Content uploaded by M. Kati Lear
Author content
All content in this area was uploaded by M. Kati Lear on Dec 29, 2021
Content may be subject to copyright.
Running head: PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT 1
NON-DISCLOSURE
Psychological Inflexibility Prospectively Predicts Client Non-Disclosure in Outpatient
Psychotherapy
Mary K. Lear and Jason B. Luoma
Portland Psychotherapy Clinic, Research, and Training Center
Christina Chwyl
Department of Psychology, Drexel University
This research was supported by internal funding from the Portland Psychotherapy Clinic,
Research, & Training Center.
Correspondence concerning this article should be addressed to Jason Luoma Ph.D., Portland
Psychotherapy, 3700 N Williams, Portland, OR 97227, USA. Email:
jbluoma@portlandpsychotherapyclinic.com
Original date submitted: 07/31/2020
Date resubmitted: 11/2/2020
Date accepted: 11/19/2020
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 2
Abstract
Psychological Inflexibility (PI) is reliably associated with adverse psychological outcomes but
little research has explored how PI may affect therapeutic process in psychotherapy. The current
study examined the longitudinal influence of PI measured at pre-treatment on likelihood of client
non-disclosure of treatment-relevant information at 15-week follow-up and evaluated PI against
an empirically supported predictor of non-disclosure, internalized shame. Sixty clients in
outpatient psychotherapy at an evidence-based treatment center (Mage= 35.38; 62% female)
completed self-report measures of PI and internalized shame at initiation of therapy, 3- and 9-
week follow-up and a dichotomous assessment of non-disclosure at 15-week follow-up. We
hypothesized that pre-treatment PI and internalized shame would uniquely and positively predict
likelihood of non-disclosure at 15-week follow-up. Results indicated that, when tested
simultaneously, only pre-treatment PI predicted likelihood of non-disclosure at follow-up (OR =
2.96). This is the first study to test the influence of PI on client non-disclosure and implicates PI
as a relevant pre-treatment individual difference variable for identifying clients at higher risk for
non-disclosure in psychotherapy.
Keywords: Psychological inflexibility, shame, disclosure, psychotherapy, secrets
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 3
Introduction
Client disclosure of distressing thoughts, feelings, and experiences is an integral part of
psychotherapy. Therapists depend on clients to accurately and candidly disclose relevant material
to develop their case formulation and direct therapy (Blanchard & Farber, 2016). In return, many
clients turn to their therapists for support in facing deeply personal and distressing experiences
they may have kept private for much of their lives. Indeed, disclosure of secrets in psychotherapy
has been associated with affirmative therapy experiences for clients including strengthened
therapeutic alliance (Hall & Farber, 2001), positive in-session emotional experiences such as
authenticity, relief, and connection with one’s therapist (Farber, Berano, & Capobianco, 2006),
and enhanced perceptions of session quality (Marks et al., 2019).
Despite the potential benefits of disclosure, a significant portion of clients in outpatient
psychotherapy endorse non-disclosure, defined here as the intentional concealment or omission
of information relevant to therapy, including life experiences, personal information, and relevant
psychological experiences (adapted from Baumann & Hill, 2016). Research indicates that
approximately 28–53% of clients in outpatient settings report having intentionally kept a secret
from their therapist1. Commonly withheld topics include relationships and intimacy, distressing
internal experiences or mental health symptoms, and stigmatized behaviors and experiences
including sexual behavior, substance use, insecurities, past abuse, and non-suicidal self-injury
and suicidal thoughts (e.g., Baumann & Hill, 2016; Hill et al., 1993; Love & Farber, 2019;
Marks et al., 2019; Pope & Tabachnick, 1994).
Concealment of this type of therapeutically rich material may impede therapeutic
progress due to incomplete understanding on the part of the therapist or missed opportunities for
1 Variations in study design seem to affect this estimate, as online sampling and checklist
assessments of non-disclosure have produced estimates as high as 84–93% (Blanchard & Farber,
2016; Love & Farber, 2019).
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 4
intervention. Approximately one-third of individuals who withheld relevant information from
their therapist in a large online sample reported they believed non-disclosure hurt their
therapeutic progress while only 4% of the sample believed non-disclosure helped their progress
(Love & Farber, 2019). Similarly, client non-disclosure of salient personal information, such as
important aspects of one’s cultural identity and distressing psychological symptoms, has been
prospectively associated with reduced symptom improvement at termination (Drinane et al.,
2018; Hook & Andrews, 2005). Identifying factors that predict non-disclosure may inform how
therapists can compassionately and effectively promote disclosure to inform treatment planning
and intervention among clients who have difficulty sharing personal or upsetting information.
Non-disclosure in therapy has been associated with trait self-concealment (SC; Larson et
al., 2015). SC is characterized as the broad predisposition to consciously hide intimate and
negatively-valanced personal information (Love & Farber, 2019) and has been conceptualized as
a form of psychological inflexibility (PI; Masuda et al., 2017). Described in the context of
Acceptance and Commitment Therapy (Hayes et al., 2012), PI refers to a rigid pattern of
behavioral responding that prioritizes the control of psychological reactions over the pursuit of
one’s chosen values (Bond et al. 2011). Applied to SC, it is theorized that some people endeavor
to avoid sharing potentially embarrassing personal information, at least partially because
disclosure often evokes aversive affect (e.g., Farber, Berano, & Capobianco, 2006). PI
significantly correlates with trait-level SC (r = .54, p < .01; Masuda et al., 2017). Qualitative
research on barriers to in-session disclosure also implicate PI, with one study reporting 80% of
clients who withheld secrets from their therapists did so to avoid discomfort (Baumann & Hill,
2016). However, no study to date has examined the influence of PI on non-disclosure in
outpatient therapy.
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 5
In addition to PI, shame has been identified a key contributor to client non-disclosure in
therapy (e.g., Hill et al. 1993; Hook & Andrews, 2005; Love & Farber, 2019; MacDonald &
Morley, 2001; Swan & Andrews, 2003). Large scale studies have found that 64–75% of clients
in outpatient therapy endorse shame as the primary reason for withholding secrets (Baumann &
Hill, 2016; Love & Farber, 2019). Similarly, shame-proneness (i.e. the dispositional tendency to
experience shame) predicted fewer retrospective reports of disclosure among clients treated for
eating disorders (Swan & Andrews, 2003) and depression (Hook & Andrews, 2005), as well as
lower self-reported willingness to disclose a potentially shameful secret to a counselor (DeLong
& Kahn, 2014). This pattern of findings suggests that withholding secrets in therapy may be
better understood as a behavior used to avoid shame specifically, rather than indicative of the
broader pattern of affective avoidance in PI.
The current study extends the present literature through two primary aims. First, we
empirically tested the prospective relation between PI at baseline and clients’ likelihood of
endorsing non-disclosure at 15-week follow-up. To our knowledge, this is the first study to
examine this association. Second, we tested whether internalized shame or PI was a stronger
predictor of non-disclosure at follow-up. We did not make a priori hypotheses about the relative
strength of PI and shame variables on non-disclosure as no longitudinal studies were available
for comparison. However, we did predict inverse associations between PI and shame and non-
disclosure.
As exploratory aims, we examined whether changes in PI and shame over the course of
treatment predicted non-disclosure based on the idea that improvements in these variables may
index meaningful shifts in the therapeutic alliance and thereby facilitate increased disclosure.
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 6
Additionally, we assessed ten reasons for client non-disclosure identified in prior research (Hook
& Andrews, 2005) to contextualize findings from this sample in the broader literature.
Method
Participants
All clients who attended intake sessions with therapists at an outpatient evidence-based
treatment center between 2015 and 2019 were informed about the opportunity to enroll in
ongoing research and were invited to elect to be contacted. Clients who expressed interest were
phoned and emailed about enrollment in this study. Eighty-two clients enrolled in the study and
completed the baseline assessment following their intake session and prior to their second
therapy session. Seven participants dropped out of the study at the 3-week assessment, six
discontinued at the 9-week assessment, and an additional two dropped out by the 15-week
follow-up. Of the remaining sixty-seven participants who completed the 15-week follow-up, five
were excluded due to missing responses on the disclosure questions at the 15-week assessment,
one for attending only couples therapy sessions during the assessment period, and one for not
attending any therapy sessions following the intake session. Our final sample consisted of 60
participants2 across 15 therapists (11 licensed clinical psychologists, 2 clinical psychology
doctoral students, and 2 licensed clinical social workers). For sample characteristics, see Table 1.
Measures
Demographics
Participants completed a questionnaire at baseline assessing demographic information.
Information regarding identity of their therapist, primary DSM-5 diagnosis, and total number of
therapy sessions was obtained via chart review.
2 Participants who completed the study (N = 60) did not differ from participants who did not
complete the study (n = 22) on PI, shame, age, sex, race, or education variables (ps > .05).
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 7
Psychological Inflexibility
Psychological Inflexibility was assessed using the Acceptance and Action Questionnaire–
2 (Bond et al., 2011; AAQ-2; αbaseline = .89; α3-week = .92; α9-week = .91), a widely used 7-item self-
report questionnaire on which higher scores are intended to reflect higher PI. The AAQ-2 was
found to have good test-retest reliability (Bond et al., 2011) and strong content validity and
internal consistency by a recent psychometric review (McAndrews et al., 2019). In the present
study, baseline AAQ-2 scores ranged from 8–49, which is consistent with scores previously
observed in transdiagnostic clinical samples (e.g., 7–49; Pinto et al., 2015).
Internalized Shame
Internalized shame was assessed using the Internalized Shame Scale (Cook, 1987; ISS; α
= .97; α3-week = .98; α9-week = .97). The ISS is comprised of 24 self-report items and has previously
demonstrated good test-retest reliability and construct validity (Cook, 1987). In the current study,
the items were rated over the last week to increase sensitivity to change over time. Baseline ISS
scores ranged from 0–91 and fell within one standard deviation of a comparative transdiagnostic
outpatient sample (Vikan et al., 2010).
Client Non-disclosure
Client non-disclosure at follow-up was assessed via two dichotomous yes/no questions
adapted from Hook and Andrews (2005): (1) “Some people find it difficult to disclose
information about themselves to their therapist. Have you at any time had any troubling
symptoms or behaviors relating to your mental health that you have not disclosed to your
therapist?” and (2) “Are there any other issues or past experiences of a distressing nature that
may be relevant to your treatment that you have not disclosed to your therapist?” A positive
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 8
endorsement of either item was coded as client non-disclosure and non-endorsement of both
items was coded as client disclosure.
Participants were also asked to endorse the relevance of ten common reasons for non-
disclosure cited in prior literature to their decisions to withhold secrets in therapy. Participants
were instructed to check as many of the reasons as they felt applied.
Procedure
Participants met with a trained research assistant for the baseline assessment session
following the intake session with their therapist. During this session, informed consent for the
study was obtained and participants completed baseline AAQ-2 and ISS assessments.
Participants were instructed to schedule the baseline assessment prior to attending their first
psychotherapy appointment and 58 participants were able to do so. There were two exceptions.
Review of electronic medical records indicated that one participant completed the baseline
assessment the same day as their first post-intake psychotherapy session; it is unclear which
appointment occurred earlier in the day. Another participant engaged in one post-intake session
prior to their baseline assessment.
All remaining assessments were completed online. Participants were prompted via email
to complete surveys at 3-, 9-, and 15-weeks after study initiation. The AAQ-2 and ISS were
completed in the baseline, 3-, and 9-week surveys and non-disclosure was assessed at 15-weeks.
Participants were compensated $15 for baseline, $15 for the 15-week, and $5 for 3- and 9-week
assessments. All study procedures were approved by an Institutional Review Board.
Analytic Plan
Analyses were conducted in SPSS version 22 or Stata version 15. Data were screened
prior to analyses to ensure statistical assumptions were met. The longitudinal influences of PI
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 9
and internalized shame on non-disclosure were tested through a series of binomial logistic
regression models using maximum likelihood estimation. Models were specified to include non-
disclosure status at 15-week follow-up as the binary outcome variable (disclosure = 0; non-
disclosure = 1). Based on research suggesting that longer therapy duration facilitates disclosure
(Farber, 2003), all models controlled for number of therapy sessions attended during the study
period. To test whether reductions in PI and shame predicted non-disclosure, a parameter
representing the linear slope in change in AAQ-2 and ISS over the course of therapy was
estimated for each individual using a Stata macro that regressed time (defined as assessment
number, not week) on the dependent variable and then exported the slope estimate for use in the
logistic models. Given the exploratory nature of this hypothesis, we also repeated models testing
quadratic slopes, but they did not add to the prediction for any models and are therefore not
reported. Cohen’s d was used as an estimation of effect size for AAQ-2 and ISS scores from
baseline to 9-week follow up. Magnitude was interpreted based on traditional benchmarks (.20 =
small, .50 = medium, .80 = large; Cohen, 1988). Odds ratios were interpreted as measures of
effect size for each variable in predicting client non-disclosure (Tabachnick & Fidell, 2013) and
are presented as log odds. AAQ-2 and ISS scores were standardized to increase interpretability
of odds ratios (Browner, 2006).
Results
Preliminary Analyses
Data screening procedures indicated no violations of statistical assumptions specific to
logistic regression models. Bivariate correlations revealed a notably strong correlation between
ISS and AAQ-2 scores measured at baseline (r = .81, p < .001). Both variables were retained in
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 10
specified models due to the theoretical distinction between global PI and internalized shame
(Bond et al., 2011; Cook et al., 1987).
Mean ISS scores were 38.86 (SD = 24.69) and 30.00 (SD = 21.72) at baseline and 9-week
follow-up, respectively (Mchange = 8.00, SD = 16.34; t = 3.73, p < .001; d = 0.38). Mean AAQ-2
scores were 28.73 (SD = 9.77) at baseline and 26.17 (SD = 1.16) at 9-week follow-up (Mchange =
2.27, SD = 6.14; t59 = 2.84, p = .006; d = .25). Forty-eight percent of participants (n = 29) were
classified as non-disclosers by endorsing at least one non-disclosure question (see Table 1 for
number of responses by question).
Logistic Regression Models
Logistic regression models are detailed in Table 2. Our first set of models assessed
whether PI predicted non-disclosure at follow-up. We first examined whether baseline PI
predicted non-disclosure at follow-up after controlling for number of sessions attended and found
higher PI predicted greater non-disclosure (β = 1.18, p = .002; OR: 3.25; 95% CI for OR [1.57,
6.74]. We then entered our variable representing each individual’s linear change in PI (i.e.,
slope) over the course of treatment and found that it did not add significantly to the model (β =
-.06, p = .839; OR: 1.06; 95% CI for OR [0.60, 1.89]).
Our second set of models tested whether shame predicted non-disclosure at follow-up.
We first examined whether baseline shame predicted non-disclosure at follow-up after
controlling for number of attended sessions and found that shame predicted higher odds of non-
disclosure (β = .81, p = .008; OR: 2.39; 95% CI for OR [1.26, 4.53]). We then entered the
variable representing each individual’s linear change in shame (i.e., slope) over the course of
treatment and found that it did not add significantly to the model (β = -.09, p = .764; OR: .92;
95% CI for OR [0.53, 1.61]).
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 11
Our final model compared whether PI or shame measured at baseline was the stronger
predictor of disclosure at follow-up after controlling for total number of sessions attended.
Results indicated that only PI continued to significantly predict client disclosure after controlling
for shame and number of sessions attended (β = 1.20, p = .046; OR: 2.96; 95% CI for OR [1.02,
8.56]).
Reasons for Non-Disclosure
Among the 29 participants who endorsed non-disclosure during therapy, the most
commonly endorsed reasons were “too ashamed” (30%), “fear of negative judgment” (23%), and
that the subject matter was “too painful to talk about” (22%) or “too private” (22%). Other
reasons included that participants did not believe the information was important to therapy (20%)
or that participants felt too guilty (15%), feared rejection (10%) or felt treatment was too short
(7%). A small percentage of participants identified therapist-related variables as the reason for
non-disclosure. Five percent reported non-disclosure due to lack of trust in their therapist and
one percent identified not disclosing due to lack of perceived empathy from their therapist. The
number of reasons reported by participants ranged from 1–7 (median = 3), with most participants
(n = 22; 76%) endorsing two or more reasons.
Discussion
This study is the first, to our knowledge, to investigate PI as a predictor of client non-
disclosure in outpatient therapy. This study extends the conceptualization of PI as relevant to SC
(e.g., Masuda et al., 2017) by establishing a temporal link between PI and specific instances of
client non-disclosure in outpatient therapy. Moreover, by examining PI over a 15-week period
and testing its influence against internalized shame, our findings guard against important
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 12
alternative explanations for the observed relation, including recall bias or facilitation of the
relation by shame, a known correlate of client disclosure.
Results were largely in line with study hypotheses. PI measured at the beginning of
therapy predicted likelihood of client non-disclosure retrospectively reported at 15-week follow-
up. Further, PI emerged as a more robust predictor of clients’ propensity for non-disclosure than
shame in this sample, which may help explain equivocal findings regarding the unique
associations between varying types of affect, such as shame, guilt, or embarrassment in client
disclosure (Farber, 2003).
Our findings also help to contextualize the larger body of qualitative research in which
the majority of reasons for non-disclosure reported by clients in past research (e.g., Farber, 2003)
and in the current study, implicate PI as a barrier to disclosing secrets in therapy. For example,
many clients endorse avoidance of various psychological reactions (e.g., embarrassment, shame,
or anxiety) or perceived undesirable outcomes (e.g., therapist judgment, hurt, or disappointment,
or concerns over change in treatment plan, such as being hospitalized or referred to a more
intensive treatment setting) as central to their decisions to withhold secrets (Blanchard & Farber,
2020; DeLong & Kahn, 2014; Farber, 2019; Love & Farber, 2019). Reasons for non-disclosure
endorsed in this study echo this pattern—the top five reasons for non-disclosure reported by
clients involved apprehension around affect (i.e., too ashamed, too guilty, too painful, too
private) or outcome (i.e., negative evaluation) associated with the disclosure. By distilling these
reasons down to a common function— psychological inflexibility—therapists may be better
poised to detect client non-disclosure in the therapy process.
The use of the AAQ-2 (Bond et al., 2011; McAndrews et al., 2019) in this study
represents an important first step to establish PI as relevant to client non-disclosure. The AAQ-2
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 13
was designed as a general measure of PI (Bond et al., 2011; McAndrews et al., 2019) and its
breadth makes it difficult to discern which PI processes (e.g., experiential avoidance versus
defusion) are most central in non-disclosure. Additionally, the AAQ-2 has been criticized over
concerns of discriminant validity due to its correlation with measures of psychological distress
(e.g., Tyndall et al., 2019) and, indeed, a high correlation was observed in this study between the
AAQ-2 and ISS, a measure of negative affect. This high correlation may have reduced the
precision of parameter estimates as observed in the wide confidence interval of the odds ratio for
the AAQ-2 in Model 5. While not eliminating broader concerns about the discriminant validity
of the AAQ-2, the AAQ-2 contributed additional predictive variance beyond the ISS mitigates
this concern somewhat. We encourage future researchers to extend our findings by clarifying the
association of PI and its sub-categories, particularly experiential avoidance, and non-disclosure
using alternative validated measures, such as the Brief Experiential Avoidance Questionnaire
(Gamez et al., 2014).
Changes on AAQ-2 and ISS scores across treatment were consistent with small and
statistically reliable effects, suggesting clients on average made small but potentially meaningful
improvements on PI and shame. In addition, changes on the AAQ-2 and ISS were smaller than
observed in studies targeting PI (Gloster et al., 2020) and shame (Judge et al., 2012; Luoma et
al., 2012), which is perhaps not surprising in our mixed treatment sample.
Despite the observed changes in AAQ-2 and ISS scores, neither changes in PI, shame,
nor the total number of therapy sessions predicted likelihood of non-disclosure during therapy.
This suggests that clients’ likelihood of keeping important secrets was not impacted by this level
of reduction in PI or shame during treatment, but was better predicted through measurement of
pre-existing PI and shame assessed at intake. Replication of this finding is merited to test
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 14
whether larger improvements on these variables could affect disclosure across treatment. It is
also possible that more proximal relational factors, such as therapeutic alliance, may precede
changes in PI and shame and could be reliable predictors of non-disclosure in the context of brief
psychotherapy. Relatedly, it is possible the null relation observed between non-disclosure and
number of sessions attended may be due to our short study duration (15-weeks), as the prior
study noting a significant relation polled clients who attended therapy an average of three years
(Farber et al., 2003). Yet, taken together, this pattern of results may imply a meaningful clinical
implication: clients who demonstrate higher levels of PI at intake may be more likely to withhold
secrets in therapy despite the duration and progress made in treatment.
Consistent with prior cross-sectional findings (Hook & Andrews, 2005; MacDonald &
Morley, 2001), feeling “too ashamed” was the most commonly endorsed reason for non-
disclosure. Participants also frequently endorsed “fear of negative evaluation,” which is
commonly associated with the affective experience of shame. Thus, shame may be a centrally
avoided emotion for some clients or topics of disclosure. Careful attention to shame-related
processes in therapy may be particularly important for clients high in shame-proneness, which
has been closely associated with SC (Larson et al., 2015) and reduced willingness to disclosure
shameful secrets in therapy (DeLong & Khan, 2014). The use of withdrawal to cope with shame
in therapy predicts impaired therapeutic alliance, which may affect potential for disclosure
(Black et al., 2013). It also seems likely that the disclosure of traditionally stigmatized topics,
such as psychological symptoms, topics of a sexual nature, or the disliked aspects of the self
(e.g., Baumann & Hill, 2016; Hook & Andrews, 2005; Love & Farber, 2019) may be particularly
threatening to shame-prone clients, meriting specific attention and care in therapy.
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 15
We hope that results from this study will encourage future research on practices intended
to promote disclosure in psychotherapy. For instance, future studies could explore whether the
assessment of PI at the beginning of therapy, or internalized shame if relevant to the case
conceptualization, could be used to inform therapist conceptualization relating to possible non-
disclosure. Our data suggest that, independent of therapy duration, for each standard deviation
(SD) increase in AAQ-2 scores there is a respective 3.25 increase in a client’s likelihood to
withhold secrets from their therapist3. Practically applied, Model 1 predicts that clients scoring
one SD above the mean on the AAQ-2 have a 73% chance of having withheld important
information, whereas clients scoring one SD below the mean had a 24% chance of having
withheld important information. While we have no prior studies for comparison, it seems that an
approximately threefold increase in the odds of non-disclosure between clients high and low in
PI may be clinical meaningful. Replication of these findings could provide a promising avenue
for improving detection of client non-disclosure at the initiation of treatment.
For clients identified as having a higher likelihood of non-disclosure, direct intervention
to build trust, enhance client willingness, and promote flexible modes of responding may be
merited, as one study found that 50% of clients who withheld secrets from their therapists
reported they may have disclosed the secret if they were directly asked, while another 40% cited
the need to build additional trust with their therapist (Love & Farber, 2017). As such, it is
possible that explicitly acknowledging secrets in treatment may invite disclosure by framing it as
a natural process within the client’s control. For example, Radically Open Dialectical Behavior
Therapy (RO-DBT; Lynch, 2018) instructs therapists to inform clients that secrets are a normal
part of therapy and offer clients permission to disclose secrets gradually as they build trust.
3 Model 3 predicted that one SD increase in ISS scores was associated with a 2.39-fold increase
in likelihood of non-disclosure. Clients scoring 1 SD above the mean on the ISS had a 69%
likelihood of non-disclosure while clients scoring 1 SD below the mean have a 30% likelihood.
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 16
Interventions found in Acceptance and Commitment Therapy (Hayes et al., 2012) could also
potentially be applied to barriers to disclosure to bolster psychological flexibility and enhance
clients’ willingness to disclose painful secrets. For clients for whom shame emerges as a barrier
to disclosure, interventions aimed to increase self-compassion may promote deeper disclosures
(Dupasquier et al., 2020).
Results of this study should be considered in the context of its methodological strengths
and limitations. As previously stated, a notable strength of the study included the use of a 15-
week longitudinal design, making it one of the few longitudinal studies on non-disclosure and
the first to establish a temporal relationship between PI and non-disclosure. However, the study
was limited by a relatively coarse assessment of non-disclosure via two items at only one time
point. Future research could meaningfully expand on these findings by including multiple
assessments using validated measures of non-disclosure across therapy. More nuanced
assessment of non-disclosure in therapy could elucidate contextual situations in which non-
disclosure in therapy may be normative and healthy versus therapy interfering.
The naturalistic context in which this study was conducted allowed for the assessment of
non-disclosure in a general therapy population at an evidence-based treatment center, which
likely increases the generalizability of our findings to other outpatient therapy contexts.
However, constraints associated with this approach are also notable. For instance, of the initial
82 participants, 27% did not have complete data at the 15-week follow-up and could not be
included in final analyses. Although study completers and non-completers did not differ on
relevant study variables, we cannot rule out the possibility that an unmeasured third variables
played a role in participants’ decisions to discontinue participation. The number of sessions
attended during the 15-week period also varied between participants, which introduced some
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 17
variance, but this was statistically controlled for in models and did not predict disclosure. Still,
future studies may benefit from examining whether findings replicate across more tightly
controlled designs in which the dose of therapy is standardized.
Finally, our design did not rule out the possibility that therapists in the study may have
addressed client non-disclosure differently as it occurred in treatment, which may have
ultimately influenced clients’ decisions to disclose. Future research may benefit from measuring
therapist effects as they relate to the relation between PI, shame, and non-disclosure. For
example, client concealment of cultural information appears to differ across dyads (Drinane et
al., 2018) and some therapists are capable of forming stronger therapeutic alliances (Del Re et
al., 2012), which has been previously linked to more self-disclosure (Hall & Farber, 2001).
Relatedly, the relative racial/ethnic homogeneity of the current sample is an important limitation,
as topics that evoke avoidance and shame may be culturally dependent. However, the current
sample did vary considerably in terms of gender, age, education, and employment, suggesting
generalizability to clients with respect to these demographics.
Conclusion
This study is the first establish a temporal relation between psychological inflexibility
and non-disclosure in outpatient therapy. Our findings indicate that PI may be a more powerful
predictor than internalized shame in predicting client non-disclosure at 15-week follow-up.
Future research would benefit from replicating current findings and examining the relevance of
PI-related assessment and intervention practices in promoting authentic client disclosure in
psychotherapy.
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 18
References
Baumann, E. C. & Hill, C. E. (2016). Client concealment and disclosure of secrets in outpatient
psychotherapy. Counselling Psychology Quarterly, 29, 53-75.
Black, R. S., Curran, D., & Dyer, K. F. (2013). The impact of shame on the therapeutic alliance
and intimate relationships. Journal of Clinical Psychology, 49, 646-654.
Blanchard M., & Farber, B. A. (2016). Lying in psychotherapy: Why and what clients don’t tell
their therapist about therapy and their relationship. Counselling Psychology Quarterly,
29, 90-112.
Blanchard, M., & Farber, B. A. (2020). “It is never okay to talk about suicide”: Patients’ reasons
for concealing suicidal ideation in psychotherapy. Psychotherapy Research, 30, 124-136.
Bond, F. W., Hayes, S. C., Baer, R. A., Carpenter, K. M., Guenole, N., Orcutt, H. K., ... & Zettle,
R. D. (2011). Preliminary psychometric properties of the Acceptance and Action
Questionnaire–II: A revised measure of psychological inflexibility and experiential
avoidance. Behavior therapy, 42, 676-688.
Browner, W. (2006). Publishing and presenting clinical research. Lippincott, Williams, &
Wilkins.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). New York:
Academic Press.
Cook, D. R., (1987). Measuring shame: The Internalized Shame Scale. Alcoholism Treatment
Quarterly, 4, 197-215.
Del Re, A. C., Flückiger, C., Horvath, A. O., Symonds, D., & Wampold, B. E. (2012). Therapist
effects in the therapeutic alliance–outcome relationship: A restricted-maximum
likelihood meta-analysis. Clinical Psychology Review, 32, 642-649.
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 19
DeLong, L. B., & Kahn, J. H. (2014). Shameful secrets and shame-prone dispositions: How
outcome expectations mediate the relation between shame and disclosure. Counselling
Psychology Quarterly, 27, 290–307.
Drinane, J. M., Owen, J., & Tao, K. W. (2018). Cultural concealment and therapy
outcomes. Journal of Counseling Psychology, 65, 239-246.
Dupasquier, J. R., Kelly, A. C., Moscovitch, D. A., & Vidovic, V. (2020). Cultivating self-
compassion promotes disclosure of experiences that threaten self-esteem. Cognitive
Therapy and Research, 44, 108-119.
Farber, B. A. (2003). Patient self-disclosure: A review of the research. Journal of Clinical
Psychology, In Session, 59, 589-600.
Farber, B. A. (2019). Client deception about substance use: Research findings and a case study.
Journal of Clinical Psychology, 76, 277-285.
Farber, B. A., Berano, K. C., & Capobianco, J. A. (2006). A temporal model of patient disclosure
in psychotherapy. Psychotherapy Research, 16, 463-469.
Gamez, W., Chmielewski, M., Kotov, R., Ruggero, C., Suzuki, N., & Watson, D. (2014). The
brief experiential avoidance questionnaire: Development and initial validation.
Psychological Assessment, 26, 35-45.
Gloster, A. T., Walder, N., Levin, M. E., Twohig, M. P., & Karekla, M. (2020). The empirical
stuatus of acceptance and commitment therapy: A review of meta-analyses. Journal of
Contextual Behavioral Science, 18, 181-192.
Hall, D. A., & Farber, B. A. (2001). Patterns of patient disclosure in psychotherapy. Journal of
the American Academy of Psychoanalysis and Dynamic Psychiatry, 29, 213-230.
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 20
Hayes, S. C., Strosahl, K. D., & Wilson, K. G. (2012). Acceptance and Commitment Therapy:
The process and practice of mindful change (2nd edition). The Guilford Press.
Hill, C. E., Thompson, B. J., Cogar, M. C., & Denman, D. W. (1993). Beneath the surface of
long-term therapy: Therapist and client report of their own and each other’s covert
processes. Journal of Counseling Psychology, 40, 278-287.
Judge, L., Cleghorn, A., McEwan, K., & Gilbert, P. (2012). An exploration of group-based
compassion focused therapy for a heterogeneous range of clients presenting to a community
mental health team. International Journal of Cognitive Therapy, 5, 420–429.
Kahn, J. H., Achter, J. A., & Shambaugh, E. J. (2001). Client distress disclosure, characteristics
at intake, and outcome in brief counseling. Journal of Counseling Psychology, 48, 203-
211.
Larson, D. G., Chastain, R. L., Hoyt, W. T., & Ayzenberg, R. (2015). Self-concealment:
Integrative review and working model. Journal of Social and Clinical Psychology, 34,
705-774.
Love, M., & Farber, B. A. (2019). Honesty in psychotherapy: Results of an online survey
comparing high vs. low self-concealers. Psychotherapy Research, 29, 607-620.
Luoma, J. B., Kohlenberg, B. S., Hayes, S. C., & Fletcher, L. B. (2012). Slow and steady wins
the race: A randomized clinical trial of acceptance and commitment therapy targeting
shame in substance use disorders. Journal of Consulting and Clinical Psychology, 80,
43–53.
Lynch, T. R. (2018). Radically open dialectical behavior therapy: theory and practice for
treating disorders of overcontrol. New Harbinger Publications.
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 21
Macdonald, J., & Morley, I. (2001). Shame and non‐disclosure: A study of the emotional
isolation of people referred for psychotherapy. British Journal of Medical
Psychology, 74, 1-21.
Marks, E. C., Hill, C. E., & Kivlighan Jr, D. M. (2019). Secrets in psychotherapy: For better or
worse?. Journal of Counseling Psychology, 66, 70-82.
Masuda, A. Tully, E. C., Drake, C. E., Tarantino, N. Ames, A. M., & Larson, D. (2017).
Examining Self-Concealment within the Framework of Psychological Inflexibility and
Mindfulness: A Preliminary Cross-Sectional Investigation. Current Psychology, 36, 184-
191.
McAndrews, Z., Richardson, J., & Stopa, L. (2019). Psychometric properties of acceptance
measures: A systematic review. Journal of Contextual Behavioral Science, 12, 261-277.
Pinto, R. A., Kienhuis, M., Slevison, M., Chester, A., Sloss, A., & Yap, K. (2017). The
effectiveness of an outpatient Acceptance and Commitment Therapy Group programme
for a transdiagnostic population. Clinical Psychologist, 21, 33-34.
Pope, K. S., & Tabachnick, B. G. (1994). Therapists as patients: A national survey of
psychologists’ experiences, problems, and beliefs. Professional Psychology: Research
and Practice, 25, 247-258.
Swan, S., & Andrews, B. (2003). The relationship between shame, eating disorders and
disclosure in treatment. British Journal of Clinical Psychology, 42, 367-378.
Tabachnick, B. G., & Fidell, L. S. (2013). Using Multivariate Statistics (6th ed.). Pearson.
Tangney, J. P. (1990). Assessing individual differences in proneness to shame and guilt. Journal
of Personality and Social Psychology, 59, 102-111.
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 22
Tyndall, I., Waldeck, D., Pancani, L., Whelan, R., Roche, B., & Dawson. (2019). The
Acceptance and Action Questionnaire-II (AAQ-II) as a measure of experiential
avoidance: Concerns over discriminant validity. Journal of Contextual and Behavioral
Science, 12, 278-284.
Vikan, A., Hassel, A. M., Rugset, A., Johansen, H. E., & Moen, T. (2010). A test of shame in
outpatients with emotional disorder. Nordic Journal of Psychiatry, 64, 196-202.
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 23
Table 1
Participant Characteristics
Range M or N SD or %
Age 19-69 35.38 12.50
Gender
Woman
Man
Transgender
Not reported
37
20
1
2
62
33
2
3
Hispanic Ethnicity
Yes
No
Race
European American
African American
Asian American
Two or more races
Other
5
55
54
1
1
3
1
8
92
90
2
2
5.0
2
Education
High School/ Some College
Associate Degree
Bachelor’s Degree
Some Graduate Coursework
Master’s Degree
Doctoral Degree
14
4
17
6
15
4
23
7
28
10
25
7
Employment
Not in labor
Unemployed
Part time employment
Full time employment
7
9
15
29
11
15
25
48
Primary DSM-5 Diagnosis
Unipolar Depressive Disorder 12 20
Anxiety Disorder 18 30
Post-traumatic Stress Disorder 5 8
Adjustment Disorder 7 12
Bipolar Mood Disorder 3 5
Alcohol Use Disorder 4 7
Borderline Personality Disorder 2 3
Obsessive Compulsive or Tic Disorder 3 5
Insomnia 3 5
Problems related to Environmental Stress 3 5
Prescribed Psych Meds 26 43
Total Sessions 1-19 9.35 4.00
Not disclosed to therapist
Both symptoms and experiences 17 28
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 24
Symptoms or behaviors only
Other issues/experiences only
7
5
12
8
Note: N = 60; Total sessions refers to total number of sessions that occurred during 15-week
study period. DSM-5 diagnoses refer to diagnoses assigned by treating clinicians; Percentages
may not add to exactly 100 due to rounding.
PSYCHOLOGICAL INFLEXIBILITY, SHAME, AND CLIENT NON-DISCLOSURE 25
Table 2
Logistic regression results predicting client non-disclosure at 15-week follow-up
Variable βS.E. Wald χ2OR [95% CI] p
Model 1
AAQ-2Baseline 1.18 0.37 10.07 3.25 [1.57, 6.74] .002
Total Sessions -0.27 0.32 0.76 0.39 [0.41, 1.42] .394
Constant 0.55 0.81 0.46 1.73 .496
Model 2
AAQ-2Baseline 1.19 0.37 10.10 3.27 [1.58, 6.80] .001
AAQ-2Slope 0.06 0.30 0.04 1.06 [0.60, 1.89] .839
Total Sessions -0.27 0.32 0.74 0.93 [0.80, 1.09] .389
Constant 0.56 0.81 0.48 1.75 .489
Model 3
ISS Baseline 0.81 0.33 7.12 2.39 [1.26, 4.53] .008
Total Sessions -0.16 0.30 0.29 0.59 [0.47, 1.54] .469
Constant 0.36 0.77 0.22 1.43 .641
Model 4
ISSBaseline 0.86 0.33 7.02 2.37 [1.25, 4.49] .008
ISSSlope -0.09 0.29 0.09 0.92 [0.53, 1.61] .764
Total Sessions -0.38 0.08 0.25 0.96 [0.83, 1.12] .617
Constant 0.33 0.78 0.18 1.40 .668
Model 5
AAQ-2Baseline 1.20 0.54 4.00 2.96 [1.02, 8.56] .046
ISSBaseline 0.08 0.50 0.03 1.08 [0.40, 2.91] .867
Total Sessions -0.25 0.32 0.60 0.94 [0.80, 1.10] .437
Constant 0.53 0.81 0.43 1.70 .514
Note: Parameter estimates derived using standardized variables in which one-unit increase
reflects one standard deviation change; Total sessions refers to total number of sessions that
occurred during 15-week study period. Bolded numerical values indicate p < .05
