The Relative Efficacy and Efficiency of Single - and Multi-Session Exposure

Behaviour Research and Therapy 159 (2022) 104203
Contents lists available at ScienceDirect
Behaviour Research and Therapy

journal homepage: www.elsevier.com/locate/brat
The relative efficacy and efficiency of single- and multi-session exposure

therapies for specific phobia: A meta-analysis
Katarzyna Odgers a, Kelly A. Kershaw a, Sophie H. Li b, Bronwyn M. Graham a, *
a
School of Psychology, University Of New South Wales, Sydney, Australia
b
Black Dog Institute, University Of New South Wales, Sydney, Australia
A R T I C L E I N F O A B S T R A C T
Keywords: Exposure therapy is the preferred treatment for specific phobia (SP), with evidence supporting its efficacy
Specific phobia whether delivered over multiple sessions or as a single session, such as One-Session Treatment. In this meta-
Exposure therapy analysis, we compared the efficiency and effectiveness of single- and multi-session exposure for SP. PsycINFO,
One-session treatment
Embase, MEDLINE, and Cochrane were systematically searched for peer-reviewed articles reporting the effects of
Single-session exposure
multi-session (k = 30) and/or single-session (k = 55) in vivo exposure on SP symptoms in clinical populations (n
= 1758 participants). A random-effects model was used to synthesise and compare the pre-post treatment effects
(Hedges’ g) on approach behaviour and self-reported SP symptoms. Mean total treatment time was significantly
longer for multi-session exposure than for single-session. There were no significant differences in the pooled
effect sizes of single-session and multi-session exposure at post-treatment and follow-up assessments; effect sizes
were large for all outcomes. Phobia subtype significantly moderated the effect size for both treatment ap
proaches, although the direction of association differed according to the outcome measures. Results suggest no
evidence for differences in the effectiveness of single- and multi-session exposure, but single-session is more time
efficient. These outcomes suggest that policies to facilitate access to single-session exposure would be beneficial.
1. Introduction approach, or in a single intense treatment session (Choy et al., 2007). A

prominent example of single-session exposure is one-session treatment
Specific phobia (SP) is characterised by an excessive and persistent (OST), originally developed by Öst (Öst, 1989; Davis et al., 2012), and
fear of a specific object or situation, which is actively avoided or involves a series of exposure tasks over a typical duration of up to 3 h. A
endured with intense distress. SP is among the most common mental large body of empirical research supports the efficacy of both
disorders in the general population, with an average lifetime prevalence multi-session treatment and OST (see Choy et al., 2007; Davis et al.,
estimate of 7.4% (Wardenaar et al., 2017). SP is associated with 2019; Wolitzky-Taylor et al., 2008; Zlomke & Davis, 2008). However,
considerable functional impairment and predicts the subsequent onset although there have been several qualitative reviews of OST literature
of a range of conditions, most notably panic disorder, generalised anx (Davis et al., 2019; Ollendick & Davis, 2013; Zlomke & Davis, 2008),
iety disorder, and obsessive-compulsive disorder (Wardenaar et al., and more recently, a study summarising OST effects for spider phobia (Li
2017). et al., 2020), treatment effects of OST, and other single-session exposure
The preferred psychological treatment for SP has long been exposure treatments, are yet to be synthesised quantitatively for all SPs across
therapy (Ollendick & King, 1998), with meta-analytic research showing adults and children. Moreover, little has been done to comprehensively
robust treatment effects superior to placebo and other psychotherapies summarise and compare the efficacy between the two approaches, with
(Wolitzky-Taylor et al., 2008). Exposure therapy encourages the indi the exception of a synthesis of four direct-comparison trials published
vidual to repeatedly and systematically confront the feared object/si before 2002 (Wolitzky-Taylor et al., 2008), which suggested a marginal
tuation to learn that the feared consequences do not occur and reduce benefit of multiple-session therapy over OST on questionnaire measures
the associated anxiety reaction (Öst, 1989; Davis et al., 2012). Exposure at follow-up, although this did not reach significance (d = 0.35, p = .06).
therapy for SP can be delivered gradually over multiple spaced sessions Interestingly, a meta-analysis of randomised controlled trials (RCTs) for
in line with the traditional cognitive behavioural therapy (CBT) youth anxiety disorders found no significant difference between
* Corresponding author. School of Psychology, University of New South Wales, Sydney, New South Wales, 2052, Australia.
E-mail address: [email protected] (B.M. Graham).
https://doi.org/10.1016/j.brat.2022.104203
Received 27 April 2022; Received in revised form 22 August 2022; Accepted 20 September 2022
Available online 21 October 2022
0005-7967/© 2022 Elsevier Ltd. All rights reserved.
K. Odgers et al. Behaviour Research and Therapy 159 (2022) 104203
treatment effects produced by brief, intensive, and concentrated treatment outcomes were not significantly moderated by phobia
CBT—which included OST—and those produced by standard CBT, both subtype, however, this analysis included all types of psychotherapy.
at post-treatment and follow-up (Öst & Ollendick, 2017). However, this Whether single- or multi-session exposure effects are moderated by
analysis included a range of anxiety disorders and combined OST with phobia subtype was thought to warrant examination.
other brief and concentrated interventions, defined as those with at least 2. Developmental stage (child/adolescent vs. adult). Exposure therapy has
50% fewer sessions or weeks of treatment compared to the standard CBT been shown to be efficacious in both adult (Wolitzky-Taylor et al.,
treatment for the disorder. Moreover, the data was restricted to youth 2008) and youth populations (Davis et al., 2019). Examining
populations. Therefore, how the single-session exposure format com developmental stage as a potential effect size moderator would help
pares with the multi-session exposure format in terms of its efficacy and determine whether adults or youth benefit more from each treatment
efficiency for the treatment of SP in children and adults is a question that approach.
remains to be addressed. 3. Intervention format (group vs. individual). Although individually
Single-session exposure has a practical benefit of being potentially delivered therapy is a more commonly studied format, the efficacy of
less disruptive to the individual’s life (Zlomke & Davis, 2008), with group therapy has been examined in both OST (e.g., Öst, 1996) and
reduced out-of-pocket costs, and therefore may be a more attractive multi-session (e.g., Teachman & Woody, 2003) exposure studies.
treatment option than multi-session exposure. Yet access to Given the purported potential benefits of group-based exposures,
single-session exposure may be limited by factors such as lack of ther such as increased service accessibility and group encouragement (Li
apist training in intensive exposure delivery, treatment providers et al., 2020), it would be beneficial to determine whether treatment
placing restrictions on the maximum duration of individual sessions, and effects are greater when therapy is delivered in a group or individual
a lack of suitable rebates for extended duration treatment sessions. An format.
empirically-based understanding of the relative efficiency and efficacy 4. Presence of researcher allegiance (for OST studies only). Whether
of multi-session and single-session exposure therapy is therefore neces researcher allegiance influenced the treatment effects reported for
sary to support the development of policies and procedures that facili single-session was thought to merit examination, given the prepon
tate the ease of access to single-session treatment. derance of single-session studies employing OST, and many of these
The primary aims of this study were to evaluate the relative effi trials were conducted by OST’s founders.
ciency and efficacy of multi- and single-session exposure therapy for the
treatment of SP. The meta-analysis was guided by the following research 2. Method
questions: What is the effect of single- and multi-session exposure on the
symptoms of SP in adults and children from pre-to post-treatment? How This meta-analysis followed the guidelines specified in the Preferred
do these two approaches compare in terms of total treatment time and Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)
treatment effects? How do the treatment effects of the two approaches statement (Moher et al., 2009) and the protocol was registered in the
compare at post-treatment follow-up assessments? Given the paucity of international prospective register of systematic reviews (PROSPERO;
RCTs directly comparing the two treatment types, we employed an in October 14, 2020).
direct approach by comparing the pooled effects of single-session with
the pooled effects of multi-session treatments. Because the topic of in 2.1. Study selection
terest was not the magnitude of exposure treatment effects relative to a
control group, a question that has been addressed in the literature (see To maximise the number of studies in our sample we initially stip
Choy et al., 2007; Wolitzky-Taylor et al., 2008), rather how single- and ulated all types of exposure therapy would be eligible for selection,
multi-session compare in reducing SP symptoms, we synthesised the including in vivo, virtual, and imaginal exposure. However, we subse
within-group effect sizes for the treated group. Arguably, this enabled a quently decided to refine the eligibility criteria to only include in vivo
fairer comparison as effect sizes were unaffected by the different types of exposures. We deviated from the registered protocol in this way to
control groups employed by different studies. Using within-group results improve the comparability of selected studies and to increase the rele
also enabled the inclusion of both pre-post and controlled trials and thus vance of findings to current clinical practice, where in vivo is the most
maximised the number of studies in the meta-analysis, an important prevalent approach and adoption of virtual reality has been slow despite
consideration given the absence of control conditions noted in the its promise (Bell et al., 2020). We applied the changes to the eligibility
Zlomke and Davis (2008) review of OST studies. criteria at the full text screen stage. Final study selection was based on
Informed by findings from previous reviews (Choy et al., 2007; the following PICOS criteria.
Wolitzky-Taylor et al., 2008; Zlomke & Davis, 2008), we hypothesised
that both single- and multi-session therapy would show large beneficial 1. Participants. Participants had to have a diagnosis of SP or be
effects on SP symptom reduction, but that there would be no significant explicitly described as a clinical sample by the authors. Samples
difference between the two approaches, i.e., both would have equivalent described as treatment-seeking individuals with SP or as outpatients
efficacy. However, based on the mean weeks of treatment reported for with SP were included. Studies with a portion of subclinical partic
standard CBT methods used in SP (Öst, 2017: 6 weeks, range 3–10 ipants were included if at least 75% of the sample met full diagnostic
weeks), we predicted that the average total treatment time would differ criteria. Merely “fearful” or predominantly subclinical samples were
across single- and multi-session approaches, such that single-session excluded, and studies were excluded if diagnosis or clinical status
would achieve treatment effects equivalent to those of multi-session was not mentioned or could not be ascertained from the sample
exposure, but with significantly fewer total treatment hours. Based on description. Participant age was not limited.
the significant heterogeneity in effect sizes reported in related 2. Intervention. Interventions were required to be in vivo exposure,
meta-analyses (Wolitzky-Taylor et al., 2008; Öst & Ollendick, 2017), we defined as the direct confrontation with the feared stimulus with the
expected to see significant dispersion in effect sizes within both treat intent to reduce anxiety, as the primary component of treatment. For
ment approaches. Therefore, we also planned to conduct exploratory single-session studies, the treatment had to comprise a single pro
analyses to investigate the sources of heterogeneity within each longed session of exposure. For multi-session studies, the treatment
approach. The following potential moderators were identified a priori for had to consist of at least two spaced exposure sessions. Both indi
analysis. vidual and group treatments were included. Treatments incorpo
rating psychoeducation and cognitive restructuring elements were
1. Phobia subtype. A meta-analysis examining psychological approaches included as long as exposure was the main element accounting for
for the treatment of SP (Wolitzky-Taylor et al., 2008) found that the majority of the content (i.e., >50% of the treatment sessions
2
comprised exposure), as these components are typically incorporated data required to calculate the pre-post and follow-up effect sizes for the
in both single-session (Öst, 2012) and multi-session (Blakey et al., BAT and self-reported phobia symptom (SPS) outcome measures (pre-
2019) exposure protocols. We excluded instances in which re and post-intervention and follow-up means and standard deviations and
searchers explicitly stated that a suboptimal exposure protocol was the sample size; otherwise, the relevant F-, t-, or p-values) and (3) in
utilised, for example, proof-of-concept studies designed to test the formation to assess the risk of bias in each study.
augmenting effect of drugs (Rothbaum et al., 2014). Also excluded Each study arm was classified according to categorical variables
were virtual reality, imaginal, and Internet-based exposures, and identified as potential moderators of the effect size and used in subse
treatments combining exposure with a drug or pill placebo or with quent meta-regression analyses. These were (a) phobia subtype, (2)
other forms of therapy (e.g., relaxation training, mindfulness) such participant age group (child/adolescent or adult), (3) the intervention
that the effects of exposure could not be independently examined. format (group or individual) and for OST studies only (4) the presence of
Systematic desensitisation (Wolpe, 1954) and eye movement researcher allegiance. SP categories were based on the four types spec
desensitisation and reprocessing (EMDR; Shapiro, 1989) were ified in the DSM-5, namely, ‘animal’, ‘situational’ (including fear of
excluded as they involve additional therapeutic components (relax flying and claustrophobia); ‘blood, injury and injection’ (BII, including
ation training and bilateral eye movements respectively) that may dental phobia), and ‘natural environment’, plus a ‘mixed’ category to
confound the exposure effects. capture studies that treated a range of different phobias. There were only
3. Comparison. No restriction was imposed on the type or use of two study arms related to natural environment phobias, both involving
control or comparison groups. the fear of heights, hence this SP category was re-named ‘heights’.
4. Outcomes. The reported outcome measures had to be a direct Studies with a maximum participant age of 17 years were categorised as
measure of treatment success, pre- and post-intervention, such as the ‘child/adolescent’ samples and studies with a minimum participant age
Behavioural Approach Task (BAT; Choy et al., 2007) or a of 18 years were categorised as ‘adult’ samples. There were seven studies
self-reported measure of the target phobia symptoms (e.g., Spider that specified an age range spanning both categories. One included
Questionnaire (SPQ); Klorman et al., 1974), as opposed to a gener participants aged 8–18 years so was classified as a ‘child/adolescent’
alised measure of anxiety or psychological distress, based on study. Two studies specified a minimum age of 16 years, and four studies
recommendations of the value of incorporating multi-method specified a minimum age of 17; these were placed in the ‘adult’ category
approaches to assessment in SP (Davis & Ollendick, 2005). The based on the sample descriptions. Fourteen studies did not specify an age
authors of studies that were published within the last 10 years with range and were classified as ‘adult’ samples based on the mean age of the
insufficient quantitative data to compute the effect size from pre-to participants and sample description. OST studies were deemed to have
post-intervention were contacted via email for the required researcher allegiance if Öst, Ollendick, or Davis was listed among the
information. If the required information was not provided or the study authors, given that single-session treatment was first developed
study was published more than 10 years ago, the study was excluded. and manualised by Öst, 1989 and later Öst & Ollendick, 2001, and Davis
5. Studies. To ensure a minimum quality standard, only published et al. (2012).
peer-reviewed articles were included. There were no language re The average treatment duration in total minutes was recorded for
strictions. We included controlled and pre-post designs, including each study arm. For multi-session studies, pre-treatment assessment
multiple-baseline-across-subjects designs if the required data was sessions (where specified) were excluded from the calculation of treat
reported for the sample. Single case or qualitative designs were ment duration and if session duration was specified as a range, we took
excluded. To ensure the independence of effect sizes, secondary an the midpoint and multiplied it by the number of sessions to estimate the
alyses of existing trials and other duplicate data were excluded. total average duration. For single-session studies that specified only the
maximum duration of the session, we took the maximum as the average
Studies were identified through a systematic database search com treatment duration. This was a conservative approach, because many of
plemented by a manual search of the reference lists of other systematic the single-session treatments included in the review incorporated both
reviews and meta-analyses. The following databases were used to the assessment and the treatment within the single session.
conduct the search: PsycINFO, Embase, MEDLINE, and Cochrane. The Data extraction was completed by KO, and KAK independently coded
first search date was September 14, 2020. Our database search strategy 13 randomly selected studies, which represented 20% of the study
used combinations of search terms relating to specific phobia (specific sample. Coding was completed for all categorical moderators tested in
phobia OR *phobia OR *phobic) and exposure therapy (exposure ther the meta-regression and for the total treatment time by study. Results
apy OR in vivo exposure OR virtual reality exposure OR imaginal showed 100% agreement on all coded variables suggesting excellent
exposure OR one session treatment OR single session treatment OR rapid reliability of data extraction.
treatment OR brief treatment OR multisession treatment OR massed
exposure) and the search was limited to peer-review articles in databases 2.3. Risk of bias assessment
with this filter function. The first database search terms reflect our
original broad inclusion criteria for type of exposure (in vivo, virtual Risk of bias was assessed using a modified version of the Quality
reality and imaginal), which was refined after the abstract screen. A Assessment Tool for Quantitative Studies, developed by the Effective
second search was performed on August 5, 2022 to identify studies Public Health Practice Project, Canada (EPHPP; Thomas et al., 2004).
published after the first search. The same search terms were used with This tool has been recommended for use in systematic reviews of
the exception that the terms ‘virtual reality exposure’ and ‘imaginal intervention studies not limited to RCT designs (Deeks et al., 2003) and
exposure’ were deleted. Studies were screened against the eligibility has acceptable construct and content validity (Thomas et al., 2004). The
criteria by two independent reviewers (KO, KAK) and any disagreements tool was modified to ensure the criteria were appropriate and relevant
were resolved through discussion with a third reviewer (BMG), who had for our particular study sample; the original tool structure and specific
the final decision. modifications are listed in Appendix A. Each study was rated for risk of
bias arising from five potential factors (1) selection bias; (2) blinding of
2.2. Data extraction outcome assessment (detection bias) (3) data collection methods, pri
marily relating to the validity and reliability of measures used (4) the
Three types of data were extracted from each study: (1) general as extent and handling of missing outcome data (attrition bias), and (5)
pects of the study (title, authors, publication year and country) and intervention integrity. Studies were rated strong (1 point), moderate (2
participant and intervention characteristics, including treatment dura points) or weak (3 points) in each factor. Risk of bias was assessed by
tion and information related to potential moderators (2) quantitative KO, and KAK scored 13 randomly selected studies constituting 20% of
3
the study sample. A weighted kappa was calculated to determine the funnel plots, which plot the effect sizes against their standard errors, and
inter-rater agreement, which is appropriate for ordinal data. There was by testing statistically using Egger’s regression (Egger et al., 1997).
substantial agreement between raters (weighted κ = 0.64). Duval and Tweedie’s (2000) ‘trim-and-fill’ method was used to estimate
the number of missing studies in the funnel plot and adjust the pooled
2.4. Data analyses effect size estimates accordingly. The resulting imputed confidence in
tervals were compared to determine whether single- and multi-session
IBM SPSS Statistics version 23 was used to conduct an independent effects were significantly different after taking into account publica
samples t-test assuming unequal variances to compare the average total tion bias.
treatment time of single-session with that of multi-session exposure. The
meta-analyses were performed using the Comprehensive Meta-Analysis 3. Results
Version 3 software. Within-group effect sizes (i.e., pre-post treatment
difference, pre-follow up difference, or post-follow up difference) were Sixty-seven studies qualified for inclusion; of these 40 reported
estimated using Hedges’ g, which corrects for potential overestimation outcomes for single-session exposure, 21 for multi-session exposure, and
when sample sizes are small (Hedges, 1981). Hedges’ g was computed six studies contributed both single and multi-session data. See Fig. 1 for
using the pre- and post-intervention means and standard deviations the PRISMA flow diagram of the study selection process.
where possible; otherwise, paired t-values and F-values were used. Given
within-group correlations are rarely reported in studies, we assumed a 3.1. Study sample characteristics
correlation value of 0.50 based on the empirically-based recommenda
tion of Balk et al. (2012) and conducted sensitivity analyses using A total of 1758 participants from 67 studies were represented in the
alternative imputation values. meta-analysis; 1199 received single-session and 559 received multi-
For studies with multiple arms, each study arm was treated as an session exposure. One study (Wright et al., 2022) contributed pre to
independent sample for which a within-group effect size was calculated, follow-up data only, hence pre to post effect size estimates and moder
however, combined results were used when reported. Effect sizes were ator analyses were based on 66 studies representing a total of 1707
calculated separately for SPS and BAT outcomes for pre-post and last participants. Single-session studies contained a total of 55 study arms
follow-up where data was available. We used the intent-to-treat results with sample sizes ranging from four to 90 participants (M = 21.80, SD =
where possible. Clinician- or parent-reported symptom severity ratings 16.63). Multi-session studies comprised a total of 30 study arms with
were used for SPS outcomes when self-report questionnaire data was not sample sizes ranging from seven to 54 participants (M = 18.63, SD =
reported. For studies using multiple measures, the mean of the effect 9.47). As shown in Table 1, the majority of single- and multi-session
sizes was used, yielding one effect size per study arm for SPS and BAT interventions involved individually-delivered therapy for adults diag
outcomes respectively. This approach is consistent with past meta- nosed with animal (predominantly spider) phobia. Single-session
analyses of exposure therapy (Carl et al., 2019; Wolitzky-Taylor et al., treatment duration ranged from 30 to 300 min (M = 163.78, SD =
2008). See Tables 2 and 3 for the synthesised pre-post and follow-up 45.06) and researcher allegiance was identified in 20 studies or 23 study
outcomes by study and Appendix B for the specific outcome measures arms (n = 629). In multi-session studies, the number of sessions ranged
used to calculate the SPS outcomes by study arm. from two to nine (M = 4.47, SD = 2.10) and treatment duration ranged
A random-effects model was used to pool the effect sizes for each from 90 to 720 min (M = 300.67, SD = 149.94). Characteristics of
treatment approach, based on the assumption that different studies do studies are presented in Table 2 for single-session and Table 3 for
not share a common effect size, a consequence of their differences in multi-session interventions.
population characteristics, aspects of the intervention and/or study
design. A random-effects model assumes there is a distribution of true 3.2. Risk of bias
effect sizes and the pooled effect size is the estimated mean of that
distribution (Borenstein et al., 2010). Heterogeneity of effect sizes was The risk of bias within our study sample varied by factor. However,
tested using the Q statistic and the I2 statistic was used to estimate the the majority of studies were rated strong to moderate in quality,
proportion of total variance due to heterogeneity, with values of 25%, meaning they had a moderate to low risk of bias. For single-session
50%, and 75% indicating a low, moderate, and large amount of het studies, the weakest areas appeared to be the extent and handling of
erogeneity respectively (Higgins & Thompson, 2002). Effects of single- missing data and the potential risk of selection bias, whereas multi-
and multi-session exposure were compared by examining the 95% session studies showed relative weakness in selection bias and the
confidence intervals of their mean effect sizes for both SPS and BAT blinding of outcome assessors. Fig. 2 summarises the quality assessment
outcomes respectively. Differences between the effects of the two ap profile for single- and multi-session studies respectively. See Appendix C
proaches were further statistically tested with a mixed-effects model for study-specific quality assessment scores.
sub-group analysis (Borenstein et al., 2009).1 Finally, as exploratory
analyses, individual random-effects meta-regressions were conducted to 3.3. Efficacy analyses
test the statistical significance of potential effect-size moderators for
single- and multi-session exposure respectively. All meta-regressions 3.3.1. Treatment duration
used the Knapp-Hartung (Knapp & Hartung, 2003) method, to hold The mean total treatment time was significantly less for single-
Type I error rates close to the nominal value. session compared to multi-session treatment, t(83) = 4.88, p < .0001,
Publication bias refers to the observation that studies with positive such that multi-session was 137 min longer than single-session on
findings are more likely to be published than studies with null findings, average. We also tested whether the same pattern of results would be
which can lead to an overestimation of the pooled effect size (Borenstein obtained when the analysis was limited to only those studies that re
et al., 2009). Presence of publication bias was investigated for single- ported the exact mean treatment duration (single-session; n = 32, M =
and multi-session study samples respectively, by visually inspecting the 141, SD = 49; multi-session treatment; n = 7, M = 203, SD = 86). Again,
mean total treatment time was significantly less for single-session
compared to multi-session treatment, t(37) = 2.62, p = .01.
1
A random-effects model was used to pool the effect sizes within single- and
multi-session respectively, and a fixed-effects model was used across the two 3.3.2. Pre to post effects
groups to test whether their mean effect size estimates differed significantly As outlined in Table 4, the mean effect sizes for both single-session
from each other as would be indicated by a significant Q-between statistic. and multi-session therapy were large and significant, indicating that
4
Table 1
Characteristics of the study arm sample for single- and multi-session exposures.
Sample characteristics Single-session Multi-session
K % n % k % n %
Phobia subtype
Animal 39 70.9 724 60.4 13 43.4 293 52.4
BII/dental 5 9.1 124 10.3 5 16.7 84 15.0
Situational 3 5.5 46 3.8 7 23.3 99 17.7
Heights – – 2 6.7 43 7.7
Mixed phobias 8 14.5 305 25.4 3 10.0 40 7.2
Age segment
Child/Adolescent 13 23.6 369 30.8 1 3.3 14 2.5
Adult 42 76.4 830 69.2 29 96.7 545 97.5
Treatment format
Individual 46 83.6 1024 85.4 24 80.8 445 79.6
Group 9 16.4 175 14.6 6 20.0 114 20.4
Total 55 100.0 1199 100.0 30 100.0 559 100.0
Note. k = number of study arms, n = number of participants.
recipients of both approaches showed a significant reduction in SP pre to post-treatment effects also apply at follow-up. Recipients of both
symptoms from pre-to post-treatment as measured by SPS and BAT treatment approaches showed a significant reduction in SP symptoms
outcomes. The numerical estimates of pre-post effect sizes were very from pre-treatment to follow-up and the overlap in the 95% confidence
close for single-session and multi-session therapy for both primary intervals of follow-up effect sizes suggests the two approaches had
outcome measurements. Furthermore, there was substantial overlap in comparable efficacy beyond the end of treatment.5 The mixed-effects
the 95% confidence intervals of effect sizes for both outcomes, sug sub-group analysis found no evidence of a significant difference be
gesting the two treatment approaches were equally effective in reducing tween single-session and multi-session follow-up effects for both SPS
SP symptoms.2 This suggestion was further supported by the results of (Q-between = 2.51, p = .113) and BAT (Q-between = 3.83, p = .050)
the mixed-effects sub-group analyses, which found no evidence of a outcomes, providing further support for this finding. Heterogeneity was
significant difference between single-session and multi-session effects on significant and large for SPS outcomes within single- and multi-session,
both SPS (Q-between = 0.20, p = .658) and BAT (Q-between = 0.04, p = and for BAT outcomes within single-session study arms (all p < .001).
.836) outcomes.3 As shown in Table 4, Q values indicated a statistically Heterogeneity was non-significant for BAT outcomes within
significant level of heterogeneity within both treatment approaches, and multi-session study arms (k = 6, p = .106), although these findings
I2 estimates, ranging from 65% to 85%, suggested that it was large in should be interpreted with caution given the Q-test is known to have
magnitude. Therefore, within both single-session and multi-session poor power when there are few studies (Hardy & Thompson, 1998).
studies, a large portion of the observed dispersion in effect sizes was
due to systematic, not random, sampling error. See Appendix D for the 3.3.4. Post to follow-up effects
forest plot graphs of single-session and multi-session pre-post effects. Post-treatment to follow-up effects, presented in Table 6, were non-
significant for all outcomes except SPS in single-session exposure, which
3.3.3. Pre to follow-up effects yielded a small and significant effect. This suggests continued modest
Twenty-nine single-session and 21 multi-session study arms reported gains in self-reported symptoms following single-session exposure.
outcomes after one or more follow-up periods, ranging from 4 weeks to However, the observed overlap in the 95% confidence intervals of the
14 months, and one single-session study (Muris et al., 1993b) reported two treatment approaches suggests there is no significant difference in
outcomes 18 months post-treatment. For comparability, we calculated their outcome trajectories following treatment completion. The mixed-
treatment effects at last follow-up 3–14 months post-treatment; this effects sub-group analysis found no evidence of a significant difference
range captured over 80% of study arms with follow-up results and between single-session and multi-session post to follow-up effects for
represented a total of 665 recipients of single-session and 260 recipients both SPS (Q-between = 2.364, p = .124) and BAT (Q-between = 2.128,
of multi-session treatment. Follow-up periods were distributed similarly p = .145) outcomes, providing further support for this finding.
across the two treatment approaches suggesting follow-up effect sizes
could be meaningfully compared.4 Results of effect size analyses, pre 3.4. Publication bias
sented in Table 5, suggest that findings of non-significant differences in
Across both single-session and multi-session studies, for all pre-post
outcomes, visual asymmetry of the funnel plots (shown in Appendix E)
2
Five outliers were observed among single-session study arms: three for SPS indicated the presence of small study effects, which may be attributed to
(g = 0.78, 3.97, 4.92) and two for BAT (g = 4.18, 5.10) outcomes. One multi- publication bias (Sterne et al., 2011). Egger’s test result was significant
session outlier was identified for SPS outcomes (g = 5.41). Excluding these across all analyses (all p < .001) providing further evidence of potential
study arms had no material impact on overall findings (single-session SPS g = publication bias. Using trim-and-fill to identify missing studies and
1.74 [1.59, 1.89], I2 = 54.64; single-session BAT g = 1.91 [1.60, 2.23], I2 = impute their effect size estimates to correct for publication bias resulted
82.64; multi-session SPS g = 1.69 [1.40, 1.98], I2 = 77.68), hence the original in a downward adjustment of all four pooled effect sizes. However, as
results are reported. shown in Table 7, overlap in the 95% confidence intervals remained,
3
Repeating analyses using alternative correlation values had no impact on suggesting there was no significant difference in the pre-post effects of
the direction of findings (For r = 0.7, SPS Q-between = 0.01, p = .930; BAT Q-
the two approaches having adjusted for publication bias.
between = 0.21, p = .649. For r = 0.3, SPS Q-between = 0.45, p = .505; BAT Q-
between < 0.01, p = .958).
4
For SPS outcomes, Fisher’s test suggested there was no significant difference
in the distribution of follow-up periods between single-session and multi-
5
session studies (3–6 months and 12–14 months, p = .742). Fisher’s test was One single-session outlier was observed for SPS outcomes (g = 0.81). There
not interpretable for BAT outcomes due to the relatively low number of studies was negligible impact on overall results after its removal (g = 2.09 [1.89, 2.29],
reporting follow-up effects. I2 = 42.08), hence the original results are reported.
5
Table 2
Characteristics of studies included in the meta-analysis: Single-session.
Authors (year) Country Study arm N Phobia subtype: Age Therapy Presence Mean total Synthesised outcomesb
Details segment format of RA treatment
Pre- Follow-up
time (mins)
post (FU period
in months)
Andersson, Waara, Sweden Live-exposure 14 Animal: Spider Adult Individual Yes 180 SPS SPS (12)
Jonsson et al. (2009)
Andersson, Waara, Sweden OST 13 Animal: Spider Adult Individual Yes 180 SPS SPS (12)
Jonsson et al. (2013)
Antony et al. (2001) Canada Focus Monitors 16 Animal: Spider Adult Individual – 120 SPS,
BAT
Focus Blunters 14 Animal: Spider Adult Individual – 120 SPS,
BAT
Arntz & Lavy, 1993 Netherlands Elaboration 19 Animal: Spider Adult Individual – 150 SPS, SPS (12)
BAT
Non-elaboration 22 Animal: Spider Adult Individual – 150 SPS, SPS (12)
BAT
Botella et al. (2016) Spain IV exposure (IVE) 31 Animal: Small Adult Individual – 180 SPS, SPS, BAT
animal BAT (6)
De Jong et al. (2000) Netherlands One-session 16 Animal: Spider Adult Individual – 180 SPS, SPS (12)
exposure (EXP) BAT
Dibbets et al. (2013) Netherlands AA-cue 8 Animal: Spider Adult Individual – 152 SPS,
BAT
AA-no cue 8 Animal: Spider Adult Individual – 116 SPS,
BAT
AB-cue 9 Animal: Spider Adult Individual – 143 SPS,
BAT
AB-no cue 8 Animal: Spider Adult Individual – 125 SPS,
BAT
Farrell et al. (2018) Australia OST + play 4 Animal: Dog Child/ Individual Yes 180 SPS,
Adolescent BAT
Goossens et al. (2007) Netherlands Phobics 16 Animal: Spider Adult Group – 300 SPS
Graham et al. (2018) Australia Combined sample 90 Animal: Spider Adult Individual Yes 180 SPS, SPS, BAT
BAT (3)
Haukebø, Skaret, & Norway One-session 20 BII/dental: Adult Individual Yes 180 SPS, SPS, BAT
Öst, 2008 a Dental BAT (12)
Heading et al. (2001) Australia Live graded 14 Animal: Spider Adult Individual – 180 SPS,
exposure (LGE) BAT
Hellstrom and Öst Sweden OST therapist 10 Animal: Spider Adult Individual Yes 180 SPS SPS (12)
(1995) directed (1-S)
Hemyari, Zomorodian, Iran Single-session 20 Animal: Rat Adult Group – 180 SPS
Shojaee, Sahraian,
and Dolatshahi
(2021) a
Leutgeb, Schäfer, and Austria Therapy group 22 Animal: Spider Adult Group – 240 SPS,
Schienle (2009) (TG) BAT
Leutgeb and Schienle Austria Therapy group 16 Animal: Spider Child/ Individual – 240 SPS,
(2012) (TG) Adolescent BAT
(Li et al., 2020)Li Australia Group OST 20 Animal: Spider Adult Group – 180 SPS,
(2019) BAT
Miloff et al. (2019) Sweden OST 49 Animal: Spider Adult Individual – 172 SPS, SPS, BAT
BAT (12)
Muris, De Jong, Netherlands Monitors 17 Animal: Spider Adult Individual – 150 SPS,
Merckelbach et al. BAT
(1993a)
Blunters 19 Animal: Spider Adult Individual – 150 SPS,
BAT
Muris et al., (1993b) Netherlands In vivo exposure 33 Animal: Spider Adult Individual 150 SPS,
(IVE) BAT
Muris and Merckelbach Netherlands Control 8 Animal: Spider Adult Individual – 150 SPS,
(1997a) BAT
Muris, Merckelbach, Netherlands Exposure-EMDR 11 Animal: Spider Child/ Individual – 90 SPS,
van Haaften et al. (EXP-EMDR)c Adolescent BAT
(1997b)
Muris, Merckelbach, Netherlands Exposure in vivo 9 Animal: Spider Child/ Individual – 150 SPS,
Holdrinet et al. (Exp-IV) Adolescent BAT
(1998)
Muskett et al. (2020) USA Combined sample 14 Mixed Child/ Individual Yes 180 SPSd SPSd (12)
Adolescent
Nielsen et al. (2016) Denmark OST 10 Mixed Child/ Individual – 180 SPSd,
Adolescent BAT
Oar et al. (2015) Australia OST 24 BII/dental: BII Child/ Individual Yes 180 SPSd
Adolescent
Ollendick et al., (2009) Sweden/ OST 85 Mixed Child/ Individual Yes 180 SPSd, SPSd (6)
USA Adolescent BAT
USA OST 43 Mixed Individual Yes 180 SPSd SPSd (6)
(continued on next page)
6
Table 2 (continued )
Authors (year) Country Study arm N Phobia subtype: Age Therapy Presence Mean total Synthesised outcomesb
Details segment format of RA treatment
Pre- Follow-up
time (mins)
post (FU period
in months)
Ollendick, Child/
Halldorsdottir, Fraire Adolescent
et al. (2015)
Parent Augmented 46 Mixed Child/ Individual Yes 180 SPSd SPSd (6)
OST (A-OST) Adolescent
Ollendick, Ryan, USA A-OST nonlocal 38 Mixed Child/ Individual Yes 180 SPSd SPSd (6)
Capriola-Hall et al. Adolescent
(2018)
Öst, 1989 Sweden Spider phobics 7 Animal: Spider Adult Individual Yes 150 SPS
Öst, Salkovskis, and Sweden/UK Therapist directed 17 Animal: Spider Adult Individual Yes 126 SPS SPS (12)
Hellstrom (1991b)
Öst, Hellstrom, and Sweden One-session 20 BII/dental: Adult Individual Yes 120 SPS, SPS (12)
Kaver (1992) a Injection BAT
Öst (1996) Sweden Small group 22 Animal: Spider Adult Group Yes 180 SPS SPS (12)
Large group 20 Animal: Spider Adult Group Yes 180 SPS SPS (12)
Öst, Brandberg, and Sweden One-session 14 Situational: Adult Individual Yes 180 SPS SPS (12)
Alm (1997) a Flying
Öst, Ferebee, and Sweden Direct treatment 16 Animal: Spider Adult Group Yes 207 SPS SPS (12)
Furmark (1997)
Öst, Alm, and Sweden One-session 15 Situational: Adult Individual Yes 180 SPS SPS (14)
Brandberg (2001) a Claustrophobia
Papalini, Lange, Bakker Netherlands Exposure therapy 25 Animal: Spider Adult Group – 180 SPS
et al. (2019)
Powers et al. (2004) USA Exposure only 17 Situational: Adult Individual – 30 SPS
(EO) Claustrophobia
Raes et al. (2011) Belgium Exposure only 15 Animal: Spider Adult Individual – 163 SPS
(EXP)
Behavioural 16 Animal: Spider Adult Individual – 166 SPS
Experiments (BE)
Rentz et al. (2003) USA In vivo exposure 26 Animal: Dog Adult Individual – 30 BATe
(IV)
Schienle et al. (2007) Austria Therapy group 14 Animal: Spider Adult Group – 240 SPS,
(TG) BAT
Schmid-Leuz et al. Germany Attention 32 BII/dental: Adult Individual – 60 SPS
(2007) Dental
Vika, Skaret, Raadal Norway One-session 28 BII/dental: Adult Individual Yes 167 SPS, SPS, BAT
et al. (2009) a Injection BAT (12)
Waters et al. (2014) Australia Attention-training 18 Mixed Child/ Individual Yes 180 SPSd SPSd (3)
control plus OST Adolescent
(ATC + OST)
Wright et al., (2022) England OST 51 Mixed Child/ Individual Yes 180 BAT (6)
Adolescent
Wrzesien et al. (2012) Spain IV Exposure 10 Animal: Small Adult Individual – 180 BAT
(IVET) animal
Note. n = number of participants; RA = Researcher allegiance; FU = Follow-up.

a
Study contributing both multi-session and single-session data.
b
See Appendix B for study-specific outcome measures included in the SPS outcome.
c
Results after treatment 1 (exposure).
d
Clinician- or parent-reported symptom severity ratings.
e
Combination of two BATs.
3.5. Exploratory analysis: effect size moderators SPS and p = .025 for BAT outcomes.6 Results suggested that recipients of
single-session exposure for animal phobia had significantly larger im
Within single-session studies, individual random effects meta re provements in SPS outcomes compared to those treated for BII/dental (p
gressions found no statistically significant relationship between pre to < .001) and mixed (p = .006) phobias, but not compared to those treated
post treatment improvement in SP symptoms and the participant age for situational phobia (p = .034). A directionally different association
segment (SPS p = .107; BAT p = .124), presence of researcher allegiance was found in BAT outcomes, with single-session recipients for BII/dental
(SPS p = .094; BAT p = .249), or whether therapy was delivered in a phobia showing significantly larger improvements in BAT scores
group or individualised format (SPS p = .148; BAT p = .365). The treated compared to those treated for animal phobia (p = .008). BAT outcomes
phobia subtype was a significant moderator of pre to post effect sizes for were significantly more improved following single-session exposure for
SPS (F(3, 48) = 7.19, p < .001) and BAT outcomes (F(2, 28) = 7.70, p = animal phobia relative to single-session exposure for mixed phobias (p
.002). We conducted pairwise comparisons using animal phobia as the = .023), although these findings are limited by the small number of
reference category and Bonferroni-adjusted alpha levels of p = .017 for study arms reporting BAT outcomes for mixed phobias (k = 2). See
6
Three pairwise comparisons were performed for SPS outcomes (animal vs.
BII/dental, mixed, and situational phobia) whereas only two were possible for
BAT outcomes (animal vs. BII/dental and mixed).
7
Table 3
Characteristics of studies included in the meta-analysis: Multi-session exposure therapy.
First author (year) Country Study arm N Phobia subtype: Age Therapy No. of Mean total Synthesised outcomes
b
Details segment format sessions treatment
time (mins)
Pre- Follow-up
post (FU period
in months)
Alpers and Sell Germany IV exposure (IV) 10 Situational: Adult Individual 6 180 SPS,
(2008) Claustrophobia BAT
Biran and Wilson USA Guided exposure 11 Mixed Adult Individual 5 250 BATc
(1981) (GE)
Blakey et al., (2019) USA Exposure with 30 Animal: Spider Adult Individual 4 240 SPS,
elimination of safety BAT
behaviour (E/ESB)
Emmelkamp et al. Netherlands In vivo (IV) 16 Heights Adult Individual 3 180 SPS, SPS (6)
(2002) BAT
Gilroy et al. (2000) Australia Live graded exposure 15 Animal: Spider Adult Individual 3 135 SPS, SPS, BAT (3)
(LGE) BAT
Haukebø et al. Norway Five-session 19 BII/dental: Adult Individual 5 300 SPS, SPS, BAT
(2008) a Dental BAT (12)
Hemyari et al. Iran Multi-session 20 Animal: Rat Adult Group 4 480 SPS
(2021) a
Michaliszyn et al. Canada In vivo (IV) 16 Animal: Spider Adult Individual 8 720 SPS, SPS, BAT (3)
(2010) BAT
Ost, Johansson, and Sweden Exposure IV: 7 Situational: Adult Individual 8 480 SPS SPS (14)
Jerremalm (1982) Behavioural reactors Claustrophobia
(E-BR)
Exposure IV: 8 Situational: Adult Individual 8 480 SPS SPS (14)
Psychological Claustrophobia
reactors (E-PR)
Öst, Lindahl, and Sweden Exposure IV (E) 9 BII/dental: Blood Adult Individual 9 473 SPS SPS (6)
Sterner (1984)
Öst, Fellenius, and Sweden Exposure IV (E) 10 BII/dental: Blood Adult Individual 5 263 SPS SPS (14)
Sterner (1991)
Öst, Hellstrom, & Sweden Five-session 19 BII/dental: Adult Individual 5 212 SPS, SPS (12)
Kaver, 1992 a Injection BAT
Öst, Brandberg, & Sweden Five-session 14 Situational: Adult Individual 5 360 SPS SPS (12)
Alm, 1997 a Flying
Öst, Alm, & Sweden Five-session 16 Situational: Adult Individual 5 300 SPS SPS (14)
Brandberg, 2001 a Claustrophobia
Preusser et al. (2017) Germany Exposure treatment 23 Animal: Spider Adult Individual 2 120 SPS
(ET)
Raeder et al. (2019) Germany Combined sample 54 Animal: Spider Adult Individual 2 103 SPS,
BAT
Rihm et al. (2016) Switzerland Wake 18 Animal: Spider Adult Group 2 240 SPS,
BAT
Rothbaum, Hodges, USA Standard exposure 15 Situational: Adult Individual 7 420 SPS SPS (6)
Smith et al. (2000) (SE) Flying
Rothbaum, USA Standard exposure 29 Situational: Adult Individual 7 420 SPS SPS (12)d
Anderson, Zimand (SE) Flying
et al. (2006)
St-Jacques et al. Canada In vivo exposure (IV) 14 Animal: Spider Child/ Individual 5 300 SPS, SPS (6)
(2010) Adolescent BAT
Steinman and USA Exposure 27 Heights Adult Individual 2 180 SPS,
Teachman (2014) BAT
Steketee et al. (1989) USA Combined sample 27 Animal: Mixed Adult Individual 2 90 SPS,
BAT
Straube et al. (2006) Germany Therapy group (TG) 13 Animal: Spider Adult Group 2 540 SPS
Teachman and USA Phobic 31 Animal: Spider Adult Group 3 270 SPS,
Woody (2003) BAT
Vansteenwegen et al. Belgium Same context 14 Animal: Spider Adult Group 3 360 SPS, SPS (12),
(2007) BAT BAT (3)
Different context 18 Animal: Spider Adult Group 3 360 SPS, SPS (12),
BAT BAT (3)
Vika et al. (2009) a Norway Five-session 27 BII/dental: Adult Individual 5 264 SPS, SPS, BAT
Injection BAT (12)
Williams et al. USA Mastery 15 Mixed Adult Individual 3 146 BAT
(1984)
Exposure 14 Mixed Adult Individual 3 155 BAT
Note. n = number of participants; FU = Follow-up.

a
indicates study contributing both multi-session and single-session data.
b
See Appendix B for study-specific outcome measures included in the SPS outcome.
c
Combination of two BATs.
d
Follow-up results included treated waitlist controls (n = 40).
8
Fig. 1. PRISMA flow diagram of study selection process (Moher et al., 2009).
Table 8 for the subgroup analysis showing the pooled mean effects of is the first comprehensive synthesis and comparison of the efficacy and
single-session exposures by phobia subtype. total therapy time of single- and multi-session exposure for the treat
Within multi-session studies, therapy format (i.e., group vs. indi ment of SP based on data from studies not limited to direct comparison
vidual) was not a significant moderator of effect sizes for SPS (p = .058) trials. Strengths of this study include its focus on clinical populations,
or BAT outcomes (p = .964) and we were unable to analyse age segment inclusion of adults and children, the synthesis of both self-report and
as a moderator because all but one multi-session studies were conducted behavioural outcome measures, and the extension of the analysis to
with adults. Phobia subtype was a significant moderator of pre-post ef include follow-up effects.
fect sizes for BAT (F(3, 14) = 7.56, p = .003), but non-significant for SPS Our results found no evidence to suggest that single- and multi-
outcomes (p = .077). Pairwise comparisons with animal phobia as the session exposure have different pre-post effects on self-reported symp
reference category and a Bonferroni-adjusted alpha level of p = .017 toms and approach behaviour. However, single-session exposure
indicated that multi-session therapies for BII/dental phobia were asso required significantly less time, suggesting it may be a more time-
ciated with significantly larger improvements in BAT scores compared efficient intervention. Indeed, single-session exposure delivered treat
to those for animal phobia (p = .012). Multi-session therapies for animal ment benefits that appeared equivalent to those of multi-session treat
phobia produced significantly larger effect sizes compared to those for ment (based on overlapping confidence intervals) with approximately 2
height phobias (p = .015), however findings must be interpreted with h less total therapy time (equating to a 45% reduction) compared to
caution as there were only two study arms reporting BAT outcomes for multi-session treatment on average. This represents a considerable
height phobias. There was no significant difference between the im reduction in direct treatment costs. As previously stated, for single-
provements in BAT outcomes observed for mixed compared to animal session studies that did not specify the mean total treatment time, we
phobia (p = .124). The pooled mean effect-sizes of multi-session expo took the maximum treatment time as a proxy (as noted, this sometimes
sure by phobia subtype are presented in Table 8. includes an assessment session, which we were able to exclude from the
As Table 8 shows, the Q-test suggested a significant level of hetero time calculations for studies of multi-session treatment). Even taking
geneity remained in the animal phobia subgroup within each treatment this conservative approach, single-session was more time efficient than
approach, and I2 indicated that this was moderate to large in size. Re multi-session therapy, a difference that remained significant in our
sults of heterogeneity analyses for other phobia subgroups varied with follow-up analysis in which we only included studies that specified the
both significant and non-significant findings. However, these findings exact mean treatment duration. Nevertheless, findings regarding dif
should be interpreted with caution given the small number of studies ferences in total treatment time should be interpreted with some
comprising these subgroups and the known limitations of heterogeneity caution. In multi-session studies, treatment duration ranged from two to
tests when the number of studies is small (Hardy & Thompson, 1998). nine sessions. It is possible that in some cases, participants might have
reached remission earlier than the standard number of sessions however
4. Discussion continued with treatment as participating in a trial. This could have
inflated the session number and mean total treatment time for the multi-
Exposure therapy has been established as the preferred treatment for session approach. Future studies tracking the time course of symptom
SP (Choy et al., 2007), one of the most common mental disorders, and a reduction in single-versus multi-session treatment are required to pro
known risk factor for subsequent mental health problems (Wardenaar vide more precise estimates of relative treatment efficiency.
et al., 2017). Recent years have seen considerable research into the ef We found no significant difference between the mean effects of sin
ficacy of single-session exposure for SP; however, how this approach gle- and multi-session exposure at follow-up assessments, and again, a
compares to multi-session exposure is not yet fully understood. We high overlap in confidence intervals of effect sizes indicating potentially
aimed to extend previous research which found a non-significant dif equivalent effects. The higher pre to follow-up effects compared to pre to
ference in the efficacy of multi-session and OST for SP across four direct post effects suggest patients continued to improve after both treatment
comparison trials (Wolitzky-Taylor et al., 2008). To our knowledge, this approaches, and our analysis of post to follow-up effects indicated this
9
Fig. 2. Quality assessment results for single- and multi-session studies.
Table 4
Pooled pre to post effect sizes by outcome for single and multi-session therapy.
Outcome k N Treatment effect Heterogeneity
Mean ES (g) 95% CI z-score p Q p I2
LL UL
SPS
Single-session 52 1112 1.78 1.61 1.94 20.96 <.001 145.88 <.001 65.04
Multi-session 27 517 1.77 1.46 2.09 11.15 <.001 132.23 <.001 80.34
BAT
Single-session 31 674 2.05 1.72 2.38 12.13 <.001 195.01 <.001 84.62
Multi-session 18 385 2.10 1.68 2.51 9.94 <.001 98.69 <.001 82.77
Note. k = total number of study arms included in effect size estimate; n = number of participants; ES = effect size (pre-post difference); CI = confidence interval; UL =
upper limit; LL = lower limit.
was the case for SPS outcomes. Our findings contradict earlier research studies, with follow-up effects for SPS outcomes pooled from 25
(Wolitzky-Taylor et al., 2008) suggesting a marginal advantage of single-session studies (including those following OST protocols) and 17
multi-session exposure over single-session exposure at follow-up for SPS multi-session study arms. However, the evidence for equivalent
outcomes and a positive relationship between the number of sessions follow-up effects on BAT outcomes is less robust given the small number
and exposure efficacy. Our results are based on substantially more of study arms used for this comparison (k = 7 and 6 for single- and
10
Table 5
Pooled pre to follow-up effect sizes by outcome for single-session and multi-session therapy.
Outcome k N Treatment effect Heterogeneity
LL UL
SPS
Single-session 24 615 2.04 1.79 2.28 16.40 <.001 64.74 <.001 64.48
Multi-session 17 250 1.73 1.35 2.12 8.85 <.001 60.67 <.001 73.63
BAT
Single-session 7 307 2.00 1.29 2.70 5.55 <.001 88.96 <.001 93.26
Multi-session 6 91 3.03 2.37 3.69 8.96 <.001 9.07 .106 44.85
Note. k = total number of study arms included in effect size estimate; n = number of participants; ES = effect size (pre-follow up difference); CI = confidence interval;
UL = upper limit; LL = lower limit.
Table 6
Pooled post to follow-up effect sizes by outcome for single- and multi-session therapy.
Outcome K N Treatment effect Heterogeneity
LL UL
SPS
Single-session 24 615 0.29 0.15 0.43 3.98 <.001 64.18 <.001 64.16
Multi-session 17 250 0.11 − 0.01 0.24 1.78 .075 16.93 .390 5.47
BAT
Single-session 6 256 0.07 − 0.10 0.23 0.80 .424 8.27 .142 39.53
Multi-session 6 91 − 0.09 − 0.29 0.10 − 0.94 .346 2.80 .730 0.00
Note. k = total number of study arms included in effect size estimate; n = number of participants; ES = effect size; CI = confidence interval; UL = upper limit; LL = lower
limit.
despite the significantly shorter therapy time? The intervention studies

Table 7
in our sample did not directly address this or offer theoretical explana
Trim-and-fill results: Adjusted pre-post effect sizes for single- and multi-session
tions. The inhibitory learning model of exposure stipulates that exposure
therapy.
to feared stimuli enables new learning to compete with or inhibit pre
Outcome k Adjusted treatment effects
viously learnt threat associations (Craske, Treanor, Conway, Zbozinek,
missing
Mean ES Change in g post 95% CI of & Vervliet, 2014). Animal literature supports the role of massed expo
(g) adjustment Mean ES sure in fear extinction learning, with rodent models showing that massed
LL UL exposure was more effective than spaced exposure in initiating extinc
SPS tion learning, however once initiated, extinction learning was better
Single- 17 1.47 − 0.321 1.28 1.65 enhanced with temporally spaced trials (Cain et al., 2003). Single ses
session sion protocols like OST potentially parallel this by inducing extinction
Multi- 10 1.24 − 0.56 0.91 1.58 learning through massed exposure and promoting further spaced expo
session
BAT
sure as part of its protocol to maintain treatment gains. Some argue that
Single- 13 1.33 − 0.72 0.98 1.67 OST exposure is conducted differently to typical in vivo exposure ther
session apy, with greater emphasis on behavioural experiments to test patients’
Multi- 6 1.51 − 0.59 1.06 1.96 catastrophic cognitions and active prevention of any behavioural or
session
cognitive avoidance during exposure (see Zlomke & Davis, 2008). The
Note. k = number of missing studies identified for funnel plot symmetry; ES = momentum created during a prolonged session may help patients
effect size; CI = confidence interval; UL = upper limit; LL = lower limit. Change progress more rapidly through their exposure tasks. Moreover, OST may
in g represents the difference between the observed pooled ES and the imputed be particularly effective for certain phobias (e.g. animal phobia) that
pooled ES estimate. require more confrontation time with the feared stimulus to adequately
test catastrophic predictions (Abramowitz, 2013), a relevant consider
multi-session respectively). Additionally, very few studies reported the ation given the overrepresentation of animal phobia in the SP literature.
extent of participants’ exposure to their feared stimulus between treat Turning to the results of our exploratory examination of effect size
ment endpoint and the follow-up assessment and we did not account for moderators, only phobia subtype was significantly related to the pre-
whether participants were encouraged to continue exposure or post effect size for both single- and multi-session treatments. This is
instructed in relapse prevention, factors which may have impacted inconsistent with previous meta-analytic research, which found no ev
long-term treatment efficacy. There was variability between studies and idence of an association between controlled effects of exposure and the
treatment approaches in their emphasis of self-directed exposure post phobia subtype (Wolitzky-Taylor et al., 2008). Our results revealed that
treatment. Twenty of the 25 single-session study arms included in the phobia subtype significantly moderated treatment effects, however,
follow-up analysis reported that they encouraged participants to findings differed according to the outcome measure. Improvement in
continue exposure or provided maintenance instructions, whereas for approach behaviour following exposure was significantly greater in
multi-session exposure, 10 of the 17 study arms included relapse patients with BII/dental phobia than in patients with animal phobia,
prevention. irrespective of the treatment approach. However, single-session expo
Our results beg the question: How does single-session exposure sure produced greater improvement in self-reported symptoms for ani
achieve treatment effects equivalent to those of multi-session exposure, mal phobia than for BII/dental phobia. The discrepancy in results
11
Table 8
Single- and multi-session pre-post effects by phobia subtype.
Sub-groups K n Stratified pre-post effects Heterogeneity
Mean ES (g) 95% CI z-score p Q p I2 (%)
LL UL
Single-session
SPS
Animal 37 688 2.02 1.84 2.20 21.59 <.001 83.89 <.001 57.09
BII/dental 5 124 1.13 0.74 1.52 5.67 <.001 6.01 .198 33.46
Mixed 7 254 1.43 1.10 1.76 8.52 <.001 6.64 .355 9.68
Situational 3 46 1.33 0.76 1.89 4.62 <.001 1.77 .412 0.00
BAT
Animal 26 511 1.97 1.68 2.26 13.29 <.001 88.97 .000 71.90
BII/dental 3 68 3.61 2.64 4.58 7.28 <.001 10.12 .006 80.24
Mixed 2 95 0.70 − 0.20 1.60 1.52 .128 0.01 .913 0.00
Multi-session
SPS
Animal 13 292 2.07 1.63 2.50 9.33 <.001 81.02 <.001 85.19
BII/dental 5 84 2.17 1.43 2.92 5.74 <.001 7.59 .108 47.28
Heights 2 42 1.09 0.08 2.10 2.11 .035 0.11 .744 0.00
Situational 7 99 1.21 0.64 1.79 4.13 <.001 13.34 .038 55.02
BAT
Animal 10 237 2.10 1.72 2.48 10.86 <.001 33.30 <.001 72.97
BII/dental 3 65 3.63 2.79 4.47 8.48 <.001 0.24 .887 0.00
Heights 2 43 0.82 0.11 1.53 2.28 .023 0.45 .501 0.00
Mixed 3 40 1.39 0.73 2.04 4.12 <.001 0.83 .659 0.00
Note. k = total number of study arms included in effect size estimate; n = total number of participants, ES = effect size; CI = confidence interval, UL = upper limit; LL =
lower limit.
between outcome measures may reflect large heterogeneity in how BATs considering the important role of mass vaccinations in the current
were administered both within and between different phobia subtypes, COVID-19 pandemic (Love & Love, 2021). Somewhat relatedly, the
with respect to aspects like task instructions (and the presence versus relative paucity of multi-session exposure studies compared to
absence of demand characteristics), and the number of task steps, as single-session, particularly in child and adolescent populations, in part
described by Castagna et al. (2017) in their review on the use of BATs in reflects that exposure is often included as an element in studies of
studies of anxious youth. Results suggesting that single-session effect generic or transdiagnostic CBT for anxiety disorders, such as the
sizes were unrelated to researcher allegiance are particularly Child-Adolescent Anxiety Multimodal Study (Walkup et al., 2008) and
note-worthy and lend confidence in overall findings, given that OST the Alleviating Specific Phobias Experienced by Children Trial (ASPECT;
manual authors were involved in over 40% of included single-session Wright et al., 2022). Such studies were not included in the present
studies. analysis as we could not demarcate the relative effects of exposure
Although we included all phobia subtypes in our study selection, the versus other elements of CBT. As such, there is a need for future
study sample was dominated by spider phobia, potentially obscuring meta-analyses or non-inferiority trials to establish the equivalency (or
different effects in other phobias. Subgroup analysis results were lack thereof) of single-session exposure to generic CBT that incorporates
examined to see whether the absence of significant differences in exposure, such as the recently completed ASPECT trial (Wright et al.,
treatment benefits between single-session and multi-session exposure 2022). This effectiveness trial demonstrated that OST (defined as one
applied to different phobia subtypes. Available data allowed us to main exposure session with an optional extra session) produced
compare the 95% confidence intervals of single- and multi-session effect non-inferior results, at a significantly reduced cost, compared to CBT on
sizes for animal, BII/dental and situational phobias. Confidence in the primary outcome measure of behavioural approach amongst chil
tervals overlapped across all possible comparisons, suggesting that dren and young people with specific phobias measured six months after
finding of equivalent effects extended to animal, BII/dental and situa treatment.
tional phobias. However, other than for animal phobia, effect size esti
mates were based on as few as three studies, hence we cannot be 4.1. Limitations
confident about the robustness of this conclusion.
The above discussion highlights a weakness in the current evidence Results should be interpreted in the context of several limitations. As
base for SP: although SP is a highly heterogenous disorder, the literature previously discussed, constraints in data quality and availability may
is dominated by animal, in particular, spider phobia intervention studies have impacted the robustness of some findings, including estimates of
leaving other phobias relatively under-examined. Furthermore, in their the mean treatment time and follow-up effects.
review of OST, Zlomke and Davis (2008) noted an absence of control In addition, there was a large amount of unexplained heterogeneity
conditions in the SP literature. Our review suggests this remains the case in almost all pooled effect sizes, suggesting studies may be estimating
with only five controlled OST studies published since 2008, and among different underlying population effects and caution is required when
these, only two large-scale trial (Ollendick et al., 2009; Wright et al., interpreting results. The few instances of non-significant heterogeneity
2022). As for multi-session exposure, approximately half of our study occurred in effects pooled from a small number of study arms, hence
sample tested exposure against a control, however, only three could not be interpreted with confidence (Hardy & Thompson, 1998).
adequately controlled trials have been published in the last 10 years. As Although phobia subtype was a significant moderator of effects, the
the vast majority of studies on exposure for SP do not include an significant heterogeneity remaining within the animal phobia subgroup
adequate control, how exposure effects compare against spontaneous suggests there are other factors driving the dispersion in effects for both
remission, or against non-specific, therapeutic elements has not been single-session and multi-session studies. While examining these mod
extensively tested. There is a need for more adequately controlled trials erators was out of scope for our meta-analysis, a recent review of 111
of exposure therapy for different types of phobias, especially BII phobia, studies investigating factors influencing the success of exposure therapy
12
for SP (Bohnlein et al., 2020) found evidence for multiple factors, Results suggest that compared to multi-session exposure, single-
including high motivation and self-efficacy and low trait anxiety session exposure is a more time-efficient treatment with no disadvan
pre-treatment; and a focus on cognitive changes and context variation tage in the size and durability of effects. In addition to the suggested
during treatment. time and cost savings, other potential advantages of single-session
A further limitation was the suggestion of potential publication bias, exposure not directly examined in this review include less disruption
which may have inflated effect size estimates. Funnel plot asymmetry to individuals’ lives (Zlomke & Davis, 2008) and a reduced risk of
may be due to publication bias or may reflect factors such as clinical or attrition (Öst & Ollendick, 2017). Given these merits, policy makers
methodological differences, other reporting biases, or chance (Sterne should ensure that healthcare rebates are structured in such a way that
et al., 2011). Regardless of the underlying reason, funnel plot asym do not disadvantage recipients of single-session exposure. Clinicians
metry was evident in both single- and multi-session samples and reas should also embrace single-session exposure as a highly effective treat
suringly, both treatment approaches showed equivalent effect sizes even ment option for SP. There is substantial evidence that clinicians are often
after respective downward adjustments for potential publication bias. reluctant to administer exposure therapy, despite its strong evidence
Finally, there may have been systematic differences between single- base, however this reluctance is modifiable via training (Farrell et al.,
session and multi-session studies on unmeasured factors that could have 2016). It is possible that better education about the merits and practi
contributed to the results, including differences in therapist training and calities of single-session exposure in clinical training could also facilitate
experience, protocol adherence, and recruitment methods (e.g., outpa increased uptake of exposure by clinicians. Finally, although this anal
tient clinics versus research participation sites, which may lead to dif ysis has focused on SP, intensive, prolonged delivery of CBT and brief or
ferences in clinical severity of the sample). As this information was not abbreviated CBT (e.g., 4 sessions) has shown promise in reducing
consistently reported amongst the included studies, we were unable to symptoms of other anxiety disorders, like panic disorder (Deacon &
test this possibility. Abramowitz, 2006) and social anxiety disorder (Jain et al., 2021). The
relative efficacy of traditional CBT compared to intensive or brief CBT
4.2. Implications for research and policy for these disorders has yet to be examined. Future research identifying
treatment formats that deliver the most time-efficient therapeutic gains
In addition to the need for more adequately controlled intervention across different diagnoses could be conducted to inform policies on the
studies for different types of phobias, this review highlighted some areas structure of healthcare benefits for the different formats.
for future research. Due to limited data, we were unable to compare
treatment efficacy beyond 14 months post treatment and most follow-up
assessments relied exclusively on subjective self-report measures, 4.3. Conclusion
without the added objectivity of BAT outcomes (Choy et al., 2007).
Longer-term effectiveness of any SP treatment is a key consideration Results suggest that single-session and multi-session therapy are both
given the common phenomenon of the return of fear and risk of relapse highly efficacious for SP, however, single-session takes significantly less
in phobia patients (Eaton et al., 2018). Longer follow-up studies with treatment time with no evidence for differential benefits. Thus, single-
behavioural measures of outcomes, alongside measures for changes in session exposure formats, such as OST, represent potential for time
diagnostic status, would be beneficial to better understand the durability and cost savings without compromising treatment efficacy.
of effects and address possible bias in self-report measures. Relatedly,
analysis of outcome measures was restricted to phobic-specific symp Funding
toms. However, it is possible that single- and multi-session exposure
differentially influence important factors beyond phobic-specific Preparation of this manuscript was supported by Interlude funding
symptoms, such as processes underlying phobias (e.g., cognitive bia from the University of New South Wales to BMG.
ses) or comorbid conditions. Indeed, there is evidence that OST leads to
reductions in symptoms of comorbid anxiety disorders in children (Ryan Declarations of interest
et al., 2017), and reductions in depressive symptoms in adults (Öst, Alm,
& Brandberg, 2001). The included studies measured a large range of None.
outcomes, from attention bias to therapeutic alliance, relapse rates and
treatment credibility. However, large heterogeneity between studies Acknowledgements
prevented these additional outcomes being included in the
meta-analysis. Future studies testing whether single- and multi-session This manuscript was submitted in partial fulfilment of the re
treatment lead to equivalent reductions in comorbid diagnoses and quirements for the degree of Master of Clinical Psychology at the Uni
other relevant factors, such as underlying mechanisms and relapse rates, versity of New South Wales 2021 by KO.
would provide additional information to ascertain the relative benefits
of each approach.
Appendix A. Supplementary data
This meta-analysis examined how single- and multi-session treat
ment compare in terms of total treatment time, a driver of treatment cost
Supplementary data to this article can be found online at https://doi.
and a potential barrier to seeking treatment. However, many individuals
org/10.1016/j.brat.2022.104203.
with SP refuse treatment for other reasons, such as fear of confronting
their feared stimulus (see Wolitzky-Taylor et al., 2008). High patient
dropout and low treatment acceptance have been highlighted as po References
tential problems for in vivo exposure-based treatments for SP (Choy Abramowitz, J. S. (2013). The practice of exposure therapy: Relevance of cognitive-
et al., 2007). The relative palatability of single- and multi-session behavioral theory and extinction theory. Behavior Therapy, 44(4), 548–558.
exposure for SP patients was not addressed in this meta-analysis and Alpers, G. W., & Sell, R. (2008). And yet they correlate: Psychophysiological activation
predicts self-report outcomes of exposure therapy in claustrophobia. Journal of
merits future examination. For example, an analysis of children and
Anxiety Disorders, 22(7), 1101–1109. Netherlands.
adolescents (aged 7–16 years) experiences of OST indicated that the vast Andersson, G., Waara, J., Jonsson, U., et al. (2009). Internet-based self-help versus one-
majority wanted to participate in treatment, felt positively toward the session exposure in the treatment of spider phobia: A randomized controlled trial.
OST therapist, and were satisfied with the format of treatment and its Cognitive Behaviour Therapy, 38(2), 114–120.
Andersson, G., Waara, J., Jonsson, U., et al. (2013). Internet-based exposure treatment
outcome (Svensson et al., 2002) suggesting that OST is a well-tolerated versus one-session exposure treatment of Snake phobia: A randomized controlled
treatment mode in youth. trial. Cognitive Behaviour Therapy, 42(4), 284–291.
13
Antony, M. M., McCabe, R. E., Leeuw, I., et al. (2001). Effect of distraction and coping Graham, B. M., Li, S. H., Black, M. J., et al. (2018). The association between estradiol
style on in vivo exposure for specific phobia of spiders. Behaviour Research and levels, hormonal contraceptive use, and responsiveness to one-session-treatment for
Therapy, 39(10), 1137–1150. spider phobia in women. Psychoneuroendocrinology, 90, 134–140.
Arntz, A., & Lavy, E. (1993). Does stimulus elaboration potentiate exposure in vivo Hardy, R. J., & Thompson, S. G. (1998). Detecting and describing heterogeneity in meta-
treatment? Two forms of one-session treatment of spider phobia. Behavioural analysis. Statistics in Medicine, 17(8), 841–856.
Psychotherapy, 21(1), 1–12. Haukebø, K., Skaret, E., Öst, L.-G., et al. (2008). One- vs. five-session treatment of dental
Balk, EM, Earley, A, Patel, K, Trikalinos, TA, & Dahabreh, IJ (November 2012). Empirical phobia: A randomized controlled study. Journal of Behavior Therapy and Experimental
Assessment of Within-Arm Correlation Imputation in Trials of Continuous Outcomes. Psychiatry, 39(3), 381–390.
Methods Research Report. (Prepared by the Tufts Evidence-based Practice Center Heading, K., Kirkby, K. C., Martin, F., et al. (2001). Controlled comparison of single-
under Contract No. 290-2007-10055-I.) AHRQ Publication No. 12(13)-EHC141-EF. session treatments for spider phobia: Live graded exposure alone versus computer-
Rockville, MD: Agency for Healthcare Research and Quality www.effectivehealthca aided vicarious exposure. Behaviour Change, 18(2), 103–113.
re.ahrq.gov/reports/final.cfm. Hedges, L. V. (1981). Distribution theory for Glass’s estimator of effect size and related
Bell, I. H., Nicholas, J., Alvarez-Jimenez, M., et al. (2020). Virtual reality as a clinical tool estimators. Journal of Educational Statistics, 6(2), 107–128.
in mental health research and practice. Dialogues in Clinical Neuroscience, 22(2), Hellstrom, K., & Öst, L.-G. (1995). One-session therapist directed exposure vs two forms
169–177. of manual directed self-exposure in the treatment of spider phobia. Behaviour
Biran, M., & Wilson, G. T. (1981). Treatment of phobic disorders using cognitive and Research and Therapy, 33(8), 959–965.
exposure methods: A self-efficacy analysis. Journal of Consulting and Clinical Hemyari, C., Zomorodian, K., Shojaee, M., Sahraian, A., & Dolatshahi, B. (2021). The
Psychology, 49(6), 886–899. effect of personality traits on cognitive behavioral therapy outcomes in student
Blakey, S. M., Abramowitz, J. S., Buchholz, J. L., et al. (2019). A randomized controlled Pharmacists with Rat phobia: A randomized clinical trial. Iranian Journal of Medical
trial of the judicious use of safety behaviors during exposure therapy. Behaviour Sciences, 46, 23–31.
Research and Therapy, 112, 28–35. Higgins, J. P. T., & Thompson, S. G. (2002). Quantifying heterogeneity in a meta-
Bohnlein, J., Altegoer, L., Muck, N. K., et al. (2020). Factors influencing the success of analysis. Statistics in Medicine, 21(11), 1539–1558.
exposure therapy for specific phobia: A systematic review. Neuroscience & Jain, N., Stech, E., Grierson, A. B., et al. (2021). A pilot study of intensive 7-day internet-
Biobehavioral Reviews, 108, 796–820. based cognitive behavioral therapy for social anxiety disorder. Journal of Anxiety
Borenstein, M., Hedges, L. V., Higgins, J. P. T., et al. (2009). Introduction to meta-analysis. Disorders, 84, Article 102473.
Hoboken: John Wiley & Sons. Incorporated. Klorman, R., Weerts, T. C., Hastings, J. E., et al. (1974). Psychometric description of
Borenstein, M., Hedges, L. V., Higgins, J. P. T., et al. (2010). A basic introduction to some specific-fear questionnaires. Behavior Therapy, 5, 401–409.
fixed-effect and random-effects models for meta-analysis. Research Synthesis Methods, Knapp, G., & Hartung, J. (2003). Improved tests for a random effects meta-regression
1(2), 97–111. with a single covariate. Statistics in Medicine, 22(17), 2693–2710.
Botella, C., Perez-Ara, M. A., Breton-Lopez, J., et al. (2016). In vivo versus augmented Leutgeb, V., Schäfer, A., & Schienle, A. (2009). An event-related potential study on
reality exposure in the treatment of small animal phobia: A randomized controlled exposure therapy for patients suffering from spider phobia. Biological Psychology, 82
trial. PLoS One, 11(2), Article e0148237. (3), 293–300.
Cain, C. K., Blouin, A. M., & Barad, M. (2003). Temporally massed CS presentations Leutgeb, V., & Schienle, A. (2012). Changes in facial electromyographic activity in
generate more fear extinction than spaced presentations. Journal of Experimental spider-phobic girls after psychotherapy. Journal of Psychiatric Research, 46(6),
Psychology: Animal Behavior Processes, 29(4), 323–333. 805–810.
Carl, E., Stein, A. T., Levihn-Coon, A., et al. (2019). Virtual reality exposure therapy for Li, S. H., Newby, J., & Graham, B. M. (2020). Day at the museum. A benchmarking and
anxiety and related disorders: A meta-analysis of randomized controlled trials. feasibility study for large group, one-session exposure treatment for spider phobia.
Journal of Anxiety Disorders, 61, 27–36. Australian Psychologist, 55(2), 121–131.
Castagna, P. J., Davis, T. E., & Lilly, M. E. (2017). The behavioral avoidance task with Love, A. S., & Love, R. J. (2021). Considering needle phobia among adult patients during
anxious youth: A review of procedures, Properties, and criticisms. Clinical Child and mass COVID-19 vaccinations. Journal of Primary Care and Community Health, 12,
Family Psychology Review, 20, 162–184. Article 21501327211007393. https://doi.org/10.1177/21501327211007393
Choy, Y., Fyer, A. J., & Lipsitz, J. D. (2007). Treatment of specific phobia in adults. Michaliszyn, D., Marchand, A., Bouchard, S., et al. (2010). A randomized, controlled
Clinical Psychology Review, 27(3), 266–286. clinical trial of in virtuo and in vivo exposure for spider phobia. Cyberpsychology,
Craske, M. G., Treanor, M., Conway, C., Zbozinek, T., & Vervliet, B. (2014). Maximising Behavior, and Social Networking, 13(6), 689–695.
exposure therapy: an inhibitory learning approach. Behaviour Research and Therapy, Miloff, A., Lindner, P., Dafgard, P., et al. (2019). Automated virtual reality exposure
58, 10–23. https://doi.org/10.1016/j.brat.2014.04.006 therapy for spider phobia vs. in-vivo one-session treatment: A randomized non-
Davis, T. E., & Ollendick, T. H. (2005). Empirically supported treatments for specific inferiority trial. Behaviour Research and Therapy, 118, 130–14.
phobia in children: Do efficacious treatments address the components of a phobic Moher, D., Liberati, A., Tetzlaff, J., et al. (2009). Preferred reporting items for systematic
response? Clinical Psychology: Science and Practice, 12, 144–160. reviews and meta-analyses: The PRISMA statement. BMJ. https://doi.org/10.1136/
Davis, T. E., Ollendick, T. H., & Öst, L. G. (2012). Intensive one-session treatment of specific bmj.b2535. British Medical Journal Publishing Group.
phobias. New York: Springer. Muris, P., De Jong, P. J., Merckelbach, H., et al. (1993a). Is exposure therapy outcome
Davis, T. E., Ollendick, T. H., & Öst, L.-G. (2019). One-Session Treatment of specific affected by a monitoring coping style? Advances in Behaviour Research and Therapy,
phobias in children: Recent developments and a systematic review. Annual Review of 15(4), 291–300.
Clinical Psychology, 15(1), 233–256. Muris, P., De Jong, P. J., Merckelbach, H., et al. (1993b). Monitoring coping style and
De Jong, P. J., Vorage, I., & Van Den Hout, M. A. (2000). Counterconditioning in the exposure outcome in spider phobics. Behavioural and Cognitive Psychotherapy, 21(4),
treatment of spider phobia: Effects on disgust, fear and valence. Behaviour Research 329–333.
and Therapy, 38(11), 1055–1069. Muris, P., & Merckelbach, H. (1997a). Treating spider phobics with eye movement
Deacon, B., & Abramowitz, J. (2006). A pilot study of two-day cognitive-behavioral desensitization and reprocessing: A controlled study. Behavioural and Cognitive
therapy for panic disorder. Behaviour Research and Therapy, 44, 807–817. Psychotherapy, 25(1), 39–50.
Deeks, J., Dinnes, J., D’Amico, R., et al. (2003). Evaluating non-randomised intervention Muris, P., Merckelbach, H., Holdrinet, I., et al. (1998). Treating phobic children: Effects
studies. Health Technology Assessment, 7(27). iii–173. Available at: https://doaj.org/a of EMDR versus exposure. Journal of Consulting and Clinical Psychology, 66(1),
rticle/867344accc294af0bc6e5178eacfd611. 193–198.
Dibbets, P., Moor, C., & Voncken, M. J. (2013). The effect of a retrieval cue on the return Muris, P., Merckelbach, H., van Haaften, H., et al. (1997b). Eye movement
of spider fear. Journal of Behavior Therapy and Experimental Psychiatry, 44(4), desensitisation and reprocessing versus exposure in vivo. A single-session crossover
361–367. study of spider-phobic children. The British Journal of Psychiatry, 171, 82–86.
Duval, S., & Tweedie, R. (2000). Trim and fill: A simple funnel-plot-based method of Muskett, A., Radtke, S. R., & Ollendick, T. (2020). A pilot study of one-session treatment
testing and adjusting for publication bias in meta-analysis. Biometrics, 56(June), for specific phobias in children with ASD traits. Journal of Child and Family Studies,
455–463. 29(4), 1021–1028.
Eaton, W. W., Bienvenu, O. J., & Miloyan, B. (2018). Specific phobias. The Lancet Nielsen, M. D., Andreasen, C. L., & Thastum, M. (2016). A Danish study of one-session
Psychiatry, 5(8), 678–686. treatment for specific phobias in children and adolescents. Scandinavian Journal of
Egger, M., Smith, G. D., Schneider, M., et al. (1997). Bias in meta-analysis detected by a Child and Adolescent Psychiatry and Psychology, 4(2), 65–76.
simple, graphical test. British Medical Journal, 315(7109), 629–634. Oar, E. L., Farrell, L. J., Waters, A. M., et al. (2015). One session treatment for pediatric
Emmelkamp, P. M. G., Krijn, M., Hulsbosch, A. M., et al. (2002). Virtual reality treatment blood-injection-injury phobia: A controlled multiple baseline trial. Behaviour
versus exposure in vivo: A comparative evaluation in acrophobia. Behaviour Research Research and Therapy, 73, 131–142.
and Therapy, 40(5), 509–516. Ollendick, T. H., & Davis, T. E. (2013). One-session treatment for specific phobias: A
Farrell, L. J., Kemp, J. J., Blakey, S. M., Meyer, J. M., & Deacon, B. J. (2016). Targeting review of Öst’s single-session exposure with children and adolescents. Cognitive
clinician concerns about exposure therapy: A pilot study comparing standard vs. Behaviour Therapy, 42(4), 275–283.
enhanced training. Behaviour Research and Therapy, 85, 53–59. Ollendick, T. H., Halldorsdottir, T., Fraire, M. G., et al. (2015). Specific phobias in youth:
Farrell, L. J., Kershaw, H., & Ollendick, T. (2018). Play-modified one-session treatment A randomized controlled trial comparing one-session treatment to a parent-
for young children with a specific phobia of dogs: A multiple baseline case series. augmented one-session treatment. Behavior Therapy, 46(2), 141–155.
Child Psychiatry and Human Development, 49(2), 317–329. Ollendick, T. H., & King, N. J. (1998). Empirically supported treatments for children with
Gilroy, L., Kirkby, K. C., Daniels, B. A., et al. (2000). Controlled comparison of computer- phobic and anxiety disorders: Current status. Journal of Clinical Child Psychology, 27
aided vicarious exposure versus live exposure in the treatment of spider phobia. (2), 156–167.
Behavior Therapy, 31(4), 733–744. Ollendick, T. H., Öst, L.-G., Reuterskiöld, L., et al. (2009). One-session treatment of
Goossens, L., Sunaert, S., Peeters, R., et al. (2007). Amygdala hyperfunction in phobic specific phobias in youth: A randomized clinical trial in the United States and
fear normalizes after exposure. Biological Psychiatry, 62(10), 1119–1125. Sweden. Journal of Consulting and Clinical Psychology, 77(3), 504–516.
14
Ollendick, T. H., Ryan, S. M., Capriola-Hall, N. N., et al. (2018). Have phobias, will therapy for posttraumatic stress disorder in Iraq and Afghanistan war veterans.
travel: Addressing one barrier to the delivery of an evidence-based treatment. American Journal of Psychiatry, 171(6), 640–648.
Behavior Therapy, 49(4), 594–603. Ryan, S. M., Strege, M. V., Oar, E. L., & Ollendick, T. H. (2017). One session treatment for
Öst, L.-G. (1989). One-session treatment for specific phobias. Behaviour Research and specific phobias in children: Comorbid anxiety disorders and treatment outcome.
Therapy, 27(1), 1–7. Journal of Behavior Therapy and Experimental Psychiatry, 54, 128–134.
Öst, L.-G. (1996). One-session group treatment of spider phobia. Behaviour Research and Schienle, A., Schafer, A., Hermann, A., et al. (2007). Symptom provocation and reduction
Therapy, 34(9), 707–715. in patients suffering from spider phobia: An fMRI study on exposure therapy.
Ost, L.-G. (2012). One-session treatment: Principles and procedures with European Archives of Psychiatry and Clinical Neuroscience, 257(8), 486–493.
adultsT. E. Davis, T. H. Ollendick, & L.-G. Öst (Eds.). Intensive One-Session Treatment Schmid-Leuz, B., Elsesser, K., Lohrmann, T., et al. (2007). Attention focusing versus
of Specific Phobias. Autism and Child Psychopathology Series, 59–95. New York, NY: distraction during exposure in dental phobia. Behaviour Research and Therapy, 45
Springer New York. (11), 2691–2703.
Ost, L.-G. (2017). Det empiriska stodet for KBT vid psykiska storningar hos barn och Shapiro, F. (1989). Eye movement desensitization: A new treatment for post-traumatic
ungdomarL.-G. Öst (Ed.). KBT inom barn- och ungdomspsykiatrin. Stockholm: Natur & stress disorder. Journal of Behavior Therapy and Experimental Psychiatry, 20(3),
Kultur. 211–217.
Öst, L.-G., Alm, T., Brandberg, M., et al. (2001). One vs five sessions of exposure and five St-Jacques, J., Bouchard, S., & Belanger, C. (2010). Is virtual reality effective to motivate
sessions of cognitive therapy in the treatment of claustrophobia. Behaviour Research and raise interest in phobic children toward therapy? A clinical trial study of in vivo
and Therapy, 39(2), 167–183. with in virtuo versus in vivo only treatment exposure. Journal of Clinical Psychiatry,
Öst, L.-G., Brandberg, M., & Alm, T. (1997). One versus five sessions of exposure in the 71(7), 924–931.
treatment of flying phobia. Behaviour Research and Therapy, 35(11), 987–996. Steinman, S. A., & Teachman, B. A. (2014). Reaching new heights: Comparing
Öst, L.-G., Fellenius, J., & Sterner, U. (1991). Applied tension, exposure in vivo, and interpretation bias modification to exposure therapy for extreme height fear. Journal
tension-only in the treatment of blood phobia. Behaviour Research and Therapy, 29 of Consulting and Clinical Psychology, 82(3), 404–417.
(6), 561–574. Steketee, G., Bransfield, S., Miller, S. M., et al. (1989). The effect of information and
Öst, L.-G., Ferebee, I., & Furmark, T. (1997). One-session group therapy of spider phobia: coping style on the reduction of phobic anxiety during exposure. Journal of Anxiety
Direct versus indirect treatments. Behaviour Research and Therapy, 35(8), 721–732. Disorders, 3(2), 69–85.
Öst, L.-G., Hellstrom, K., & Kaver, A. (1992). One versus five sessions of exposure in the Sterne, J. A. C., Sutton, A. J., Ioannidis, J. P. A., et al. (2011). Recommendations for
treatment of injection phobia. Behavior Therapy, 23(2), 263–282. examining and interpreting funnel plot asymmetry in meta-analyses of randomised
Öst, L.-G., Johansson, J., & Jerremalm, A. (1982). Individual response patterns and the controlled trials. BMJ, 343, d4002.
effects of different behavioral methods in the treatment of claustrophobia. Behaviour Straube, T., Glauer, M., Dilger, S., et al. (2006). Effects of cognitive-behavioral therapy
Research and Therapy, 20(5), 445–460. on brain activation in specific phobia. NeuroImage, 29(1), 125–135.
Öst, L.-G., Lindahl, I.-L., Sterner, U., et al. (1984). Exposure in vivo vs applied relaxation Svensson, L., Larsson, A., & Öst, L.-G. (2002). How children experience brief-exposure
in the treatment of blood phobia. Behaviour Research and Therapy, 22(3), 205–216. treatment of specific phobias. Journal of Clinical Child and Adolescent Psychology, 31
Öst, L.-G., & Ollendick, T. H. (2001). Manual for the one-session treatment of specific (1), 80–89.
phobias in children and adolescents. Blacksburg: Virginia Polytechnic University. Teachman, B. A., & Woody, S. R. (2003). Automatic processing in spider phobia: Implicit
Unpublished manuscript available from authors. fear associations over the course of treatment. Journal of Abnormal Psychology, 112
Öst, L.-G., & Ollendick, T. H. (2017). Brief, intensive and concentrated cognitive (1), 100–109.
behavioral treatments for anxiety disorders in children: A systematic review and Thomas, B. H., Ciliska, D., Dobbins, M., et al. (2004). A process for systematically
meta-analysis. Behaviour Research and Therapy, 97, 134–145. reviewing the literature: Providing the research evidence for public health nursing
Ost, L.-G., Salkovskis, P. M., & Hellstrom, K. (1991b). One-session therapist-directed interventions. Worldviews on Evidence-Based Nursing, 1(3), 176–184.
exposure vs. self-exposure in the treatment of spider phobia. Behavior Therapy, 22(3), Vansteenwegen, D., Vervliet, B., Hermans, D., et al. (2007). Verbal, behavioural and
407–422. physiological assessment of the generalization of exposure-based fear reduction in a
Papalini, S., Lange, I., Bakker, J., et al. (2019). The predictive value of neural reward spider-anxious population. Behaviour Research and Therapy, 45(2), 291–300.
processing on exposure therapy outcome: Results from a randomized controlled trial. Vika, M., Skaret, E., Raadal, M., et al. (2009). One- vs. five-session treatment of intra-oral
Progress in neuro-psychopharmacology & biological psychiatry, 92, 339–346. injection phobia: A randomized clinical study. European Journal of Oral Sciences, 117
Powers, M. B., Smits, J. A. J., & Telch, M. J. (2004). Disentangling the effects of safety- (3), 279–285.
behavior utilization and safety-behavior availability during exposure-based Walkup, J. T., Albano, A. M., Piacentini, J., et al. (2008). Cognitive behavioral therapy,
treatment: A placebo-controlled trial. Journal of Consulting and Clinical Psychology, 72 sertraline, or a combination in childhood anxiety. New England Journal of Medicine,
(3), 448–454. 359(26), 2753–2766.
Preusser, F., Margraf, J., & Zlomuzica, A. (2017). Generalization of extinguished fear to Wardenaar, K. J., Lim, C. C. W., Al-Hamzawi, A. O., et al. (2017). The cross-national
untreated fear stimuli after exposure. Neuropsychopharmacology, 42(13), 2545–2552. epidemiology of specific phobia in the World Mental Health Surveys. Psychological
Raeder, F., Heidemann, F., Schedlowski, M., et al. (2019). No pills, more skills: The Medicine, 47(10), 1744–1760.
adverse effect of hormonal contraceptive use on exposure therapy benefit. Journal of Waters, A. M., Farrell, L. J., Zimmer-Gembeck, M. J., et al. (2014). Augmenting one-
Psychiatric Research, 119, 95–101. session treatment of children’s specific phobias with attention training to positive
Raes, A. K., Koster, E. H. W., Loeys, T., et al. (2011). Pathways to change in one-session stimuli. Behaviour Research and Therapy, 62, 107–119.
exposure with and without cognitive intervention: An exploratory study in spider Williams, S. L., Dooseman, G., & Kleifield, E. (1984). Comparative effectiveness of guided
phobia. Journal of Anxiety Disorders, 25(7), 964–971. mastery and exposure treatments for intractable phobias. Journal of Consulting and
Rentz, T. O., Powers, M. B., Smits, J. A. J., et al. (2003). Active-imaginal exposure: Clinical Psychology, 52(4), 505–518.
Examination of a new behavioral treatment for cynophobia (dog phobia). Behaviour Wolitzky-Taylor, K. B., Horowitz, J. D., Powers, M. B., et al. (2008). Psychological
Research and Therapy, 41(11), 1337–1353. approaches in the treatment of specific phobias: A meta-analysis. Clinical Psychology
Rihm, J. S., Sollberger, S. B., Soravia, L. M., et al. (2016). Re-presentation of olfactory Review, 28(6), 1021–1037.
exposure therapy success cues during non-rapid eye movement sleep did not increase Wolpe, J. (1954). Reciprocal inhibition as the main basis of psychotherapeutic effects.
therapy outcome but increased sleep spindles. Frontiers in Human Neuroscience, 10, Archives of Neurology and Psychiatry, 72(2), 205–226.
340. Wright, B., Tindall, L., Scott, A. J., et al. (2022). One session treatment (OST) is
Rothbaum, B. O., Anderson, P., Zimand, E., et al. (2006). Virtual reality exposure therapy equivalent to multi- session cognitive behavioral therapy (CBT) in children with
and standard (in vivo) exposure therapy in the treatment of fear of flying. Behavior specific phobias (ASPECT): Results from a national non-inferiority randomized
Therapy, 37(1), 80–90. controlled trial. Journal of Child Psychology and Psychiatry. https://doi.org/10.1111/
Rothbaum, B. O., Hodges, L., Smith, S., et al. (2000). A controlled study of virtual reality jcpp.13665 (in press).
exposure therapy for the fear of flying. Journal of Consulting and Clinical Psychology, Wrzesien, M., Burkhardt, J.-M., Botella, C., et al. (2012). Evaluation of the quality of
68(6), 1020–1026. collaboration between the client and the therapist in phobia treatments. Interacting
Rothbaum, B. O., Price, M., Jovanovic, T., et al. (2014). A randomized, double-blind with Computers, 24(6), 461–471.
evaluation of D-cycloserine or alprazolam combined with virtual reality exposure Zlomke, K., & Davis, T. E. (2008). One-session treatment of specific Pphobias: A detailed
description and review of treatment efficacy. Behavior Therapy, 39(3), 207–223.
15

The Relative Efficacy and Efficiency of Single - and Multi-Session Exposure

Uploaded by

Copyright:

Available Formats

The Relative Efficacy and Efficiency of Single - and Multi-Session Exposure

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Relative Efficacy and Efficiency of Single - and Multi-Session Exposure

Uploaded by

Copyright:

Available Formats

Behaviour Research and Therapy 159 (2022) 104203

Contents lists available at ScienceDirect

Behaviour Research and Therapy

The relative efficacy and efficiency of single- and multi-session exposure

1. Introduction approach, or in a single intense treatment session (Choy et al., 2007). A

Note. k = number of study arms, n = number of participants.

Note. n = number of participants; RA = Researcher allegiance; FU = Follow-up.

Note. n = number of participants; FU = Follow-up.

Fig. 2. Quality assessment results for single- and multi-session studies.

Mean ES (g) 95% CI z-score p Q p I2

Mean ES (g) 95% CI z-score p Q p I2

Mean ES (g) 95% CI z-score p Q p I2

despite the significantly shorter therapy time? The intervention studies

Mean ES (g) 95% CI z-score p Q p I2 (%)

You might also like