Decreasing treatment dropout by addressing expectations for treatment length

Abstract

Therapy dropout or premature termination is a significant problem which impedes the delivery of psychotherapy services. In this study, a method aimed to reduce the occurrence of premature termination by addressing clients' duration expectations was examined. Sixty-three adult clients seeking psychotherapy services from a psychology department training clinic were randomized into control (n=32) and education groups (n=31). On average, those clients in the education group, who were provided information about the dose-effect model prior to their intake appointment, were found to stay in treatment significantly longer (d=0.55) and were more likely to be classified as therapy completers (RR=3.55) when compared to clients in the control group.

Full text

Decreasing treatment dropout by addressing expectations for
treatment length
JOSHUA K. SWIFT
1
, & JENNIFER L. CALLAHAN
2
1
University of Alaska Anchorage, Psychology, Anchorage, Alaska, USA &
2
University of North Texas, Psychology, Denton,
Texas, USA
(Received 1 September 2010; revised 7 October 2010; accepted 8 November 2010)
Abstract
Therapy dropout or premature termination is a significant problem which impedes the delivery of psychotherapy services. In
this study, a method aimed to reduce the occurrence of premature termination by addressing clients’ duration expectations
was examined. Sixty-three adult clients seeking psychotherapy services from a psychology department training clinic were
randomized into control (n32) and education groups (n31). On average, those clients in the education group, who were
provided information about the dose-effect model prior to their intake appointment, were found to stay in treatment
significantly longer (d0.55) and were more likely to be classified as therapy completers (RR3.55) when compared to
clients in the control group.
Keywords: dropout; duration; expectations; premature termination; pre-therapy education
Therapy dropout is a significant problem that fre-
quently impedes the delivery of psychotherapy ser-
vices. It has been defined by others as occurring when
a ‘‘client has left therapy before obtaining a requisite
level of improvement or completing therapy goals’’
(Hatchett & Park, 2003), and is referred to in the
literature by a number of different names, including
attrition, early withdrawal, premature termination,
and client-initiated unilateral termination. While
researchers have disagreed as to how exactly this
phenomenon should be operationalized (Hatchett &
Park, 2003; Pekarik, 1985b; Swift, Callahan, &
Levine, 2009; Wierzbicki & Pekarik, 1993), there is
a clear consensus concerning the deleterious effects
dropout has on both clients (exhibiting poorer
treatment outcomes) and service providers (experi-
encing a loss of revenue, underutilization of time, and
demoralization) (Barkham et al., 2006; Barrett,
Chua, Crits-Christoph, Gibbons, & Thompson,
2008; Pekarik, 1985a; Reis & Brown, 2006).
Furthermore, the problem of client dropout is
common and widespread in therapy; reviews have
estimated that between 30% and 60% of all clients
discontinue therapy prematurely (Clarkin & Levy,
2004; Garfield, 1994; Wierzbicki & Pekarik, 1993),
with even higher rates observed in some settings
(Callahan, Aubuchon-Endsley, Borja, & Swift, 2009;
Swift et al., 2009; Tryon, 1999).
In an effort to better understand therapy dropout,
researchers have attempted to identify the variables
related to its occurrence. In an early review of the
dropout literature, Baekeland and Lundwall (1975)
found only slight evidence that the client demographic
variables of age, gender, and socioeconomic stat us can
predict who will terminate prematurely. In a more
recent review, Garfield (1994) concluded that there
were no consistent study findings linking prema-
ture termination to any client demographic variables.
Similarly, in a 1993 meta-analysis Wierzbicki and
Pekarik found that attrition was not significantly
related to client age, gender, or marital status, and
was only modestly related to socioeconomic status.
Other reviews of the literature have likewise reported a
lack of a relation between dropout and client demo-
graphic variables (Barrett et al. 2008; Clarkin & Levy,
2004; Reis & Brown, 1999).
In contrast to the null or weak findings linking
dropout to demographic variables, each of the re-
views previously mentioned and more recent research
have found that premature termination is strongly
linked to unmet client expectations (Arnkoff, Glass,
& Shapiro, 2002; Aubuchon-Endsley, & Callahan,
2009; Greenberg, Constantino, & Bruce, 2006;
Correspondence concerning this article should be addressed to Joshua K. Swift, Ph.D., Department of Psychology, University of Alaska
Anchorage, 3211 Providence Drive SSB214, Anchorage, AK 99508, USA. Email: Joshua.Keith.Swift@gmail.com
Psychotherapy Research
2011, 18, iFirst article
ISSN 1050-3307 print/ISSN 1468-4381 online #2011 Society for Psychotherapy Research
DOI: 10.1080/10503307.2010.541294
Downloaded By: [University of North Texas] At: 20:28 25 January 2011

Hansen, Hoogduin, Schaap, & de Haan, 1992; Nock
& Kazdin, 2001; Orlinsky, Grawe, & Parks, 1994;
Reis & Brown, 1999; Walitzer, Dermen, & Connors,
1999). This relation is particularly strong for expec-
tations concerning treatment length in which most
clients expect therapy to be very brief. For example,
Pekarik (1991) in one study found that 20% of clients
surveyed expected to attend two sessions or less, and
roughly 50% expected five sessions or less. Unfortu-
nately, the duration expectations that clients hold are
typically much lower than expectations held by
therapists (Mueller & Pekarik, 2000; Pekarik &
Finney-Owen, 1987; Pekarik & Wierzbicki, 1986;
Tryon, 1999) and are also lower than what is needed
for therapy to be effective (Swift & Callahan, 2008).
Unrealistically low duration expectations become
a problem in therapy because clients typically do not
attend more sessions than they originally expect. For
example, Scamardo, Bobele, and Biever (2004)
found that of 74 clients seen in a psychology
department’s clinic, only four underestimated the
number of sessions they actually attended. A number
of other studies have found similar results (Mueller
& Pekarik, 2000; Pekarik, 1985a, 1991; Pekarik &
Wierzbicki, 1986; Tryon, 1999). Given that many
clients hold unrealistically low expectations for
treatment duration and that expected duration has
been found to be the best predictor of actual
duration, it is perhaps not surprising that studies
have found that clients who hold unrealistic duration
expectations are more likely to be classified as
treatment dropouts (Mueller & Pekarik, 2000;
Tryon, 1999).
Addressing Duration/Attendance Expectations
With unrealistic duration expectations being linked
to premature termination, one might hypothesize
that efforts aimed at altering clients’duration and
attendance expectations could result in reduced
occurrences of therapy dropout. In an effort to
reduce rates of premature termination, Sheeran,
Aubrey, and Kellett (2007) asked clients to build
attendance intentions by repeating three times ‘‘As
soon as I feel concerned about attending my
appointment, I will ignore that feeling and tell myself
this is perfectly understandable!’’ In this study, those
clients who were randomized to repeat this statement
were significantly more likely to show up to their
initial appointment compared to clients randomized
to a control group (75% versus 63% attendance).
More recently, Buckner et al. (2009) compared
attendance rates between clients who were asked to
spend 10 minutes imagining attending therapy ses-
sions and control group clients who were merely
provided 10 minutes of information about the clinic.
Although a significant manipulation effect was not
found for the entire sample of clients, those in the
imagination group with anxiety disorders attended
close to double the number of sessions compared to
control group clients with anxiety disorders.
Others have attempted to address clients’unrea-
listic duration and attendance expectations by pro-
viding clients with pre-therapy training or education.
In one study, Reis and Brown (2006) provided a
group of adult outpatient clients the opportunity to
estimate and discuss their expectations for treatment
duration with their therapists. Despite Reis and
Brown’s‘‘strong theoretical’’ belief that this proce-
dure would be effective in reducing premature
termination, they found no differences in dropout
rates between clients who did and did not discuss
their expectations with their therapists. Unfortu-
nately, methodological limitations with this study
prevent us from drawing any major conclusions from
the results. Clients in the discussion group of this
study were first asked to estimate how many sessions
they planned to attend. If the initial estimates were
deemed unrealistic, their therapists were then in-
structed to help them develop more realistic expec-
tations. In this study the definition of a realistic
duration estimate was three or more sessions. How-
ever, expected recovery in three sessions is still very
unrealistic for most clients. Thus, even in the
discussion group many clients with unrealistic
expectations were not provided any realistic duration
education.
In a second study that attempted to use education
to alter unrealistic duration expectations, Swift and
Callahan (2008) provided 155 participants with
education about the general effectiveness of therapy
based on the dose-effect literature (i.e., chances of
recovery are greater as one attends more therapy
sessions and in general it takes about 13 to 18
sessions for 50% of clients to recover). Participants
were then asked to state the number of sessions that
they would expect to attend if they were seeking
treatment. Compared to participants randomized to
a control group, those who received the duration
education expected to attend over twice the number
of sessions (17 sessions versus eight sessions
expected by control participants). Although signifi-
cant results were found, because Swift and Callahan
used a student population the results may not fully
generalize to actual clinical settings. Additionally,
although this study found that expectations could be
altered, it did not examine whether pre-therapy
education could improve actual treatment atten-
dance or decrease the rate of therapy dropout.
Given the limitations of these previous studies, we
felt that further research was needed to examine the
effectiveness of pre-therapy duration education. In
2J. K. Swift & J. L. Callahan
Downloaded By: [University of North Texas] At: 20:28 25 January 2011

this study we aimed to examine whether pre-therapy
education based on the dose-effect literature would
have an influence on (1) clients’expectations for
treatment length, (2) actual treatment length, and (3)
rates of premature termination. Due to the relatively
high attrition rates that have previously been docu-
mented in training clinic settings (Aubuchon-
Endsley & Callahan, 2009; Swift et al., 2009; Tryon,
1999), we believed such a setting would be ideal
for testing our hypotheses. Given that Swift and
Callahan (2008) found this type of pre-therapy
education to be useful in creating more realistic
duration expectations, and expectations have been
found to be related to actual attendance and therapy
dropout, we hypothesized that the duration educa-
tion would have a significant impact in each of the
three areas of examination.
Method
Client Participants
Participants in this study were 60 adult clients who
sought individual therapy services from a univer-
sity-based psychology department clinic (see
Figure 1 for participant flow). This particular
clinic offers reduced fee services to both students
enrolled in the university and individuals from the
larger community. During the study time period
there were a total of 69 clients who presented at
the clinic; however, four clients were not offered
participation by their intake therapists, two clients
refused participation, and three did not complete
any of the study measures. Of the 60 participating
clients, 62% were female, 54% were single, and
78% were Caucasian. Another 6.3% of partici-
pants self-identified as African American, 1.6%
Hispanic, 11.1% Native American, and 3.2% as
Bi/Multi-Racial American. The average age of
these clients was M28.68 (SD11.12) ranging
from 18 to 65 years old. Roughly half of the
participants were current university students,
12.7% had not graduated from high school,
15.9% had graduated from high school only, and
11.1% had obtained a college degree. The majority
of these clients were given a diagnosis of a mood,
anxiety, or adjustment-based disorder at intake.
About 41% of the sample had previous therapy
experience. The average score on the Outcome
Questionnaire 45.2 (OQ-45.2: Lambert et al.,
1996) for participating clients at intake was M
74.02 (SD25.41), falling in the clinical range for
the measure.
Clients agreeing to study
participation (N = 63)
Randomization
Control group (N = 32) Education group (N = 31)
Started the study (N = 31) Started the study (N = 29)
Assigned a therapist according to normal intake procedures
Dropped out
of therapy
(N = 24)
Dropped out
of therapy
(N = 9)
Completed
therapy
(N = 7)
Completed
therapy
(N = 20)
Adult clients presenting at
the clinic for individual
therapy during the study
period (N = 69)
• Not recruited due to
clinician error (N = 4)
•Refused participation
(N = 2)
Figure 1. Flow of participants through the study.
Addressing treatment length expectations 3
Downloaded By: [University of North Texas] At: 20:28 25 January 2011

Procedure and Measures
Prior to the start of their intake appointments clients
were consented for participation. Those 63 clients
who agreed to participate were then asked to
complete a short survey that included demographic
information and questions that were used for other
ongoing research projects in the clinic. Clients were
next randomized into one of two conditions: control
(n32) and education groups (n31). One client
in the control group and two clients in the education
group failed to complete any of the remaining parts
of the study and thus were dropped from further
analyses. Clients in both groups were then provided
a handout with some basic information about
therapy services at the clinic (e.g., sessions typically
last 1 hour and occur once per week) and those in
the education group were provided additional in-
formation covering the dose-effect model of therapy.
The dose-effect model (Howard, Kopta, Krause, &
Orlinsky, 1986) indicates that there is a positive, but
negatively accelerated relationship between treat-
ment length and the percentage of clients who
recover. For example, Lambert, Hansen, and Finch
(2001) analyzed data from over 10,000 clients and
found that approximately 20% of clients had recov-
ered by five sessions, 35% by 10 sessions, 50% by 20
sessions, and 70% by 45 sessions of therapy. In a
review of seven dose-effect studies, Hansen,
Lambert, and Forman (2002) concluded that it
takes roughly 1318 sessions for 50% of clients to
show a clinically significant change. Based on this
information, clients in the education group were
provided the following script: ‘‘An extensive body of
research looking at psychotherapy outcomes indi-
cates that on average it takes approximately 1318
therapy sessions for 50% of clients to recover. Some
clients may recover in a shorter amount of time while
other clients may take longer, depending on a
number of factors. However, on average it takes
approximately 1318 sessions for 50% of clients to
recover.’’ Clients in both groups were then asked to
indicate how many sessions they expected to attend.
After this point, the remainder of therapy services
proceeded as usual according to regular clinic pro-
cedures without study interference. Clients met with
an intake therapist for roughly 2 hours, were placed
on a waitlist, and then were assigned to a therapist by
the student clinic director according to caseload and
client fit. Therapy was provided by one of 17 trainees.
Each was a graduate student enrolled in an APA-
approved clinical psychology scientist-practitioner
PhD program at the same university and each
received weekly supervision on his/her clinic
cases by a member of the psychology department’s
clinical faculty. In general, this program’s theoretical
orientation can be described as cognitive-behavioral;
however, trainees in this program are encouraged to
use evidence-based interventions when working with
clients, rather than sticking to one set orientation.
The therapists were unaware of the purposes of this
study and were blind to the study conditions of the
clients. After the termination of therapy for each
participating client, charts were reviewed to deter-
mine the total number of sessions that were attended.
In addition, therapists were asked to classify each of
their clients as therapy ‘dropouts’(client-initiated
unilateral terminators) or therapy ‘completers’.
Results
Randomization
Clients were randomized into control and education
groups. To check the randomization procedures and
to ensure comparability between conditions, partici-
pating clients from the two groups were first
compared on demographic (age, gender, race, mar-
ital status, student status, and highest level of
education) and pre-therapy variables (previous ther-
apy experience and OQ-45.2 scores). No significant
differences were found between the two client
groups (control and education) on any of the
variables of comparison.
Expected Duration
Given the results previously found by Swift and
Callahan (2008), we were first interested in examin-
ing whether the duration-based education would
influence the number of therapy sessions clients
expected to attend. Of the 60 participating clients,
10 failed to provide an expected duration estimate.
Many of these clients either filled in the space
provided with question marks or wrote ‘‘I don’t
know.’’ Interestingly, all 10 of these clients were from
the control group. An additional two participants
(one from each condition) reported duration expec-
tations that were greater than three standard devia-
tions from the mean (e.g., expecting 100 sessions).
Although some may not consider this an unrealistic
expectation, in terms of data management these two
scores were extreme compared to the averages.
Thus, these scores were considered outliers and
were not included in this analysis. However, data
from these two clients were maintained for the
remaining analyses.
Clients in the control group (n20) on average
expected to attend M6.43 (SD4.86) sessions of
therapy. In contrast, clients in the education group
(n28) on average expected to attend M15.16
(SD3.50) sessions of therapy. An independent
samples t-test indicated that these expectations
4J. K. Swift & J. L. Callahan
Downloaded By: [University of North Texas] At: 20:28 25 January 2011

were significantly different from each other [t(46) 
7.25, pB.001], demonstrating a large effect (d
2.14).
Actual Duration
We were next interested in examining whether the
duration-based education would influence the actual
number of therapy sessions attended by clients.
While clients in the control group (n31) attended
an average of M5.9 (SD7.3) therapy sessions,
clients in the education group (n29) attended an
average of M10.59 (SD9.91) sessions. An
independent samples t-test indicated that the num-
ber of actual sessions attended was significantly
different between the two groups [t(58) 2.09,
pB.05], demonstrating a medium-sized effect (d
0.55).
Premature Termination
Although we found that on average clients who
received the duration-based education attended
more sessions compared to clients who did not receive
the education, this information does not tell us
whether the duration-based education had an influ-
ence on therapy dropout. The dose-effect literature
has generally found that with an increased number of
sessions attended there is a greater likelihood of
improvement and recovery (Hansen et al., 2002;
Howard et al., 1986); however, a number of research-
ers have found that treatment length does not always
correspond to treatment success (Baldwin, Berkeljon,
Atkins, Olsen, & Nielsen, 2009; Barkham et al., 2006;
Hatchett & Park, 2003; Swift et al., 2009). We were
thus further interested in examining whether the
duration-based education would have an influ-
ence on eventual therapy dropout.
Of the 31 control group clients, seven were
identified by their therapists as ‘completers’while
24 were identified as ‘dropouts’. In contrast, of
the 29 clients who received the duration-based
education, 20 were classified as treatment ‘comple-
ters’and only nine were classified as treatment
‘dropouts’. A chi-squared test based on the 2 2
cross-classification table found this difference to be
significant [x
2
(1, 60)13.03, pB.001]. Calculation
of the Relative Risk ratio (representing the prob-
ability of event occurrence between groups) indi-
cated that clients who received the duration-based
education were more than three and a half times as
likely (RR 3.55) to be identified as treatment
‘completers’compared to the clients who did not
receive the education.
Discussion
In this study we sought to examine the effects of
providing clients with pre-therapy duration educa-
tion based on the dose-effect literature. In summary,
we found that clients who received the brief educa-
tion prior to their first therapy session predicted to
attend a greater number of therapy sessions (d
2.14), actually attended more sessions of therapy
(d0.55), and were more likely to be classified by
their therapists as therapy ‘completers’rather than
‘dropouts’(RR 3.55). While previous research has
demonstrated the importance of addressing clients’
role expectations prior to the start of therapy
(Acosta, Yamamoto, Evans, & Skilbeck, 1983;
Orlinsky et al., 1994; Reis & Brown, 2006; Walitzer
et al., 1999), these findings demonstrate the value of
providing clients with pre-therapy education that
also addresses treatment duration and outcome
expectations.
To our knowledge this study was the first to
attempt to alter clients’duration expectations by
providing them with education about the general
effectiveness of therapy. Although novel, the results
of this study fit well with the existing research on
duration expectations. First, we found that clients’
expectations for treatment length could be altered by
providing them an anchor with which they could
base their own expectations. Swift and Callahan
(2008) found very similar results with their under-
graduate student population. It is thought that
clients come to therapy with low duration expecta-
tions not because they believe that they are different
from the norms and thus will recover much quicker
than the average client, but because they are not
aware of what the norms are for treatment length.
The fact that clients may not be aware of what to
expect was demonstrated by the finding that while
10 of the participants (one-third) who did not
receive education were not able to provide an
expectation, all of the clients who were given a
reference point were able to state a prediction of
treatment length. Interestingly, clients in the control
group on average made a duration estimate (M
6.43) similar to what has been found in previous
research [Pekarik (1991) found 50% expected five
sessions or less]. In contrast, clients in the education
group made estimates (M15.16 sessions) that
were almost identical to the education they were
given (1318 sessions). These results further support
the idea that clients used the education as a reference
point on which they based their own expectations.
A number of previous studies have found that
clients’expected duration for treatment is closely
related to their actual duration (Mueller & Pekarik,
2000; Pekarik, 1985a, 1991; Pekarik & Wierzbicki,
Addressing treatment length expectations 5
Downloaded By: [University of North Texas] At: 20:28 25 January 2011

1986; Tryon, 1999). Thus, the finding that clients in
our education group attended more sessions of
therapy may be due to the fact that we were able
to alter their expectations. Additionally, comparing
the expectations of our clients to their actual
attendance, we found that in both groups clients
attended fewer sessions than they originally
expected. This finding matches what has been
observed by others (Scamardo et al., 2004; Tryon,
1999).
In contrast to Reis and Brown’s (2006) findings,
the pre-therapy duration education used in this
study did have an impact on rates of premature
termination between the education and control
groups. The main difference between ours and
Reis and Brown’s study is the manner in which
clients were provided duration education. Whereas
in Reis and Brown’s study clients were only asked to
state their expectations and then had a discussion
with their therapists if the expectations seemed
unrealistic, in our study we provided clients with
factual statements based on the dose-effect litera-
ture. Clients’expectations may be influenced to a
greater degree when information based on research
is presented to them (our study) rather than merely
discussing expectations with their therapists (Reis
and Brown’s study). In our study no client-therapist
discussion of expectations, including duration ex-
pectations, was systematically included (although
this may have occurred in some cases for both study
conditions as part of ‘‘treatment as usual’’). Addi-
tionally, our studies differed in our definitions of
‘‘realistic’’ duration expectations. In Reis and
Brown’s study, three sessions was deemed an appro-
priate treatment length and only clients who stated
they expected two sessions or less were encouraged
to attend more. In our study we provided clients with
an anchor of 1318 sessions for 50% of clients to
recover; an anchor much higher than Reis and
Brown’s cut off, and we believe it is a more
‘‘realistic’’ estimate of what is needed for therapy
to be successful. This higher estimate may explain
why we observed an effect on treatment dropout
while they did not.
Study Considerations
In considering the applications of the findings from
this study, some limitations should first be consid-
ered. First, this study took place in a psychology
department training clinic. It should be noted that
the rate of dropout observed in the control group
was high (77.42% prematurely terminated in
this group). Although this rate is higher than what
has been estimated in general [Wierzbicki and
Pekarik’s (1993) meta-analysis found an average
rate of approximately 47%], this rate is similar to
what has been found in other training clinic settings
[Aubuchon-Endsley & Callahan (2009) found a
67.92% client dropout rate in the same clinic used
in this study, but with a different sample; Swift et al.
(2009) found 77.04% of their clients from a different
psychology department clinic dropped out; Tryon
(1999) reported 71.21% of clients at a university
counseling center either dropped out after only one
appointment or unilaterally terminated]. Given the
high base rate of termination in the training clinic
setting, it may have been easier for our pre-therapy
manipulation to have had an effect on reducing
dropout.
Although it is believed that the duration education
would still have an effect on reducing dropout in
other settings, the sample of clients used in this study
may also limit the generalizability of the results.
First, even though the sample size was large enough
to detect differences, it was relatively small (only 20
participants in the control group for one of the
comparisons). Also, the clients who were included in
this sample were primarily Caucasian, female, single,
and were seeking treatment for a mood, anxiety, or
adjustment-based disorder. Additionally, although
on average the intake OQ-45.2 scores of this sample
were in the clinical range, scores indicated less severe
disturbance than what has been reported in
the measure’s manual for other various clinic set-
tings. Further research is needed to test whether the
duration education used in this study would have
similar effects in other settings and with other types
of client populations.
Additionally, the clients were all seen in a setting
that generally espouses a cognitive-behavioral model.
Although clinicians were encouraged to use whatever
evidence-based interventions were deemed most
appropriate for their individual clients, the focus
was on briefer courses of therapy. The expectations
found in this setting may be different from those
found in settings that encourage longer durations of
treatment. However, this clinic was not known in the
community as a ‘‘cognitive-behavioral’’ clinic, and
clients seeking services at this clinic did not generally
request a specific type of treatment. Thus it is
thought that the pre-therapy expectations that these
clients held were not overly influenced by a specific
mode of therapy that they were expecting to receive.
Another consideration that should be kept in mind
when interpreting the results of this study is the
method we used to assess treatment dropout. Pre-
mature termination has been operationalized in a
number of different ways, including failure to attend
a given number of sessions, attending an intake
appointment only, missing the last scheduled ap-
pointment without returning to the clinic, and by
6J. K. Swift & J. L. Callahan
Downloaded By: [University of North Texas] At: 20:28 25 January 2011

therapist judgment. Studies have shown that differ-
ent results are obtained depending on the operatio-
nalization used (Hatchett & Park, 2003; Pekarik,
1985b; Swift et al., 2009; Wierzbicki & Pekarik,
1993). In this study we measured dropout based on
therapist judgment, which historically has been
regarded as the most accurate method of classifying
premature termination (Pekarik, 1985b). However,
studies have found that therapists sometimes classify
their clients as dropouts erroneously*when clients
themselves believe that they have completed treat-
ment (Beckham, 1992; Scamardo et al., 2004) or
when outcome measures have shown clinically sig-
nificant change (Swift et al., 2009). Swift et al.
(2009) recommend an operationalization for drop-
out based upon whether the client has made a
clinically significant change prior to discontinuing
therapy. However, because in this clinic therapists
did not routinely use an outcome measure with their
clients, we were left to base our results on therapist
judgment. Future studies may want to examine the
effect of pre-therapy duration-based education on a
measure of dropout based on clinically significant
change or more generally on treatment outcomes.
Clinical Recommendations and Future
Directions
In our study we found that providing clients with
pre-therapy duration-based education was effective
in altering duration expectation, leading clients to
attend more sessions of therapy and being less likely
to be rated by their therapists as treatment dropouts.
Based on these results we would recommend that
duration expectations be discussed with all clients
prior to treatment. This education should include
some type of education addressing a recommended
number of sessions for recovery to occur. In our
study the education was very brief (a few sentences
only), was provided in paper format, was based on
research findings from the dose-effect literature, and
was provided prior to the first therapy appointment.
These procedures could easily be implemented in
other clinic settings without much effort on the part
of staff or therapists or interference of client care.
More detailed discussions, different educational
formats (e.g., via the computer, therapist discus-
sions, or video presentations), and different recom-
mendations for treatment length (perhaps therapists
working from an orientation espousing a more long-
term model or perhaps those working with more
seriously impaired clients may want to help their
clients develop expectations for even longer treat-
ment durations) may also prove useful in addressing
clients’duration expectations. It is also recom-
mended that discussions of treatment length should
occur in conjunction with routine outcome monitor-
ing throughout the course of therapy. This would
perhaps be the best approach to preventing clients
from dropping out of treatment. Further research is
needed to study each of these possibilities. Addi-
tionally, due to the relatively small sample size and
limitations in the generalizability of this study,
replication is warranted.
Acknowledgements
Portions of this article were completed while the first
author was a graduate student at Oklahoma State
University.
References
Acosta, F.X., Yamamoto, J., Evans, L.A., & Skilbeck, W.M.
(1983). Preparing low-income Hispanic, Black, and White
patients for psychotherapy. Evaluation of a new orientation
program. Journal of Clinical Psychology, 39, 872877.
doi:10.1002/1097-4679(198311)39:6B872::AID-JCLP22703
906103.0.CO;2-X.
Arnkoff, D.B., Glass, C.R., & Shapiro, S.J. (2002). Expectations
and preferences. In J.C. Norcross (Ed.), Psychotherapy relation-
ships that work: Therapist contributions and responsiveness to
patients (pp. 335356). New York, NY: Oxford University
Press.
Aubuchon-Endsley, N.L. & Callahan, J.L. (2009). The hour of
departure: Predicting attrition in the training clinic from role
expectancies. Training and Education in Professional Psychology,
3, 120126. doi:10.1037/a0014455.
Baekeland, F. & Lundwall, L. (1975). Dropping out of treatment:
A critical review. Psychological Bulletin, 82, 738783.
doi:10.1037/h0077132.
Baldwin, S.A., Berkeljon, A., Atkins, D.C., Olsen, J.A., & Nielsen,
S.L. (2009). Rates of change in naturalistic psychotherapy:
Contrasting dose-effect and good-enough level models of
change. Journal of Consulting and Clinical Psychology, 77, 203
211. doi:10.1037/a0015235.
Barkham, M., Connell, J., Stiles, W.B., Miles, J.N.V., Margison,
F., Evans, C., & Mellor-Clark, J. (2006). Dose-effect relations
and responsive regulation of treatment duration: The good
enough level. Journal of Consulting and Clinical Psychology, 74,
160167. doi:10.1037/0022-006X.74.1.160.
Barrett, M.S., Chua, W.J., Crits-Christoph, P., Gibbons, M.B., &
Thompson, D. (2008). Early withdrawal from mental health
treatment: Implications for psychotherapy practice. Psychother-
apy: Theory, Research, Practice, Training, 45, 247267.
doi:10.1037/0033-3204.45.2.247.
Beckham, E.E. (1992). Predicting patient dropout in psychother-
apy. Psychotherapy: Theory, Research, Practice, Training, 29, 177
182. doi:10.1037/0033-3204.29.2.177.
Buckner, J.D., Cromer, K.R., Merrill, K.A., Mallott, M.A.,
Schmidt, N.B., Lopez, C., Holm-Denoma, J.M., & Joiner,
T.E., Jr. (2009). Pretreatment intervention increases treatment
outcomes for patients with anxiety disorders. Cognitive Therapy
Research, 33, 126137. doi:10.1007/s10608-007-9154-x.
Callahan, J.L., Aubuchon-Endsley, N., Borja, S.E., & Swift, J.K.
(2009). Pretreatment expectancies and premature termination
in a training clinic environment. Training & Education in
Professional Psychology, 3, 111119. doi:10.1037/a0012901.
Clarkin, J.F., & Levy, K.N. (2004). Influence of client variables
on psychotherapy. In M.J. Lambert (Ed.), Handbook of
Addressing treatment length expectations 7
Downloaded By: [University of North Texas] At: 20:28 25 January 2011

psychotherapy and behavior change (5th ed) (pp. 194226). New
York: Wiley & Sons.
Garfield, S.L. (1994). Research on client variables in psychother-
apy. In A.E. Bergin & S.L. Garfield (Eds.), Handbook of
psychotherapy and behavior change (4th edn) (pp. 190228).
New York: Wiley & Sons.
Greenberg, R.P., Constantino, M.J., & Bruce, N. (2006). Are
patient expectations still relevant for psychotherapy process and
outcome? Clinical Psychology Review, 26, 657678. doi:10.1016/
j.cpr.2005.03.002.
Hansen, A.M., Hoogduin, C.A., Schaap, C., & de Haan, E.
(1992). Do dropouts differ from successfully treated obsessive-
compulsives? Behavior Research and Therapy, 30, 547550.
doi:10.1016/0005-7967(92)90040-N.
Hansen, N.B., Lambert, M.J., & Forman, E.M. (2002). The
psychotherapy dose-response effect and its implications for
treatment delivery services. Clinical Psychology: Science and
Practice, 9, 329343. doi:10.1093/clipsy/9.3.329.
Hatchett, G.T. & Park, H.L. (2003). Comparison of four
operational definitions of premature termination. Psychother-
apy: Theory, Research, Practice, Training, 40, 226231.
doi:10.1037/0033-3204.40.3.226.
Howard, K.I., Kopta, S.M., Krause, M.S., & Orlinsky, D.E.
(1986). The dose-effect relationship in psychotherapy. Amer-
ican Psychologist, 41, 159164. doi:10.1037/0003-066X.41.2.
159.
Lambert, M.J., Hansen, N.B., & Finch, A.E. (2001). Patient-
focused research: Using patient outcome data to enhance
treatment effects. Journal of Consulting and Clinical Psychology,
69, 159172. Doi:10.1037//0022-006X.69.2.159.
Lambert, M.J., Hansen, N.B., Umpress, V., Lunnen, K., Okiishi,
J., & Burlingame, G.M. (1996). Administration and scoring
manual for the OQ-45.2. Stevenson, MD: American Profes-
sional Credentialing Services LLC.
Mueller, M. & Pekarik, G. (2000). Treatment duration prediction:
Client accuracy and its relationship to dropout, outcome, and
satisfaction. Psychotherapy: Theory, Research, Practice, Training,
37, 117123. doi:10.1037/h0087701.
Nock, M.K. & Kazdin, A.E. (2001). Parent expectancies for child
therapy: Assessment and relation to participation in treatment.
Journal of Child and Family Studies, 10, 155180. doi:10.1023/
A:1016699424731.
Orlinsky, D.E., Grawe, K., & Parks, B.K. (1994). Process and
outcome in psychotherapy: Noch einmal. In A.E. Bergin & S.L.
Garfield (Eds.), Handbook of psychotherapy and behavior change
(4th ed) (pp. 270376). New York: Wiley & Sons.
Pekarik, G. (1985a). Coping with dropouts. Professional Psycho-
logy: Research and Practice, 16, 114123. doi:10.1037/0735-
7028.16.1.114.
Pekarik, G. (1985b). The effects of employing different termina-
tion classification criteria in dropout research. Psychotherapy,
22,8791. doi:10.1037/h0088531.
Pekarik, G. (1991). Relationship of expected and actual treatment
duration for adult and child clients. Journal of Clinical Child
Psychology, 20, 121125. doi:10.1207/s15374424jccp2002_2.
Pekarik, G. & Finney-Owen, K. (1987). Outpatient clinic
therapist attitudes and beliefs relevant to client dropout.
Community Mental Health Journal, 23, 120130. doi:10.1007/
BF00757166.
Pekarik, G. & Wierzbicki, M. (1986). The relationship between
clients’expected and actual treatment duration. Psychotherapy:
Theory, Research, Practice, Training, 23, 532534. doi:10.1037/
h0085653.
Reis, B.F. & Brown, L.G. (1999). Reducing psychotherapy
dropouts: Maximizing perspective convergence in the psy-
chotherapy dyad. Psychotherapy: Theory, Research, Practice,
Training, 36, 123136. doi:10.1037/h0087822.
Reis, B.F. & Brown, L.G. (2006). Preventing therapy dropout in
the real world: The clinical utility of videotape preparation and
client estimate of treatment duration. Professional Psychology:
Research and Practice, 37, 311316. doi:10.1037/0735-
7028.37.3.311.
Scamardo, M., Bobele, M., & Biever, J.L. (2004). A new
perspective on client dropouts. Journal of Systemic Therapies,
23,2738. doi:10.1521/jsyt.23.2.27.36639.
Sheeran, P., Aubrey, R., & Kellett, S. (2007). Increasing
attendance for psychotherapy: Implementation intentions and
the self-regulation of attendance-related negative affect. Journal
of Consulting and Clinical Psychology, 75, 853863. doi:10.1037/
0022-006X.75.6.853.
Swift, J.K. & Callahan, J.L. (2008). A delay-discounting measure
of great expectations and the effectiveness of psychotherapy.
Professional Psychology: Research and Practice, 39, 581588.
doi:10.1037/0735-7028.39.6.581.
Swift, J.K., Callahan, J.L., & Levine, J.C. (2009). Using clinically
significant change to identify premature termination. Psy-
chotherapy: Theory, Research, Practice, Training, 46, 328335.
doi:10.1037/a0017003.
Tryon, G.S. (1999). Counseling dropout relative to client attrac-
tiveness, disturbance, and expected number of sessions. Coun-
selling Psychology Quarterly, 12, 285291. doi:10.1080/
09515079908254098.
Walitzer, K.S., Dermen, K.H., & Connors, G.J. (1999). Strategies
for preparing clients for treatment: A review. Behavior Mod-
ification, 23, 129151. doi:10.1177/145445599231006.
Wierzbicki, M. & Pekarik, G. (1993). A meta-analysis of
psychotherapy dropout. Professional Psychology: Research and
Practice, 24, 190195. doi:10.1037/0735-7028.24.2.190.
8J. K. Swift & J. L. Callahan
Downloaded By: [University of North Texas] At: 20:28 25 January 2011