ORIGINAL RESEARCH

Are Dependency and Self-Criticism Risk Factors for Major Depressive Disorder?

Ari E Zaretsky, MD^1,2, Maurizio Fava, MD^3,4, Katharine G Davidson, BA⁵, Joel A Pava, PhD⁵, John Matthews, MD⁵, Jerrold F Rosenbaum, MD⁶

Objective: To determine whether dependent and self-critical personality traits are associated with specific types of life events and whether these traits change with pharmacotherapy.

Method: Overall, 142 depressed outpatients completing 8 weeks of fluoxetine treatment were administered the Life Experiences Survey (LES) at baseline and the Dysfunctional Attitude Scale (DAS) and Hamilton Depression Rating Scale (HDRS) at baseline and endpoint.

Results:The DAS dependency subscale, but not the self-criticism subscale, showed significant correlations with life events regardless of congruency. Baseline HDRS scores were positively correlated with both DAS subscales and total score. The DAS subscales, the total DAS score, and the HDRS all improved significantly with treatment.

Conclusions: These results confirm a growing body of research that has found an association between sociotropic or dependent personality traits and life events.

(Can J Psychiatry 1997;42:291–297)

Key Words: dysfunctional attitudes, dependency, self-criticism, depression, life events

Over the past few decades, psychiatrists and psychologists have conceptualized the relationship between Axis I and Axis II disorders in different ways (1–9). In particular, Beck hypothesized that 2 cognitive-personality traits are important in depression: sociotropy and autonomy (10,11). A sociotropic individual is one who defines his or her self-worth in terms of contact with and approval from others. Highly sociotropic individuals are strongly invested in positive interchanges with other people and are very concerned about the possibility of disapproval or rejection. They often act in ways designed to please others in order to secure their attachments. When experiencing interpersonal loss, sociotropic individuals are likely to present with a specific clinical picture. This includes a sense of deprivation as well as a symptom profile traditionally associated with the concept of anxious depression or reactive depression (12,13).

In contrast, Beck described an autonomous individual as one who is very invested in preserving his or her freedom and independence and achieving meaningful goals. Highly autonomous individuals are extremely concerned about the possibility of personal failure and often act in ways to maximize their control over the environment, thereby reducing the probability of failure. When experiencing personal failure, autonomous individuals tend to present with a different clinical picture compared with sociotropic persons. For example, the former will present with a strong theme of defeat, environmental withdrawal, and a symptom profile traditionally characteristic of endogenous (13) or endogenomorphic depression (14). Beck also suggested that people with sociotropic depression would respond better to a therapeutic approach emphasizing support, whereas depressed autonomous individuals would respond preferentially to a problem-solving approach (10).

In the last decade, a number of studies have tested the construct, criterion, and predictive validity of Beck’s hypotheses about sociotropy and autonomy with mixed results (15–30). Although there is good evidence for several aspects of construct validity for the Sociotropy scale, including relations with depression, in response to specific matching events, and relations to specific depressive symptoms, there is much weaker evidence concerning the Autonomy scale, with only one study suggesting a specific vulnerability to achievement-related events (22). The purpose of our study was to investigate Beck’s hypotheses that different cognitive-personality dimensions are associated with different environmental factors in depression onset and that these same personality subtypes can also predict the symptom profiles of patients once they actually become depressed. Although there are some important differences in conceptualization, Segal and colleagues (31) note that there is a growing convergence of opinion among theorists from differing orientations regarding the association between affiliative or achievement personality traits and depression (10,32–34) and that many different psychological instruments exist to help researchers achieve such classifications of subjects. We therefore elected to examine the related concepts of dependency and self-criticism in a sample of depressed outpatients. Dependency and self-criticism are constructs that partly reflect the sociotropy–autonomy dichotomy and are derived from the DAS (35), an instrument frequently used as a vulnerability measure in cognitive studies of depression (36). We were able to study different psychological variables of a large sample of depressed outpatients who were assessed while medication-free during the acute episode as well as following treatment. No studies to date have examined the relationship between interpersonal and achievement concerns and 2 DSM-IV depressive subtypes that are consistent with Beck’s conceptualization of a “sociotropy depression” and “autonomy depression”: atypical and melancholic depression. Given the widespread acceptance of DSM-IV nosology, studies exploring the relation between these cognitive-personality traits and atypical versus melancholic depression may be of special relevance to the current practising clinician.

We hypothesized that 1) both dependent and self-critical traits would be highly correlated with the number of trait-congruent life events occurring in the previous 12 months, 2) these personality traits would be relatively independent from severity of depression and would remain fairly stable regardless of change in the severity of depression following antidepressant treatment, and 3) dependent patients would be more likely to present with atypical depression compared with self-critical patients, and conversely, self-critical patients would be more likely to present with melancholic depression compared with dependent patients.

Method

Population

The 142 subjects (83 women and 59 men; mean age 40.4 years ± 10.8 years) who participated in this study were drawn from 176 patients consecutively enrolled in and completing the open phase of a treatment study with fluoxetine conducted at the Depression Research Program of the Clinical Psychopharmacology Unit of the Massachusetts General Hospital. Patients ranged in age from 18 to 65 years, suffered from major depressive disorder (MDD) diagnosed with the Structured Clinical Interview for DSM-III-R—Patient Edition (SCID-P) (37), and had an initial 17-item HDRS ³ 16 (38). Thirty-four patients (18 women and 16 men; mean age 40.9 years ± 10.7 years) were excluded from the analyses because they did not correctly complete all questionnaires before and after treatment. There were no significant differences in mean age (t = –0.26, P > 0.05), gender distribution (c² = 0.15, P > 0.05), or baseline HDRS scores (t = 1.6, P > 0.05) between subjects included (n = 142) and excluded (n = 34) from analyses. The mean HDRS score for patients included in analyses was 19.8 ± 3.3.

Exclusion criteria included a history of organic mental disorders, history of seizure disorder, serious or unstable medical illness, substance use disorders (including alcohol) active within the last 12 months, serious suicidal risk, pregnancy, lactation, schizophrenia, delusional disorder, psychotic disorder not elsewhere classified, bipolar disorder, significant antisocial personality disorder, history of multiple adverse drug reactions or allergy to study drugs, nonresponse to an adequate course of antidepressant treatment, concomitant use of birth control pills or nonstudy psychotropic medication, and clinical or laboratory evidence of hypothyroidism. As part of this study, patients were treated with fluoxetine 20 mg/day for 8 weeks. Patients were asked to volunteer to participate in this add-on study, and informed written consent was obtained.

Procedures

1. DAS. All 142 subjects completed the DAS Form A (35) before and after treatment. This self-rating questionnaire consists of 40 items about specific contingencies of self-worth. Beck and colleagues (11) have shown that 2 broad personality groupings can be discerned from the DAS: one based on the need for approval (dependent) and the other related to performance evaluation (self-critical). The individual DAS items used in the selection of these 2 personality groupings were derived from a factor analysis (39) that yielded 2 groups of items which overlap in content with other vulnerability subtypes proposed, such as Blatt’s (34) anaclitic and introjective depressive subtypes and the sociotropic and autonomous subtypes derived from Beck’s Sociotropy Autonomy Scale (SAS) (Beck and others 1983, unpublished observations). The DAS was chosen because its items were constructed from clinical material of depressed patients, it has shown reasonable levels of interrelation with other measures of affiliation and achievement in depression (40,41), and it is frequently used as a vulnerability measure in cognitive studies of depression (36). In addition, the DAS has been shown to identify dependent and self-critical personality traits independent of depressive symptom severity. Finally, the DAS has been shown to exhibit good internal consistency and stability over time, with coefficient alpha ranging from 0.84 to 0.92 and a test–retest correlation over an 8-week period of 0.84 (41). Rather than classify subjects as either dependent or self-critical on the basis of cutoff scores derived from the DAS subscales, we used these 2 traits as continuous variables in our analyses. This approach minimizes the risk of misclassification for subjects with scores near an arbitrary cutoff point.

2. LES and life event classification. The procedure for tracking stressful life events involved the use of the LES (42), a standardized, self-report inventory of recent life events filled out by patients at baseline. The LES is composed of 57 life events that represent a number of general activity categories. Each life event was scored by the patients on a scale from –3 to +3, with –3 representing an extremely negative event and +3 representing an extremely positive event. A rating of 0 would suggest no impact. In order to classify life events in terms of their congruency or matching with either personality trait (dependent or self-critical), all of the items in the LES were examined by one author (AZ) and each life event categorized (a priori) as being primarily related to interpersonal loss (therefore congruent with dependency), primarily related to achievement failure (therefore congruent with self-criticism), or congruent with neither (indeterminate). Of the 57 life events, 25 items were selected as representing interpersonal loss, and 24 items were selected as primarily representing failure concerns. An example of items in the interpersonal loss category was a life event like “death of a close family member.” In the failure category were life events such as “being fired from a job.” Ten items were deemed to be “indeterminate” (not primarily representative of self-critical or dependent concerns). An example of such a life event is pregnancy.

3. Evaluation of depressive subtype. Major depressive episode with melancholic features (melancholia) was diagnosed by using DSM-III-R criteria and the SCID-P (37). Major depression with atypical features (atypical depression) was diagnosed by using the Columbia Atypical Depression Diagnostic Scale (43). Of 142 subjects, 6 met criteria for melancholia alone and 44 met criteria for atypical depression alone.

Data Analyses

Simple linear regressions were used to assess the relationship between interpersonal loss or failure life events and the dependency and self-criticism scales of the DAS. Simple linear regressions were also used to assess the relationship between depression severity and both DAS and LES, as well as between the changes pre- and posttreatment in DAS and HDRS. Two-tailed paired t tests were used to assess the differences between pre- and posttreatment scores on the HDRS and on the dependency and self-criticism scales of the DAS. Two-tailed unpaired t tests were used to test for differences between patients with and without atypical depression or melancholia on the DAS.

Results

Table 1 describes the relationship between depression severity as measured by the HDRS and both DAS and LES. There was a statistically significant positive correlation between depression severity and the individual subscales of the DAS as well as the total DAS score. In addition, the relationship between the degree of negativity of the life events and the HDRS was also statistically significant. Only 2 of these variables are retained in a stepwise multiple regression model: negative life events (LES negative) and self-criticism (DAS self-critical). Each contributed equal weight, and together, both variables accounted for 9.8% of the variance in baseline depression severity (HDRS baseline). As shown in Table 2, the DAS dependency subscale showed statistically significant correlations with both interpersonal loss and failure life events, whereas the DAS self-criticism subscale was not significantly correlated with either life event subtype. It should be noted that these findings were observed even when we controlled for the potentially confounding effects of the severity of depression. The correlations between changes in the HDRS and changes in the DAS are shown in Table 3. Changes in depression severity were positively correlated with change in the individual subscales of the DAS and the total DAS score. Finally, Table 4 shows that the DAS self-critical subscale, the DAS dependency subscale, the total DAS score, and the HDRS all improved significantly with treatment.

In our sample, patients with atypical depression (n = 44) had significantly higher scores on the DAS self-critical subscale than those without atypical depression (n = 80; note that 12 of the subjects had mixed typology and could not be classified) (t = 2.7, P < 0.008), while no difference in DAS dependency scores were observed (t = 1.2, P > 0.05). Patients with melancholia (n = 6) showed no significant difference from patients without melancholia (n = 136) on either the self-critical (t = 0.12, P > 0.05) or dependent (t = –0.66, P > 0.05) subscales of the DAS. There were no significant gender differences in the dependency (t = –0.96, P > 0.05) or self-critical subscales of the DAS (t = 1.3, P > 0.05) or in interpersonal loss life events (t = 0.14, P > 0.05) or failure-related life events (t = 1.3, P > 0.05).

Discussion

In this study, we investigated Beck’s hypothesis (10) regarding potential cognitive vulnerabilities for depression and the corresponding subtypes of depression that would be manifested. We predicted that depressed patients possessing certain character traits or personality styles such as dependency or self-criticism would be more likely to have recently experienced life events that are directly related to these vulnerabilities compared to life events that are unrelated to these vulnerabilities. We also predicted that patients scoring high in terms of dependency traits on the DAS would manifest atypical depressive symptoms, whereas patients scoring high in terms of self-critical traits would manifest melancholic depressive symptoms. Our results only partially validated this congruency hypothesis. Dependent subjects were significantly more likely to have experienced life events within the previous 12 months, but the type of life event was nonspecific. We did, however, detect that the correlation with interpersonal loss life events was much stronger than with the failure life events. This indicates that dependent individuals with depression may be more likely to have experienced affiliative life events compared with achievement-related life events in the previous year. In contrast, we did not detect any relationship between self-critical personality traits and frequency of either type of life event within the previous year.

Although it is commonly believed that women tend to base self-esteem more on affiliation than on achievement (44,45), we did not detect any differences between men and women in scores on the dependency or self-critical subscales of the DAS. One possible explanation for these results is that although “normal” women may be, on average, more affiliative than men, depression per se may increase the levels of pathological dependency in both sexes, thereby eliminating possible gender differences.

Our results confirm the findings of our earlier study (46) that dysfunctional attitudes are significantly correlated with depression severity and that improvement in these measures is significantly associated with reduction in severity of depression. It should be noted, however, that fluoxetine treatment was followed by a much greater decrease (46%) in depressive symptomatology than in self-critical (12%) and dependent (9%) personality traits as measured by the DAS.

Furthermore, we found a statistically significant relationship between depression severity and patients’ subjective ratings of the negativity of their recent life events, but not the actual number of life events. Since the majority of the life events used in the interpersonal loss and failure categories may be construed as negative and/or threatening, these results may suggest that an individual’s interpretation of events has a greater relevance in predicting psychological distress than the “objective” nature of the events alone. This is consistent with Brown, Bifulco, and Harris’s (47) contextual threat model of life events and depression.

In addition, we found that patients with atypical depression had significantly higher scores on the self-critical subscale of the DAS compared with those who displayed nonatypical depression. In our sample, melancholic depression did not appear to be associated with higher or lower levels of either personality trait. This contradicts Peselow and others’ (24) finding that sociotropy was related to nonendogenous depression, whereas autonomy was related to endogenous depression. It also is at variance with Robins and others’ (48) recent prospective study of college students, which reported support for Beck’s symptom-specificity hypothesis for both sociotropy and autonomy. Our findings, however, must be considered tentative because of the extremely small sample size of patients with melancholia. In addition, our study evaluated the presence of atypical and melancholic depressive symptoms, whereas Robins and colleagues evaluated a depressive symptom composite derived from the Beck Depression Inventory.

Our results are at variance with a number of previous investigations of Beck’s congruency hypothesis. As Segal and others (31) note, however, the pattern of matching between life events and personality style has often varied not only across studies but even within studies. For example, 3 studies (22,25,49) indicated a greater magnitude of depression for dependent subjects who experienced interpersonal stress, whereas 2 other studies (21,31) reported a relationship between self-critical subjects who experienced achievement stress and depression. Our findings are more similar to those of Robins and others (48), who reported that both sociotropy and autonomy were associated with increases in depression level with both types of life events.

Population sampling differences may partially explain this discrepancy, since the first 3 studies involved college students, whereas the last 2 studies involved clinical populations. An even more compelling explanation, however, is that these inconsistent results are due to unresolved problems in the scales used to measure dependent and self-critical personality styles. Four assessment measures have been used in the studies previously mentioned: the DAS, the SAS, the Depressive Experiences Questionnaire (DEQ) (50), and the Revised Depressive Experiences Questionnaire (RDEQ) (51). Although all 4 measures show adequate levels of interrelation (40,41,52,53), inconsistencies may have resulted from the lack of standardization in the classification of subjects and in the analysis of data. The recently developed Personal Style Inventory (PSI) (54) is intended to provide greater differentiation between affiliative and achievement personality dimensions than the DAS, DEQ, RDEQ, or the SAS, but an even more attractive strategy to increase homogeneity suggested by Blaney and Kutcher (16) would be to use both self-report and behavioural measures of these personality dimensions and to retain only subjects who satisfied criteria derived from both of these concurrent assessment methods.

One other explanation of our discrepant results is that dependent patients’ personality characteristics may foster a response style which inflates estimates of life stress (perhaps by lowering the threshold for perceiving and acknowledging distress), thereby leading to greater reporting of life events compared with their self-critical counterparts. In addition, as noted by Segal and colleagues (31), achievement and interpersonal events may vary in terms of time of onset and effect in that interpersonal events are experienced as having a more immediate impact, whereas achievement events are experienced as more insidious and cumulative in their effects. Given that our study, like many of studies in this field (18,25,55), employs a cross-sectional design and self-report life history questionnaires, systematic bias accentuating interpersonal events may have occurred. Conducting personal interviews with subjects to gauge the impact of the stressors in their lives would have been a more valid way to gather this type of data; the self-report inventory may not distinguish whether the reported life event represents an objective threat or one that is idiosyncratically perceived as threatening by the depressed subject (21). As Robins and colleagues note (48), another important issue that needs to be considered in this type of research is the real possibility that depression or personality traits themselves may influence the occurrence of specific life events, not just the reaction to them.

In conclusion, these results are consistent with a growing body of research that has found an association between dependent personality traits and life events. Our results also support earlier studies reporting a correlation between depression severity and dysfunctional cognitive style. It should be noted, however, that these conclusions are based on relatively small correlations (maximum 0.33). Whether these findings have general clinical relevance to the practitioner, therefore, is still open to question.

Although dependency itself may not explain a lot of the variance in the entire depressed sample, it may be relevant for a particular subset of the depressed population. These dependent individuals may have a particular diathesis to life stresses and be especially likely to report negative life events. It may therefore be important for the clinician to consider specifically the interactions of these psychosocial factors as part of the dependent patient’s comprehensive treatment.

References

1. Cloninger CR. A systematic method for clinical description and classification of personality variants. Arch Gen Psychiatry 1987;44:573–88.

2. Cloninger CR, Svrakic DM, Pryzbeck TR. A psychobiological model of temperament and character. Arch Gen Psychiatry 1993;50:975–90.

3. Downs NS, Swerdlow NR, Zisook S. The relationship of affective illness and personality disorders in psychiatric outpatients. Ann Clin Psychiatry 1992;4:87–94.

4. Joffe RT, Regan JJ. Personality and response to tricyclic antidepressants in depressed patients. J Nerv Ment Dis 1989;177:745–9.

5. Joyce PR. Predictors for treatment response and treatment selection. Current Opinion in Psychiatry 1994;7:26–9.

6. Joyce PR, Mulder RT, Cloninger CR. Temperament predicts clomipramine and desipramine response in major depression. J Affect Disord 1994;30:35–46.

7. Shea MT, Widiger TA, Klein MH. Comorbidity of personality disorders and depression: implications for treatment. J Consult Clin Psychol 1993;60:857–68.

8. Svrarkic DM, Whitehead C, Pryzbeck TR, Cloninger CR. Differential diagnosis of personality disorders by the seven factor model of temperament and character. Arch Gen Psychiatry 1993;50:991–9.

9. Tyrer P, Seivewright N, Ferguson B, Murphy S, Johnson AL. The Nottingham study of neurotic disorder: effect of personality status on response to drug treatment, cognitive therapy and self-help over two years. Br J Psychiatry 1993;162:219–26.

10. Beck AT. Cognitive therapy of depression: new perspectives. In: Clayton PJ, Barrett JE, editors. Treatment of depression: old controversies and new approaches. New York: Raven Press; 1983. p 265–90.

11. Beck AT, Epstein N, Harrison R. Cognitions, attitudes and personality dimensions in depression. British Journal of Cognitive Psychotherapy 1983;162:118–21.

12. Gillespie RD. The clinical differentiation of types of depression. Guys Hospital Reports 1929;9:306–44.

13. Kiloh LG., Garside RF. The independence of neurotic depression and endogenous depression. Br J Psychiatry 1963;109:451–63.

14. Klein DF. Endogenomorphic depression: a conceptual and terminological revision. Arch Gen Psychiatry 1974;31:447–54.

15. Bagby RM, Cox BJ, Schuller D, Levitt AJ, Swinson RP, Joffe RT. Diagnostic specificity of the dependent and self-critical personality dimensions in major depression. J Affect Disord 1992;26:59–64.

16. Blaney PH, Kutcher GS. Measures of depressive dimensions: are they interchangeable? J Pers Assess 1991;56:502–12.

17. Cappeliez P. The relationship between Beck’s concepts of sociotropy and autonomy and the NEO-Personality Inventory. Br J Clin Psychol 1993;32:78–80.

18. Clark DA, Beck AT, Brown GK. Sociotropy, autonomy and life event perceptions in dysphoric and nondysphoric individuals. Cognitive Therapy and Research 1992;16:635–52.

19. Franchel RL, Robson K. Self-criticism and interpersonal dependency as vulnerability factors to depression. Cognitive Therapy and Research 1992;16:419–35.

20. Gilbert P, Reynolds S. The relationship between the Eysenck Personality Questionnaire and Beck’s concept of sociotropy and autonomy. Br J Clin Psychol 1990;29:319–25.

21. Hammen C, Ellicott A, Gitlin M, Jamison KR. Sociotropy/autonomy and vulnerability to specific life events in patients with unipolar depression and bipolar disorders. J Abnorm Psychol 1989;98:154-160.

22. Hammen C, Marks T, Mayol A, de Mayo R. Depressive self-schemas, life stress, and vulnerability to depression. J Abnorm Psychol 1985;94:308–19.

23. Moore RG, Blackburn IM. Sociotropy, autonomy and personal memories in depression. Br J Clin Psychol 1993;32:460–2.

24. Peselow ED, Robins CJ, Sanfilipio MP, Block P, Fieve RR. Sociotropy and autonomy: relationship to antidepressant drug treatment response and endogenous-nonendogenous dichotomy. J Abnorm Psychol 1992;101:479–86.

25. Robins CJ, Block P. Personal vulnerability, life events and depressive symptoms: a test of a specific interactional model. J Pers Soc Psychol 1988;54:847–52.

26. Robins CJ, Luten AG. Sociotropy and autonomy: differential patterns of symptom presentation in unipolar depression. J Abnorm Psychol 1991;100:74–7.

27. Robins CJ, Block P, Peselow ED. Cognition and life events in major depression: a test of the mediation and interaction hypotheses. Cognitive Therapy and Research 1990;14:299–313.

28. Robins CJ, Block P, Peselow ED. Endogenous and nonendogenous depressions: relations to life events, dysfunctional attitudes and event perceptions. Br J Clin Psychol 1990;29:201–7.

29. Robins CJ, Block P, Peselow ED. Relations of sociotropic and autonomous personality characteristics to specific symptoms in depressed patients. J Abnorm Psychol 1989;98:86–8.

30. Sahin N, Ulusoy M, Sahin N. Exploring the sociotropy-autonomy dimensions in a sample of Turkish psychiatric inpatients. J Clin Psychol 1993;49:751–63.

31. Segal ZV, Shaw BF, Vella DD, Kratz R. Cognitive and life stress predictors of relapse in remitted unipolar depressed patients: test of the congruency hypothesis. J Abnorm Psychol 1992;101:26–36.

32. Arieti S, Bemporad J. The psychological organization of depression. Am J Psychiatry 1980;137:1360–5.

33. Bowlby J. The making and breaking of affectional bonds, I: aetiology and psychopathology in light of attachment theory. Br J Psychiatry 1977;130:201–10.

34. Blatt SJ. Levels of object representation in anaclitic and introjective depression. Psychoanal Study Child 1974;29:107–58.

35. Weissman AN, Beck AT. Development and validation of the Dysfunctional Attitudes Scale: a preliminary investigation. In: Proceedings of the meeting of the American Educational Research Association; 1978; Toronto, ON.

36. Williams JMG, Healy D, Teasedale JD, and others. Dysfunctional attitudes and vulnerability to persistent depression. Psychol Med 1990;20:375–81.

37. Spitzer RL, Williams JBW, Gibbon M, First MB. User’s guide for the Structured Clinical Interview for DSM-III-R (SCID). Washington (DC): American Psychiatric Press; 1990.

38. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960;23:56–61.

39. Cane DB, Olinger LJ, Gotlib IH, Kuiper NA. Factor structure of the dysfunctional attitudes scale in a student population. J Clin Psychol 1986;42:307–9.

40. Barnett PA, Gotlib IH. Psychosocial functioning and depression: distinguishing among antecedents, concomitants and consequences. Psychol Bull 1988; 104:97–126.

41. Cane DB, Gotlib IH. Assessing dependent and self-critical dimensions of depression: a preliminary inquiry. Vancouver: Canadian Psychological Association; 1987.

42. Sarason IG, Johnson JH, Siegel JM. Assessing the impact of life changes: development of the Life Experiences Survey. J Consult Clin Psychol 1978;46:932–46.

43. Stewart J, McGrath P, Quitkin F, and others. Relevance of DSM-III depressive subtype and chronicity of antidepressant efficacy in atypical depression. Arch Gen Psychiatry 1991;48:319–23.

44. Stiver IP. The meanings of “dependency” in male-female relationships. In: Jordan JV, Kaplan JAG, Miller JB, Stiver IP, Surrey JL, editors. Women’s growth in connection. New York: Guilford; 1991. p 143–61.

45. Wong MM, Csikszentmihalyi M. Affiliation motivation and daily experiences: some issues on gender differences. Journal of Personality and Social Psychology 1991;60:154–64.

46. Fava M, Bless E, Otto MW, Pava JA, Rosenbaum JF. Dysfunctional attitudes in major depression: changes with pharmacotherapy. J Nerv Ment Dis 1994;182:45–9.

47. Brown GW, Bifulco A, Harris T.O. Life events vulnerability and onset of depression: some refinements. Br J Psychiatry 1987;150:30–42.

48. Robins CJ, Hayes AM, Block P, and others. Interpersonal and achievement concerns and the depressive vulnerability and symptom specificity hypotheses: a prospective study. Cognitive Therapy and Research 1995;19:1–20.

49. Zuroff DC, Mongrain M. Dependency and self-criticism: vulnerability factors for depressive affective states. J Abnorm Psychol 1987;96:14–22.

50. Blatt SJ, D’Afflitti JP, Quinlan DM. Experiences of depression in normal young adults. J Abnorm Psychol 1976;85:383–9.

51. Welkowitz J, Lish JD, Bond RN. The Depressive Experiences Questionnaire: revision and validation. J Pers Assess 1985;49:89–94.

52. Klein DN. The Depressive Experiences Questionnaire: a further evaluation. J Pers Assess 1989;53:705–15.

53. Riley WT, McCranie EW. The Depressive Experiences Questionnaire: validity and psychological correlates in a clinical sample. J Pers Assess 1990;54:523–33.

54. Robins CJ, Ladd JS, Welkowitz J, and others. The Personal Style Inventory: preliminary validation studies of new measures of sociotropy and autonomy. Journal of Psychopathology and Behavioral Assessment 1994;16:277–300.

55. Robins CJ. Congruence of personality and life events in depression. J Abnorm Psychol 1990;99:393–7.

Résumé

Objectif : Déterminer si les traits de personnalité de dépendance et d’autocritique sont liés à des types spécifiques d’événements de la vie et si ces traits sont modifiés par la pharmacothérapie.

Méthode : Dans l’ensemble, 142 patients externes déprimés ont suivi un traitement à la fluoxétine pendant 8 semaines.  Au départ, ils ont été évalués grâce à l’échelle d’expériences vécues (Life Experiences Survey [LES]) et, au départ et à la fin, grâce à l’échelle des attitudes dysfonctionnelles (Dysfunctional Attitude Scale [DAS]) et à l’échelle d’évaluation de la dépression de Hamilton (Hamilton Depression Rating Scale [HDRS]).

Résultats : La sous-échelle de dépendance de la DAS, mais non pas la sous-échelle d’autocritique, a révélé des corrélations significatives avec les événements de la vie, sans égard à la congruence.  Les scores de départ de la HDRS étaient corrélés négativement avec les sous-échelles de la DAS et le score total.  Le traitement a permis l’amélioration significative des sous-échelles de la DAS, du score total de la DAS et de la HDRS.

Conclusions : Ces résultats confirment un nombre croissant de travaux de recherche où l’on a constaté un lien entre les traits de personnalité sociotrope ou dépendant et les événements de la vie.

Manuscript received July 1996, revised January 1997.

¹Assistant Professor, Department of Psychiatry, University of Toronto, Toronto, Ontario.

²Head, Cognitive Behaviour Therapy Clinic, Mount Sinai Hospital, Toronto, Ontario.

³Assistant Professor of Psychiatry, Harvard Medical School, Cambridge, Massachusetts.

⁴Director, Depression Clinical and Research Program, Massachusetts General Hospital, Boston, Massachusetts.

⁵Depression Clinical and Research Program, Massachusetts General Hospital, Boston, Massachusetts.

⁶Associate Professor of Psychiatry, Harvard Medical School, Cambridge, Massachusetts.

Address for correspondence: Dr AE Zaretsky, Head, Cognitive Behaviour Therapy Clinic, Department of Psychiatry, Mount Sinai Hospital, Suite 941A, 600 University Avenue, Toronto, ON M5G 1X5

Can J Psychiatry, Vol 42, April 1997