JOURNAL OF PERSONALITY ASSESSMENT, 1990, 55(3&4), 618-629 Copyright 1990, Lawrence Erlbaum Associates, Inc.

Emotional Reliance and Social Loss: Effects on Depressive Symptomatology James C. Overholser Case Western Reserve University

A reactive form of dependence has been proposed to occur when a person is undergoing a period of substantial stress and change. The present study assessed 114 psychiatric inpatients categorized according to the presence or absence of social loss and their level of emotional reliance on others. Both emotional reliance and social loss were related to a variety of depressive symptoms. A significant interaction was observed between emotional reliance and social loss on depression severity as measured by the Beck Depression Inventory (BDI). In general, subjects high in emotional reliance but experiencing no social loss displayed higher levels of depression than emotionally reliant subjects who had undergone a social loss. Patients reporting high emotional reliance on others, in the aftermath of a social loss, may be reacting to the loss and suffer from less-severe and less-chronic pathology. Subjects reporting excessive emotional reliance in the absence of any precipitating exit event may be displaying more of a trait-like pathology. Personality disorder pathology should occur with such frequency and intensity so it can be observed even when obvious eliciting stimuli are absent.

Attachment and dependency issues remain important throughout the life span (Ainsworth, 1989). Traditional views of personality consider unfulfilled dependency needs t o be a cause of anxious and depressed moods. Building o n Spitz's earlier work o n anaclitic depression (Spitz & Wolf, 1946), Blatt (1974) argued that depression is related t o the excessive dependency placed o n others for support, reassurance, and gratification. In a similar way, Bowlby (1960, 1973, 1977a, 197713, 1982) refined the concept of separation anxiety as a n important element in dependency. When children are uncertain about the accessibility of their primary caretaker, they become anxious over such tenuous attachment, and this lays the foundation for dependency. Because the child's primary caretaker can be viewed as a safety signal, limited or uncertain availability to the caretaker can imply a threat of danger. Furthermore, the separation process initially results in anxiety, but is soon followed by depression and despair. This relationship between separation and depression has been supported in both

EMOTIONAL RELIANCE AND SOCIAL LOSS

61!3

clinical reports (Berlin, 1986) and research investigations (Tennant, Hurry, 6 r Bebbington, 1982). Whether considered a long-standing personality trait or a transient emotional state, the concept of dependency appears closely tied to depression (Birtchnell, 1984, 1988). Previous research has found a strong relationship between depression and dependency in adult clinical samples (Hirschfeld et al., 1983). For example, many patients simultaneously meet criteria for a major depressive episode and the dependent ~ e r s o n a l i disorder t~ (Charne~, Nelson, & Quinlan, 1981; Koenigsberg, Ka~lan,Gilmore, & Cooper, 1985; Shea, Glass, Pilkonis, Watkins, & do chert^, 1987).Also, patients assessed after recovery from depression continue to display significantlyhigher levels of dependency as compared to controls (see Barnett & Gotlib, 1988),even though this may involve a significant drop as compared to their own personality description when depressed Uoffe dr Regan, 1988). Such research suggests that dependent persons are at risk for depression (Birtchnell, I 988). Dependent persons are thought to become clinically depressed when the person upon whom they depend leaves through death, divorce, or relocatioin (Malinow, 1981; Millon, 1981). When safely attached to a nurturant caretaker, they have less reason to experience emotional distress. Although dependency cannot account for all types or causes of depression, it does seem that many depressives have a fragile self-esteem due to their dependency on others for support and nurturance (Hirschfeld, Klerman, Chodoff, Korchin, & Barretr, 1976).This is in agreement with the behavioral literature which has documented self-reinforcement deficits in a subsample of depressives (Heiby, 1983a, 1983b, 1986; Heiby, Campos, Remick, & Keller, 1987).Excessive reliance on others far reinforcement makes the person vulnerable to social loss. When an important social loss occurs, the dependent person loses a vital source of reinforcement and is incapable of compensating with self-reinforcement. The Diagnostic and Statistical Manual of Mental Disorders (3rd ed., rev.; DSM-Ill-R) diagnostic criteria (American Psychiatric Association, 1987) acknowledge that depression is a common complication of the dependent personality disorder, describing the dependent personality as feeling "devastated or helpless when close relationships end" (see Table 1). However, a temporary period of dysphoria may be a natural consequence of the separation process (Bowlby, 1973). Research on the effects of life stress has shown that major stressors or minor "hassles" can have significant effects on one's emotional and physiological state (Eckenrode, 1984). Although feelings of dependency may be caused by specific threatening events, the notion of a personality disorder suggests that the disorder occurs with such frequency and intensity that it I S observed in the absence of any obvious triggering stimuli (Birtchnell, 1988). Such a person is likely to experience depression as a way of life (Gross, 1981). In contrast, self-reliant individuals have been found more resistant to the effects of life stress (Funch & Marshall, 1984).

TABLE 1

Diagnostic Criteria for the Dependent Personality Disorder A pervasive pattern of dependent and submissive behavior, beginning by early adulthood and present in a variety of contexts, as indicated by at least five of the following: 1. Is unable to make everyday decisions without an excessive amount of advice or reassurance from others. 2. Allows others to make most of hi or her important decisions (e.g., where to live, what job to take). 3. Agrees with people even when he or she believes they are wrong, because of fear of being rejected. 4. Has difficulty initiating projects or doing things on his or her own. 5. Volunteers to do things that are unpleasant or demeaning in order to get other people to like him or her. 6. Feels uncomfortable or helpless when alone or goes to great lengths to avoid being alone. 7. Feels devastated or helpless when close relationships end. 8. Is frequently preoccupied with fears of being abandoned. 9. Is easily hurt by criticism or disapproval. Note. Based on DSM-111-R criteria (American Psychiatric Association, 1987).

The notion of a reactive form of dependency was proposed in the theoretical literature (Parens & Saul, 1971), but has not been empirically examined. Essentially, a reactive form of dependency occurs when a person is undergoing a period of substantial stress and readjustment. Early studies (e.g., Schacter, 1959) have shown that stressful times serve to increase affiliation tendencies in normal adults. Apparently, emotional distress produces an increased desire to affiliate with others (Greenberg & Bornstein, 1988). This may be a natural process, but excessive interpersonal dependency occurring in the absence of stress may be indicative of pathological processes. The present study was designed to explore the relationship between dependency and social loss in psychiatric inpatients. Because both of these variables appear related to depressive symptomatology, only patients receiving a diagnosis of a depressive disorder were involved in the study. An attempt was made to identify patients high in dependency following a social loss by comparing them to dependent patients who experienced no social loss. Also, depressive control groups displaying low levels of dependency with and without experiencing a social loss, were used. Depression severity, problem-solving styles, and attribution and coping variables were assessed.

METHOD Subjects One hundred fourteen psychiatric inpatients at a private, university-affiliated psychiatric hospital were assessed. All patients had received a primary diagnosis

EMOTIONAL RELIANCE AND SOCIAL LOSS

62'1

of depression (major depression, single or recurrent episode; dysthymic disorder; or adjustment disorder, depressed) from an independent team of treatinlg psychiatrists who were blind to the purposes of the study. Subjects were excluded from the study if they received a secondary diagnosis of schizophrenia, schizoaffective disorder, bipolar disorder, any anxiety disorder, mental retardaltion, or organic brain syndrome. All subjects were between the ages of 20 and 55 and were tested during the first week of hospitalization. Subjects were excluded from the study if their education level was less than the eighth-grade level, if they had difficulties with the English language, or if they displayed evidence of chronic alcohol or substance abuse problems.

Independent Variables The Interpersonal Dependency Inventory (Hirschfeld et al., 1977) contains three subscales, assessing emotional reliance on other people, level of social self-confidence, and assertion of autonomy. The Emotional Reliance scale, reflecting attachment-related dependency, appears central to the current definition of the dependent personality. A median split on the Emotional Reliance scale was used to classifiysubjects as high or low in emotional reliance. The Life Experiences Survey (Sarason, Johnson, & Siegel, 1978) was used to assess the presence and severity of stressful life events experienced during the previous 6 months. For the purposes of this study, the Life Experiences Survey was used to assess the presence of social losses or exit events whereby an important person in the patient's life left them through death, divorce, or relocation. The objective occurrence of social loss was used instead of their subjective impact ratings in order to reduce any possible confound between emotional reliance scores and social loss. Furthermore, stress researchers often use the objective occurrence of specific events because the subjective ratings do not contribute much additional information (Monroe, Bellack, Hersen, fii Himmelhoch, 1983; Ross & Mirowsky, 1979; Shrout, 1984). Also, ratings of thLe objective occurrence of specific events are more reliable (Neugebauer, 1981) and less affected by emotional and interpretive biases (Monroe, 1982; Paykel, 1983). Such biases become especially prominent when patient samples are used (Gerst, Grant, Yager, & Sweetwood, 1978; Paykel, 1983). Thus, the use of objective occurrence of events ignored subjects' interpretations of the events, but allowed for the scoring of social loss separate from emotional reliance.

Dependent Measures The BDI (Beck, Ward, Mendelson, Mock, & Erbaugh, 1961) is a 21-item scalle designed to assess depression severity. This scale has a wealth of data docu-

menting its clinical and research utility (Beck, Steer, & Garbin, 1988; Kendall, Hollon, Beck, Hammen, & Ingram, 1987). The Dysfunctional Attitudes Scale (DAS; Weissman & Beck, 1978) includes 40 statements to be rated on a 7-point scale ranging from totally agree to totally disagree. The statements reflect maladaptive beliefs and attitudes commonly endorsed by depressed patients. Possible scores range from 40 to 280, with scores above 120 reflecting clinical levels of depressive attitudes. The Problem-Solving Inventory (PSI; Heppner & Petersen, 1982) includes 35 statements describing one's typical response tendencies when confronted with an unspecified problem. Items are rated on a 6-point scale ranging from strongly agree (1) to strongly disagree (6). This scale contains three factorial-derived subscales: Problem-Solving Confidence, Personal or Self-control Over Problems, and Approach-Avoidance Tendencies. Higher scores reflect more pathology in the specified area. A series of attributional ratings were collected following the procedure established by Gong-Guy and Hammen (1980). These measures ask the subject to select the four most stressful events that occurred over the previous 6 months. Subjects were then asked to rate each event according to its perceived importance, its ability to affect one's mood, its ability to cause depression in the subject, and its ability to affect other areas of the person's life. Also, subjects rated their own and other people's ability to prevent similar events in the future and how much subjects blamed themselves for the occurrence of the event. Based on the work of Moos, Cronkite, Billings, and Finney (1983), subjects were asked to rate the frequency with which they used 33 different coping responses when attempting to manage the most stressful event they had experienced. These items were then grouped into 7 subscales: Active Cognitive Coping, Active Behavioral Coping, Avoidance Coping (relying on passive, indirect coping strategies, e.g., eating, drinking, complaining, or smoking), Logical Analysis, Information Seeking, Problem-Solving, Affective Regulation, and Emotional Discharge. These coping scales have demonstrated adequate psychometric properties in previous research (Billings & Moos, 1981).

RESULTS A preliminary analysis was conducted in order to rule out the possible effects of marital status on emotional reliance to control for the social and emotional support typically provided by one's spouse. Subjects were categorized according to the presence or absence of a spouse (or cohabitating partner). No differences across groups were seen in emotional reliance, t(115) = .75, ns, thus ruling out marital status as a possible confounding variable. Next, differences in emotional reliance were examined across subjects categorized by the presence or absence of exit events. Most subjects experiencing a social loss reported only one (52%)or

two (31%)losses. Thus, social loss was a discontinuous variable, not amenable to multiple regression ~rocedures.A t test across social loss (presendabseni:) revealed significant differences in emotional reliance, t(115) = 2.25, P < .03. Subjects reporting the presence of one or more social loss (n = 79) obtained significantly higher emotional reliance scores (M = 48.33) as compared to subjects reporting no exit events (n = 35, M = 44.24). Because of the relevance and interrelatedness of emotional reliance and social loss, subjects were classified into four groups based on a median split on the emotional reliance scale in combination with information about the presence or absence of social loss. Initial analyses examined differences between groups on basic demographic and clinical variables (see Table 2). Results showed the groups did not differ in terms of age, F(3,110) = .24, ns; race (all subjects were White); sex, X2(3)= 1.49, ns; yearly income, F(3,99) = 1.85, ns; or educational background, F(1, 100) == .31, ns. They were also similar on clinical variables such as the number of previous psychiatric hospitalizations, F(1, 110) = .98, ns; the number of days TABLE 2

Demographic and Clinical Variables Across the Four Groups Low Emotional Reliance Variable

No Social Loss

Social Loss

-

High Emotional Reliance No Social Loss

Social Lass -

n Age Race (% White) Sex (% Female) Marital status % Single % Married % W/S/D Years education Yearly income (in thousands)

# Previous psychiatric hospitalizations Positive family history of psychiatric problems (%)

# Days in hospital Suicidal ideation (% present)

Note. W = widowed; S = separated; D = divorced.

-

spent in the hospital, F(3, 110) = .12, ns; the family history of psychiatric problems, X2(3)= 1.56, ns; and the presence of suicidal ideation at the time of hospitalization, x2(3) = 4.13, ns. Also, there were no differences in terms of social support as measured by the number of friends they had, F(3, 102) = .40, ns. Thus, we were able to rule out the possible confounding effects of a variety of relevant variables. In terms of depression severity (see Table 3), a significant interaction between emotional reliance and social loss was found on BDI scores, F(1, 110) = 4.15, p < .05. At low levels of emotional reliance, the presence of social loss had no effect on depression severity (keeping in mind that all subjects were depressed inpatients). However, at high levels of emotional reliance, significantly higher depression scores were seen in subjects reporting no social loss. Emotional reliance was also associated with higher levels of dysfunctional attitudes, F(1, 110) = 23.34, P < .0001, reflecting the presence of depressogenic attitudes and cognitions. When examining subscales of the Problem-Solving Inventory, significant effects were found for both Emotional Reliance and Social Loss. Emotional reliance was related to a lack of control over problems, F(1, 110) = 4.95, p < .03. Social loss was significantlyrelated to a lack of problem-solving confidence, F(1, 110) = 4.55, p < .05, and feelings of inadequate personal control over the problems, F(1, 110) = 6.57, p < .01. Specific problem-solving and attributional variables were examined as related to the most stressful, "non-exit" event reported by the patient. A multivariate

TABLE 3 Test Data Across the Four Groups Low Emotional Reliance Variable BDIa

DAS~ Problem solving inventory Problem solving confidencec Personal controlb9' Avoidance of autonomy

High Emotional Reliance

N o Social Loss

Social Loss

N o Social Loss

Social Loss

OD) M

19.95 (11.70) 129.23 (37.42)

20.20 (11.83) 125.50 (35.93)

32.15 (6.68) 175.31 (36.99)

23.53 (9.54) 160.10 (43.52)

M (SD) M (SD)

35.14 (11.32) 22.95 (3.97)

30.97 (9.89) 19.93 (5.75)

36.69 (13.15) 25.62 (3.88)

39.79 (11.45) 22.55 (6.53)

M (SD)

51.95 (12.75)

50.13 (15.57)

55.46 (17.15)

53.61 (15.58)

M

"Significant interaction effect. bSignificant main effect for emotion reliance. 'Significant main effect for social loss.

EMOTIONAL RELIANCE AND SOCIAL LOSS

62!5

analysis of variance (MANOVA) revealed significant effects for both emotional reliance, F(11, 55) = 3.13, p < .003, and social loss, F(11,55) = 2.07, P < .04, across the attributional ratings. Univariate analyses of variance (ANOVAs) showed that emotional reliance was associated with attributions suggesting a feeling that the event greatly affected their mood, F(l, 65) = 4.59, P < .05, and was especially likely to increase their depression, F(1, 65) = 16.87, p < .0001. Furthermore, emotional reliance was associated with a reduced belief in the ability to avoid stressful events, F(1,65) = 4.12, p < .05; an increased tendency to blame oneself for the occurrence of negative events, F(1,65) = 6.52, p < .Ol; and the belief that other people would be more effective than they were in managing the emotional consequences of the event, F(1,65) = 13.30, p < .001. Social loss was associated with attributions suggesting the stressful event had a strong influence on their depression, F(l, 65) = 4.27,p < .05, as well as other areas of their life, F(1, 65) = 3.72, p < -05. A significant interaction between social loss and emotional reliance was observed a n subjects' beliefs in their ability to prevent negative events in their lives, F(1, 65) = 4.50, p < .05. A variety of differences, mainly limited to nonsignificant trends, were seen in the manner subjects attempted to cope with the most stressful event, as seen in problem-solving, emotional reliance, F(l, 68) = 3.26, p < .075; social loss, F(1, 68) = 4.62, p < .05; avoidance coping, emotional reliance, F(l, 68) = 3.12, p

Emotional reliance and social loss: effects on depressive symptomatology.

A reactive form of dependence has been proposed to occur when a person is undergoing a period of substantial stress and change. The present study asse...
698KB Sizes 0 Downloads 0 Views