understanding depression and effective treatment

B e l i e f s A b o u t t h e C a u s e s o f D e p r e s s i o n a n d T r e a t m e n t P r e f e r e n c e s �

Shabad-Ratan Khalsa,1 Kevin S. McCarthy,1 Brian A. Sharpless,1

Marna S. Barrett,1 and Jacques P. Barber1,2

1 University of Pennsylvania School of Medicine 2 Philadelphia VA Medical Center

The relation between patients’ beliefs about the causes of their depression, treatment preferences,

and demographic variables was studied in a sample of 156 patients in a randomized controlled trial for

depression (supportive-expressive psychotherapy vs. medication vs. placebo). No gender differences

were found in beliefs or preferences. Racial differences were found for causes endorsed, but not

preferences. Treatment experience predicted endorsement of characterological and biological causes.

Psychotherapy experience predicted preference for medication. Finally, patients preferring psychotherapy

endorsed childhood and complex causes more than those preferring medication, but the groups

did not differ in other reasons endorsed. Implications of findings are discussed. & 2011 Wiley Periodicals,

Inc. J Clin Psychol 67:539–549, 2011.

Keywords: psychotherapy; major depressive disorder; gender; patient preferences; psychodynamics

Patients may presently choose from a variety of efficacious psychotherapeutic and pharmaco- logical treatments for depression. Psychotherapies such as cognitive, interpersonal, and behavioral

therapies have been established as empirically supported treatments (e.g., Chambless & Ollendick, 2001), and brief dynamic therapy is gaining in support (e.g., Leichsenring, Rabung, & Leibing, 2004). Antidepressant medications have been shown to be effective as well, with selective

serotonin reuptake inhibitors being the most widely used (e.g., Cipriani et al., 2005) especially with more severe depression (e.g., Fournier et al., 2010). However, despite the availability of these efficacious treatments, there remains a need to understand the reasons why patients prefer certain treatments. Is it because of how different approaches conceptualize psychopathology?

Understandably, different therapeutic approaches employ different etiological models in their conceptualization and treatment of depression, with psychotherapy tending towards psycho- social conceptions, and pharmacotherapy more biological ones (Goldstein & Rosselli, 2003).

Therefore, it is of interest to investigate whether patients’ beliefs about the causes of their depression are leading them to prefer theoretically consistent treatments.

Beliefs About the Causes of Depression

Studies on beliefs about mental illness have identified a variety of causes that patients endorse including interpersonal problems (e.g., poor social skills, bad romantic relationships,

mistreatment by others), developmental events (e.g., childhood problems, unresolved issues with one’s family), personality or cognitive causes (e.g., irrational concerns, emotional upset, depressive ways of thinking), biological factors (e.g., genetics, biochemical imbalances),

environmental factors (e.g., stress, illness of self or others), and religious causes (e.g., bad luck, fate, God’s will) among others (Addis, Truax, & Jacobson, 1995; Atkinson, Worthington,

� This article was reviewed and accepted under the editorship of Beverly E. Thorn.

Work was conducted at the Center for Psychotherapy Research and the Mood Disorder Section of the Department of Psychiatry, University of Pennsylvania School of Medicine. Written with support from National Institute of Mental Health grant R01 MH 061410. The sertraline and the placebo pills were provided by a grant from Pfizer Corp. Registered in clinicaltrials.gov.

Correspondence concerning this article should be addressed to: Jacques P. Barber, Suite 648, Center for Psychotherapy Research, Department of Psychiatry, University of Pennsylvania School of Medicine, 3535 Market Street, Philadelphia, PA 19104-3309; e-mail: [email protected]

JOURNAL OF CLINICAL PSYCHOLOGY, Vo l . 6 7 ( 6 ) , 5 3 9 -- 5 4 9 ( 20 1 1 ) & 2011 Wiley Periodicals, Inc. Published online in Wiley Online Library (wileyonlinelibrary.com/journal/jclp). D O I : 1 0 . 1 0 0 2 / j c l p . 2 0 7 8 5

Dana, & Good, 1991; Brown et al., 2001; Foulks, Persons, & Merkel, 1986). In addition to identifying some of the specific causes that patients endorse, Addis and Carpenter (1999) have

suggested that the complexity of patients’ beliefs about the causes of their depression (operationalized as the number of different beliefs that patients have about their depression as well as the intensity with which they endorse those beliefs), may be conceptually important as

well to psychotherapy process and outcome. Some researchers have sought to determine whether demographic and cultural variables

including gender, race, and previous treatment experience affect patients’ beliefs about the causes of depression. It has been found by several researchers that men and women differ in the causes they

endorse. Although differences have been found, studies have varied in what those particular differences are as they have utilized different measures of beliefs (Angst et al., 2002; Atkinson et al., 1991; Robbins & Tanck, 1991; Schweizer et al., 2009). The findings on racial differences have

been more consistent, with African Americans found to be less accepting of genetic or biological explanations for mental illness and depression and more accepting of spiritual explanations than Caucasians (Givens, Houston, Van Voorhees, Ford, & Cooper 2007; Millet, Sullivan, Schwebel, &

Myers, 1996; Schnittker, Freese, & Powell, 2000). Although there have been exceptions (e.g., Atkinson et al., 1991), the majority of studies support the idea that ethnic minority and Caucasian patients differ in their beliefs about the origins of their mental illness.

With regards to prior treatment experience, etiology beliefs have been shown to change after receiving treatment, with endorsement of treatment-incongruent reasons for depression diminishing after successful treatment (Leykin, DeRubeis, Shelton, & Amsterdam, 2007). It is possible that patients may change their beliefs as to what caused their depression to be more

consistent with the treatment they received (Leykin, DeRubeis, Shelton et al., 2007) or with the beliefs of their counselors (Atkinson et al., 1991). Only Schweizer et al. (2009) seem to have examined the relationship between receiving prior treatment for one’s current episode of

depression and beliefs about the causes of depression (they reported that patients who endorsed biological and characterological causes more highly were more likely to have received previous treatment for their current episode of depression). The present study is the

first to examine the ways in which experience with psychotherapy or medication including that for previous episodes of depression or other problems might affect these beliefs.

Treatment Preferences

Patients often come to treatment with a preference for the treatment they expect to be most helpful for them [e.g., psychotherapy generally (‘‘talking treatment’’), a specific type of

psychotherapy (‘‘cognitive–behavioral’’), or medication]. Some of the variables most commonly associated with preference include gender, race, and previous treatment experience (Churchill et al., 2000; Dwight-Johnson, Sherbourne, Liao, & Wells, 2000). In surveys of primary care

practice patients, women appear to be more accepting of treatment for depression in general, and are slightly more likely to prefer counseling as compared to men (Churchill et al., 2000; Dwight-Johnson et al., 2000). Racial differences are more pronounced than gender differences

in treatment preferences. African Americans appear to be less accepting of treatment in general, than Caucasians or Hispanics, and are especially reluctant to use antidepressants. When given a choice, African Americans have been found to be more likely to choose counseling over

medication as compared to Caucasians (Cooper et al., 2003; Dwight-Johnson et al., 2000; Givens et al., 2007; Schnittker, 2003). These racial differences in treatment preference appear to be fairly consistent in primary care settings, but have received less research focus in psychiatric and psychological practice and in randomized controlled trials (RCTs).

Prior psychotherapy experience has also been associated with treatment preference. Previous experience with counseling was associated with a preference for counseling over ‘‘trying to pull themselves together; seeing a psychiatrist; taking tablets’’ (Churchill et al.,

2000, p. 905). However, previous experience with medication was not related to treatment preference. In another study, those who had a greater knowledge of counseling were more likely to prefer counseling over medication. A lack of experience with medication was also

related to preference for counseling (Dwight-Johnson et al., 2000).

540 Journal of Clinical Psychology, June 2011

Relation of Reasons for Depression and Treatment Preference

Putting these two lines of research together, it is possible patients’ beliefs about the causes of their depression and their preferences for treatment might be related. Specifically, patients’ explanations for their depression might inform their beliefs as to which treatment they expect

to help them best and, therefore, prefer to receive. Alternatively, patients may initially prefer a treatment (e.g., favoring medications because they are more heavily marketed or are less costly or time-consuming than psychotherapy) and then may find beliefs about their depression

congruent with that particular treatment. Understanding the potential relation between beliefs about the causes of depression and

treatment preferences is important because it may potentially inform psychotherapy. For

instance, congruence between a patient’s beliefs about depression and being provided a theoretically consistent treatment (such as interpersonal psychotherapy for patients endorsing that their depression is a result of interpersonal problems) has been shown to increase perceived treatment credibility, greater motivation to engage in treatment, and greater satisfaction with

that treatment (e.g., Addis & Carpenter, 1999; Addis & Jacobson, 1996; Atkinson et al., 1991; Meyer & Garcia-Roberts, 2007; for an exception, see Goldstein & Rosselli, 2003). Complexity of beliefs about the causes of one’s depression (or endorsing more reasons for depressions at higher

levels) has been found to relate to treatment preferences and expectations about treatment. More specifically, patients with higher complexity of beliefs rated an ‘‘activation-oriented’’ (but not an ‘‘insight-orientated’’) psychotherapy rationale as presented in videotaped or written

format as less credible and less likely to be helpful. Addis and Carpenter (1999) suggest that one possible explanation for this finding is that perhaps people who believe that their depression has many complex causes may be less accepting of a treatment they may view as too simple.

Only one study has examined beliefs about the causes of depression as they relate directly to treatment preference. This study, conducted by Atkinson and colleagues (1991), found that beliefs were unrelated to preference for counseling orientation (‘‘thinking,’’ ‘‘feeling,’’ or ‘‘acting’’). Schweizer and colleagues (2009) found that patient beliefs about the causes of their depression

affected treatment assignment (which was influenced by patient preferences). More specifically, patients scoring more highly on ‘‘intraindividual’’ items on the Reasons for Depression Questionnaire (Addis et al., 1995) were more likely to be assigned to cognitive–behavioral therapy,

whereas those endorsing biological causes more strongly were more likely to be assigned to psychopharmacological treatment. Although patient preference was taken into account for treatment assignment, treatment type was chosen in conjunction with a therapist (Schweizer et al.,

2009). The present study is the first we are aware of to examine beliefs about the causes of depression and their relation to preference between medication and psychotherapy in a clinical setting.

Present Study

Therefore, the purpose of this study was twofold: (a) to examine how beliefs about the causes of depression and treatment preferences (i.e., pharmacotherapy or psychotherapy) relate to each other, and (b) to determine what factors influence beliefs and preferences. We utilize data from a RCT comparing psychotherapy versus medication versus pill-placebo for depression that was able

to recruit a relatively diverse sample in terms of gender, race, and previous treatment experience. Based on the previous literature we hypothesized that treatment preferences would vary based on beliefs about the causes of depression. Specifically, patients preferring medication were expected to

more strongly endorse biological causes, whereas those preferring therapy would be less likely to endorse biological causes and more likely to endorse personal or situational factors. Further, we will investigate how race, gender, and prior treatment experience relate to beliefs and preferences.

Methods

Participants

Data were drawn from a sample of 156 patients taking part in a RCT comparing supportive-

expressive psychotherapy to sertraline/venlafaxine to pill-placebo (authors). Inclusion criteria

541Beliefs About Depression

for the study were an age between 18 and 70, a primary DSM-IV diagnosis of major depressive disorder (MDD) according to the Diagnostic and Statistical Manual of Mental Disorders,

Fourth Edition (DSM-IV; American Psychiatric Association, 1994), and a minimum score of 14 on the 17-item Hamilton Rating Scale for Depression (Hamilton, 1960). Excluded from the study were patients with a history of psychotic disorder, bipolar disorder, substance

dependence in the last 6 months, seizure disorders, or current suicide risk. Patients were also excluded if they had a previous nonresponse to a trial of sertraline or venlafaxine of adequate dose and duration in the last year or had any significant medical conditions (e.g., heart disease, pregnancy) that would interfere with participation in the study.

Of the 156 patients who entered the study, 92 (59%) were women. Eighty-one (52%) patients identified themselves as having a minority racial background with 70 (45%) African American, 3 (2%) Asian, and 8 (5%) Latino/a. Patients in this study were primarily of lower

socioeconomic level (i.e., n 5 119, 76%; with incomes under $30,000) and reporting a mean of 13.5 (SD 5 3.5) years of education. Comorbidity was common, and 76% (n 5 119) of the patients met criteria for at least one other Axis I disorder in addition to MDD, and 47%

(n 5 73) met criteria for at least one Axis II personality disorder.

Procedures

Following a detailed phone screen, potential patients attended an initial evaluation interview during which the study was explained and an informed consent was signed. Once agreeing to participate in the study, patients met with a trained diagnostician who conducted a formal

diagnostic interview. Patients meeting the inclusion criteria for the study then completed several self-report measures related to beliefs about depression, treatment preferences, previous experience with mental health treatment, and attitudes and expectations for

treatment.

Measures

Reasons for Depression Questionnaire. The Reasons for Depression Questionnaire (RFD; Addis et al., 1995) assesses patient’s subjective understanding of the causes of their depression, including characterological factors, achievement, intimacy, interpersonal conflict, existential, childhood difficulties, physical problems, and relationship factors (Addis et al.,

1995). In this study, a shortened version of the RFD was used (Leykin, DeRubeis, Shelton et al., 2007) that included 13 items from the original RFD with an additional item (‘‘I was born to be this way’’). Prior to treatment assignment, patients were asked to rate how much

they believed each item was responsible for their depression using a 4-point Likert scale (0 5 definitely not a reason, 1 5 probably not a reason, 2 5 probably a reason, 3 5 definitely a reason).

The RFD was completed by 145 patients. To group RFD items into meaningful subscales, a principal axis factor analysis with a promax rotation was conducted. Four factors explained 53% of the variance among items: relationship (a 5 .90), childhood (a 5 .74), characterological (a 5 .67), and biological (a 5 .53). Two items were excluded as they loaded on more than one factor with less than the recommended 0.1 difference between the highest and second highest factor loading (Swisher, Beckstead, & Bebeau, 2004). Leykin and colleagues (Leykin, DeRubeis, Shelton et al., 2007) found that these two items formed a fifth factor, which they

called the intimacy factor, but the factor analysis conducted with this dataset did not support a fifth factor. Relationship, childhood, characterological, and biological subscale scores were created by computing the means of the items that the factor analysis identified as belonging to

the discrete factors. The RFD complexity score consisted of the mean of all 12 RFD items that were retained after our factor analysis (a 5 .71).

Attitudes and Expectations Questionnaire. This questionnaire (adapted from Elkin et al., 1985; Sholomskas, 1990) is a face-valid measure of patients’ attitudes toward different

types of treatments and interventions. Two items on this measure were of interest to this study.

542 Journal of Clinical Psychology, June 2011

The first served as our measure of treatment preferences and was a forced-choice dichotomous question of whether the patient preferred to receive either psychotherapy or medication

(‘‘In coming in for treatment of depression, I would prefer to receive [circle one answer]: [a] drug treatment or [b] talking treatment’’). The other item served as our measure of previous experience in treatment. Patients were asked to indicate the number of previous times in their

life they had previously undergone psychotherapy (‘‘How many times have you tried psychotherapy/counseling [at least two sessions] to help you with your problems?’’) and, separately, the number of times they had tried antidepressant medication (‘‘How many times have you been placed on medication to help with your problems?’’).

Data Analysis

For the purposes of this article, analyses were conducted on available data. Therefore, the

number of subjects varied between analyses based on how many patients had completed each measure. Independent sample t tests were used to assess possible differences in the any of the RFD factors based on gender, race, or treatment preference. Independent sample t tests were also used to determine whether the number of previous courses of psychotherapy differed

based on patients’ treatment preference. Regression analyses were employed to test whether previous experience with psychotherapy predicted greater endorsement of psychosocial causes of depression.

Results

Beliefs About the Causes of Depression

As can be seen in Table 1, across the sample the RFD complexity score and subfactors

averaged between 1 and 1.5, indicating that the factor was rated as somewhere between probably not a reason (1.0) and probably a reason (2.0). We used independent sample t tests to assess for gender differences on any of the RFD factors. Contrary to expectations, no

significant differences were found between males and females on any of the factors or the complexity scores (all ps4.25).

We then used independent sample t tests to determine whether there were any differences in RFD factors by minority status (see the last two columns of Table 1). As predicted, ethnic

minorities endorsed biological (d 5 0.39) and characterological (d 5 0.42) causes of depression at significantly lower levels than Caucasian individuals with all other comparisons nonsignificant. We also tested for an interaction between race and gender in an ANOVA

predicting each of the RFD factors and complexity scores separately, but all interaction tests were nonsignificant (ps4.40)

We then conducted regression analyses to test whether previous experience with

psychotherapy predicted greater levels of endorsement of psychosocial causes of depression.

Table 1 Descriptive Statistics for Reasons for Depression Factors for the Entire Sample and by Gender and Minority Status

Gender Minority status

RFD factor Total Women Men Minority Caucasian

Complexity 1.32 (0.49) 1.32 (0.53) 1.31 (0.41) 1.28 (0.51) 1.36 (0.45)

Relationship 0.96 (0.97) 0.94 (0.98) 0.99 (0.98) 1.00 (0.97) 0.91 (0.98)

Childhood 1.55 (0.82) 1.61 (0.87) 1.46 (0.71) 1.67 (0.81) 1.43 (0.81)

Characterological 1.42 (0.67) 1.39 (0.70) 1.46 (0.62) 1.28 (0.66) 1.56 (0.66)

Biological 1.29 (0.80) 1.34 (0.81) 1.21 (0.79) 1.14 (0.84) 1.45 (0.74)

Note. RFD 5 Reasons for depression. Table values are mean scores, and values in parentheses are

standard deviations.

543Beliefs About Depression

Patients had been in psychotherapy an average of 1.65 (SD 5 2.20) times in their lifetime. Number of previous courses of psychotherapy did not predict endorsement of relation-

ship causes of depression, r 5 .14, F(1, 130) 5 2.66, po.11; childhood causes, r 5 .06, F(1, 131) 5 .496, po.48; nor complexity scores, r 5 .12, F(1, 131) 5 1.92, po.17. However, greater number of previous courses with psychotherapy did predict greater levels of

characterological causes, r 5 .21, F(1, 131) 5 5.83, po.02, and biological causes r 5 .24, F(1, 131) 5 7.57, po.007. We also tested whether previous experience with medication predicted endorsement of biological causes of depression. Patients reported an average of 1.17 (SD 5 1.55) previous trials with medication. Greater number of previous courses of

medication did not predict relationship, r 5 .02, F(1, 135) 5 .03, po.86; or childhood, r 5 .12, F(1, 136) 5 1.87, po.17, causes of depression. However, greater number of previous courses of medication significantly predicted greater endorsement of biological, r 5 .22,

F(1, 136) 5 6.74, po.01, and of characterological causes of depression, r 5 .20, F(1, 136) 5 5.66, po.02. Greater number of previous courses of medication also predicted higher complexity scores, r 5 .19, F(1, 136) 5 5.20, po.02.

Treatment Preferences

At intake, 151 patients indicated their preference for either psychotherapy or medication.

Fifty-nine percent (n 5 89) of patients in this subsample preferred psychotherapy whereas 41% (n 5 62) preferred medication. Broken down by gender, 58% (n 5 53) of the 91 women in the subsample and 60% (n 5 36) of the 60 men preferred psychotherapy over medication. However, contrary to predictions, there were no significant gender differences in treatment

preference, w2(1) 5 0.05, ns. Sixty-two percent (n 5 48) of the 77 individuals of minority status preferred psychotherapy

over medication, whereas 55% (n 5 41) of the 74 Caucasian individuals preferred

psychotherapy over medication. This was not significantly different, w2(1) 5 0.75, ns. We also tested for an interaction between gender and race to predict treatment preference in a logistic regression; however, this was nonsignificant as well, exp(b) 5 .97, w2(1) 5 .002, po.97.

The relation between treatment preference and previous experience with psychotherapy was examined. Contrary to predictions, patients preferring psychotherapy reported fewer previous courses of psychotherapy (M 5 1.23, SD 5 1.74) than did patients preferring medication (M 5 2.30, SD 5 2.64; d 5 0.49, df 5 135, po.005).

We also examined the relation between treatment preference and previous experience with medication. Patients preferring medication reported an average of 1.43 (SD 5 1.79) courses of medication in the past, whereas patients preferring psychotherapy reported an average of 0.99

(SD 5 1.34) courses of medication; however, this difference was not significant (d 5 0.29, df 5 142, po.09).

Relation of Beliefs About the Causes of Depression and Treatment Preferences

Table 2 presents the mean RFD factor and complexity scores for patients preferring psychotherapy and those preferring pharmacotherapy. Consonant with our predictions, those

patients who preferred psychotherapy endorsed childhood reasons for depression more strongly than those who preferred medication (d 5 0.47, df 5 140, po.008). However, no differences were found between those preferring psychotherapy and those preferring medication on the relationship (d 5 0.25, df 5 139, ns), characterological (d 5 0.12, df 5 140,

ns), or biological (d 5 0.06, df 5 140, ns) factors. As one of the two items of the biological factor relates to genetic causes of depression (which may not be seen as affected by medication), a t test was conducted with the biochemical item alone. This test approached

significance (d 5 0.32, df 5 132, po.07), such that those preferring medication endorsed the biochemical imbalance item more strongly than those preferring psychotherapy. As predicted, we found that patients preferring psychotherapy endorsed more complex causes of depression

than those preferring psychopharmacotherapy (d 5 0.39, df 5 140, po.02).

544 Journal of Clinical Psychology, June 2011

Discussion

This study represents one of the first examinations of the association between patient beliefs about the causes of their depression and their preferences for treatment. Further, with more

than 50% of the sample comprised of ethnic minorities, we were able to examine racial differences in beliefs about the causes of depression and treatment preferences, which is not possible in many clinical settings.

Beliefs About the Causes of Depression

In contrast to previous studies (Angst et al., 2002; Atkinson et al., 1991; Robbins & Tanck,

1991; Schweizer et al., 2009), no differences were found between women and men in terms of beliefs about the causes of their depression. As previous examinations of the relation between gender and beliefs about the causes of depression have used different measures and populations, it is perhaps not so surprising that our findings differed from theirs. Our finding

that minority patients were less likely to endorse biological reasons is consistent with the published literature (e.g., Givens et al., 2007; Schnittker et al., 2000). However, we did not predict our finding that minority status individuals would endorse characterological reasons

for depression at lower levels than Caucasian individuals. Previous experience with psychotherapy or medication both significantly predicted greater

endorsement of the characterological and biological causes, but not relationship or childhood

causes. Interestingly, Schweizer and colleagues found a very similar relationship, such that patients endorsing biological and characterological beliefs more highly were more likely to have previously received some sort of treatment for their current episode of depression (Schweizer et al., 2009). Speculating as to why this relation was found, patients who greatly

endorse these two factors may represent a group that has tried different treatments for depression, failed to improve, and accepted characterological and biological explanations, which are somewhat more fundamental to their sense of self than the possibly more tractable

relationship and childhood causes.

Treatment Preferences

Neither gender nor minority status were related to treatment preference. There are several possible explanations for these findings. Other studies that have found racial differences in preference for treatment (e.g., Cooper et al., 2003) were conducted in a primary care or

otherwise naturalistic setting. Considering there is some evidence that minority individuals are less likely to seek out specialty mental health services than Caucasians (e.g., Cooper-Patrick, Crum, & Ford, 1994), our patients may differ in some way from the population at large.

Furthermore, our patients may not have had strong preferences (as strength of preference was

Table 2 Descriptive Statistics for Reasons for Depression Factors by Treatment Preferences

Treatment preference

RFD factor Psychotherapy Medication

Complexity 1.40 (0.46) 1.21 (0.50)

Relationship 1.07 (1.01) 0.82 (0.93)

Childhood 1.71 (0.79) 1.33 (0.84)

Characterological 1.46 (0.67) 1.38 (0.67)

Biological 1.32 (0.83) 1.27 (0.77)

Note. RFD 5 Reasons for depression. Table values are mean scores, and values in parentheses are

standard deviations.

545Beliefs About Depression

not measured), and those who had a strong preference may not have enrolled in the study with the possibility of assignment to a nonpreferred treatment modality.

The finding that patients preferring medication and those preferring psychotherapy differed in previous experience with psychotherapy but not previous experience with medication could be due to several factors. Perhaps patients are more likely to accept that the medication that

they would be receiving as part of the study would be sufficiently different from the medication they had tried in the past. For patients who do not know much about psychotherapy, all different types of therapies may be viewed as similar. The patients who preferred medication after several attempts at psychotherapy may represent a group of people who have tried

therapy in the past, failed to improve, and decided that therapy was not effective for them. Unfortunately, we did not have the data available as to whether previous treatment courses resulted in improvement or not.

Relation of Beliefs About the Causes of Depression and Treatment Preferences

Our hypothesis that patients preferring medication and those preferring psychotherapy would differ in beliefs about the causes of depression was partially supported. It makes intuitive sense that one’s beliefs about the causes of mental illness would affect one’s preference for

treatment, and there is some evidence to support the idea that when etiological beliefs are congruent with a treatment rationale, the treatment is rated as more helpful (Iselin & Addis, 2003). Of course, there may not be perfect synchrony between beliefs about the causes of one’s depression and preferences for treatment. For example, many patients may believe that their

depression has a biological cause, but may be wary of the side effects common in pharmacotherapy and therefore prefer psychotherapy.

Patients preferring psychotherapy endorsed childhood reasons for depression more strongly

than those preferring medication. As some forms of psychotherapy directly target such problems (e.g., psychodynamic therapies such as object relations therapy), patients may view such therapeutic orientations as potentially more beneficial than medication. Whereas we

expected that depressed patients who preferred medication would endorse biological causes of depression more strongly than those who preferred psychotherapy, this was not found to be the case. Perhaps this was because there are biological causes for depression that can be

treated by medication (such as biochemical imbalance), whereas other biological causes (such as genetic contributions) are not treated in such a manner. In fact, when we examined whether the biochemical item alone predicted treatment preference, the test approached significance, providing some evidence as to why the biochemical factor as a whole was not associated with

treatment preference. Contrary to predictions, those preferring psychotherapy did not endorse relationship and

characterological causes of depression more than those preferring medication. This is

somewhat surprising, as many types of psychotherapy seek to ameliorate problems in these domains. One possible explanation is that patients may have a stereotypical view of psychotherapy in which the patients’ childhood is the only focus, and current relationships and

ways of thinking are ignored. It is also possible that patients simply did not feel that those causes of depression were amenable to change. Unfortunately, the study did not collect detailed information on what patients expected to happen in psychotherapy.

Finally, as expected, patients preferring psychotherapy endorsed more complex beliefs

about the causes of depression (as represented by endorsing more items more strongly on the RFD) than those preferring medication. People who feel that their depression has a more complicated etiology may be less likely to believe that their underlying problems cannot be

solved with just medication.

Limitations

There are a number of limitations to the present study, and as such, this should be considered a preliminary investigation in a relatively newly studied area. As a number of analyses were

conducted, one must interpret significant findings with caution, as type II errors are of

546 Journal of Clinical Psychology, June 2011

concern. With regards to the factor analyses of the RFD, three of the four subscales had low to very low alphas. It must also be noted that mean scores on the subscales generally fell

between 1 and 2 (with the exception of the biological subcale, which had a mean a little under 1). Thus, most patients felt that the reasons were in between ‘‘probably not a reason’’ and ‘‘probably a reason.’’ Therefore, although some patients did find the reasons listed on the scale

to describe the reasons they were suffering from depression, many did not. These mean scores were similar to those found by Leykin, DeRubeis, Shelton et al. (2007) who also used the shortened version of the RFD. Future research could explore the relationships between variables found in the present study utilizing a measure listing more comprehensive reasons (as

the full version of scale does) and with a more nuanced Likert scale that would allow for a clearer picture as to what patients thought was causing their depression.

One must also be cautious in the interpretation of complexity score. Thus far, there has

been little study of what endorsing a greater number of reasons for depression at higher levels might actually represent. Complexity is simply one possible explanation, and our index (mean RFD item scores) was only a crude measure of complexity that did not distinguish

between the number of separate causes of depression endorsed and the intensity with which each reason was endorsed (for discussions of the measurement of complexity, see Locke, 2003; McCarthy, Connolly Gibbons, & Barber, 2008). Endorsing numerous causes may represent a

realistic assessment of stressors in one’s life, or may be excuse giving and reluctance to change (Addis & Carpenter, 1999).

Future Directions

In closing, it is apparent that future studies could benefit by examining in more detail the strength of patients’ preferences, what patients know about therapy, and the quality and duration of previous treatment as they relate to patients’ beliefs about the causes of their

depression and their preferences for treatment. Both the specific causes endorsed and the complexity of causes endorsed have been previously shown to affect various aspects of psychotherapy process such as homework compliance, attendance, and outcome, making

these beliefs of potential importance to psychotherapy researchers (Addis & Carpenter, 1999; Addis & Jacobson, 1996; Foulks et al., 1986). More investigation into whether higher mean score on measures of patient beliefs about the causes of their depression truly represents complexity is also necessary, as current conclusions regarding this construct are fairly

speculative. Studies have shown that preference can affect both the process and outcome of

psychotherapy (e.g., Iacoviello et al., 2007; Leykin, DeRubeis, Gallop et al., 2007; Lin

et al., 2005). One recent meta-analysis found a small, but significant effect of treatment preference on outcome. Specifically, patients matched to their preferred treatment were found to experience a greater chance of improvement than those who were not matched to their

preferred treatment (Swift & Callahan, 2009). Previous research examining patient treatment preference has tended to focus on demographic variables that are not modifiable. As beliefs about the causes of depression may be amenable to change, future research could examine

whether therapists who address these beliefs may be able to counter the effect of a belief- treatment incongruence. Greater study of how treatment preferences and beliefs about the causes of depression may interact to affect psychotherapy process and outcome, and how amenable they are to change is warranted.

References

Addis, M.E., & Carpenter, K.M. (1999). Why, Why, Why?: Reason-giving and rumination as predictors of

response to activation- and insight-oriented treatment rationales. Journal of Clinical Psychology, 55,

881–894.

Addis, M.E., & Jacobson, N.E. (1996). Reasons for depression and the process and outcome of cognitive-

behavioral psychotherapies. Journal of Consulting and Clinical Psychology, 64, 1417–1424.

547Beliefs About Depression

Addis, M.E., Truax, P., & Jacobson, N.S. (1995). Why do people think they are depressed?: The Reasons

for Depression Questionnaire. Psychotherapy, 32, 476–483.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.).

Washington, DC: Author.

Angst, J., Famma, A., Gastpar, M., Lépine, J.-P., Mendlewicz, J., & Tylee, A. (2002). Gender differences

in depression: Epidemiological findings from the European DEPRES I and II studies. European

Archives of Psychiatry and Clinical Neuroscience, 252, 201–209.

Atkinson, D.R., Worthington, R.L., Dana, D.M., & Good, G.E. (1991). Etiology beliefs, preferences for

counseling orientations, and counseling effectiveness. Journal of Counseling Psychology, 38, 258–264.

Brown, C., Dunbar-Jacob, J., Palenchar, D.R., Kelleher, K.J., Bruehlman, R.D., Sereika, S., et al. (2001).

Primary care patients’ personal illness models for depression: A preliminary investigation. Family

Practice, 18, 314–320.

Chambless, D.L., & Ollendick, T.H. (2001). Empirically supported psychological interventions:

Controversies and evidence. Annual Review of Psychology, 52, 685–716.

Churchill, R.K.M., Gretton, V., Chilvers, C., Dewey, M., Duggan, C., & Lee, A. (2000). Treating

depression in general practice: Factors affecting patients’ treatment preferences. British Journal of

General Practice, 50, 905–906.

Cipriani, A., Brambilla, P., Furukawa, T.A., Geddes, J., Gregis, M., Hotopf, M., et al. (2005). Fluoxetine

versus other types of pharmacotherapy for depression (review). The Cochrane Database of Systematic

Reviews, 4. Art No CD004185.pub2.

Cooper, L.A., Gonzales, J.J., Gallo, J.J., Rost, K.M., Meredith, L.S., Rubenstein, L.V., et al. (2003). The

acceptability of treatment for depression among African-American, Hispanic, and White primary care

patients. Medical Care, 41, 479–489.

Cooper-Patrick, L., Crum, R.M., & Ford, D.E. (1994). Characteristics of patients with major

depression who received care in general medical and specialty mental health settings. Medical Care,

32, 15–24.

Dwight-Johnson, M., Sherbourne, C.D., Liao, D., & Wells, K.B. (2000). Treatment preferences among

depressed primary care patients. Journal of General Internal Medicine, 15, 527–534.

Elkin, I., Parloff, M.B., Hadley, S.W., & Autry, J.H. (1985). NIMH treatment of depression collaborative

research program: Background and research plan. Archives of General Psychiatry, 42, 305–316.

Foulks, E.F., Persons, J.B., & Merkel, R.L. (1986). The effect of patients’ beliefs about their illnesses on

compliance in psychotherapy. American Journal of Psychiatry, 143, 340–344.

Fournier, J.C., DeRubeis, R.J., Hollon, S.D., Dimidjian, S., Amsterdam, J.D., Shelton, R.C., et al. (2010).

Antidepressant drug effects and depression severity: A patient-level meta-analysis. Journal of the

American Medical Association, 303, 47–53.

Givens, J.L., Houston, T.K., Van Voorhees, B.W., Ford, D.E., & Cooper, L.A. (2007). Ethnicity and

preferences for depression treatment. General Hospital Psychiatry, 29, 182–191.

Goldstein, B., & Rosselli, F. (2003). Etiological paradigms of depression: The relationship between

perceived causes, empowerment, treatment preferences, and stigma. Journal of Mental Health, 12,

551–563.

Hamilton, M. (1960). A rating scale for depression. Journal of Neurological and Neurosurgical Psychiatry,

23, 56–62.

Iacoviello, B.M., McCarthy, K.S., Barrett, M.S., Rynn, M., Gallop, R., & Barber, J.P. (2007). Treatment

preferences affect the therapeutic alliance: Implications for randomized controlled trials. Journal of

Consulting and Clinical Psychology, 75, 194–198.

Iselin, M.-G., & Addis, M.E. (2003). Effects of etiology on perceived helpfulness of treatments for

depression. Cognitive Therapy and Research, 27, 205–222.

Leichsenring, F., Rabung, S., & Leibing, E. (2004). The efficacy of short-term psychodynamic

psychotherapy in specific psychiatric disorders: A meta-analysis. Archives of General Psychiatry, 61,

1208–1216.

Leykin, Y., DeRubeis, R.J., Gallop, R., Amsterdam, J.D., Shelton, R.C., & Hollon, S.D. (2007). The

relation of patients’ treatment preferences to outcome in a randomized clinical trial. Behavior Therapy,

38, 209–217.

Leykin, Y., DeRubeis, R.J., Shelton, R.C., & Amsterdam, J.D. (2007). Changes in patients’ beliefs about

the causes of their depression following successful treatment. Cognitive Therapy and Research, 31,

437–449.

548 Journal of Clinical Psychology, June 2011

Lin, P., Campbell, D.G., Chaney, E.F., Liu, C.-F., Heagerty, P., Felker, B.L., et al. (2005). The influence

of patient preference on depression treatment in primary care. Annals of Behavioral Medicine, 30,

164–173.

Locke, K.D. (2003). H as a measure of complexity of social information processing. Personality and Social

Psychology Review, 7, 268–280.

McCarthy, K.S., Connolly Gibbons, M.B., & Barber, J.P. (2008). The relation of rigidity across

relationships with symptoms and functioning: An investigation with the revised Central Relationship

Questionnaire. Journal of Counseling Psychology, 55, 346–358.

Meyer, B., & Garcia-Roberts, L. (2007). Congruence between reasons for depression and motivations for

specific interventions. Psychology and Psychotherapy: Theory, Research, and Practice, 80, 525–542.

Millet, P.E., Sullivan, B.F., Schwebel, A.I., & Myers, L.J. (1996). Black Americans’ and White Americans’

views of the etiology and treatment of mental health problems. Community Mental Health Journal, 32,

235–242.

Robbins, P.R., & Tanck, R.H. (1991). Gender differences in the attribution of causes for depressed

feelings. Psychological Reports, 68, 1209–1210.

Schnittker, J. (2003). Misgivings of medicine? African Americans’ skepticism of psychiatric medication.

Journal of Health and Social Behavior, 44, 506–524.

Schnittker, J., Freese, J., & Powell, B. (2000). Nature, nurture, neither, nor: black-white differences in

beliefs about the causes and appropriate treatment of mental illness. Social Forces, 78, 1101–1132.

Schweizer, S., Peeters, F., Huibers, M., Roelofs, J., van Os, J., & Arntz, A. (2010). Does illness attribution

affect treatment assignment in depression? Clinical Psychology and Psychotherapy, 17, 418–426.

Sholomskas, D.E. (1990). Interviewing methods. In B.B. Wolman & G. Stricker (Eds.), Depressive

disorders: Facts, theories, and treatment methods (pp. 231–247). Oxford: Wiley.

Swift, J.K., & Callahan, J.L. (2009). The impact of client treatment preferences on outcome: A meta-

analysis. Journal of Clinical Psychology, 65, 368–381.

Swisher, L.L., Beckstead, J.W., & Bebeau, M.J. (2004). Factor analysis as a tool for survey analysis using

a professional role orientation inventory as an example. Physical Therapy, 84, 784–799.

549Beliefs About Depression