghausen, 1989 ). Even though negative aspects of relationships appear so important, relatively little empirical attention has been paid to positive aspects of relationships, such as social support. Social support has been shown to have robust effects on a broad range of psychiatric and biological outcomes outside of bipolar disorder. Within unipolar depression, Brown and his colleagues have shown that women experiencing a severe event without support from a confidant had a 40% risk of developing depression; in contrast, those with a confidant's support had a 4% risk ( Brown &Andrews, 1986 ). Other studies have replicated the importance of social support, both directly and as a stress buffer, within depression as well as other psychiatric disorders ( Kessler, Price, & Wortman,1985 ; Monroe & Johnson, 1992 ). The biological vulnerability to bipolar disorder makes the physiological effects of social support particularly intriguing. Social support is strongly linked with both
View Full Document
Unlocking...