Relationship difficulties and agoraphobia

Relationship difficulties and agoraphobia

Ckntcal Pcyrholo~-yRevmo. Vol. 5. pp. 581-595, Printed in the USA. All rights reserved. 1985 027%7358185 $3.00 + .oo Copyright 0 1985 Pergamon Press...

1MB Sizes 3 Downloads 12 Views

Ckntcal Pcyrholo~-yRevmo. Vol. 5. pp. 581-595, Printed in the USA. All rights reserved.


027%7358185 $3.00 + .oo Copyright 0 1985 Pergamon Press Ltd.


Universily; Kings ton, Ontario


Problems of an interpersonal nature such as unassertiveness and dependence are considered as they affect features of the relationships of agoraphobics, such as spousal satisfaction and sexual functioning. Theoretical perspectives are reviewed followed by an appraisal of the evidence bearing on these issues. It is concluded that the incidence and nature of relationship problems in agoraphobia remains unclear, although such problems are clearly absent in the development and maintenance of this disorder in some patients. The relevance of these issues for treatment is similarly unclear, although the evidence indicates that involving partners in therapy andlor employing procedures which deal with relationship problems, appear to enhance treatment effectiveness.

Clinical disorders are complex phenomena usually involving several areas of dysfunction. Diagnoses, however, focus on the most prominent features, sometimes ignoring other important correlates of the disorder. Moreover, the labels applied to the clinical phenomena have a profound effect on methods applied to their investigation and treatment. Lately, it has been argued that agoraphobia may not be a phobic disorder, if the term phobia is used in the sense of fear attached to a discrete set of cues (Hallam, 1978). This doubt rests on the fact that agoraphobia is rarely an isolated set of fears which exist in an otherwise well-functioning individual (Chambless, 1982). In this vein, several theoretical analyses have hypothesized that understanding the agoraphobic’s interpersonal relationships may be of critical importance in understanding the disorder and its treatment. Emmelkamp (1980) distinguishes two types of interpersonal problems which may be experienced by agoraphobics: unassertiveness and dissatisfaction with the spouse (partner). While this distinction has advantages for research, in practice the two appear as interdependent features of relationship difficulties. Similarly, other authors (e.g., Andrews, 1966; Shafar, 1976) have stressed the importance of a Requests for reprints Queen’s


should Kingston,

be addressed Ontario,

to Dr. W. Marshall,

Canada, 581




of Psychology,

dependent interpersonal style in the deveiopment and, particularly, the maintenance of agoraphobia. Again this seems to be another feature of relationship problems and may be understood as an alternative way of describing unassertiveness. In the present paper we will refer to these problems as “interpersonal” or “relationship” difficulties. We prefer to discuss relationships generally rather than simply marriages because whether or not the relationships of agoraphobics have legal status, seems to be irrelevant to the issue. THEO~CAL


Psychoanalytic writers (e.g., Ruddick, 1961) stress what they see as the “extremely dependent, orally fixated, (and) clinging character structure” of agoraphobics. In this view, agoraphobia is said to be an adult version of “separation anxiety,” that is, anxiety that the infant feels at the absence of the mother. The agoraphobic adult is thought to regress emotionally to infantile dependence, and their partner in a relationship is said to substitute for the mother and allay this anxiety and dependency (Dixon, Monchaux, SCSanders, 1957). The phobic person is considered to be extremely dependent upon the spouse, who, in turn, is fortified by such dependence f Fry, 1962). The relationship is therefore, seen as complementary: one is endlessly supportive, while this supportive partner is understood to lack self-confidence, a deficiency which is diminished by hisiher role in the relationship (Mittfeman, 1956). Behavioral perspectives have been expanded in recent years to etnbrace similar features of interpersonal functioning, although these theorists have not abandoned entirely the view that classical and operant conditioning processes play a role in the etiology of agoraphobia. Goldstein and Chambless (Goldstein, 1970; Goldstein & Chambless, 1978) in particular have stressed the importance of the agoraphobic’s interpersonal relationships in the development and maintenance of their problems. For example, Goldstein (1970) claimed that most agoraphobics are in a relationship from which they wish to flee but cannot because they fear independence. In their reanalysis of agoraphobia, Goldstein and Chambless (1978) distinguish “complex” from “simple” agoraphobia, where the former is considered to have as its central element a “fear of fear,” which involves concern about the consequences of feeling afraid {e.g., losing control, fainting, heart attacks). This f-ear is said to develop in individuals with low levels of self-suf~ciency, mainly during periods of interpersol~ai conflict. Consequently, agoraphobics are described as nonassertive, pervasively fearfuf individuals, who perceive themselves to be incapable of functioning independently. This causes them to suffer from social anxiety and marked fears about responsibility as well as experiencing conflict between the desire for individuatization and the longing to remain in a familiar and secure environment. According to Goldstein & Chambless, when this interpersonal conflict persists long enough, or is worsened by other events (e.g., illness, birth of a child), the pre-agoraphobic person is likely to experience panic attacks which precipitate the entire agoraphobic syndrome. Fodor (1974) proposes a sex-role theory of agoraphobia, based on the finding that most agoraphobics are women. She believes that these women have been reinforced for an extreme version of stereotypic female behavior, so that they become especially helpless, dependent, and fearful. Such women are likely to seek out partners who will protect them from the necessity to be independent and



assertive. However, this may only be a temporary solution, and the woman may gradually become more dependent, more fearful, and finally agoraphobic and very possibly resentful of her partner. While Goldstein and Chambless see relationship difficulties as the source of conflict which causes agoraphobia, Fodor sees traditional stereotypic sex roles as inadvertently increasing the likelihood that these women will develop both agoraphobia and dependent relationships. Theoretical analyses, from various perspectives, then, emphasize a dependent personality and unsatisfactory interpersonal relationships as central to either the etiology and/or the maintenance of agoraphobia. We now turn to a consideration of the empirical bases for these views. EVIDENCEFOR RElATIONSHIP DIFFICULTIES Despite the fact that many authors have claimed that relationship difficulties and agoraphobia are intertwined, very little in the way of controlled research has been carried out. From the propositions outlined above, we might expect research to have determined the presence or absence of relationship turmoil in agoraphobics and their partners. In this respect we need to know whether these problems are apparent prior to the onset of the disorder or if they only appear after agoraphobia becomes established. With respect to the specifics of these relationship problems, agoraphobics should display one or another of the following features: dependent styles of interrelating; hostility toward, or dissatisfaction with their partners; sexual maladjustment; adherence to stereotypic sex-roles; or dominant-submissive features to their relationships. It would also be valuable to have an account of either the impact of treatment on any of these problems, or the difficulties which these problems present for treatment. Unfortunately, the evidence does not always clearly address these issues, and much of it comes from either uncontrolled or poorly controlled studies. Clinical Reports

Fry (1962) examined the spouses and marital interactions of an unspecified number of patients with phobic anxiety states having mainly agoraphobic symptoms. The spouses were described as having feelings of inferiority and inadequacy which affected their marital functioning and their life in general. Moreover, the spouses reluctantly revealed a history of symptoms very similar to those of the patients. These symptoms became more obvious only upon treatment of the “ill partner.” Fry concluded that the patients and spouses resembled one another, and that the spouses had a vested interest in maintaining the phobia in the partner. Forty-one married women who had become housebound by their phobic anxiety symptoms were interviewed and had their case records reviewed by Roberts (1964). Only nine of these patients (22%) reported that they “were other than compatibly and reasonably happily married.” The incidence of sexual problems among them, however, was quite high (53%). Given the high incidence of sexual problems, it seems possible that some of these women were denying marital distress since sexual difficulties have been reported to be highly correlated with marital happiness (Brady, 1976; Frank, Anderson, 8c Kupfer, 1976). Lazarus (1966) described an agoraphobic woman whose husband needed her

L. Klrincr and W. L. Murshall


pathological dependency because of his own insecurity and fears of rejection. Lazarus concluded that individuals cannot develop and maintain agoraphobic behavior patterns unless someone in their interpersonal field submits to their demands and supports their dependency. Sixteen of Goldstein’s (1973) 20 female agoraphobic patients felt strong urges to escape their relationship but were the phobia. In his view, agoraphobics

unable to do so at the time of the onset of typically have marital conflicts and wish to

flee the marriage but cannot do so because of their fears of being alone. Goodstein and Swift (1977) report that an analysis of the marriages of three agoraphobic women revealed maladaptive relationships where the woman was dependent and insecure. These authors argued that the women had deliberately sought parental protectors in their husbands. Conversely, the husbands were said to display a dominating and obsessive nature, and felt that control and authority over their wives was essential to their masculine self-image. Goodstein and Swift concluded that the problems inducing the phobia were mainly related to dominance and submission within the marital interaction. Similarly, Everaerd, Rijken, and Emmelkamp (1973) reported that 45% of their agoraphobics requested a discussion of family problems during the course of exposure in vivo treatment. The interpretations offered by the authors of these rather anecdotal clinical reports, then, provided the basis for the development of current views, which suggest that agoraphobia is associated with relationship conflict. It is, however, difficult to sort out observation from interpretation in these studies, so that at best these reports are suggestive, and controlled studies are required before firm conclusions

can be made.

Con trolled Studies We have been generous in including a number of the following studies in this section, since the controls exercised in many of them are rather loose and hardly meet satisfactory standards for the objective evaluation of the issues. For instance, some of these studies use retrospective recall as the basis for inferring problems, which presents all manner of difficulties in interpretation, not the least of which are a self-serving bias and a bias introduced by current distressed status. Nevertheless, it is proper to note that for some aspects of the “relationship difficulties” hypothesis (e.g., the etiological process), there is no reasonable alternative strategy available; long-term prospective studies are expensive and, in any case, it is hard to see how one could choose an “at risk” population to make such a study feasible. Similarly, a number of studies have chosen to analyze available case notes and compare the data derived from these concerning relationship difficulties with those extracted from the case notes of matched patients having other disorders. Case notes detail unsystematic observations by many and varied staff members, usually over an extended period which embraces changing views of problems and is contaminated by momentary sentiments regarding the nature of the disorders in question. These caveats must be kept in mind when considering the value of the following reports. In what appears to be the first attempt at a systematic, partially controlled study of agoraphobia, Webster (1953) analyzed the case notes of 25 agoraphobics, all married women, and compared them with the case records of two other groups, each consisting of 25 married women diagnosed as “conversion hysteria” and “anx-


iety neurosis.”



of the latter

been a bit unfortunate in retrospect, whether anxiety neuroses constitutes



for comparative


may have

since there is nowadays some debate about a distinct clinical entity from agoraphobia

(Hallam, 1978). In any case, Webster found that the case notes recorded four features which differentiated the agoraphobics from the other two groups. The agoraphobics were described as: (a) lacking adequate fathering, because of the father’s withdrawal from active involvement in family life; (b) being subject to maternal domination and overprotection; (c) having at adulthood, abnormal fears of bodily injury, pregnancy and childbirth; and (d) being sexually unresponsive to their husbands. Webster concluded that agoraphobic women lack self-confidence and coping skills and readily move from dependency on a mother to dependency on a husband. He also suggested that these patients tend to choose immature men with more than usual personal problems, who somehow encourage their wife’s agoraphobic symptoms. Perhaps the observation of Webster which is most in line with the view that agoraphobia is related to relationship conflict, is that these patients were almost invariably sexually unresponsive to their husbands. As noted earlier, sexual problems are significantly related to marital conflict. Another early controlled study by Agulnik (1970), although it did not directly address the issue of marital conflict, nevertheless produced evidence that is relevant to this issue. He found that the spouses of agoraphobics were, for the most part, satisfactorily adjusted and both partners were accurate in their perceptions of one another. These observations suggest a low rather than a high rate of marital conflict. Torphy and Measey (1974) studied the marital interactions of 28 married women who were members of the Open Door Association, a voluntary British association for agoraphobics. The women and their husbands completed questionnaires which measured mutual perceptions, using eight bipolar scales (e.g., unintelligent-bright; generous-selfish). The couples were also asked to rate their marital satisfaction on a five-point scale. Based on the couples’ combined ratings, the marriages were divided into “good marriages” (n = 16) and “poor marriages” (n = 12). Partners in the poor marriages tended to misperceive each other, with the wives overevaluating the toughness and stability of their husbands. In the good marriages, however, partners tended to perceive each other quite accurately and positively. The authors suggest that in some cases the demands made by the agoraphobics on the relationship, as well as the typical misperceptions, may have led to an unsatisfactory marriage. Indeed, their observation that approximately 43% of the agoraphobic women in the sample reported some degree of marital dissatisfaction, confirms the relevance of this issue and suggests that such problems may be common in agoraphobia. In recent years, no one has been a more persistent advocate of the importance of relationship problems in agoraphobia than Hafner (1982). In one study (Hafner, 1977a) he examined the symptoms and personalities of 33 agoraphobic women and their husbands, by interviewing all patients, both alone and with their spouse, to obtain information about attitudes towards the marriage, the phobia, and other aspects of domestic life. Different questionnaires were completed by both partners to assess the phobic symptoms, as well as other psychoneurotic symptoms, general hostility, and satisfaction with their self and their spouse. Agoraphobics’ husbands were not demonstrably neurotic, and there was no significant overall relationship between whatever symptoms the husband did display

L. Kleiner and W L. Marshall


and the duration

of severity

of the woman’s



wives experienced


as much self-dissatisfaction than did the husbands, and they were also significantly more dissatisfied with themselves than their husbands were with them. Hafner took these observations to imply a resistance to symptom relief or change on the part of the husbands. Regarding the hostility scores and their direction, two distinct types of interaction emerged: (a) extrapunitive women with high hostility scores were married to relatively intrapunitive men with normal hostility; while (b) intrapunitive women with low hostility were married to extrapunitive men with relatively high hostility. Hafner considers these complementary hostility profiles to indicate a form of assortative mating. In assortative mating the partners choose one another on the basis of perceived desired attributes. Unfortunately, despite Hafner’s persistent interest in marital problems in agoraphobia, he did not directly measure conflict. We are, therefore, left with an unclear picture of the nature or degree of marital dissatisfaction in either these patients or their partners. Given the description of their relationship, it would be surprising if satisfaction direct evidence it would

were high amongst these couples, but in the absence be unwise to extrapolate beyond the data.


In contradiction to some of the earlier findings, Buglass, Clarke, Henderson, Kreitman, and Presley (1977) could find no indications of marital difficulties in agoraphobics. These authors compared 30 married agoraphobic women who were referred to their clinic, with matched nonpsychiatric controls selected from general practice records. All subjects were interviewed alone and with their partners to gather information concerning current symptomatology, early childhood experiences, perception of the partner, allocation of household tasks, and social activities. “manifest interaction” ratings were made on the basis of a tapeIn addition, recording of the conjoint interview. All participants completed questionnaires to evaluate symptomatology (Cornell Medical Index), personality (Eysenck Inventory), and perception of self and spouse (Semantic Differential). Agoraphobic women did not differ from controls in their sexual adjustment prior to regarded in both course. married reported

the onset of the phobia. A similar proportion (not specified) in both groups their sexual relations as satisfactory and, conversely, the same proportion groups (about 25%) reported never experiencing orgasm during interWith the onset of the phobia, however, 16 of the 24 patients who were already, reported a marked loss of sexual drive. Moreover, many patients continuing intercourse principally to please their husbands. These ob-

servations clearly suggest that, insofar as disturbed sexual relations are an indication of marital difficulties, such difficulties are a product of, rather than a cause of, agoraphobia. Consistent with this suggestion is the report of Buglass that the husbands of agoraphobics complained of adverse effects upon them resulting from their wife’s disorder. Despite these adverse effects, about 60% of the husbands saw themselves as sympathetic; however, only about half of the wives agreed with this view, and some 10% considered their husbands to be remote, silent, and perplexed. This study, then, does not support an etiological role in agoraphobia for relationship difficulties, although, while the majority of the marriages appeared to be essentially normal, a minority were clearly troubled. In their review of this study, Mathews, Gelder, and Johnston (198 1) suggest that these observations are consistent with their own clinical experience. Mathews et al. claim that the reason many clinicians believe there to be a strong agoraphobia, is that these clinicians

relationship between marital difficulties and remernber vividly the dramatic but rare in-



stances of such problems and forget the more common but mundane cases where marital satisfaction and spouse support were normal. Despite the generally sound design of the Buglass et al. study, its conclusions have been criticized by several authors. Chambless (1982), for example, claimed that since depression and marital conflict are highly related in women, and markedly depressed agoraphobic women were excused from the study, the sample was biased against finding marital dissatisfaction. In an analysis which did not concern itself with design issues, Hafner (1982) proposed a different interpretation of the findings of Buglass et al. Pointing to the inconsistencies between the reports of the patients and their spouses, Hafner claimed that this indicated the husbands were either emotionally immune to the amount of conflict within their marriage, or were disguising or denying their real feelings and attitudes. For example, the husband’s behavior was independent of the degree of cooperation in the marriage, while amongst the phobic women affection and cooperation were positively related and, the less assertive they were, the more cooperative the marriage was. Moreover, we have already noted the discrepancies between the partners concerning their views of the husband’s sympathy toward their wives. The apparent similarity between the agoraphobic and normal couples on several variables, Hafner suggests, might not so much imply a trouble-free marriage in the agoraphobics, but rather a tendency to not reveal real difficulties due to lack of assertiveness or fear of the consequences of facing their marital distress. While there seems to be little doubt that agoraphobics are generally unassertive (Goldstein 8c Chambless, 1978), this position taken by Hafner, appears dangerously close to being excluded from disconfirmation. For example, Hafner seems to interpret a denial of marital dissatisfaction as meaning that the agoraphobic may be covering up real problems. While this is possible, his ready acceptance of admissions of dissatisfaction as valid, suggests that disconfirmation of this hypothesis may be hard to come by. Similarly, Hafner’s interpretation of the meaning of the discrepancy in reports between agoraphobics and their partners concerning the husband’s sympathy, appears to reflect a readiness to accept reports if they agree with the hypothesis while rejecting those which disagree. While Hafner’s analysis quite correctly indicates possible problems with the interpretation of self-reports, for his arguments to have credibility, he must provide alternative strategies for the evaluation of marital satisfaction; to date he has not done so. Goldstein and Chambless (1978) supported their reanalysis of agoraphobia by data obtained from 32 agoraphobics (25 women) and 36 clients with phobias of external specific stimuli (24 women). Information was gathered from the client’s records, which included three scales: Bernreuter Self-Sufficiency Scale, Willoughby Emotional Maturity Scale, and the Fear Survey Schedule. When possible the therapists rated their clients at the beginning of treatment on a variety of seven-point scales. Unfortunately, we are not told how frequently this was done. However, the authors report that compared to other phobics, the clients defined as “complex” agoraphobics were characterized by significantly less emotional maturity, more social anxiety, less self-sufficiency, and marked fears of responsibility, decisionmaking, disapproval and criticism. Moreover, the onset of agoraphobia was typically reported by the patients to have occurred during times of high interpersonal conflict and in the absence of specific traumatic events. Although incidence of marital conflict at the time of the study was not reported, Goldstein and Chambless pointed to the remarkably consistent picture of nonassertive, pervasively fearful individuals

who perceive themselves to be incapable of functioning independently and who usually depend WI a partner. In the case of a bad marriage, they claim that a conflict and crisis develops between the impulse to leave and fears of being alone. Although consistent with the clinical reports presented earlier, this analysis needs a more objective evaluation. Both their reliance on clinical records and retrospective reports, as well as an unclear distinction between observation and interpretation, make the study by Goldstein and Chambless less valuable than it might otherwise have been, although one cannot deny the impetus their study provided for subsequent research. To sum up, no clear picture has emerged regarding the incidence of relationship difXiculties in agoraphobia. It does seem to be clear that some agoraphobics have problems in their relationships, but we do not know how many suffer from these difficulties. Nor do we know whether these problems preceded or followed the development of the agoraphobia. Interestingly, when compared to “normal” samples (Agulnik, 1970; Buglass et al., 1977) the marital relations of agoraphobics were not found to be disrupted. Yet, when marital dissatisfaction was assessed in agoraphobic groups independently (Hafner, 1977a; Torpy & Measey, 1974) or when compared to other clinical groups (Goldstein & Chambless, 1978; Webster, 1953), marital conflict appeared to be frequently associated with the disorder. It is reasonable, therefore, to speculate that although agoraphobic patients do not differ from normals in their overall interpersonal relations, the interaction between their personal characteristics (e.g., dependency, lack of assertiveness, etc.) and marital dissatisfaction might nevertheless be important in the development and maintenance of the phobia. With respect to this latter issue, the impact of treatment on an ongoing relationship and the problems this presents for implementing effective treatment, may be revealing. We now turn to this issue. RFJATIONSHIPDIFFICULTKSAND TREATMENT Recent research (e.g., Emmelkamp, 1982) has found in vivo exposure techniques to be the treatment of choice for phobic disorders. These powerful behavioral interventions often produce rapid and effective changes in agoraphobic symptomatology (Marks, 1979). Exposure treatment for agoraphobia, however, fails to produce benefits in a sufficient number of clients, and this warrants concern. In fact, outcome data from three recent reports (Barlow, Mavissakalian, & Hay, 1981; Emmelkamp & Kuipers, 1979; McPherson, Brougham, & McLaren, 1980) indicate that between 30% and 40% of agoraphobics who complete exposure treatment fail to improve. Furthermore, of those patients who do improve, a significant proportion fail to display satisfactory levels of functioning at followup (McPherson et al., 1980). Finally, drop-out rates are high in any treatment program for agoraphobics, with approximately 12% of those patients withdrawing from behavioral treatments (Jansson & Ost, 1982), and 25% to 40% withdrawing from treatments which include drugs (Zitrin, Klein, & Woerner, 1978, 1980). It may be that interpersonal factors, particularly relationship problems, play a role in these failures or drop-outs. Interpersonal factors may interact with the patient’s response to treatment in two ways: (a) relationship problems can interfere with the direct treatment of the phobic symptoms, resulting in treatment failure; or (b) significant treatment induced-changes in phobic symptomatology can have either adverse or positive



effects on the patient’s relationships. Several studies have attempted to examine these possibilities. For example, Hudson (1974) divided 18 agoraphobic patients into three categories on the basis of the stability and adjustment of their families. Seven families were classified as “well adjusted,” seven as “anxious-patients” families (i.e., family members responded to the patient as a highly anxious individual), while four families were said to be “sick’ (i.e., families in which severe pathology was manifested with the patients’ symptoms being part of this pathology). Hudson treated all patients with short-term drug-assisted exposure in vivo. Treatment effects were unfortunately evaluated simply by the therapist’s global ratings, which render any conclusions tentative at best. However, improvements were shown by all patients in the well-adjusted families. Patients from the other two family groups displayed only partial improvement or no improvement at all. The poorest results were obtained by patients in the “sick” family group. Hand and Lamontagne (1976) treated 25 agoraphobic clients (16 women) with in vivo group exposure for 12 hours, spread over 3 days. Fifteen of the clients improved satisfactorily based on a composite score derived from a psychiatrist’s ratings. Seven clients, however, had acute crises in their marital relationships shortly after phobia removal, and asked for, or were offered, marital therapy. The authors emphasized that 14 out of 2 1 married patients in the group were aware of chronic marital problems before the treatment began; but all preferred treatment for their phobia rather than marital therapy. Even after the exposure treatment, only three out of the seven couples experiencing acute marital problems agreed to participate in joint marital counselling. Two different patterns of interaction between the phobia and marital conflict were found. In half of the patients, improvement in phobic symptoms was followed by an exacerbation of the marital problems. In the rest, however, phobia removal did not provoke any acute marital crises, and in some cases seemed to lead to an improvement in the marital relationship. Although this study did not use systematic measures of marital problems, and the data presented are anecdotal, it does suggest that in a number of agoraphobics, relationship difficulties may interact negatively with the treatment of their phobia. However, it is also important to note that in some cases, improvement in phobic symptoms enhanced the relationship. In his exposure treatment of 30 married agoraphobic women, Hafner (197713) found differences in response to treatment depending on their scores on a measure of hostility. At 12-month followup patients in the most hostile group showed a decrease in hostility while the husbands showed an increase; these patients also displayed less improvement in phobic symptoms than those categorized as least hostile. In the least hostile group, both husbands and wives showed a decrement in their hostility scores, with the wives responding satisfactorily to treatment. Furthermore, husbands in the most hostile group showed an increase in neurotic symptoms and self-dissatisfaction scores 3 months after treatment just when their wives appeared most improved. Hafner concluded that these patterns of changes in the hostile group suggest that a proportion of husbands are adversely affected by the initial symptomatic improvement in their wives. He also claimed that case details suggested that these husbands had attempted to sabotage treatment. Unfortunately, these illustrations of so-called disruptive attempts by the husbands (e.g., one case of a husband’s suicidal attempt; two cases of reactive depression in the husbands, and four cases

where the husband displayed abnormal jealousy reactions), do not appear to be clear instances of deliberate sabotage. However, these behaviors by the husbands do appear to support the notion that these men were profoundly affected by healthful improvements in their wives, suggesting some serious problems in the relationships which perhaps served to maintain the agoraphobia. This latter possibility is further supported by Hafner’s observation that in some 25% of the cases, symptomatic improvement was followed by increased marital discord. However, given that improvements in agoraphobic symptoms often represent a radical change in the patient.‘s behavior, it is perhaps no surprise that in some unprepared partners there are disruptive effects which may be transitory. unfortunately, Hafner did not measure marital satisfaction objectively and he did not indicate whether or not he attempted to prepare the husbands for the radical change which symptomatic improvement can bring. We have elsewhere (Marshall & Gauthier, 1983) argued that the proper preparation of patients and their families improves the effectiveness of exposure-based treatments. In a subsequent treatment program, Milton and Hafner (1979) classified 14 agoraphobic women, who completed both treatment and a 6-month followup, as either maritally satisfied or dissatisfied. Again, unfortunately, no objective basis was provided for this classification. Both groups showed significant improvement in phobic and neurotic symptoms following treatment; however, by followup the maritally dissatisfied patients indicated partial relapse, while patients who were satisfied with their marriage displayed continued improvement. Six of the patients who relapsed indicated that their marital adjustment had deteriorated since they entered treatment. Milton and Hafner were convinced by the remarks of those relapsed patients that prior to treatment the phobic behavior of these women either served to hold their marriage together or to cover up the husband’s problems or inadequacies. However, the bases of these conclusions by Milton and Hafner are questionable and are at least in need of careful empirical validation. On the basis of pre-treatment scores on a measure of marital satisfaction, Bland and Hallam (198 1) divided their agoraphobic patients into those who had a “good marriage” and those who had a “poor marriage.” Patients in the good marriage group responded well to treatment and maintained their improvement in both phobic and general neurotic symptoms at 3-month followup. In contrast, agoraphobics in the poor marriage group improved initially after treatment but later lost most of their gains and reported no changes in general neurotic symptoms. Interestingly, treatment had no impact on levels of dissatisfaction with their spouses, for either patients or their husbands in the poor marriage group. Furthermore, effective treatment did not lead to a deterioration in marital satisfaction nor to attempts by the husband to undermine treatment. On the other hand, level of satisfaction prior to treatment clearly predicted successes and failures. Although Emmelkamp (1980) conducted a well-designed study to examine the effects of more defined interpersonal problems on response to in vivo treatment, he failed to address the apparent implication of the earlier studies; namely that the negative implications of poor marital adjustment do not become obvious until some months after treatment is terminated. At 1 month after treatment termination, Emmelkamp found no effects on outcome resulting from the pre-treatment measures of either marital problems or lack of assertiveness in the patient. It is unfortunate that such a good study did not involve a longer followup; 1 month is



probably too short a time for the negative effects to be apparent as Emmelkamp himself admits (Emmelkamp 8c Van der Hout, 1983). The treatment studies reviewed so far have not included attempts to involve the spouse or to deal with the marital problems directly. Three such studies have been reported. The first of these was reported by Cobb, McDonald, Marks, and Stern (1980). They compared behavioral marital therapy with exposure in vivo for the treatment of 11 couples with marital and mixed phobic-obsessive problems; unfortunately, it is not clear how many patients were agoraphobic. The couples were randomly assigned to one of two treatment orders in a crossover design. One group received 10 sessions in vivo exposure for phobic-obsessive targets, followed by 10 sessions of behavioral marital therapy. The other group received marital therapy first and then exposure treatment. Marital therapy involved contracting, clarification of difficulties and goals, and, where needed, sexual skills training. Couples were treated individually, with spouses present during both exposure and marital treatments. Phobic-obsessive symptoms and marital problems were assessed by self-reports, an interview and an independent assessor. Exposure led to significant improvements in phobic-obsessive targets, although there were no gains on two structured marital adjustment scales. In contrast, marital therapy led to significant improvements only in the relationship difficulties, and even here the benefits were no greater than those produced by exposure. Cobb et al concluded that exposure is the treatment of choice for phobias and obsessive-compulsive problems even when severe marital problems co-exist with the anxiety disorder. They suggest that marital therapy may be appropriate if marital conflict still exists following exposure treatment. However, it is important to remind the reader that for the most part the marital therapy employed in this study, produced little in the way of benefits on marital problems. This observation calls into question the effectiveness of this treatment procedure and to that extent reflects on the value of this study in estimating the impact of marital therapy on exposure-based treatments of agoraphobia. Furthermore, the disadvantage of the cross-over design is that it does not permit the long-term evaluation of each treatment component. This issue, as we have observed earlier, may be critical to the evaluation of the interaction of marital difficulties and the treatment of agoraphobia. Barlow et al., (1981), examined the effects of a behavioral program which focused only on the phobic problems, but did include the patient’s spouse. Their rationale for this strategy was that since spouses appear to occasionally undermine therapy, including them in treatment may secure their cooperation as well as facilitate greater practice during exposure sessions. Reflecting our earlier remarks, we might also add that including the patient’s partner more properly prepares them to deal with the changes produced by treatment (Marshall 8c Gauthier, 1983). Six agoraphobic women and their spouses participated in 10 to 13 sessions of a group therapy program consisting of exposure and cognitive restructuring, in which the husbands acted as co-therapists. Ratings of phobic severity and marital satisfaction were filled out independently by both partners throughout the course of therapy and up to 12-month followup. All clients showed improvements with respect to their phobic behavior at posttreatment, while the outcome data revealed two patterns of relationship between marital satisfaction and severity of the phobia. In four couples, ratings of marital


I,. Kleiner and U! L. Alarshull

satisfaction demonstrated a parallel, positive relationship with ratings of the phobia, that is, as the phobia improved, marital satisfaction increased. For two couples the inverse relationship was noted, that is, improvements in the phobia were correlated with decreases in marital satisfaction. These patterns were not related to the absolute levels of marital satisfaction prior to therapy, but in the couples showing a positive relationship, there was a marked agreement between the husband’s and wife’s perceptions of the severity of the phobia. On the other hand, in the two couples showing an inverse relationship, the husbands rated their wives’ phobias as considerably less of a problem than did the women themselves. This may suggest a lack of empathy and understanding on the part of the spouse. The authors concluded that regardless of the degree of marital satisfaction, all cliem : responded well to the behavioral intervention, and in the majority, the treatment appeared to improve the marriage or at least have no substantial ill-effect on it. While the data reported in this study are correlational and descriptive, and no statistical evaluations were provided, it did serve as a preliminary step for a larger scale, more sophisticated study. Barlow, O’Brien, and Last (1984) report the behavioral treatment of 28 agoraphobic women, half of whom were treated with their husbands as co-therapists, and half without their husbands. All husbands were willing to cooperate so that differences between the groups could not be ascribed to this factor. Treatment in this study involved cognitive restructuring and directions to engage self-initiated exposure. These treatment endeavors and directions were dealt with in a group format where discussions of procedural details, practice requirements, and problems were encouraged. In many respects this program is similar to that offered by Mathews, Gelder, and Johnston (1981). It is, however, important to note that “couples in the spouse group were also encouraged to communicate more effectively during periods of anxiety and panic” (p. 47), and this may have been useful in providing a model for resolving other conflicts in the relationships. Barlow et al. found that clients in the spouse group showed greater improvements across a variety of measures than did the nonspouse group. Similarly, benefits for the patients on measures of social and family functioning were more rapid in the spouse group, although by the end of treatment the nonspouse group had caught up. While this study clearly points to the value of including spouses in the treatment of agoraphobics, the failure to provide long-term followup data reflects on the value of the report given the earlier observation that pre-treatment marital dissatisfaction does not effect outcome until several months after treatment is complete. Presumably, however, such long-term followup data will be forthcoming. It would have been valuable if the authors had chosen to determine the relationship between pre-treatment measures of marital satisfaction and outcome to see if these measures had predictive value, but again perhaps this is best left to a later report so that long-term outcome can be evaluated. Similarly, the definition of “social and family” functioning could have been more clearly specified to determine whether marital satisfaction, assertiveness, dependency, and so on, were related to outcome or were responsive to treatment. Nevertheless, this is one of the more informative studies of the relationship between interpersonal conflicts and treatment for agoraphobics. In summary, then, the results of examining the effects of treatment on both phobic symptomatology and relationship difficulties suggest an important relationship between these two spheres of problems. Relationship difficulties prior to





to predict




or failures,



specific nature of these relationship difficulties needs to be more clearly defined. Including partners (perhaps not just the spouses of those who are married) as cotherapists in exposure-based treatments for agoraphobics appears to enhance the value of the program considerably, although long-term effects still need to be evaluated. And again, the impact of these procedures on other aspects of interpersonal functioning (e.g., lack of assertiveness, dependent life style, stereotypic sex role behaviors, etc.) needs to be determined. CONCLUSIONS In conclusion, then, the incidence and nature of relationship conflict in agoraphobia remains unclear. Several studies report that such conflict is more common among many agoraphobics (approximately 45% appear to suffer such problems) than it is among patients from other clinical groups. Marital conflict is clearly not always involved in the development and maintenance of agoraphobia, as not all these patients are married and, of those who are, not all are dissatisfied with their marriage. The interaction, however, between interpersonal difficulties and agoraphobits’ personal characteristics (e.g. dependency, lack of assertiveness, etc.) might nevertheless be important in the development and, particularly, the maintenance of the phobia. Indeed, our evaluation of treatment studies led us to conclude that involving partners in therapy, and/or employing treatment components which deal with relationship problems, enhances the effectiveness of the intervention program. More detailed analyses of the various features of interpersonal difficulties which may cause disharmony in relationships among agoraphobics are necessary before more precise conclusions can be made, but the evidence strongly suggests that these are issues to which we must give research attention. REFERENCES

Agulnik, P. (1970). The spouse of the phobic patient. British Journal of Psychiatry, 117, 59-67. Andrews, J, D. W. (1966). Psychotherapy of phobias. Psychological Bulletin, 66, 455-480. Barlow, D. H., Mavissakalian, M., & Hay, L. (1981). Couples treatment of agoraphobia: Changes in marital satisfaction. Behauiour Research and Therapy, 19, 245-255. Barlow, D. H., O’Brien, G. T., & Last, C. G. (1984). Couples treatment of agoraphobia. Therapy, 15,4158.

Bland, K., & Hallam, R. (1981). Relationship between response to graded exposure and marital satisfaction in agoraphobics. Behauiour Research and Therapy, 19, 335-338. Brady, J. P. (1976). Behavior therapy and sex therapy. American Journal ofpsychiatq, 133, 896-899. Buglass, D., Clarke, J., Henderson, A., Kreitman, N., & Presley, A. (1977). A study of agoraphobic housewives. Psychological Medicine, 7, 73-86. Chambless, D. (1982). Characteristics of agoraphobics. In D. Chambless, & A. Goldstein (Eds.), Agoraphobia: Mulliple perspectives on theory and treatment. New York: Wiley Interscience. Cobb, J., McDonald, R., Marks, I., & Stern, R. (1981). Marital versus exposure therapy: Psychological treatments of co-existing marital and phobic-obsessive problems. Behavioural Analysis Modification, 4, 3-17. Dixon, J., Monchaux, C., & Sanders, J. (1957). Patterns of anxiety: The phobias. BriC.shJoumal ofMedical Psychology, 30, 34-39.

Emmelkamp, l? M. G. (1980). Agoraphobics’ interpersonal problems: Their role in the effects of exposure in viva therapy. Archives of General Psychiatry, 37, 1303 - 1306.

L. KltGrt~r and


Emmelkamp, York:

P. M. G. (1982).



P M. G., & Kuipers, of Psych&q, (Eds.),

W., Rijken,

approximation’ Fodor,

New York:


phobic C.,



1, 305-3




R. (1977b).




In E. B. Foa & P M. G.

& Sons.

A comparison

of ‘flooding’

& V. Burtle

and ‘successive


Research and Therapy, (Eds.),



Women in therapy,

R. (1978).




sex therapy



1, 245-252.

Family Process,



in understanding



of Behavior




of agoraphobia. with phobic

of treatment.


A plea

for empi-

Association for Behavior




of Psychotherapy.


9, 47-59.





31, 285-292.

or pathogenic



of the agoraphobic




American Journal

of agoraphobic 130,


of the European

& Schawarzenberg.

A reanalysis


of Psychi&?,


and their influence

on treatment




of the agoraphobic


A critical


In D. Chambless

New York:

Multiple perspective,\ on theory and trmtment.



of agoraphobia.


D. (1978).

‘l‘he marital


of couples


& W Tunner

of Psychiatq,

R. (1982).



and focus


Brztlsh Journal

Brotish Journal Hafner,




of symptoms

R. (1977a).


study four years after


In V. Franks,



of an anxiety

R., & Swif-t, K. (1977).


D. (1976).


and Modification,

the source

research and practice.




in women.


A., & Chambless,


in treating

P. M. G. (1973).

of Psychiatq,


A. (1973).




In ,J. C. Brengelmann


A follow-up


of agoraphobia.

& Kupfer,


A. (1970).




Fry, W. (1962).


A. (1983).

H. M., & Emmelkamp,

American Journal



in behavior therapy. New York:


E., Anderson,




in the treatment

I. (1974).


A. (1979).


P M. G., & VanderHout,

Emmelkamp Everaerd,

Phobic and obsrssive-compulszvr


British Journal Emmelkamp,

VV L. Marshall

of the concept.


Briti.yh Journal

& A. Goldstein


of Psych&y,



319. Hand,

I., & Lamontagne,

removal. Hudson,

Y. (1976).


B. (1974).



of interpersonal

and Practice,

13, 405-4

Theon’, Research



of agoraphobics







by behavior


Brutish Journal

of Social

Work. 4, 51-.59. Jansson,

L., & <>st, L. (1982).



for agoraphobia:

An evaluative



PsyrholoRy Rpuieuj, 2, 3 11-337. Lazarus,

A. (1966).

Research Marks,

I. (1979).


and care

A. M., Gelder,





of neurosis.

J. (1983).

P.~yrhological Medicine,


M. G., &Johnston,

E M., Brougham, treated

in flooding. Wiley

9, 629-660. (Eds.),

& Sons.

D. W. (1981).

L., & McLaren,

by behavioral

E, & Hafner,

R. (1979).


A. (1964).

of Psych&y, Ruddick,

methods: The

of Grneral





and treatment.

New York:

S. (1980).


A four-year


of improvement Brhaviour


in agoraphobic and Therapy,



of behavior


36, 807-811.



A follow-up

for agoraphobia

in family

in relation


study of phobic-anxiety

to marital

In V. Eisenstein state.

British Journal


B. (1961).


S. (1976).


P.\ychology, 49,

2 1 l-236.



In E. B. Foa & P. M. G. Emmelkamp

Mittleman, B. (1956). Analysis of reciprocal neurotic patterns (Ed.), Neurotzc interactton iu marriage.. London: Tavistock.


of agoraphobia.






McPherson, patients


in behavior therapy. New York: John





4, 95-97.

W. L., & Gauthier,




and Therapy,

D., 8c Measey,


of phobic

1.. (1974).



Journal A study


of PsychoanalyGs, of 90


in agoraphobia.

42, 537-543. cases.


B&w-h Journal

of Medical

of Cltnical Psychology,


351-353. Webster, 18.

A. (1953).



of phobias

in married








Zitrin, C. M., Klein, D. F., & Woerner, M. G. (1978). Behavior therapy, supportive psychotherapy, imipramine and phobias. Archives of General Psych&y, 35, 307-316. Zitrin, C. M., Klein, D. E, & Woerner, M. G. (1980). Treatment of agoraphobics with group exposure in vivo and imipramine. Archives of General Psychiatry, 37, 63-72.