Psychiatric Annals


The Role of the Family in Major Depressive Illness

Gabor I Keitner, MD; Ivan W Miller, PhD; Christine E Ryan, PhD


To neglect the social environment in which major psychiatric illness in general, and affective disorders in particular, exists, limits therapeutic effectiveness and likely prolongs the illness.


To neglect the social environment in which major psychiatric illness in general, and affective disorders in particular, exists, limits therapeutic effectiveness and likely prolongs the illness.

Hsychopathology exists within a social context. Most patients, especially those with affective disorders, live with their families. Not surprisingly, patients' illnesses exert a major influence on significant others in their social field. Conversely, the ways in which these significant others respond to patients' illnesses has a measurable and clinically significant impact on the symptomatology and course of those illnesses. It is increasingly evident that to neglect the social environment in which major psychiatric illness in general, and affective disorders in particular, exists, limits therapeutic effectiveness and likely prolongs the illness.

Given the reality of managed care and increasing pressure for efficient and rapid treatment programs, it is important to explore a variety of new approaches that may facilitate recovery. The availability of effective psyehopharmacologic agents is a significant step in that direction. Recent descriptions and empirical investigations of short-term psychotherapies, such as interpersonal psychotherapy and cognitive behavioral therapies, further enhance our abilities to provide effective, efficient, and relatively rapid treatments. Nonetheless, pharmacotherapy and psychotherapy do not deal directly with the patient's psychosocial environment. Some type of family intervention may also be needed. The purpose of this article is to review current findings regarding the importance of the role of the family in major depressive illness and to highlight its treatment implications.


Studies over the past 10 years have consistently shown that an episode of major depression is associated with significant family dysfunction.1 This is true for both inpatients and outpatients. Depressed inpatients were found to have significantly more marital maladjustment than a diagnostically mixed inpatient comparison group.2,3 Depression may also have a greater impact on marital life than does rheumatoid arthritis or cardiac illness.4

A wide range of family functions are disturbed during the acute episode of major depression. Communications, problem solving, and role functioning within the family are especially likely to be dysfunctional. Communication refers to the effectiveness, extent, clarity, and directness of information exchange in the family. Problem solving reflects the family's ability to resolve instrumental and affective problems. Roles reflect the efficacy with which family tasks are allocated and accomplished. Tasks include those associated with the provision of resources, nurturance and support, the development of life skills, and management of the family system.


Comorbidity is a complicating factor in assessing and dealing with the social pressures of depressed patients. Major depression tends to coexist with other illnesses. In our own research we have referred to major depression that coexists with any other Axis I, II, or ?? disorder as compound depression.5 Comorbidity has major clinical and research implications. We found, for instance, that families of patients with compound depression6 reported worse family functioning, especially in their ability to solve problems and in maintaining acceptable standards of behavior, both during the acute depressive episode and at six-month follow-up, than patients with pure major depression. The presence of comorbidity makes it even more important for the clinician to be alert to the impact of the illnesses on family members.

An additional major issue is the recognition of the burden that depressive illness places on family members. It is well recognized that living with a depressed person has significant impact on others in the immediate social environment.7"9 The negative symptoms of depression, such as worrying, fatigue, hopelessness, anxiety, anergia, and lack of interest in social life, are the most upsetting to relatives as they appear to be on a continuum with normal behavior and seemingly more under the patient's control. On the other hand, florid symptoms are more easily attributed to the illness and, therefore, seem less upsetting to family members.


It is important to consider whether the family dysfunction noted during the acute episode is a persistent feature of the depressive illness or whether it is related to the stress that accompanies the acuteness of the onset of a depressive episode. The treatment implications are clearly different if one has to deal only with a difficult adjustment to an unusual set of circumstances, or if there is some underlying persistent family pathology that needs to be more systematically explored and treated.

In general, the literature suggests that although there is improvement in family functioning as the depressive episode remits, families of patients with major depression still evidence more problematic family functioning at remission of the depression than do control families.10-14

The above findings suggest, but do not prove, that family dysfunction appearing during the acute episode is not just a reaction to the acute stress brought about by the intensity of the illness. The clinical implication of these findings is that depressed patients and their familles need continued monitoring, support, and intervention even after the acute episode has subsided.


Few studies have addressed the important clinical question of the relationship between family functioning and the course of the depressive illness. It is important to find out to what extent the social environment affects the course of the depression in order to determine the kind of therapeutic interventions that are most effective in bringing about a rapid resolution of the depression.

Our research group14 found that patients whose general family functioning improved over the course of the depressive illness had a significantly shorter time to recovery (4.1 months) than patients of families who did not improve (8.1 months) in their general functioning. Positive changes in overall family functioning during the course of the illness were associated with faster recovery times.

In a study of life stressors, social resources, and the four-year course of depression, Swindle et al15 noted that life stressors remained stable or changed very little Over the course of the four-year follow-up period as did medical conditions and the negative home environment. They noted that pre-intake medical conditions and family conflicts consistently predicted poorer long-term outcomes. Persisting family conflicts had an enduring negative effect on symptoms.

A growing number of studies have documented the significant impact of family functioning and social support on the outcome of a major depressive episode. Social support and family functioning are complementary concepts. Family functioning is one type of social support and perhaps one of the most important ones, if family is defined as those living together in the same home. The quality of interpersonal relationships is an important component of social support.16 These studies consistently highlight the importance of taking into consideration the family and social environment of patients in order to maximize treatment effectiveness and increase the probability of a positive outcome.

George et al16 found that both the size of the social network and subjective social support were significant predictors of depressive symptoms at six- to 32month follow-up for 150 middle-aged and elderly adults with a diagnosis of major depression. Their study supports the hypothesis that social support affects the outcome of depressive illness.

In a one-year follow-up of 265 unipolar depressed outpatients, Moos17 found that pre-intake medical conditions, family conflict, and lack of family support predicted a poorer treatment outcome. Conversely, patients who had a close confidante and less family conflict showed better outcome with brief therapy, while patients who lacked a confidante and had more family conflicts needed longer treatment to reach a positive outcome. Moos17 also noted that post-treatment stressors and lack of social resources were associated with poorer treatment outcome. This study also emphasized the importance of dealing with family dysfunction and social disorganization in order to resolve the depressive episode.

In a one-year follow-up study of untreated major depression in the community, Sargeant et al18 found that, for women in particular, an unstable marital history was associated with poor outcome. Goering et al19 also noted that at six-month follow-up only 51% of 47 depressed married women had recovered. Recoverywas predicted by the depressed woman's rating of her current marital relationship and by her husband's rating of the premorbid relationship. The strongest predictor of post-hospital symptom course was a depressed woman's perception of the quality of the support she received from her spouse. Other family factors related to recovery were the depressed woman's level of criticism toward her spouse, and her satisfaction with his communication, affection, and relationship to their children.

Goering et al19 also found that patients' perceptions, for the most part, were accurate appraisals of marital support. She noted that women who had been married for more than seven years and had unsupportive marriages were least likely to recover. She suggested that the depressed woman's perception of marital support may be a better prognostic indicator of symptom course than the clinical characteristics of the depressive illness.

We attempted to further refine our understanding of which clinical and psychosocial factors contributed to the likelihood of recovery in 78 inpatients with major depression, who were followed for a 12-month period.20 By the 12th month following hospitalization, 48.6% of the patients had recovered from their depression. Logistic regression analyses identified five risk factors that were associated with nonrecovery: longer inpatient hospital stay; earlier age of depression onset; poor family functioning; greater number of previous hospitalizations; and comorbidity. Of these five risk factors associated with nonrecovery, only two appear to be amenable to clinical intervention -comorbidity and poor family functioning. This finding highlights the important role of family functioning in the likelihood of recovery from depression and is a powerful argument for including family interventions as part of the treatment plan.

Finally, in assessing speed of recovery for major depression, McLeod21 followed a sample of 119 depressed, married subjects for a 12-month period. Significant predictors of recovery in this study included comorbidity, social support, age, and education. Interestingly, social support in the form of spouse's positive responses to the depression predicted rapid recovery, whereas the perception that friendships were eonflietual predicted slow recovery. In particular, spouse's negative reactions to the depression became particularly important predictors of recovery as the episode became longer. Social support was not only associated with faster recovery, but was found to have particularly significant meaning for the depressed patient at different phases of the illness.


Persisting family dysfunction, particularly in the form of high levels of criticism, appears to be strongly related to the likelihood of relapse. A number of studies22"24 have shown that depressed patients whose family members had high levels of criticism and hostility toward the patient were much more likely to relapse during a follow-up period. Depressed patients also tend to relapse at lower levels of criticism than do schizophrenic patients. Marital distress, in combination with perceived criticism, is strongly related to a tendency to relapse.


We do not know the causal sequence between depression and problematic family functioning. Does problematic family functioning predispose to or facilitate the emergence of depressive illness, or does the illness and its impact on patients' interpersonal styles create family difficulties in coping? Barnett and Gotlib,25 in attempting to distinguish among antecedents, concomitants, and consequences of psychosocial functioning and depression, concluded that research provides considerable support for the notion that disturbances in intimate relationships are both antecedents and consequences of depression.

In an attempt to further explore this question, we26 divided the families of depressed patients into "functional* and "dysfunctional" groups. Families were classified as dysfunctional if both the family members and an external rater using standardized subjective (Family Assessment Device27) and objective (McMaster Clinical Rating Scale28) assessment instruments agreed that the family had significant problems. Families who did not meet these criteria were classified as functional. We assessed how clinical and psychosocial characteristics of depressed patients from dysfunctional families compared with those from functional families at the acute episode and at one-year follow-up.

Apart from the patients' neuroticism, a dysfunctional family was not significantly associated with severity of illness or other parameters of the patients' depression. Depressed patients with dysfunctional families did not differ significantly from depressed patients with functional families in severity of depression, previous depressive history, depression subtype, other nonaffective psychiatric diagnoses, or neuroendocrine functioning. The presence or absence of a dysfunctional family appeared to be independent of the characteristics of the patients' depressive disorder.

Depressed patients who were living with dysfunctional families did have a significantly poorer course of illness over a 12-month follow-up, with higher levels of depression, a lower percentage of recovery, and poorer overall functioning. This study provided further support for the recognition that the presence of a dysfunctional family is an important prognostic factor for depressed patients. Findings also suggested that the social environment in which the depression evolves, independent of the depressive symptoms and the severity of depression, has a significant modifying effect on the evolution ofthat depressive episode.

There is no reason, however, to assume a single linear relationship between depression and family functioning. A more useful model recognizes the mutually reinforcing pattern of interactions between patient vulnerability and the family's way of coping with the patient's illness.

According to this model, a patient with vulnerability to major depression develops a depressive episode. The depressive episode may have been precipitated by a variety of factors including genetic predisposition, early life experiences, personality variables, current life events, or severe family conflicts. Regardless of the etiology of the episode, the patient's family and/or significant others are put in a position of having to respond to and deal with the depression. If the family and/or social support system is able to respond effectively, the depressive illness may last a relatively short time and may remit more readily.

Conversely, if the family is unable to respond adequately to the patient's affective illness because of its own difficulties, then the evidence suggests that the illness will be more prolonged, the patient less likely to recover, and more likely to relapse into subsequent episodes. The family's competence to respond to the illness is also influenced by a variety of factors including its socioeconomic level, family composition, its overall functioning, and current family stresses. Patient vulnerability and family competence are seen as mutually reinforcing forces that may act either to further the depressive illness or provide a way of lessening its impact and tendency to recur. It is important to emphasize that the factors affecting either patient vulnerability or family competence can be constructive and supportive or destructive leading to further dysfunction.


To date, there have been very few controlled studies of family interventions for patients with major depression. While a number of such studies are currently in progress, results are still years away.

McLean et al29 used a behavioral approach for the treatment of depressed patients and their spouses with particular emphasis on behavioral contracts and modification of verbal interactional styles. The group receiving the behavioral marital treatment showed a more significant reduction of problematic behaviors, depressive symptoms, and negative actions and reactions than did a group that received standard community treatment.

O'Leary and Beach30 compared the effectiveness of marital therapy with cognitive therapy and a waiting-list control condition for depressed women with marital dysfunction. Both active treatments were effective in alleviating symptoms of the depression, but only the marital therapy improved both the depression and the marital dissatisfaction.

Another study also compared cognitive behavioral therapy with behavioral marital therapy and the combination of the two for depressed women outpatients.31 This study found that behavioral marital therapy was helpful for maritally distressed depressed women but had little impact on those depressed women who had satisfying marriages.

In one of the few controlled studies comparing the effectiveness of antidepressant drugs or marital therapy as well as the combination of the two in depression, Friedman32 found that the combination of the drug and marital therapy was more effective than either treatment administered separately. Drug therapy was faster and more effective in relieving symptoms and clinically improving the depression, while marital therapy was more effective in improving performance and perception of the marital relationship.

One other study has compared the effectiveness of adding psychoeducational family therapy during hospitalization to standard pharmacotherapy for patients with schizophrenia and affective disorders.33-35 At discharge from hospital, this research group found a positive effect for family intervention, but only for female patients with affective disorders and their families. By the 18-month follow-up, the therapeutic effect was restricted to female patients with schizophrenia or "major affective disorders" while the effect on male patients was minimal or "slightly negative."

Our research group is in the process of assessing the effectiveness of adding family therapy and/or cognitive behavioral psychotherapy to standard pharmacotherapy for patients with major depression. We are interested in the question of which patients, with what kind of deficits, can most benefit from which combination of treatments. Increasingly, the critical question from a clinical point of view is not whether family therapy or psychotherapy or pharmacotherapy is most effective, but rather how we can determine which combination of the above treatments is most effective for which particular patient at which point in time.

Many therapists and family theoreticians see significant differences between family therapy and marital therapy. Our approach is to meet with all of the people living together in the same home and/or with other significant people in the patient's social field. Often this means a meeting with only one other person; many times we meet with larger groups of involved family members and close friends. Our treatment principles and approaches are very similar in both situations.

The treatment model that we use is the Problem-Centered Systems Therapy of the Family.36 This is a short-term (average eight to 12 sessions) treatment that requires an active role by both the family and the therapist. The basic principles of the treatment provide an outline of the approach used.

We emphasize major stages of treatment that need to be followed sequentially rather than the specific interventions and styles of a therapist. There is active collaboration with the family and a focus on the family's responsibility for change. We deal with current problems in an open, direct, and clear fashion. A proper assessment of multiple dimensions of family functioning precedes treatment. During treatment, we monitor progress through the expectation of behavioral change by family members. Finally, we try as much as possible to identify and build on the strengths of each family rather than dissecting their weaknesses.


The evidence to date seems convincing that the social environment in which a depressive episode evolves has a significant impact on the onset, duration, course, outcome, relapse potential, and response to treatment of the depressive illness.

In order to make clinical sense of empirical findings and in an attempt to guide our thinking about therapeutic options, we conceptualize an interactive system, as outlined above, which attempts to integrate knowledge about a patient's vulnerability for major depression and the family's competence in dealing with the depression.

This model of patient vulnerability and family competence requires that both the patient and the family be thoroughly assessed and treatment interventions be provided to deal with areas of difficulty in all parts of the system. The patient's vulnerability can be minimized with the aid of individual psychotherapy and psychopharmacological agents. The family's competence can be strengthened through the use of family intervention techniques, including psychoeducational groups, family therapy, and self-help support programs.

Not all depressed patients and their families require family therapy. A significant proportion do not. Current evidence, nonetheless, suggests that it is important, at the minimum, to assess both patients and their social support systems, particularly their family, in order to be more responsive to the broad range of needs of depressed patients. For some patients, a family assessment will be sufficient to rule out the need for further intervention and to identify and reinforce strengths. For other patients, a competent family assessment may provide all the help the family needs to adjust to the illness. The remaining patients and their families are likely to benefit from ongoing family intervention. This family intervention, however, should be provided as an integrated part of a total treatment program. The combination of pharmacotherapy, psychotherapy, and family therapy as indicated represents current state-of-the-art treatment for major depression.


1. Keitner GI, Miller IW. Family functioning and major depression: an overview. Am J Psychiatry. 1990; 147:1128-1137.

2. Crowther JH. The relationship between depression and marital maladjustment: a descriptive study. J Nerv Mental Dis. 1985; 173:227-231.

3. Miller IW, Kabacoff RI, Keitner GI, Epstein NB, Bishop DS. Family functioning in the families of psychiatric patients. Compr Psychiatry. 1986; 27:302-312.

4. Bouras N, Vanger P, Bridges PK. Marital problems in chronically depressed and physically ill patients and their spouses. Compr Psychiatiy. 1986;27:127-130.

5. Keitner GI, Ryan CE, Miller IW, Kohn R, Epstein NB. 12-Month outcome of patients with major depression and comorbid psychiatric or medical illness (compound depression). Am J Psychiatry. 1991; 148:345-350.

6. Keitner GI, Miller IW, Ryan CE, Epstein NB, Bishop DS. Compounded depression and family functioning during the acute episode and at 6-month follow-up. Compr Psychiatry. 1989; 30:512-521.

7. Coyne JC, Kessler RC, Tal M, Turnbull J, Wortman CB, Greden J. Living with a depressed person. J Consult Clin Psychol. 1987; 55:347-352.

8. Jacob M1 Frank E, Kupfer D, Carpenter L. Recurrent depression: an assessment of family burden and family attitudes. J Clin Psychiatry. 1987; 48:395-400.

9. Fadden G, Bebbington P, Kuipers L. Caring and its burdens: a study of the spouses of depressed patients. Br J Psychiatry. 1987; 151:660-667.

10. Merikangas RK, Prusoff BA, Kupfer DJ, Frank E. Marital adjustment in major depression, J Affect Disord. 1985;9:5-11.

11. Hinchliffe M, Vaughan PW, Hooper D, Roberts FJ. The melancholy marriage: an inquiry into the interaction of depression; 50: expressiveness. Br J Med Psychol. 1977;50:125-142.

12. Billings AG, Moos RH. Psychosocial processes of remission in unipolar depression: comparing depressed patients with matched community controls. J Consul Clin Psychol. 1985; 53:314-325.

13. Krantz SE, Moos RH. Functioning and life context among spouses of remitted and nonremi tted depressed patients. J Consult Clin Psychol. 1987; 55:353-360.

14. Keitner GI, Miller TW, Epstein NB, Bishop DS, Fruzzetti AE. Family functioning and the course of major depression. Compr Psychiatry. 1987; 28:54-64.

15. Swindle RW, Cronkite RC, Moos RH. Life stressors, social resources, coping and the 4-year course of unipolar depression. J Abnorm Psychol. 1989; 98:468-477.

16. George LK, Blazer DG, Hughes DC, Fowler N. Social support and the outcome of major depression. BrJ Psychiatry. 1989; 154:478-485.

17. Moos RF. Depressed outpatients' life contexts, amount of treatment, and treatment outcome. J Nerv Ment Dis. 1990; 178:105-112.

18. Sargeant JK, Bruce ML, Florio LP, Weissman MM. Factors associated with 1-year outcome of major depression in the community. Arch Gen Psychiatry. 1990; 47:519-526.

19. Goering PN, Lancee WJ, Freeman SJJ. Marital support and recovery from depression. Br J Psychiatry. 1992; 160:76-82.

20. Keitner GI, Ryan CE, Miller IW, Norman WH. Recovery and major depression: factors associated with twelve-month outcome. Am J Psychiatry. 1992; 149:93-99.

21. McLeod JD, Kessler RC, Landis KR. Speed of recovery from major depressive episodes in a community sample of married men and women. J Abnorm Psychol. 1992; 101:277-286.

22. Vaughn CE, Leff JP. The measurement of expressed emotion in the families of psychiatric patients. Br J Soc Clin Psychol. 1976; 15:157-165.

23. Hooley JM, Orley J, Teasdale JD. Levels of expressed emotion and relapse in depressed patients. Br J Psychiatry. 1986; 148:642-647.

24. Hooley JM, Teasdale JD. Predictors of relapse in unipolar depressives: expressed emotion, marital distress and perceived criticism. J Abnorm Psychol. 1989; 98:229-235.

25. Barnett AP, Gotlib EH. Psychosocial functioning and depression: distinguishing among antecedents, concomitants, and consequences. Psychol Bull. 1988; 104:97-126.

26. Miller TW, Keitner GI, Wishman MA, Ryan CE, Epstein NB, Bishop DS. Depressed patients with dysfunctional families: description and course of illness. J Abnorm Psychol. In press.

27. Miller TW, Epstein NB, Bishop DS1 Keitner GI. The McMaster Family Assessment Device: reliability and validity. Journal of Marital and Family Therapy. 1985; 11:345-356.

28. Miller IW, Kabacoff RI, Epstein NB, Bishop DS, Keitner GI. The development of a clinical rating scale for the McMaster Model of Family Functioning. Fam Process. In press.

29. McLean PD, Ogston K, Graves L. A behavioral approach to the treatment of depression. J Behau Ther Exp Psychiatry. 1973; 4:323-330.

30. O'Leary KD, Beach RH. Marital therapy: a viable treatment for depression and marital discord. Am J Psychiatry. 1990; 147:183-186.

31. Jacobson NS, Dobson K, Fruzzetti AE, Schmaling KB, Salusky S. Marital therapy as a treatment for depression. J Consult Clin Psychol. 1991; 59:547-557.

32. Friedman AS. Interaction of drug therapy with marital therapy in depressive patients. Arch Gen Psychiatry. 1975; 32:619-637.

33. Haas G, Gliele I, Clarkin J, et al. Inpatient family intervention: a randomized clinical trial; II: results at hospital discharge. Arch Gen Psychiatry. 1988; 45:217-224.

34. Spencer J, Glick I, Haas G. A randomized clinical trial of inpatient family intervention; III: effects at 6-month and 18-month follow-ups. Am J Psychiatry. 1988; 145:1115-1121.

35. Glick A), Clarkin JF, Haas GL, Spencer JH Jr, Chen CL. A randomized clinical trail of inpatient family intervention; VI: mediating variables and outcome. Fam Process. 1991; 30:85-99.

36. Epstein NB, Bishop DS. Problem-centered systems therapy of the family. In: Gurman AS, Kniskern DP, eds. Handbook of Family Therapy. New York, NY: Brunner/Mazel; 1981:444-482.


Sign up to receive

Journal E-contents