ACT Techsupported Chronicconditions Meta BRAT 2022

Behaviour Research and Therapy 148 (2022) 103995
Contents lists available at ScienceDirect
Behaviour Research and Therapy

journal homepage: www.elsevier.com/locate/brat
Technology-supported Acceptance and Commitment Therapy for chronic

health conditions: A systematic review and meta-analysis
Matthew S. Herbert a, b, c, *, Cara Dochat a, d, Jennalee S. Wooldridge a, b, c, Karla Materna a,
Brian H. Blanco a, Mara Tynan a, d, Michael W. Lee a, Marianna Gasperi a, b, c, Angela Camodeca a,
Devon Harris a, Niloofar Afari a, b, c
a
VA San Diego Healthcare System, San Diego, CA, USA
b
Department of Psychiatry, University of California, San Diego, San Diego, CA, USA
c
VA Center of Excellence for Stress and Mental Health, San Diego, CA, USA
d
San Diego State University/University of California, San Diego Joint Doctoral Program in Clinical Psychology, San Diego, CA, USA
A R T I C L E I N F O A B S T R A C T
Keywords: Chronic health conditions (CHCs) are common and associated with functional limitations. Acceptance and
Chronic disease commitment therapy (ACT) shows promise in improving functioning, quality of life, and distress across several
Chronic illness CHCs. The purpose of this study was to conduct a systematic review of technology-supported ACT for CHCs and
Disability
perform a meta-analysis on functioning and ACT process outcomes. Multiple databases were systematically
Functioning
Acceptance
searched for randomized controlled trials. A total of 20 unique studies with 2,430 randomized participants were
included. CHCs addressed in these studies were chronic pain (k = 9), obesity/overweight (k = 4), cancer (k = 3),
hearing loss (k = 1), HIV (k = 1), multiple sclerosis (k = 1), and tinnitus (k = 1). Internet and telephone were the
most used technology platforms. All studies included therapist contact with considerable heterogeneity between
studies. Random effects meta-analyses found medium effect sizes showing technology-supported ACT out
performed comparator groups on measures of function at post-treatment (Hedges’ g = − 0.49; p = 0.002) and
follow-up (Hedges’ g = − 0.52; p = 0.02), as well as ACT process outcomes at post-treatment (Hedges’ g = 0.48; p
< 0.001) and follow-up (Hedges’ g = 0.44; p < 0.001). Technology-supported ACT shows promise for improving
function and ACT process outcomes across a range of CHCs. Recommendations are provided to optimize
technology-supported ACT for CHCs. PROSPERO registration number: CRD42020200230.
1. Introduction high rates of depression, anxiety, and poor quality of life (Clarke &
Currie, 2009; Megari, 2013).
Chronic health conditions (CHC), also referred to as chronic diseases Acceptance and commitment therapy (ACT) is a cognitive-
or chronic illnesses, are medical conditions lasting one year or longer behavioral approach that has shown promise for improving func
that result in functional limitations and/or require ongoing care (U.S. tioning, quality of life, and distress across various CHCs, including HIV,
Department of Health & Human Services, 2019). Approximately 60% of cancer, epilepsy, and chronic pain (Gloster et al., 2020; Graham et al.,
U.S. adults have at least one CHC and 42% have multiple CHCs (Buttorff 2016). ACT is based on the psychological flexibility model, a unified
et al., 2017). CHCs present significant challenges at the societal and model of behavior change that focuses on six core treatment processes:
individual level. CHCs are associated with reduced workplace produc present moment awareness, acceptance, defusion, self-as-context,
tivity and absenteeism (Collins et al., 2005) and significant healthcare values, and committed action (Hayes et al., 2006). Briefly, present
burden (Dieleman et al., 2016). Individuals living with CHCs often moment awareness refers to the ongoing non-judgmental contact with
experience difficulties adjusting to functional restrictions and making here-and-now experiences. Acceptance entails willingness to feel un
necessary behavioral modifications (e.g., health monitoring, diet, exer pleasant private experiences (e.g., thoughts, emotions, body sensations)
cise, medication adherence) to best manage the condition(s) (Eton et al., without making counterproductive attempts to change or avoid these
2013). Further, CHCs affect overall well-being and are associated with experiences, particularly in the context of pursuing goals. Defusion is a
* Corresponding author. 3350 La Jolla Village Drive, San Diego, CA 92161, USA.
E-mail address: [email protected] (M.S. Herbert).
https://doi.org/10.1016/j.brat.2021.103995
Received 28 June 2021; Received in revised form 1 November 2021; Accepted 6 November 2021
Available online 12 November 2021
0005-7967/© 2021 Elsevier Ltd. All rights reserved.
M.S. Herbert et al. Behaviour Research and Therapy 148 (2022) 103995
process of de-literalizing thoughts, or seeing thoughts as thoughts rather 2. Materials and methods
than absolute truth, and decreasing the influence thoughts have over
actions. Self-as-context represents a transcendent aspect of the self that This systematic review with meta-analysis was registered in PROS
is separate from the content (e.g., thoughts) of the self. Values refer to PERO (ID: CRD42020200230) and conducted in accordance with the
chosen life directions that are defined as important and meaningful at Preferred Reporting Items for Systematic Reviews and Meta-Analyses
the individual level, while committed action is the active engagement of (PRISMA) guidelines (Liberati et al., 2009). A protocol paper for the
values-consistent behavior. These processes are targeted in ACT treat review was not prepared.
ment with the overall goal of increasing psychological flexibility, or the
ability to persist in or stop behavior in the service of values regardless of 2.1. Eligibility criteria
unwanted private experiences (Hayes et al., 2006).
ACT interventions are particularly suitable for CHCs. A core tenant of We used the five PICOS components (participants, interventions,
the ACT model is that maladaptive behavior typically occurs as a result comparators, outcomes, and study design) to design our research ques
of experiential avoidance, or the attempt to eliminate or control the tion and eligibility criteria (Moher et al., 2010). Studies were required to
form, frequency, or sensitivity of unwanted private experiences, even meet the following inclusion criteria: (P) sample was adults, 18 years
when doing so causes harm (Hayes et al., 2006). Experiential avoidance and older, with a CHC; (I) delivered an ACT-based intervention partially
can manifest in different ways among individuals with CHCs. For or completely using technology; (C) included a comparison condition
example, people may over- or under-utilize the healthcare system to including other active treatments, treatment as usual, and waitlist con
reduce worry about their condition, engage in substance use or trol; (O) included a quantitative measure of functioning consistent with
over-eating to manage comorbid depression or anxiety, or fail to initiate the World Health Organization’s International Classification of Func
and/or maintain disease self-management behaviors such as injections tioning, Disability, and Health (Jette, 2006) and/or a measure of ACT
or medications due to physical discomfort. ACT addresses experiential processes; and (S) used a randomized controlled design. Studies were
avoidance by situating the client in here-and-now experiences, noticing excluded if: (P) the condition or population was primarily mental
and distancing from unwanted thoughts that fuel experiential avoidance health-related, including insomnia; (I) intervention delivery did not
(e.g., catastrophizing about the condition, ruminating about the past), include technology, technology was used for the purposes of data
offering experiential acceptance as an alternative to avoidance, and collection only, or the intervention was not primarily ACT-based; (O)
promoting values-aligned goal setting to motivate behavior change. In outcomes of interest were not reported or only qualitatively assessed; or
this way, ACT shifts the focus of treatment from getting rid of unpleasant (S) study design was cross-sectional, case study, case series, or used
private experiences to living a rich, meaningful life regardless of the non-random assignment to treatment group.
presence of such experiences. For this reason, the primary outcome of
interest in ACT interventions is improved functioning (Feliu-Soler et al., 2.2. Information sources and search strategy
2018), rather than decreased mental health symptoms or
condition-specific clinical outcomes (e.g., HbA1c levels or pain The online databases of PubMed, PsycINFO, and Web of Science
severity). were systematically searched in February 2021. No search limitations or
Similar to other cognitive-behavioral approaches, ACT is tradition filters were imposed. Only peer-reviewed manuscripts published in
ally delivered in-person, either individually or in groups, across several English were included. No lower limit to year of publication was
consecutive weekly sessions. This creates significant barriers to treat imposed. Due to the large number of possible CHCs, we did not specify
ment, as many persons with CHCs already attend numerous healthcare conditions. Of note, overweight and obesity were included because these
appointments, may be immunocompromised, have mobility limitations, conditions are recognized by the American Medical Association and
or live in remote areas, and thus may not have the time, money, or National Institutes of Health as a CHC (Kyle et al., 2016). Below is an
ability to attend weekly clinic appointments (Brundisini et al., 2013). example search in PubMed:
Recently, this problem has been compounded by the COVID-19
pandemic and restrictions on non-emergency in-person visits, high (((((Acceptance [Title/Abstract]) AND (Commitment [Title/Ab
lighting the need for virtual healthcare delivery (Van Daele et al., 2020). stract])) OR (Acceptance-based [Title/Abstract])) OR (contextual
Technology-supported ACT (i.e., ACT delivered partially or completely behav*[Title/Abstract])) OR (contextual cognitive behav*[Title/
with the use of technology, including telephone, internet, or smartphone Abstract])) AND ((((((((((((((((((((((((((internet [Title/Abstract]) OR
components) has the potential for increasing the accessibility of ACT for (internet-*[Title/Abstract])) OR (smartphone [Title/Abstract])) OR
CHC populations, either as a replacement of or supplement for in-person (mhealth [Title/Abstract])) OR (m-health [Title/Abstract])) OR
treatment. Technology-supported ACT for CHCs also is potentially less (ehealth [Title/Abstract])) OR (e-health [Title/Abstract])) OR
resource-intensive and more cost-effective for healthcare systems (electronic [Title/Abstract])) OR (online [Title/Abstract])) OR (app*
(Elbert et al., 2014). [Title/Abstract])) OR (video*[Title/Abstract])) OR (web*[Title/
While several individual studies have supported the utility of Abstract])) OR (web-*[Title/Abstract])) OR (*phone [Title/Ab
technology-supported ACT for CHCs and there is some support for stract])) OR (phone-*[Title/Abstract])) OR (mobile [Title/Ab
technology-supported ACT for mental health conditions (Brown et al., stract])) OR (mobile-*[Title/Abstract])) OR (blended [Title/
2016; Thompson et al., 2021), there has been no systematic review or Abstract])) OR (tele*[Title/Abstract])) OR (tech*[Title/Abstract]))
meta-analysis to describe existing studies or examine pooled treatment OR (computer [Title/Abstract])) OR (computer-*[Title/Abstract]))
effects. Given promising results of ACT for CHCs (Dochat et al., 2021; OR (e− *[Title/Abstract])) OR (virtual [Title/Abstract])) OR (digital
Graham et al., 2016) and high relevancy of technology-supported [Title/Abstract])) OR (cyber*[Title/Abstract]))
treatments, the purpose of this study was to conduct a systematic re
view and meta-analysis of technology-supported ACT interventions for Additionally, reference lists of included manuscripts were inspected
CHCs with an emphasis on functioning and ACT process outcomes. Our as well as databases on the Association for Contextual and Behavioral
aims were to (1) describe the design and methodology of Science website.
technology-supported ACT interventions for CHCs, including use of
specific technology modality (ies) and degree of therapist involvement; 2.3. Study selection
(2) quantitatively examine efficacy using meta-analysis; and (3) provide
recommendations for future research. Study selection proceeded in three stages and was independently
performed by two study authors (K.M. and B.H.B.). Disagreements were
2
resolved by consensus and consultation with the first author. In stage 1 duplicates” feature and by hand. Study coding and data extraction
(screening), all manuscripts returned from database searches were im occurred in Excel and was accomplished by the study team. M.W.L. and
ported into a reference management software (EndNote ×8). These A.C. extracted sample characteristics, K.M. and B.M.B. extracted inter
manuscripts received title/abstract review. Studies that clearly failed to vention details, and C.D. and M.T. extracted outcomes, including group
meet inclusion criteria or met exclusion criteria were removed. In stage means and standard deviations, statistical significance test results, and
2 (selection), remaining studies received a full-text review to determine effect sizes when reported. M.S.H. oversaw and double-checked all data
inclusion status. Ineligible studies were removed and categorized ac extraction procedures. Results from intention-to-treat (ITT) analyses and
cording to exclusion reason. In stage 3 (hand-searching), the reference unadjusted means and standard deviations were extracted when avail
section of studies selected for inclusion were reviewed to identify able. Study authors were contacted for information as needed. Addi
additional potential manuscripts not previously identified through tional information was obtained from protocol papers published prior to
database searches. The titles, abstracts, and full texts of these manu the included manuscript.
scripts were examined as necessary. Ineligible studies were removed and
categorized according to exclusion reason (see Fig. 1).
2.5. Risk-of-bias assessment
2.4. Data extraction and management Study quality was assessed by J.S.W. and M.G. using version 2 of the
Cochrane risk-of-bias (RoB) tool (RoB-2) for randomized trials (Sterne
EndNote ×8 was used to store results from database and hand et al., 2019). RoB-2 assesses five domains of study design and reporting:
searches, sort manuscripts, and categorize according to exclusion randomization, deviations from intended intervention, missing outcome
criteria. Duplicates were removed using the EndNote ×8 “remove data, measurement of the outcome, and selection of the reported result.
Fig. 1. Flow diagram.
3
Each domain is comprised of questions rated as yes, probably yes, prob conditions that provided an active intervention including ACT or CBT
ably no, no, or no information. Domain-specific algorithms are then used were classified as active intervention comparators and were not used in
to generate a domain-specific RoB judgment (low, high, some concerns). meta-analyses. Comparing technology-supported ACT to other ACT-
Domain-specific RoB judgements are synthesized to generate an overall based comparators would be more appropriate for other study designs,
RoB judgement. Studies are rated low risk if all respective domains are such as non-inferiority (e.g., when comparing to in-person delivery) or
rated as low risk, some concerns if at least one domain is rated some comparative effectiveness (e.g., when comparing two or more
concerns and no domains are rated high risk, and high risk if at least one technology-supported ACT conditions with different treatment compo
domain is rated high or if multiple domains are rated some concerns in a nents). Further, technology-supported ACT would not be expected to
way that substantially lowers confidence in the result. Funnel plots, outperform a CBT-based comparator. This is supported by a recent re
contour-enhanced funnel plots, and Eggers’ test were used to assess view of meta-analyses that found that ACT is generally not superior to
publication bias across studies included in the meta-analysis. CBT (Gloster et al., 2020). For studies with two comparator conditions,
active controls were chosen over inactive controls.
2.6. Meta-analysis Sensitivity analyses were conducted by removing studies with effects
determined to be outliers, as indicated when the 95% confidence in
Quantitative analysis of treatment efficacy was conducted using terval of an individual study did not overlap with the 95% confidence
between-group random effects meta-analysis. Random effects models interval of the pooled treatment effect, as well as studies deemed to be
are better suited for meta-analysis in the context of between study het high RoB. We also conducted exploratory post hoc subgroup meta-
erogeneity (Field & Gillett, 2010), which was anticipated in the present analyses to explore technology modality (internet versus other), thera
study. Analyses were conducted in R version 3.6.1 using the metafor and pist involvement (in-person, telephone, or video versus asynchronous),
dmetar packages (Harrer et al., 2019), using the inverse variance type of condition (chronic pain versus other), and comparator type
method and Hedges’ g as the standardized mean difference with (active versus inactive). Due to the number of studies and availability of
Knapp-Hartung adjustments to calculate the pooled effect size confi data at follow-up, these exploratory analyses focused on functioning and
dence interval (Knapp & Hartung, 2003). Study effects were weighted by ACT process outcomes at post-treatment only. Further, due to the
size. This was not imposed, but rather was the natural weighting based exploratory nature of subgroup analyses, an alpha value of 0.10 was
on study size, reflecting the relative contribution of each to the pooled used to warrant inspection of individual subgroups.
effect. Hedges’ g values of 0.2 were considered a small effect, 0.5 a
medium effect, and 0.8 a large effect. Statistical heterogeneity was 3. Results
assessed using I2, Cochran’s Q-statistic, and τ2 (using DerSimonian-Laird
estimator) (Higgins et al., 2003). I2 is the percentage of variability in 3.1. Literature search
effect sizes due to heterogeneity rather than sampling error (as specified
in the Cochrane handbook: 0–40% might not be important; 30–60% may Fig. 1 shows the number of manuscripts identified throughout the
represent moderate heterogeneity; 50–90% may represent substantial screening, hand-searching, and selection phases. After removal of du
heterogeneity; 75–100%: considerable heterogeneity). The Q-statistic is plicates, titles and abstracts of 1,510 articles were examined. Of these,
the weighted sum of squared differences between individual study ef 66 underwent full-text review. A total of 45 were deemed ineligible for
fects and the pooled effect, from which I2 is derived. The Q-statistic the following reasons: population did not have a CHC (k = 1), not an
chi-squared significance test is known to be low-powered for analyses ACT-based intervention (k = 4), intervention was not technology sup
with few studies and should be interpreted with caution (Higgins et al., ported (k = 7), no comparison condition (k = 14), conditions were not
2003). τ2 is another metric of between-study variance in effect sizes randomized (k = 1), or article did not include relevant outcomes or was
(Deeks et al., 2019). A prediction interval, which accounts for a secondary analysis that did not include unique outcomes of interest (k
between-study variance and is less sensitive to number of studies than = 18). Two manuscripts deemed eligible were from the same study
standard heterogeneity estimates, was also calculated (Harrer et al., (Kristjánsdóttir et al., 2013a, 2013b). Both were included in the quali
2019). Prediction intervals provide a range in which future study effects tative synthesis as the latter described results from the final follow-up
are predicted to fall based on present evidence in the meta-analysis. period. This resulted in a total of 21 manuscripts included in the qual
Meta-analyses were conducted on functioning and ACT process itative synthesis representing 20 unique studies (Buhrman et al., 2013;
outcomes at post-treatment and the first follow-up period. While ACT Hawkes et al., 2014; Herbert et al., 2017; Hesser et al., 2012; Ishola &
processes refer to proposed mechanisms underlying treatment response, Chipps, 2015; Kristjánsdóttir et al., 2013a, 2013b; Levin et al., 2020; Lin
they are also frequently assessed as outcome measures in ACT studies. et al., 2017; Molander et al., 2018; MMosher et al., 2018, 2019; Potts
When studies included more than one measure of functioning, general et al., 2020; Proctor et al., 2018; Rickardsson et al., 2021; Sairanen et al.,
measures were chosen over condition-specific measures to reduce het 2017; Scott et al., 2018; Simister et al., 2018; Thorsell et al., 2011;
erogeneity. When studies included more than one measure of an ACT Trompetter et al., 2015; Weineland et al., 2012).
process outcome, we chose measures that were most reflective of the
psychological flexibility model to reduce heterogeneity. For example, 3.2. Risk-of-bias assessment
the Acceptance and Action Questionnaire (AAQ-II), which is a general
measure of psychological inflexibility, would be chosen over the Chronic Fig. 2 shows the RoB for the included studies as judged by study
Pain Acceptance Questionnaire (CPAQ), which consists of subscales authors (J.S.W., M.G.). RoB was assessed with information garnered
capturing acceptance of pain and engagement in activities despite pain. from study manuscripts, pre-registration websites (i.e., clinicaltrials.gov
However, the CPAQ would be chosen over questionnaires capturing or similar; 13 out of 20), study protocol papers (Hawkes et al., 2009;
individual components of the model (e.g., Cognitive Fusion Question Lappalainen et al., 2014; Lin et al., 2015; Molander et al., 2015) and
naire). To maintain consistency in scoring direction, measures were email correspondence with study authors (9 out of 20). Because of
reverse coded as needed. For function measures, lower scores reflect inconsistent reporting across included studies and inadequate informa
greater functional improvement/less functional impairment. For ACT tion, it was often not possible to determine whether RoB criteria were
process measures, higher scores reflect greater psychological flexibility. met. Thus, the overall RoB for many of the included studies (11 out of
Comparator conditions were considered inactive control compara 20) indicates ‘some’ RoB concerns. Of the included studies, five were
tors if they provided treatment as usual, waitlist, or no intervention, or judged to have a ‘low’ RoB, and the remaining four were judged to have
active control comparators if they provided materials and activities that a ‘high’ RoB. Across studies, potential bias was identified most
controlled for time and attention given to participants. Comparator commonly in the domains of ‘bias due to missing outcome data,’ and
4
Fig. 2. Risk of bias domains.
‘bias in selection of the reported result.’ RoB was generally deemed low 17.8%–100%). Thirteen studies did not report race/ethnicity. Across
for ‘bias arising from the randomization process,’ and ‘bias due to de studies that reported race/ethnicity, 16.3% of participants were iden
viations from intended interventions.’ All studies were deemed low for tified as non-white (range: 0%–53%). Chronic pain was the most
‘bias in measurement of the outcome’ according to the RoB-2 criteria. frequently targeted CHC (k = 9), followed by overweight/obesity (k =
Examination of funnel plots, contour-enhanced funnel plots, and Eggers’ 4), cancer (k = 3), hearing loss (k = 1), HIV (k = 1), multiple sclerosis (k
test did not indicate any strong evidence of publication bias. = 1), and tinnitus (k = 1).
Studies were conducted in various North American and European
countries, as well as one study in Nigeria. Regarding study design, 14 out
3.3. Study characteristics
of 20 were RCTs and six out of 20 were pilot RCTs. All studies included a
post-treatment assessment and 15 out of 20 included at least one follow-
Table 1 shows characteristics of included studies. There was a total of
up assessment, ranging from six weeks to one year. The majority of
2,430 participants across 20 unique studies with a range of 27–410 per
studies reported ITT results, with the exception of Ishola and Chipps
study. The mean age was 51.8 years (SD = 14.3; range: 31.6–66.4 years)
(2015), Kristjánsdóttir et al. (2013a, 2013b), and Levin et al. (2020),
based on 19 studies that reported mean and standard deviation for their
which reported completer results only. Attrition regarding completion of
sample. Female gender distribution across studies was 69.7% (range:
5
Table 1
Characteristics of included studies (K = 20).
Study (country) Participant Assessment schedule Technology-supported ACT n Attrition⸸ Functioning ACT Process
characteristics conditions and comparators (post-tx, F/U) Outcome Outcome
Chronic Pain
Buhrman et al, 2013 N = 76 Baseline, post-tx, 6-month F/U A. ACT guided internet A. A. 24%, 24% MPI-I CPAQ*
(Sweden) Mage = 49.1 B. Moderated online discussion 38 B. 16%, NA
Female: 59.2% B.
Non-white = NR 38
Herbert et al, 2017 N = 129 Baseline, mid-tx, post-tx, 3- A. ACT video-teleconferencing A. A. 47%, 55% BPI-I CPAQ
(USA) Mage = 52.0 month FU, 6-month F/U B. ACT in-person 65 B. 23%, 28%
Female: 17.8% B.
Non-white = 53% 64
Kristjansdottir et al, N = 140 Pre-inpatient, Post-inpatient/ A. ACT smartphone app A. A. 33%, 47% FIQ^ CPAQ*^
2013a,b (Norway) Mage = 44.2 Baseline, post-tx, 5-month F/ B. Non-interactive website for 70 B. 47%, 43%
Female: 100% U, 11-month F/U pain self-management w/ACT- B.
Non-white = NR based content 70
Lin et al, 2017 N = 302 Baseline, post-tx, A. ACT guided internet A. A. 29%, 46% MPI-I*^ AAQ-II
(Germany) Mage = 51.7 4-month F/U B. ACT unguided internet 100 B. 33%, 45%
Female: 75.0% C. Waitlist control ± B. C. 11%, 26%
Non-white = NR 101
C.
101
Rickardsson et al, N = 113 Baseline, post-tx, 3-month F/ A. ACT guided internet A. A. 19%, 25% PII* PIPS*
2021 (Sweden) Mage = 49.5 U, 6-month F/U, 12-month F/ B. Waitlist control 57 B. 4%, NA
Female: 84.1% U B.
Non-white = NR 56
Scott et al, 2018 (U. N = 63 Baseline, post-tx, 6-month F/U A. ACT guided internet A. A. 26%, 26% WSAS CPAQ-8
K.) Mage = 42.5 B. Treatment as usual 31 B. 22%, 19%
Female: 63.5% B.
Non-white = 32
19.1%
Simister et al, 2018 N = 67 Baseline, post-tx, 3-month F/U A. ACT guided internet A. A. 18%, 24% FIQ-R*^ CPAQ*^
(Canada) Mage = 39.7 B. Treatment as usual 33 B. 9%, 26%
Female: 95% B.
Non-white = NR 34
Thorsell et al., 2011 N = 115 Baseline, post-tx, 6- month F/ A. ACT self-help w/telephone A. A. 54%, 56% Ö MPQ*^ CPAQ*^
(Sweden) Mage = 46.0 U, 12-month F/U sessions 61 B. 50%, 52%
Female: 64.4% B. Telephone-based applied B.
Non-white = NR relaxation 54
Trompetter et al., N = 238 Baseline, post-tx, 3-month F/U A. ACT guided internet A. A. 28%, 36% MPI-I*^ PIPS*^
2015 (Netherlands) Mage = 52.8 B. Expressive writing ± 82 B. 35%, 37%
Female: 76% C. Waitlist control B. C. 19%, 17%
Non-white = NR 79
C.
77
Overweight and Obesity
Levin et al, 2020 N = 79 Baseline, post-tx, 2-month F/U A. ACT guided internet A. A. 13%, 8% – AAQ-W*
(USA) Mage = 39.6 B. Waitlist control 39 B. 5%, NA
Female: 82.3% B.
Non-white = 40
7.6%
Potts et al, 2020 N = 55 Baseline, post-tx A. ACT self-help w/telephone A. A. 29% – CompACT
(USA) Mage = 39.6 sessions 17 B. 40%
Female: 82.3% B. ACT self-help w/email prompts B. C. 28%
Non-white = C. Waitlist control ± 20
7.6% C.
18
Sairanen et al, 2017 N = 254 Baseline, post-tx, 6-month F/U A. ACT unguided smartphone app A. A. 12%, 14% – AAQ-II
(Finland) Mage = 49.5 B. ACT in-person 85 B. 26%, 29%
Female: 84.5% C. No intervention ± B. C. 20%, 21%
Non-white = 0% 84
C.
85
Weineland et al, 2012 N = 39 Baseline, post-tx A. ACT guided internet A. A. 21% – AAQ-W*
(Sweden) Mage = 43.1 B. Treatment as usual 19 B. 10%
Female: 89.7% B.
Non-white = NR 20
Cancer
Hawkes et al, 2014 N = 410 Baseline, post-tx, 6-month F/U A. ACT telephone A. A. 17%, 22% FACIT-F AAQ-II*
(Australia) Mage = 66.4 B. Treatment-as-usual 205 B. 14%, 20%
Female: 46.1% B.
Non-white = NR 205
Mosher et al., 2018 N = 47 Baseline, post-tx, 1-month F/U A. ACT telephone A. A. 22%, 26% MDASI –
(USA) Mage = 56.2 B. Telephone-based education/ 23 B. 12%, 17%
Female: 100% support B.
Non-white = 11% 24
(continued on next page)
6
Table 1 (continued )
Study (country) Participant Assessment schedule Technology-supported ACT n Attrition⸸ Functioning ACT Process
characteristics conditions and comparators (post-tx, F/U) Outcome Outcome
Mosher et al, 2019 N = 50 Baseline, post-tx, 1-month F/U A. ACT telephone A. A. 20%, 20% MDASI –
(USA) Mage = 62.6 B. Telephone-based education/ 25 B. 28%, 28%
Female: 44% support B.
Non-white = 14% 25
Hearing Loss
Molander et al, 2018 N = 61 Baseline, post-tx A. ACT guided internet A. A. 13% HHIE-S* HAQ*
(Sweden) Mage = 58.7 B. Waitlist control 31 B. 3%
Female: 67.2% B.
Non-white = NR 30
HIV
Ishola and Chipps, N = 66 Baseline, post-tx A. ACT text messages A. A. 0% – AAQ-II*
2015 (Nigeria) Mage = 31.6 B. Post-HIV counseling 33 B. 15%
Female: 100% B.
Non-white = NR 33
Multiple Sclerosis
Proctor et al, 2018 (U. N = 27 Baseline, post-tx A. ACT self-help w/telephone A. A. 7% MSIS-P AAQ-II
K) Mage = 45.9 sessions 14 B. 8%
Female: 88.9% B. Treatment as usual B.
Non-white = NR 13
Tinnitus
Hesser et al, 2012 N = 99 Baseline, post-tx, 1-year F/U A. ACT guided internet A. A. 6%, 6% THI* TAQ*
(Sweden) Mage = 48.5 B. CBT guided internet 35 B. 6%, 14%
Female: 43.4% C. Moderated online discussion ± B. C. 0%, 0%
Non-white = NR 32
C.
32
Note: AAQ-II = Acceptance and Action Questionnaire; AAQ-W = Acceptance and Action Questionnaire for Weight; BPI-I = Brief Pain Inventory – Pain Interference
subscale; CompACT = Comprehensive Assessment of Acceptance and Commitment Therapy Processes; CPAQ = Chronic Pain Acceptance Questionnaire; CPAQ-8 = 8-
item Chronic Pain Acceptance Questionnaire; FACIT-F = Functional Assessment of Chronic Illness Therapy-Functional Wellbeing subscale; FIQ = Fibromyalgia Impact
Questionnaire; FIQ-R = Fibromyalgia Impact Questionnaire Revised; F/U = follow-up; HAQ=Hearing Acceptance Questionnaire; HHIE-S=Hearing Handicap In
ventory for the Elderly-S; MDASI = MD Anderson Symptom Inventory; MPI-I = Multidimensional Pain Inventory – Pain Interference Subscale; MSIS-P = Multiple
Sclerosis Impact Scale-Physical Health subscale; NR = not reported; ÖMPQ = 5 aggregated function items from the Örebro Musculoskeletal Pain Questionnaire;
PII=Pain Interference Index; PIPS= Psychological Inflexibility in Pain Scale; post-tx = post-treatment; TAQ = Tinnitus Acceptance Questionnaire; THI = Tinnitus
Handicap Inventory; WSAS=Work and Social Functioning Scale.
⸸ Attrition is in regard to completed assessments; F/U refers to first follow-up timepoint; * significant improvement favoring technology-supported ACT at post
treatment; ^ significant improvement favoring technology-supported ACT condition at first follow-up timepoint. ± denotes comparison condition. Underlined text
signifies outcomes that were included in meta-analysis.
assessments in technology-supported ACT conditions ranged from 0% to Intervention design and content. A summary of technology-
54% at post-treatment, and 6%–56% at the first follow-up assessment, supported ACT intervention design and content is detailed in Table 2.
with 11 out of 20 and 12 out of 15 studies reporting 20% or greater Active treatment duration ranged from four weeks to six months, over
attrition at post-treatment and follow-up, respectively. which five to 10 modules of intervention content were delivered. Con
A total of 13 out of 20 studies provided information on the amount of tent was delivered through combinations of in-person and/or
intervention completed; however, reported values varied considerably technology-supported sessions with a therapist, written feedback, audio
across studies. For example, 10 out of 20 reported the percentage of files, video files, texts, informational websites, and self-help readings.
participants that completed all treatment sessions/modules, but it was Intervention descriptions consistently included acceptance, present
often unclear if provided values were derived from the number of par moment awareness, defusion, connection with values, and committed
ticipants that completed the intervention or the number of participants action as ACT process outcomes. Most studies (k = 18) explicitly re
randomized. Only four out of 20 studies provided a clear definition of ported the use of experiential exercises and/or metaphors to target ACT
“completers” or “per protocol” and associated percentages of partici processes. “Creative hopelessness” is a common ACT treatment method
pants meeting criteria. Scott et al. (2018), which consisted of two to examine the limitations of control strategies that was explicitly
in-person or telephone sessions and eight internet modules, defined mentioned in five studies.
completers as those who completed seven out of 10 treatment sessions Therapist involvement. Several studies used a combination of
(61%). Trompetter et al. (2015) defined completers as those who methods for therapist involvement. The majority of studies (k = 15) used
completed six out of nine internet modules (72%). Lin et al. (2017) in-person and/or telephone contact, while the remaining five studies
specified per protocol as completing five out of seven internet modules used methods that did not require real-time contact with a therapist (i.e.,
(Guided Internet: 43%; Unguided: 30%). Rickardsson et al. (2021) asynchronous). Specific methods included in-person (k = 6), telephone
defined completers as completing 50% of content (77%). (k = 11), video-teleconferencing (k = 1), asynchronous communication
via internet (e.g., email; k = 8), SMS text messaging (k = 1), and no
therapist involvement (k = 1). Only Lin et al. (2017) directly compared
3.4. Technology-supported ACT
an intervention with and without therapist involvement. Two studies
included group (Sairanen et al., 2017) or dyadic (Mosher et al., 2019)
Type of technology used. A range of technology modalities were
interventions while the majority (k = 18) were delivered on an indi
used, either as stand-alone means or in combination with other tech
vidual basis. Study therapists predominantly had graduate training
nology or in-person formats (see Tables 1 and 2). Regarding the primary
ranging from master’s level psychology students and social workers to
technology modality, internet-based content was most frequently used
clinical psychologists with several years of ACT experience. Two studies
(k = 10), followed by telephone (k = 6), smartphone application (k = 2),
lacked therapist descriptions (Ishola & Chipps, 2015; Weineland et al.,
SMS text-messaging (k = 1) and video-teleconferencing (k = 1).
7
Table 2
Intervention details of technology-supported ACT treatments for chronic health conditions (K = 20).
Study Primary ACT Intervention Delivery Intervention Content Therapist Involvement Therapist Characteristics
Method
Chronic Pain
Buhrman et al, 7 weekly guided internet modules Behavioral approach to chronic pain; Provided weekly feedback on Graduate students in final term
2013 with information, metaphors and creative hopelessness; willingness and homework via online platform; after of five-year clinical psychology
assignments; MP3 files with acceptance; defusion and goal setting; module 3 and at 7 weeks, completed program receiving weekly
mindfulness and experiential self-as-context; values and committed structured telephone calls lasting less supervision from clinical
exercises action; willingness in accordance with than 30 min to motivate participation psychologist
values; maintenance of learned and answer questions
strategies; mindfulness exercises
Herbert et al, 2017 8 weekly video-teleconferencing Limits of control; focus on experience; Delivered weekly individual At least master’s level
sessions with metaphors and values; cognitive defusion; acceptance; manualized ACT sessions, each 60 psychology graduate students
experiential exercises; at-home mindfulness; committed action; min long with weekly group supervision
assignments continued action in support of values
Kristjánsdóttir 1 in-person session; 4 weeks of 3 daily Cognitive defusion; mindfulness; Provided 1 in-person individual Therapists had experience in
et al., 2013a diary entries via smartphone; audio values and values-based action; session; daily, personalized, health care sciences (nursing
Kristjánsdóttir files with mindfulness exercises; acceptance vs avoidance situational feedback via secure and/or psychology) and ACT
et al., 2013a informational website website; final feedback provided in a training; content of feedback
summary was supervised
Lin et al, 2017 A. 7 weekly guided internet modules A. Pain psychoeducation; creative A. “E-Coaches” provided weekly A. Psychologists (eCoaches)
with information, assignments, hopelessness; mindfulness; control and personalized and standardized trained and supervised by an
metaphors and mindfulness exercises, acceptance; primary and secondary feedback by e-mail via a secure web- experienced clinical
video and audio files; automatically suffering; defusion; self-as-context; based platform after module psychologist
generated text-messages (SMS) to thoughts, emotions, and goal setting; completion; approximately 2 h total B⋅N/A
support integration of concepts into values; willingness and committed per participant
daily life; action; summary of program and B. No therapist involvement
B. 7 weekly unguided internet maintenance
modules with information, B. Same content as condition A
assignments, metaphors and
mindfulness exercises, video and
audio files; automatically generated
text-messages (SMS) to support
integration of concepts into daily life
Rickardsson et al, 8 weeks of daily internet content in Acceptance; defusion; present moment Via text messages (SMS) provided 3 licensed psychologists and 2
2021 “microlearning” format with text, awareness; exposure; behavior feedback, support, clarifications, intern psychologists who
audio, illustrations, experiential analysis; pain education; values encouragement, and reminders to worked at a tertiary pain clinic
exercises and value-oriented exposure engage in treatment; responded
within 48 h; at least one weekly
contact with each participant; phone
support upon request; averaged 12.5
min per week per participant
Scott et al, 2018 2 in-person or telephone sessions; 8 Creative hopelessness; values; Provided in-person or telephone Master’s level psychologist
guided internet video sessions (twice openness; cognitive defusion; values- individual sessions; after each session with supervised ACT
weekly for 3 weeks, once weekly for 2 based action; awareness; self-as- provided individualized feedback experience; 3 experienced
weeks) with experiential exercises, context; committed action; reviewed within 24–72 h via email. doctoral level psychologists;
metaphors and questions; completed progress and planned maintenance of weekly meetings and regular
in 10–12 weeks gains supervision
Simister et al, 2018 7 guided internet modules self-paced Creative hopelessness; acceptance; Reviewed homework and provided Master’s level therapist under
over a 2-month period with MP3s, values; psychoeducation; cognitive written feedback via online platform guidance of a registered
videos and pdfs containing defusion; contact with present psychologist with ACT
metaphors, vignettes and experiential moment; self-as-context; willingness expertise
exercises and committed action
Thorsell et al, 2011 2 in-person sessions, 7 weeks using Psychoeducation; avoidance-suffering Provided initial and concluding in- Psychology interns trained in
self-help manual with 6 weekly cycle; values; defusion; mindfulness; person 90-min individual sessions; ACT and supervised
telephone sessions; CD with willingness and acceptance; 30-min weekly telephone sessions;
supplementary exercises (Telephone) committed action; action plan and responded to participant email
obstacles questions as needed
Trompetter et al., 9 guided internet modules with text, Psychoeducation; avoidance of pain; Provided weekly structured Graduated psychology
2015 experiential exercises and metaphors values; committed action; pain personalized response on students trained and
over 9–12 weeks; audio files with acceptance; cognitive defusion; self-as- participants’ progress via enclosed supervised by a registered CBT
mindfulness exercises; personal diary context; pain, social context & and encrypted web-based system therapist experienced with
communication; experiences of ACT
previous ACT-participants
Overweight and Obesity
Levin et al, 2020 8 weekly guided internet modules Nutrition and physical activity Monitored online learning Doctoral student in clinical/
with text, videos and interactive education; strategies to increase management system usage and counseling psychology
exercises physical activity and improve diet; provided 5–10-min weekly coaching
weight stigma; being stuck; defusion; telephone calls to increase adherence
acceptance; mindfulness; values; and provide support in implementing
committed action; recommitting after program
slips
Potts et al, 2020 A. 7 book chapters completed over 8 A. "The Diet Trap” book by Lillis et al., A. Initial and final 30-min coaching A. Advanced clinical/
weeks; weekly telephone sessions; 2014, which covers key ACT skills and call; 6 weekly 5–10 min calls to counseling psychology
weekly online chapter quiz; concepts to reduce harm from weight motivate participation and generalize
(continued on next page)
8
Table 2 (continued )
Study Primary ACT Intervention Delivery Intervention Content Therapist Involvement Therapist Characteristics
Method
journaling self-stigma and increase motivators for skills to daily life; weekly email doctoral student
B. 7 book chapters completed over 8 engaging in meaningful health and reminder of tasks and brief, tailored, B. Same as condition A
weeks; weekly online chapter quiz; quality of life improving behaviors supportive statements; contact time
journaling B. Same content as condition A approximately 120 min over 8 weeks
per participant
B. Weekly email reminder of tasks
and brief tailored, supportive
statements; if needed 2 supportive
email reminders; contact and
preparation time approximately 90
min over 8 weeks per participant
Sairanen et al, 2017 8-week intervention; one in-person Values clarification; acting according Presented an overview of ACT Trained psychologist
meeting; Smartphone app with 41 with values; mindfulness skills; intervention and the smartphone app
exercises with text, audio and video observing self; and acceptance skills. during one in-person group meeting
components Focus was on ACT skills, but minor
parts of mindful eating, relaxation, and
everyday physical activity included
Weineland et al, 2 in-person sessions; 6 weekly internet Creative hopelessness; defusion; self- Provided 2 in-person sessions Not described
2012 modules with texts, mindfulness as-context; acceptance; committed including behavior analysis of
audio files, written exercises and action; contact with the present avoidance and weekly telephone
videos moment; values support of ACT content
Cancer
Hawkes et al, 2014 11 individual telephone sessions over Cancer psychoeducation; values; Provided manualized individual Degrees in nursing,
6-month period (10 bi-weekly mindfulness; defusion; acceptance; health coaching sessions via psychology, or health
sessions for 5 months and final session committed action; motivational telephone promotion and 5 years of
4 weeks later); handbook; postcards; interviewing; problem solving; action experience; 6 weeks of study
pedometer planning; goal setting; review and training; weekly supervision
monitoring health behaviors with investigators
Mosher et al., 2018 6 weekly telephone-based sessions; Mindfulness; suffering and control; Provided 50–60-min telephone Master’s level social worker
handouts; CD perspective taking; cognitive defusion; sessions; assessed participant and with experience in ACT trained
acceptance; transcendent sense of self; strategies for managing symptoms; and supervised by 2
values clarification and committed covered week’s topic; assessed home psychologists
action in face of distress/symptoms; practice and skills; discussed home
responding more effectively to practice
symptoms
Mosher et al., 2019 6 weekly telephone-based sessions Mindfulness; control vs non-control Provided 50–60 min dyadic and Master’s level social worker
(dyadic for sessions 1 and 4–6, strategies; perspective taking; individual telephone sessions with experience in ACT trained
individual for sessions 2 and 3); cognitive defusion; acceptance; values regarding coping, mindfulness, and and supervised by 2
handouts; CD clarification; committed action other skills; discussed home practice psychologists
for the week ahead
Hearing Loss
Molander et al, 8 weekly internet-based modules with Psychoeducation about hearing, Provided weekly feedback on 1 licensed psychologist with
2018 printable text files and audio files; hearing strategies and devices; exercises and answered questions ACT experience and 4
homework; feedback suffering; values; relaxation; through online platform supervised master-level
acceptance; defusion; mindfulness; students
mindful communication; maintenance
of gains; experiential avoidance;
summary
HIV
Ishola, 2015 1 in-person session; weekly texts Acceptance; cognitive defusion; being Delivered 1 in-person ACT session Not described
(SMS) over a 3-month period present; self as context; values;
committed action; post HIV test
counseling
Multiple Sclerosis
Proctor et al, 2018 8 weekly telephone sessions with a “Get Out of Your Mind and Into Your Provided weekly, theoretically Trainee clinical psychologist
self-help book Life” book which covers all oriented support calls that averaged with supervision from an
components of the ACT model 14 min per participant experienced ACT practitioner-
researcher
Tinnitus
Hesser et al, 2012 8 weekly internet-based modules as Mindfulness; cognitive defusion; Monitored homework; between each 1 licensed psychologist and 6
downloadable PDF’s; homework; identifying personal values and goals; module provided feedback, guidance master’s level students; trained
online messages exchanged with willingness in context of value-based and support via secure encrypted web in CBT and internet-based ACT
therapist; MP3s behavior change; information about page; averaged 9 min per week per for tinnitus; therapists
tinnitus; as needed addressed common participant supervised weekly
and specific problems; skill
maintenance
9
2012) and another mentioned psychology, nursing, or health promotion (MDASI) global symptom interference subscale (k = 2), Multiple Scle
degrees without specifying the degree level (Hawkes et al., 2014). rosis Impact Scale (MSIS) physical subscale (k = 1), Hearing Handicap
Inventory for the Elderly-short (HHIE-S; k = 1), the Functional Assess
3.5. Comparator conditions ment of Chronic Illness Therapy – Functional Wellbeing subscale (k = 1),
and level of function as defined by the Örebro Musculoskeletal Pain
All studies included at least one non-ACT comparator condition, with Questionnaire (ÖMPQ; k = 1), which was computed by aggregating 5
the exception of Herbert et al. (2017), which was a non-inferiority RCT function items (ability to carry out light work, walk for an hour, com
comparing video-teleconferencing delivered ACT to in-person delivery. plete household chores, shop for groceries, and sleep).
Inactive control comparators included treatment as usual (k = 5), no Functioning outcomes from Herbert et al. (2017) and Hawkes et al.
intervention (k = 1), and waitlist (k = 6). Active control comparators (2014) were excluded from meta-analyses due to the in-person ACT
included moderated online discussion forum (k = 2), telephone-based comparator and insufficient data, respectively. Thirteen effects were
therapist-delivered applied relaxation (k = 1), self-guided online included in the post-treatment meta-analysis. Between-group effect sizes
expressive writing (k = 1), telephone-based therapist-delivered educa for included studies ranged from small (k = 6), to medium (k = 3), to
tion and support (k = 2), post-HIV counseling (k = 1), and a large (k = 4). Study effects displayed moderate to substantial hetero
non-interactive website for pain self-management with ACT-based geneity (I2 = 61.8%; Q (12) = 31.42, p = 0.001; τ2 = 0.10).
strategies and exercises (k = 1). Active intervention comparators Technology-supported ACT significantly outperformed comparator
included therapist-guided internet CBT (k = 1), in-person ACT (k = 2), groups at post-treatment with a medium pooled effect size (mean Hed
unguided internet-based ACT (k = 1), and ACT self-help intervention ges’ g = − 0.49, 95% CI [− 0.76, − 0.22], p = 0.002) (Fig. 3a). No outliers
with email prompts (k = 1). were detected. When studies deemed high RoB were excluded (k = 2),
Because ACT- or CBT-based comparators were not included in meta- heterogeneity slightly decreased (I2 = 58.4%; Q (10) = 24.06, p = 0.01;
analyses, we briefly describe findings related to function and ACT pro τ2 = 0.08), and results remained significant with a medium pooled effect
cess outcomes in these studies. In Herbert et al. (2017), non-inferiority (mean Hedges’ g = − 0.57, 95% CI [− 0.86, − 0.28], p = 0.002).
between video-teleconferencing and in-person delivered ACT was sup A total of eight effects were included in the follow-up meta-analysis.
ported on the function outcome (Brief Pain Inventory (BPI) pain inter Between-group effect sizes ranged from small (k = 4), to medium (k = 3)
ference subscale) at posttreatment and 6-month follow-up. Pooling both to large (k = 1). Study effects displayed moderate to substantial het
conditions together, large effect sizes were observed at both time points erogeneity (I2 = 64.2%; Q (7) = 19.55, p = 0.01; τ2 = 0.12). Technology-
(within-group Cohen’s d = 0.81 and 0.84, respectively). Further, supported ACT significantly outperformed comparator groups at follow-
non-inferiority was supported on the ACT process outcome (CPAQ) at up with a medium pooled effect size (mean Hedges’ g = − 0.52, 95% CI
post-treatment, but not at follow-up. Pooling both conditions together, [− 0.93, − 0.10], p = 0.02) (Fig. 3b). No outliers were detected. When
large effect sizes were reported at both time points (within-group studies deemed high RoB were excluded (k = 2), heterogeneity
Cohen’s d = 1.19 and 1.01, respectively). Lin et al. (2017) compared increased (I2 = 71.3%; Q (5) = 17.44, p = 0.004; τ2 = 0.15) and pooled
guided internet-based ACT, unguided internet-based ACT, and waitlist effect effects remained medium but fell outside of statistical significance
control. The guided ACT condition, but not the unguided ACT condition, (mean Hedges’ g = − 0.52, 95% CI [− 1.12, 0.07], p = 0.07).
showed significantly greater improvement on the function outcome ACT Processes. A total of 18 studies included an ACT process
(Multidimensional Pain Inventory (MPI) pain interference subscale) measure, all of which were self-report. As shown in Table 1, these
compared to the control group at posttreatment (Cohen’s d = 0.58) and measures included the CPAQ (k = 6), AAQ-II (k = 5), Psych Inflexibility
4-month follow-up (Cohen’s d = 0.58). There were no significant dif in Pain Scale (PIPS; k = 2), Acceptance and Action Questionnaire for
ferences between guided and unguided conditions. Further, neither Weight-Related-Difficulties (AAQ-W; k = 2), TAQ (k = 1), Hearing
guided or unguided conditions were associated with significant Acceptance Questionnaire (HAS; k = 1), and the CompACT (k = 1).
improvement on the ACT process outcome (AAQ-II) relative to waitlist Similar to functioning outcomes, ACT process outcomes from Her
control. Sairanen et al. (2017) compared an unguided ACT smartphone bert et al. (2017) and Hawkes et al. (2014) were excluded from
app, in-person ACT, and a control group that received no intervention. meta-analyses due to the in-person ACT comparator and insufficient
Function was not assessed and both ACT treatments failed to exhibit data, respectively, leaving 16 effects that were included in the
significant improvement on the ACT process outcome (AAQ-II) post-treatment meta-analysis. At post-treatment, between-group effect
compared to the control group. Potts et al. (2020) compared self-help sizes for included studies ranged from small (k = 7), to medium (k = 4),
ACT plus telephone coaching, self-help ACT plus email prompting, and to large (k = 5). Study effects displayed moderate to substantial het
waitlist control. Function was not assessed and there were no significant erogeneity (I2 = 58.2%; Q (15) = 35.90, p = 0.001; τ2 = 0.08).
differences on the ACT process outcome (Comprehensive assessment of Technology-supported ACT significantly outperformed comparator
Acceptance and Commitment Therapy processes; CompACT) across the groups at post-treatment with a medium pooled effect size (mean Hed
three groups. Hesser et al. (2012) compared internet-delivered ACT, ges’ g = 0.48, 95% CI [0.26, 0.71], p < 0.001) (Fig. 3c). One study effect
internet-delivered CBT, and a monitored internet discussion group. At was determined to be an outlier (Molander et al., 2018). When this study
posttreatment, both ACT and CBT conditions were associated with sig was removed, heterogeneity decreased (I2 = 34.0%; Q (14) = 21.21, p =
nificant improvements on the function outcome (Tinnitus Handicap 0.10; τ2 = 0.03), and results remained significant with a medium pooled
Inventory; THI) (Cohen’s d: ACT = 0.68; CBT = 0.70) and ACT process effect favoring ACT (mean Hedges’ g = 0.54, 95% CI [0.38, 0.71], p <
outcome (Tinnitus Acceptance Questionnaire; TAQ) (Cohen’s d: ACT = 0.001). Further, when studies deemed high RoB were excluded (k = 4),
0.59; CBT = 0.45) compared to the monitored internet discussion group. heterogeneity was similar (I2 = 58.1%; Q (11) = 26.27, p = 0.006; τ2 =
There were no differences between ACT and CBT conditions. 0.07), and a significant small pooled effect was found favoring ACT
(mean Hedges’ g = 0.40, 95% CI [0.13, 0.66], p = 0.01).
3.6. Outcomes and meta-analyses A total of seven effects were included in the follow-up meta-analysis.
Between-group effect sizes ranged from small (k = 4), medium (k = 2),
Functioning. A total of 15 studies included a measure of func to large (k = 1). Study effects displayed minimal heterogeneity (I2 =
tioning, all of which were self-report. As shown in Table 1, these mea 0.0%; Q (6): 2.77, p = 0.383; τ2 = 0.0). Technology-supported ACT
sures were the MPI pain interference subscale (k = 3), BPI pain significantly outperformed comparator groups with a small pooled effect
interference subscale (k = 1), Fibromyalgia Impact Questionnaire (FIQ; size (mean Hedges’ g = 0.44, 95% CI [0.30, 0.58], p < 0.001) (Fig. 3d).
k = 2), Pain Interference Index (PII; k = 1), Work and Social Functioning No outlier effects were identified. When studies deemed high RoB were
Scale (WSAS; k = 1), THI (k = 1), MD Anderson Symptom Inventory excluded (k = 2), heterogeneity remained minimal (I2 = 0.0%; Q (4) =
10
Fig. 3a. Between-group meta-analysis results and forest plot for function outcomes at post-treatment.
Fig. 3b. Between-group meta-analysis results and forest plot for function outcomes at follow-up.
Fig. 3c. Between-group meta-analysis results and forest plot for ACT process outcomes at post-treatment.
11
Fig. 3d. Between-group meta-analysis results and forest plot for ACT process outcomes at follow-up. Note: Error bars are 95% confidence intervals; dashed line is
pooled effect size; red line is prediction interval. SD = standard deviation. SMD = standardized mean difference, calculated as Hedges’ g. For function outcomes,
negative SMD values indicate greater functional improvement/less functional impairment. For ACT process outcomes, positive SMD values indicate greater indicators
of psychological flexibility. Unadjusted means and SDs were obtained for Mosher et al. (2018, 2019) and Scott et al. (2018). (For interpretation of the references to
colour in this figure legend, the reader is referred to the Web version of this article.)
1.43, p = 0.84; τ2 = 0.00), and pooled effects were medium and sig population (chronic pain vs. other) and comparator type (active vs.
nificant (mean Hedges’ g = 0.46, 95% CI [0.30, 0.62], p = 0.001). inactive) did not moderate pooled effect sizes on function outcomes at
post-treatment. Further, technology modality, therapist contact method,
clinical population, and comparator type did not moderate pooled effect
3.7. Post hoc exploratory subgroup analyses sizes on ACT process outcomes.
Table 3 shows findings of exploratory subgroup analyses. Technol 4. Discussion

ogy modality (internet vs. other) and therapist contact method (in-per
son/telephone vs. asynchronous) moderated the pooled effect sizes on To the best of our knowledge, this is the first systematic review and
function outcomes at post-treatment using the alpha level of 0.10. meta-analysis to examine the impact of technology-supported ACT on
Specifically, internet-based studies (k = 8) showed significant medium functioning and ACT process outcomes across multiple CHCs. A total of
effects (mean Hedges’ g = − 0.63, 95% CI [− 0.97, − 0.28], p = 0.004) 21 manuscripts describing 20 unique studies met the inclusion criteria,
and non-internet-based studies (k = 5) showed non-significant small of which only four were rated as having high RoB, attesting to the
effects. Both internet- and non-internet-based studies showed moderate overall acceptable quality of the studies. Nearly half of the studies were
to substantial heterogeneity. Studies with in-person/telephone therapist focused on chronic pain Internet and telephone were the most used
contact (k = 8) showed small effects that fell outside of significance technology modalities and nearly all technology-supported ACT condi
(mean Hedges’ g = − 0.31, 95% CI [− 0.65, − 0.03], p = 0.07) and studies tions included therapist involvement. Meta-analyses found significant
utilizing asynchronous contact (k = 5) showed significant medium ef medium pooled effects favoring technology-supported ACT for func
fects (mean Hedges’ g = − 0.74, 95% CI [− 1.23, − 0.24], p = 0.01). Both tioning outcomes and ACT process outcomes at post-treatment and small
studies with in-person/telephone therapist contact and asynchronous to medium pooled effects at follow-up. Taken together, this review
contact showed moderate to substantial heterogeneity. Clinical
Table 3
Post-hoc exploratory sub-group analyses at post-treatment.
Outcome Subgroup k Hedges’ g 95% CI Heterogeneity (I2) Test for differences
Technology Modality
Function Internet 8 − 0.63 (-0.97, − 0.28) 60.9% Q (1) = 3.09, p = 0.08
Other 5 − 0.22 (-0.73, 0.29) 46.3%
ACT process Internet 10 0.48 (0.16, 0.80) 63.9% Q (1) = 0.01, p = 0.94
Other 6 0.49 (0.07, 0.92) 54.4%
Therapist Contact Method
Function In-person/phone 8 − 0.31 (-0.65, 0.03) 54.3% Q (1) = 3.42, p = 0.06

Asynchronous 5 − 0.74 (-1.23, − 0.24) 63.4%
ACT process In-person/phone 11 0.57 (0.34, 0.79) 39.6% Q (1) = 0.80, p = 0.37
Asynchronous 5 0.32 (-0.38, 1.03) 75.8%
Clinical Population
Function Chronic pain 8 − 0.54 (-0.87, − 0.22) 58.6% Q (1) = 0.47, p = 0.49
Other 5 − 0.35 (-1.05, 0.35) 71.4%
ACT process Chronic pain 8 0.53 (0.33, 0.72) 8.2% Q (1) = 0.39, p = 0.53
Other 8 0.39 (-0.10, 0.87) 74.5%
Comparator Type
Function Active 7 − 0.42 (-0.86, 0.01) 65.0% Q (1) = 0.30, p = 0.58

Inactive 6 − 0.56 (-1.04, − 0.09) 60.9%
ACT process Active 6 0.62 (0.37, 0.87) 0.0% Q (1) = 1.82, p = 0.18
Inactive 10 0.37 (0.03, 0.72) 67.3%
12
demonstrates the growing literature on technology-supported ACT for conducted both sensitivity analyses removing outliers and studies with
CHCs and provides initial support of its efficacy compared to inactive high RoB, as well as exploratory post hoc subgroup analyses on tech
and active control conditions for outcomes most relevant to ACT and nology modality, therapist contact type, clinical population, and
CHCs. comparator type. Sensitivity analyses did not substantially alter find
Despite significant advances in technology, the majority of studies ings, with the exception of function outcomes at follow-up, which
included in this review relied on relatively older technology, including became non-significant after removing two studies deemed high RoB. In
internet-based content and telephone, while only two studies utilized post hoc subgroup analyses, we found that both technology modality
smartphone technology. This is somewhat surprising given the interest (internet vs. other) and therapist contact type (in-person/telephone vs.
in smartphone applications generally and advantages of smartphone asynchronous) moderated functioning outcomes at post-treatment, with
technology within research settings (e.g., accessibility, real-time data results demonstrating larger effects for internet-based studies and
tracking). Previous reviews have demonstrated the efficacy of smart studies utilizing asynchronous therapist contact. It should be noted that
phone applications for CHCs (Alwashmi et al., 2016; Wang et al., 2014), four of the five non-internet-based studies were telephone-based, and
as well as the efficacy of mindfulness- and acceptance-based smartphone that these four telephone-based studies also comprised half of the in-
applications (Linardon, 2020). ACT-based smartphone applications person/telephone therapist contact sub-group. Although these results
have been developed and tested for improving diet and exercise (Levin must be interpreted with caution, findings suggest using internet as a
et al., 2017) and chronic pain (Gentili et al., 2020), and therefore future mode of technology with asynchronous therapist contact is likely an
RCTs of ACT-based smartphone applications for CHCs should be effective strategy to deliver ACT content for CHCs and potentially better
forthcoming. than low technology methods like telephone. Additionally, because of
All of the studies included in this review included some level of the strong existing evidence for ACT for chronic pain, it is not surprising
therapist involvement, making it difficult to systematically examine the that the majority of technology-supported ACT trials have been on
specific impact of therapists in technology-supported ACT for CHCs. chronic pain. Although slightly larger pooled treatment effects in func
Only one study directly compared guided and unguided interventions tion and ACT process outcomes were observed in chronic pain relative to
and did not find differences on outcomes but showed lower attrition other conditions, between group analyses did not approach significance.
with the guided compared to the unguided intervention (Lin et al., Thus, results provide support for the use of technology-supported ACT
2017). Further, we found considerable heterogeneity in the extent of for chronic pain as well as CHCs other than chronic pain. Further, there
therapist involvement and modality of communication (e.g., phone, were no differences in pooled effects between studies comparing to
email, messaging board). Only one study directly compared modality of inactive or active controls, suggesting that technology-supported ACT
therapist contact (telephone versus email) and found differences on may hold promise for improving functioning in CHCs in comparison to
relevant outcomes (e.g., eating and physical activity) but not on ACT some active control conditions. More research with appropriate research
process outcomes (psychological flexibility) (Potts et al., 2020). These designs and active intervention comparisons are needed to examine
limited findings are similar to what was found in previous reviews of whether technology-supported ACT is as effective or more effective than
internet-based CBT for psychiatric and somatic disorders (Carlbring other evidence-based active interventions for improving functioning in
et al., 2017) and anxiety and depression in individuals with CHCs CHCs.
(Mehta et al., 2019). Further, evidence has accumulated that
technology-based interventions, including ACT, with therapist involve 4.1. Recommendations for future research
ment tend to be more effective than unguided interventions for mental
health conditions or symptoms (Baumeister et al., 2014; Thompson Technology-supported interventions hold promise for increasing
et al., 2021). However, results from technology-based interventions for access to treatment. This is particularly relevant in the context of the
CHCs are mixed, in part because many studies do not directly compare COVID-19 pandemic and increase in telehealth appointments. This re
guided to unguided formats (Beatty & Lambert, 2013). Thus, view provides initial support for the utility of technology-supported ACT
well-designed and large studies of technology-supported ACT in for CHCs. However, additional high-quality research is needed to un
terventions with CHCs are needed to directly examine the method, derstand how to optimize its delivery and maximize its efficacy. Below
amount, intensity, and role of therapist involvement. are several recommendations to accomplish this task.
The results of our meta-analyses showed that technology-supported
ACT was efficacious in improving both functioning and ACT process 1) Expand research to include other CHC populations. The majority of
outcomes at post-treatment and follow-up across multiple CHCs. Our included studies focused on individuals with chronic pain, followed
findings are broadly consistent with previous reviews of mobile and by overweight/obesity, and cancer. There were no technology-
internet technologies for disease management and distress in CHCs supported trials of ACT for other common CHCs such as asthma,
(Beatty & Lambert, 2013; Ebuenyi et al., 2021) as well as reviews of ACT diabetes, cardiovascular disease, and chronic kidney disease that met
interventions for chronic pain (Hughes et al., 2017) and multiple CHCs the inclusion criteria for this review. Additionally, multimorbidity of
(Dochat et al., 2021). The medium-sized pooled effects found in our CHCs is common, which further impacts functioning and quality of
analyses were larger than the effect sizes found in previous life and increases health care burden (Buttorff et al., 2017). Pop
meta-analyses of internet-supported ACT for mental health outcomes ulations with multiple co-occurring CHCs are ideal for
like depression and anxiety (Brown et al., 2016; Thompson et al., 2021). technology-supported ACT interventions yet are not adequately
This is possibly a result of heterogeneity due to differing populations, represented in the literature.
technologies, comparator conditions, and other characteristics of the 2) Improve methodological rigor of RCTs. The primary methodological
studies included in this review. Alternately, it is possible ACT in concerns among included studies were lack of clarity regarding an
terventions to improve functioning in CHCs are well-suited for delivery alyses and missing data/attrition. To reduce reporting bias and
via technology methods. Our findings also are consistent with previous promote open science, we recommend pre-specifying data analyses
ACT intervention research showing that post-treatment gains are often using a standard framework, such as Pre-SPEC (Kahan et al., 2020).
maintained between post-treatment and follow-up (Gifford et al., 2004; Further, attrition in behavioral trials of individuals with CHCs is a
Lee et al., 2015). Further, the small-to medium-sized pooled effects for well-documented concern in the broader literature (Davis & Addis,
ACT process outcomes remained robust across timepoints, suggesting 1999), including digital interventions, particularly when in
that materials designed to address ACT processes for CHCs can be terventions are self-guided (Macea et al., 2010). In this review, 11
learned through technology means. out of 20 (55%) studies at post-treatment and 12 out of 15 (80%)
To address the heterogeneity observed among included studies, we studies at follow-up reported 20% or greater attrition, which is a
13
common cut-off to indicate potentially problematic attrition (Dum documented sex differences in CHCs, including chronic pain (Bartley
ville et al., 2006). Methods to reduce attrition such as email and text & Fillingim, 2013) and diabetes risk (Ding et al., 2006). Future
message reminders were used by some included studies and are research is encouraged to recruit a more balanced women-to-men
recommended. At minimum, investigators should undertake sensi ratio that reflects the prevalence of the specific CHC and examine
tivity analyses to determine the impact of attrition on study out sex differences.
comes. 6) Leverage unique technology-based opportunities including ecolog
While the RoB was deemed low for all studies per the Cochrane ical momentary assessment and just-in-time interventions. Life with
RoB-2 criteria for ‘bias in the measurement of outcome,’ not all CHCs occurs outside of medical and mental healthcare visits.
outcome measures were ideal. For example, Thorsell et al. (2011) Symptoms, behaviors, and psychological processes pertinent to these
used select items from the ÖMPQ to assess function, which was conditions fluctuate across and within days, and therefore may not
designed to predict long-term disability and work absenteeism and is be adequately captured by standard self-report measures or
not a measure of functioning per se. Further, the dynamic and adequately addressed during discrete healthcare appointments. The
context-dependent nature of psychological flexibility makes it an inclusion of real-time, ambulatory assessment methods such as
inherently difficult construct to assess using static, global self-report accelerometry and ecological momentary assessment may better
measures that do not consider temporality or situational context. inform our understanding of the lived experiences and treatment
Although measures such as the AAQ-II and CPAQ have been vali needs of people with CHCs. These data can in turn inform adaptive
dated and are widely-used, these measures also have been criticized just-in-time interventions (Nahum-Shani et al., 2018) that can pro
for their content validity (Van Ryckeghem, 2020; Wolgast, 2014). vide tailored ACT intervention components at times when in
Newer measures of psychological flexibility that address some of dividuals can benefit most (Levin et al., 2019).
these limitations include the Multidimensional Psychological Flexi
bility Inventory, which was able to distinguish psychological flexi 4.2. Strengths and limitations
bility from distress (Landi et al., 2021), and the Psy-Flex measure, a
6-item measure conducive to repeated sampling that includes situ This study has several strengths. We provide a systematic review and
ational and temporal specifiers (Gloster et al., 2021). meta-analysis of technology-supported ACT for functioning and ACT
3) Increase clarity of reporting treatment engagement and strategies to process outcomes across multiple CHCs. We included only RCTs and
increase treatment engagement. It is important to understand the examined only between-group effects, which allows stronger conclu
amount of intervention delivered through technology (e.g., number sions to be drawn about intervention efficacy than does examination of
of hours or extent of content) in order to help design interventions single-arm trials and within-group effects. In addition to a quantitative
that deliver adequate exposure to the intervention. Future in approach to examining functioning and ACT process outcomes at post-
vestigations also are encouraged to systematically quantify and treatment and follow-up, qualitative detailed descriptions of interven
report the amount and level of intervention exposure by both treat tion content and therapist involvement are helpful for the clinical
ment completers and non-completers. Further, future studies are guidance and application of technology-supported ACT. Further, rec
encouraged to include strategies to help maximize treatment ommendations for future research are provided to inform the develop
engagement. This may include integrating positive reinforcement (e. ment and evaluation of future RCTs of technology-supported ACT.
g., praise and rewards for completing tasks), ensuring that system Nonetheless, our study has shortcomings. Not all studies reported
components are familiar and attractive to users, and considering functioning and ACT process outcomes or reported outcomes at both
sociodemographic information known to affect response rates, such post-treatment and follow-up time points, which limited the number of
as gender, age, and severity of the CHC (Karekla et al., 2019). studies that could be included in the quantitative analyses. Additionally,
4) Directly compare technology-supported ACT for CHC using different many prominent CHCs were not represented, including diabetes and
technology modalities and varying levels of therapist involvement. cardiovascular disease. Thus, it is uncertain if our results generalize to
Given the changes made to delivery of healthcare as a result of the these conditions. Studies that focus on additional populations are
COVID-19 pandemic, technology-supported interventions will likely needed to better assess the effectiveness of technology-supported ACT
become even more prevalent and necessary. Thus, there is a great for CHCs. Gray literature was not included in our search. While there is
need to optimize their impact and determine the extent of required ongoing debate on the impact of excluding gray literature in meta-
therapist involvement. Once optimized, large-scale non-inferiority analyses (Schmucker et al., 2017), it is possible that publication bias
trials comparing virtual to in-person delivery and comparative may have influenced study results. Although we examined several
effectiveness trials examining technology-supported ACT with established indicators of publication bias, these methods are themselves
differing treatment components are needed to firmly establish the limited, particularly when analyzing a small number of treatment ef
efficacy of technology-supported ACT for CHCs. fects. Clinical heterogeneity in terms of intervention format, treatment
5) Address lack of diversity. The majority of included studies did not intensity/dose, clinical population, level of clinician involvement, spe
report race/ethnicity. Across studies that did, only 16% of samples cific technology modality used, and comparator condition characteris
were non-White. This shortcoming is important to address in both tics likely contributed to the observed statistical heterogeneity and may
technology-supported interventions and ACT-based interventions have impacted conclusions about efficacy. While we examined some
generally (Woidneck et al., 2012). While access to the internet has sources of heterogeneity, others should be quantitatively explored in
steadily increased across the globe, the “digital divide” persists. For future reviews using subgroup or network meta-analysis. Further, the
example, Black and Latinx adults are less likely to use technology to majority of studies included condition-specific measures of function
assess health management websites and search for health informa which are considered more sensitive to change. Future studies and larger
tion compared to White adults (Mitchell et al., 2019). Further, meta-analyses are encouraged to include both condition-specific and
questions remain on how to best deliver mindfulness- and general measures of functioning. Finally, we did not conduct
acceptance-based interventions among underserved populations that meta-analyses with active intervention comparators. This should be kept
may be facing adversity (Sobczak & West, 2013). There is a need to in mind when contextualizing the pooled effects from this study.
increase the inclusion and reporting of diverse racial and ethnic
populations in technology-supported ACT for CHCs to avoid creating 4.3. Conclusion
additional healthcare access inequities. Similarly, the majority of
participants in these studies were female. While we are unaware of Technology-supported ACT shows promise to improve functioning
any literature that shows sex differences in ACT, there are and ACT process outcomes across a range of implementation methods,
14
CHC populations, and comparator conditions. Included studies used a Baumeister, H., Reichler, L., Munzinger, M., & Lin, J. (2014). The impact of guidance on
Internet-based mental health interventions—a systematic review. Internet
variety of technology modalities to deliver ACT content and nearly all
Interventions, 1(4), 205–215. https://doi.org/10.1016/j.invent.2014.08.003
studies involved a therapist to motivate participation and/or facilitate Beatty, L., & Lambert, S. (2013). A systematic review of internet-based self-help
content comprehension. Meta-analyses showed that technology- therapeutic interventions to improve distress and disease-control among adults with
supported ACT outperformed comparator conditions in improving chronic health conditions. Clinical Psychology Review, 33(4), 609–622. https://doi.
org/10.1016/j.cpr.2013.03.004
functioning and ACT process measures at post-treatment and follow-up. Brown, M. G. A., Hoon, A. E., & John, A. (2016). Effectiveness of web-delivered
Additional high-quality research is needed to demonstrate short- and acceptance and commitment therapy in relation to mental health and well-being: A
long-term efficacy compared to other active interventions and to inform systematic review and meta-analysis. Journal of Medical Internet Research, 18(8).
https://doi.org/10.2196/jmir.6200. e221–e221.
treatment optimization. Continued research in this area holds promise Brundisini, F., Giacomini, M., DeJean, D., Vanstone, M., Winsor, S., & Smith, A. (2013).
for improving function and quality of life among millions of individuals Chronic disease patients’ experiences with accessing health care in rural and remote
living with CHCs worldwide. areas: A systematic review and qualitative meta-synthesis. Ontario Health Technology
Assessment Series, 13(15), 1–33. https://doi.org/10.1016/j.brat.2013.02.010
Buhrman, M., Skoglund, A., Husell, J., Bergström, K., Gordh, T., Hursti, T., Bendelin, N.,
Credit Author Statement Furmark, T., & Andersson, G. (2013). Guided internet-delivered acceptance and
commitment therapy for chronic pain patients: A randomized controlled trial.
Behaviour Research and Therapy, 51(6), 307–315.
M.S.H.: Conceptualization, Data curation, Methodology, Project Buttorff, C., Ruder, T., & Bauman, M. (2017). Multiple chronic conditions in the United
administration, Supervision, Investigation, roles/Writing – original States. In Multiple chronic Conditions in the United States. Rand. https://doi.org/
draft, Writing – review & editing. C.D.: Conceptualization, Data cura 10.7249/tl221
Carlbring, P., Andersson, G., Cuijpers, P., Riper, H., & Hedman-Lagerlöf, E. (2017).
tion, Formal analysis, Investigation, Visualization, roles/Writing – Internet-based vs. Face-to-face cognitive behavior therapy for psychiatric and
original draft. J.S.W.: Conceptualization, Data curation, Methodology, somatic disorders: An updated systematic review and meta-analysis. Taylor &
Investigation, roles/Writing – original draft. K.M.: Conceptualization, Francis, 47(1), 1–18. https://doi.org/10.1080/16506073.2017.1401115
Clarke, D. M., & Currie, K. C. (2009). Depression, anxiety and their relationship with
Data curation, Investigation, roles/Writing – original draft. B.H.B.: chronic diseases: A review of the epidemiology, risk and treatment evidence. Medical
Conceptualization, Data curation, Investigation, roles/Writing – original Journal of Australia, 190(7 SUPPL), S54–S60. https://doi.org/10.5694/j.1326-
draft. M.T.: Data curation, roles/Writing – original draft. M.W.L.: Data 5377.2009.tb02471.x
Collins, J. J., Baase, C. M., Sharda, C. E., Ozminkowski, R. J., Nicholson, S.,
curation, roles/Writing – original draft. M.G.: Data curation, roles/ Billotti, G. M., Turpin, R. S., Olson, M., & Berger, M. L. (2005). The assessment of
Writing – original draft. A.C.: Data curation, roles/Writing – original chronic health conditions on work performance, absence, and total economic impact
draft. D.H.: Data curation, roles/Writing – original draft. N.A.: for employers. Journal of Occupational and Environmental Medicine, 47(6), 547–557.
https://doi.org/10.1097/01.jom.0000166864.58664.29
Conceptualization, Funding acquisition, Project administration, Re Davis, M. J., & Addis, M. E. (1999). Predictors of attrition from behavioral medicine
sources, Supervision, Writing – review & editing treatments. Annals of Behavioral Medicine, 21(4), 339–349.
Deeks, J. J., Higgins, J. P. T., & Altman, D. G. (2019). Analysing data and undertaking
meta-analyses. In Cochrane handbook for systematic reviews of interventions (pp.
Support 241–284). Wiley. https://doi.org/10.1002/9781119536604.ch10.
Dieleman, J. L., Baral, R., Birger, M., Bui, A. L., Bulchis, A., Chapin, A., Hamavid, H.,
Dr. Herbert is supported by Veterans Affairs Rehabilitation Research Horst, C., Johnson, E. K., Joseph, J., Lavado, R., Lomsadze, L., Reynolds, A.,
Squires, E., Campbell, M., DeCenso, B., Dicker, D., Flaxman, A. D., Gabert, R., &
and Development Service under Career Development Award Murray, C. J. L. (2016). US spending on personal health care and public health,
1IK2RX002807. Dr. Herbert, Ms. Dochat, Dr. Materna, Mr. Blanco, Ms. 1996-2013. JAMA - Journal of the American Medical Association, 316(24), 2627–2646.
Tynan, Mr. Lee, Ms. Camodeca, Ms. Harris, and Dr. Afari are partially https://doi.org/10.1001/jama.2016.16885
Ding, E. L., Song, Y., Malik, V. S., & Liu, S. (2006). Sex differences of endogenous sex
supported by National Institute of Diabetes and Kidney Disease under hormones and risk of type 2 diabetes: A systematic review and meta-analysis. Jama,
R01DK106415. Dr. Wooldridge is supported by the VA Office of Aca 295(11), 1288–1299. https://doi.org/10.1001/jama.295.11.1288
demic Affiliates advanced fellowship in women’s health. Dr. Gasperi is Dochat, C., Wooldridge, J. S., Herbert, M. S., Lee, M. W., & Afari, N. (2021). Single-
session acceptance and commitment therapy (ACT) interventions for patients with
supported by Veterans Affairs Clinical Science Research and Develop chronic health conditions: A systematic review and meta-analysis. Journal of
ment Service under Career Development Award 1IK2CX002107. The Contextual Behavioral Science, 20, 52–69. https://doi.org/10.1016/j.
views expressed in this paper are those of the authors and do not reflect jcbs.2021.03.003
Dumville, J. C., Torgerson, D. J., & Hewitt, C. E. (2006). Reporting attrition in
the official policy or position of the funding agency, Department of
randomised controlled trials. British Medical Journal, 332(7547), 969–971. https://
Veterans Affairs, the United States Government, or any institutions with doi.org/10.1136/bmj.332.7547.969
which the authors are affiliated. Ebuenyi, M. C., Schnoor, K., Versluis, A., Meijer, E., & Chavannes, N. H. (2021). Short
message services interventions for chronic disease management: A systematic
review. Clinical EHealth. https://doi.org/10.1016/j.ceh.2020.11.004
Disclosure statement Elbert, N. J., Van Os-Medendorp, H., Van Renselaar, W., Ekeland, A. G., Hakkaart-Van
Roijen, L., Raat, H., Nijsten, T. E. C., & Pasmans, S. G. M. A. (2014). Effectiveness and
cost-effectiveness of ehealth interventions in somatic diseases: A systematic review
None of the authors have any conflicts of interest to disclose. of systematic reviews and meta-analyses. Journal of Medical Internet Research, 16(4).
https://doi.org/10.2196/jmir.2790. e110–e110.
Eton, D., Elraiyah, T. A., Yost, K., Ridgeway, J., Johnson, A., Egginton, J., Mullan, R. J.,
Declaration of interest statement
Murad, M. H., Erwin, P., & Montori, V. (2013). A systematic review of patient-
reported measures of burden of treatment in three chronic diseases. Patient Related
All authors declare no conflicts of interest. Outcome Measures, 4. https://doi.org/10.2147/prom.s44694, 7–7.
Feliu-Soler, A. M. F., Gutierrez-Martinez, O., Scott, W., McCracken, L. M., &
Luciano, J. V. (2018). Current status of acceptance and commitment therapy for
chronic pain: A narrative review. Journal of Pain Research, 11. https://doi.org/
Declaration of competing interest 10.2147/JPR.S144631, 2145–2145.
Field, A. P., & Gillett, R. (2010). How to do a meta-analysis. British Journal of
All authors declare no conflicts of interest. Mathematical and Statistical Psychology, 63(3), 665–694. https://doi.org/10.1348/
000711010X502733
Gentili, C., Zetterqvist, V., Rickardsson, J., Holmström, L., Simons, L. E., & Wicksell, R. K.
References (2020). ACTsmart: Guided smartphone-delivered acceptance and commitment
therapy for chronic pain—a pilot trial. Pain Medicine, 22(2), 315–328. https://doi.
org/10.1093/pm/pnaa360
Alwashmi, M., Hawboldt, J., Davis, E., Marra, C., Gamble, J.-M., & Ashour, W. A. (2016).
Gifford, E. V., Kohlenberg, B. S., Hayes, S. C., Antonuccio, D. O., Piasecki, M. M.,
The effect of smartphone interventions on patients with chronic obstructive
Rasmussen- Hall, M. L., & Palm, K. M. (2004). Acceptance-based treatment for
pulmonary disease exacerbations: A systematic review and meta-analysis. JMIR
smoking cessation. Behavior Therapy, 35(4), 689–705. https://doi.org/10.1016/
MHealth and UHealth, 4(3), Article e5921. https://doi.org/10.2196/mhealth.5921
S0005-7894(04)80015-7
Bartley, E. J., & Fillingim, R. B. (2013). Sex differences in pain: A brief review of clinical
Gloster, A. T., Block, V. J., Klotsche, J., Villanueva, J., Rinner, M. T., Benoy, C.,
and experimental findings. British Journal of Anaesthesia, 111(1), 52–58. https://doi.
Walter, M., Karekla, M., & Bader, K. (2021). Psy-flex: A contextually sensitive
org/10.1093/bja/aet127
15
measure of psychological flexibility. Journal of Contextual Behavioral Science. https:// Levin, M., Petersen, J., Durward, C., Bingeman, B., Davis, E., Nelson, C., & Cromwell, S.
doi.org/10.1016/j.jcbs.2021.09.001 (2020). A randomized controlled trial of online acceptance and commitment therapy
Gloster, A. T., Walder, N., Levin, M., Twohig, M., & Karekla, M. (2020). The empirical to improve diet and physical activity among adults who are overweight/obese.
status of acceptance and commitment therapy: A review of meta-analyses. Journal of Translational Behavioral Medicine, 1–25. https://doi.org/10.1093/tbm/ibaa123
Contextual Behavioral Science. https://doi.org/10.1016/j.jcbs.2020.09.009 Levin, M. E., Pierce, B., & Schoendorff, B. (2017). The acceptance and commitment
Graham, C. D., Gouick, J., Krahé, C., & Gillanders, D. (2016). A systematic review of the therapy matrix mobile app: A pilot randomized trial on health behaviors. Journal of
use of Acceptance and Commitment Therapy (ACT) in chronic disease and long-term Contextual Behavioral Science, 6(3), 268–275. https://doi.org/10.1016/j.
conditions. Clinical Psychology Review, 46, 46–58. https://doi.org/10.1016/j. jcbs.2017.05.003
cpr.2016.04.009 Liberati, A., Altman, D. G., Tetzlaff, J., Mulrow, C., Gøtzsche, P. C., Ioannidis, J. P. A.,
Harrer, M., Cuijpers, P., Furukawa, T. A., & Ebert, D. D. (2019). Doing meta-analysis in R: Clarke, M., Devereaux, P. J., Kleijnen, J., & Moher, D. (2009). The PRISMA statement
A hands-on guide. PROTECT Lab Erlangen. for reporting systematic reviews and meta-analyses of studies that evaluate health
Hawkes, A. L., Pakenham, K. I., Chambers, S. K., Patrao, T. A., & Courneya, K. S. (2014). care interventions: Explanation and elaboration. PLoS Medicine, 6(7). https://doi.
Effects of a multiple health behavior change intervention for colorectal cancer org/10.1371/journal.pmed.1000100
survivors on psychosocial outcomes and quality of life: A randomized controlled Linardon, J. (2020). Can acceptance, mindfulness, and self-compassion be learned by
trial. Annals of Behavioral Medicine, 48(3), 359–370. https://doi.org/10.1007/ smartphone apps? A systematic and meta-analytic review of randomized controlled
s12160-014-9610-2 trials. Behavior Therapy, 51(4), 646–658. https://doi.org/10.1016/j.
Hawkes, A. L., Pakenham, K. I., Courneya, K. S., Gollschewski, S., Baade, P., beth.2019.10.002
Gordon, L. G., Lynch, B. M., Aitken, J. F., & Chambers, S. K. (2009). A randomised Lin, J., Luking, M., Ebert, D., Buhrman, M., …G. A.-I., & 2015, U. (2015). Effectiveness
controlled trial of a tele-based lifestyle intervention for colorectal cancer survivors and cost-effectiveness of a guided and unguided internet-based Acceptance and
(’CanChange’): Study protocol. BMC Cancer, 9. https://doi.org/10.1186/1471-2407- Commitment Therapy for chronic pain: Study protocol for a. Internet Interventions,
9-286, 286–286. 2(1), 7-16. https://doi.org/10.1016/j.invent.2014.11.005.
Hayes, S. C., Luoma, J. B., Bond, F. W., Masuda, A., & Lillis, J. (2006). Acceptance and Lin, J., Paganini, S., Sander, L., Lüking, M., Daniel Ebert, D., Buhrman, M., Andersson, G.,
commitment therapy: Model, processes and outcomes. Behaviour Research and & Baumeister, H. (2017). An Internet-based intervention for chronic pain—a three-
Therapy, 44(1), 1–25. https://doi.org/10.1016/j.brat.2005.06.006 arm randomized controlled study of the effectiveness of guided and unguided
Herbert, M. S., Afari, N., Liu, L., Heppner, P., Rutledge, T., Williams, K., Eraly, S., acceptance and commitment therapy. Deutsches Arzteblatt International, 114(41),
VanBuskirk, K., Nguyen, C., & Bondi, M. (2017). Telehealth versus in-person 681–688. https://doi.org/10.3238/arztebl.2017.0681
acceptance and commitment therapy for chronic pain: A randomized noninferiority Macea, D. D., Gajos, K., Calil, Y. A. D., & Fregni, F. (2010). The efficacy of web-based
trial. The Journal of Pain, 18(2), 200–211. https://doi.org/10.1016/j. cognitive behavioral interventions for chronic pain: A systematic review and meta-
jpain.2016.10.014 analysis. The Journal of Pain, 11(10), 917–929. https://doi.org/10.1016/j.
Hesser, H., Gustafsson, T., Lundén, C., Henrikson, O., Fattahi, K., Johnsson, E., jpain.2010.06.005
Westin, V. Z., Carlbring, P., Mäki-Torkko, E., & Kaldo, V. (2012). A randomized Megari, K. (2013). Quality of life in chronic disease patients. Health Psychology Research,
controlled trial of Internet-delivered cognitive behavior therapy and acceptance and 1(3). https://doi.org/10.4081/hpr.2013.e27, 27–27.
commitment therapy in the treatment of tinnitus. Journal of Consulting and Clinical Mehta, S., Peynenburg, V. A., & Hadjistavropoulos, H. D. (2019). Internet-delivered
Psychology, 80(4), 649. https://doi.org/10.1037/a0027021 cognitive behaviour therapy for chronic health conditions: A systematic review and
Higgins, J. P. T., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). Measuring meta-analysis. Journal of Behavioral Medicine, 42(2), 169–187. https://doi.org/
inconsistency in meta-analyses. BMJ, 327(7414), 557–560. https://doi.org/ 10.1007/s10865-018-9984-x
10.1136/bmj.327.7414.557 Mitchell, U. A., Chebli, P. G., Ruggiero, L., & Muramatsu, N. (2019). The digital divide in
Ishola, A. G., & Chipps, J. (2015). The use of mobile phones to deliver acceptance and health-related technology use: The significance of race/ethnicity. The Gerontologist,
commitment therapy in the prevention of mother–child HIV transmission in Nigeria. 59(1), 6–14. https://doi.org/10.1093/geront/gny138
Journal of Telemedicine and Telecare, 21(8), 423–426. https://doi.org/10.1177/ Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2010). Preferred reporting items for
1357633X15605408 systematic reviews and meta-analyses: The PRIsMa statement. International Journal
Jette, A. (2006). Toward a common language for function, disability, and health. Physical of Surgery, 8(5), 336–341. https://doi.org/10.1136/bmj.b2535
Therapy, 86(5), 726–734. https://doi.org/10.1093/ptj/86.5.726 Molander, P., Hesser, H., Weineland, S., Bergwall, K., Buck, S., Hansson-Malmlöf, J.,
Kahan, B. C., Forbes, G., & Cro, S. (2020). How to design a pre-specified statistical Lantz, H., Lunner, T., & Andersson, G. (2015). Internet-based acceptance and
analysis approach to limit p-hacking in clinical trials: The Pre-SPEC framework. BMC commitment therapy for psychological distress experienced by people with hearing
Medicine, 18(1). https://doi.org/10.1186/s12916-020-01706-7 problems: Study protocol for a randomized controlled trial. American Journal of
Karekla, M., Kasinopoulos, O., Neto, D. D., Ebert, D. D., Van Daele, T., Nordgreen, T., Audiology, 24(3), 307–310. https://doi.org/10.1044/2015_AJA-15-0013
Höfer, S., Oeverland, S., & Jensen, K. L. (2019). Best practices and recommendations Molander, P., Hesser, H., Weineland, S., Bergwall, K., Buck, S., Jäder Malmlöf, J.,
for digital interventions to improve engagement and adherence in chronic illness sufferers. Lantz, H., Lunner, T., & Andersson, G. (2018). Internet-based acceptance and
European Psychologist. https://doi.org/10.1002/sim.1482 commitment therapy for psychological distress experienced by people with hearing
Knapp, G., & Hartung, J. (2003). Improved tests for a random effects meta-regression problems: A pilot randomized controlled trial. Cognitive Behaviour Therapy, 47(2),
with a single covariate. Statistics in Medicine, 22(17), 2693–2710. 169–184. https://doi.org/10.1080/16506073.2017.1365929
Kristjánsdóttir, Ó. B., Fors, E. A., Eide, E., Finset, A., Stensrud, T. L., Van Dulmen, S., Mosher, C. E., Secinti, E., Hirsh, A. T., Hanna, N., Einhorn, L. H., Jalal, S. I., Durm, G.,
Wigers, S. H., & Eide, H. (2013a). A smartphone-based intervention with diaries and Champion, V. L., & Johns, S. A. (2019). Acceptance and commitment therapy for
therapist-feedback to reduce catastrophizing and increase functioning in women symptom interference in advanced lung cancer and caregiver distress: A pilot
with chronic widespread pain: Randomized controlled trial. Journal of Medical randomized trial. Journal of Pain and Symptom Management, 58(4), 632–644. https://
Internet Research, 15(1). https://doi.org/10.2196/jmir.2249. e2249–e2249. doi.org/10.1016/j.jpainsymman.2019.06.021
Kristjánsdóttir, Ó. B., Fors, E. A., Eide, E., Finset, A., Stensrud, T. L., Van Dulmen, S., Mosher, C. E., Secinti, E., Li, R., Hirsh, A. T., Bricker, J., Miller, K. D., et al. (2018).
Wigers, S. H., & Eide, H. (2013b). A smartphone-based intervention with diaries and Acceptance and commitment therapy for symptom interference in metastatic breast
therapist feedback to reduce catastrophizing and increase functioning in women cancer: a pilot randomized trial. Supportive Care in Cancer, 26(6). https://doi.org/
with chronic widespread pain. Part 2: 11-Month follow-up results of a randomized 10.1007/s00520-018-4045-0
trial. Journal of Medical Internet Research, 15(3). https://doi.org/10.2196/jmir.2442. Nahum-Shani, I., Smith, S. N., Spring, B. J., Collins, L. M., Witkiewitz, K., Tewari, A., &
e2442–e2442. Murphy, S. A. (2018). Just-in-time adaptive interventions (JITAIs) in mobile health:
Kyle, T. K., Dhurandhar, E. J., & Allison, D. B. (2016). Regarding obesity as a disease: Key components and design principles for ongoing health behavior support. Annals
Evolving policies and their implications. Endocrinology and Metabolism Clinics of of Behavioral Medicine, 52(6), 446–462. https://doi.org/10.1007/s12160-016-9830-
North America, 45(3), 511. https://doi.org/10.1016/j.ecl.2016.04.004 8
Landi, G., Pakenham, K. I., Crocetti, E., Grandi, S., & Tossani, E. (2021). The Potts, S., Krafft, J., & Levin, M. E. (2020). A pilot randomized controlled trial of
multidimensional psychological flexibility inventory (MPFI): Discriminant validity acceptance and commitment therapy guided self-help for overweight and obese
of psychological flexibility with distress. Journal of Contextual Behavioral Science, 21, adults high in Weight self-stigma. Behavior Modification. https://doi.org/10.1177/
22–29. https://doi.org/10.1016/j.jcbs.2021.05.004 0145445520975112, 014544552097511–014544552097511.
Lappalainen, R., Sairanen, E., Järvelä, E., Rantala, S., Korpela, R., Puttonen, S., Proctor, B. J., Moghaddam, N. G., Evangelou, N., & das Nair, R. (2018). Telephone-
Kujala, U. M., Myllymäki, T., Peuhkuri, K., Mattila, E., Kaipainen, K., Ahtinen, A., supported acceptance and commitment bibliotherapy for people with multiple
Karhunen, L., Pihlajamäki, J., Järnefelt, H., Laitinen, J., Kutinlahti, E., Saarelma, O., sclerosis and psychological distress: A pilot randomised controlled trial. Journal of
Ermes, M., & Kolehmainen, M. (2014). The effectiveness and applicability of Contextual Behavioral Science, 9, 103–109. https://doi.org/10.1016/j.
different lifestyle interventions for enhancing wellbeing: The study design for a jcbs.2018.07.006
randomized controlled trial for persons with metabolic syndrome risk factors and Rickardsson, J., Gentili, C., Holmström, L., Zetterqvist, V., Andersson, E., Persson, J.,
psychological distress. BMC Public Health, 14(1). https://doi.org/10.1186/1471- Mats, L., Ljótsson, B., Rikard, & Wicksell, K. (2021). Internet-delivered acceptance
2458-14-310 and commitment therapy as microlearning for chronic pain: A randomized
Lee, E. B., An, W., Levin, M. E., & Twohig, M. P. (2015). An initial meta-analysis of controlled trial with 1-year follow-up. European Journal of Pain, 25(5), 1012–1030.
Acceptance and Commitment Therapy for treating substance use disorders. Drug and https://doi.org/10.1002/ejp.1723
Alcohol Dependence, 155, 1–7. https://doi.org/10.1016/j.drugalcdep.2015.08.004 Sairanen, E., Tolvanen, A., Karhunen, L., Kolehmainen, M., Järvelä-Reijonen, E.,
Levin, M. E., Haeger, J., & Cruz, R. A. (2019). Tailoring acceptance and commitment Lindroos, S., Peuhkuri, K., Korpela, R., Ermes, M., Mattila, E., & Lappalainen, R.
therapy skill coaching in the moment through smartphones: Results from a (2017). Psychological flexibility mediates change in intuitive eating regulation in
randomized controlled trial. Mindfulness, 10(4), 689–699. https://doi.org/10.1007/ acceptance and commitment therapy interventions. Public Health Nutrition, 20(9),
s12671-018-1004-2 1681–1691. https://doi.org/10.1017/S1368980017000441
16
Schmucker, C. M., Blümle, A., Schell, L. K., Schwarzer, G., Oeller, P., Cabrera, L., … Trompetter, H. R., Bohlmeijer, E. T., Veehof, M. M., & Schreurs, K. M. (2015). Internet-
consortium, O. (2017). Systematic review finds that study data not published in full based guided self-help intervention for chronic pain based on acceptance and
text articles have unclear impact on meta-analyses results in medical research. PLoS commitment therapy: A randomized controlled trial. Journal of Behavioral Medicine,
One, 12(4), Article e0176210. https://doi.org/10.1371/journal.pone.0176210 38(1), 66–80. https://doi.org/10.1007/s10865-014-9579-0
Scott, W., Chilcot, J., Guildford, B., Daly-Eichenhardt, A., & McCracken, L. M. (2018). U.S. Department of Health & Human Services. (2019). About chronic diseases | CDC. In
Feasibility randomized-controlled trial of online Acceptance and Commitment National center for chronic disease prevention and health promotion (NCCDPHP).
Therapy for patients with complex chronic pain in the United Kingdom. European https://www.cdc.gov/chronicdisease/about/index.htm.
Journal of Pain, 22(8), 1473–1484. https://doi.org/10.1002/ejp.1236 Van Daele, T., Karekla, M., Kassianos, A. P., Compare, A., Haddouk, L., Salgado, J., …
Simister, H. D., Tkachuk, G. A., Shay, B. L., Vincent, N., Pear, J. J., & Skrabek, R. Q. Van Assche, E. (2020). Recommendations for policy and practice of
(2018). Randomized controlled trial of online acceptance and commitment therapy telepsychotherapy and e-mental health in Europe and beyond. Journal of
for Fibromyalgia. The Journal of Pain, 19(7), 741–753. https://doi.org/10.1016/j. Psychotherapy Integration, 30(2), 160. https://doi.org/10.1037/int0000218
jpain.2018.02.004 Van Ryckeghem, D. (2020). Acceptance is not acceptance, but acceptance. Wiley Online
Sobczak, L. R., & West, L. M. (2013). Clinical considerations in using mindfulness-and Library, 25(1), 3–4. https://doi.org/10.1002/ejp.1672
acceptance-based approaches with diverse populations: Addressing challenges in Wang, J., Wang, Y., Wei, C., Yao, N., Yuan, A., Shan, Y., & Yuan, C. (2014). Smartphone
service delivery in. Cognitive and Behavioral Practice, 20(1), 13–22. https://doi.org/ interventions for long-term health management of chronic diseases: An integrative
10.1016/j.cbpra.2011.08.005 review. Liebertpub.Com, 20(6), 570–583. https://doi.org/10.1089/tmj.2013.0243
Sterne, J. A., Savović, J., Page, M. J., Elbers, R. G., Blencowe, N. S., Boutron, I., … Weineland, S., Arvidsson, D., Kakoulidis, T. P., & Dahl, J. (2012). Acceptance and
Eldridge, S. M. (2019). RoB 2: A revised tool for assessing risk of bias in randomised commitment therapy for bariatric surgery patients, a pilot RCT. Obesity Research &
trials. BMJ, 366. https://doi.org/10.1136/bmj.l4898 Clinical Practice, 6(1), e21–e30. https://doi.org/10.1016/j.orcp.2011.04.004
Thompson, E. M., Destree, L., Albertella, L., & Fontenelle, L. F. (2021). Internet-based Woidneck, M. R., Pratt, K. M., Gundy, J. M., Nelson, C. R., & Twohig, M. P. (2012).
acceptance and commitment therapy: A transdiagnostic systematic review and meta- Exploring cultural competence in acceptance and commitment therapy outcomes.
analysis for mental health outcomes. Behavior Therapy, 52(2), 492–507. https://doi. Professional Psychology: Research and Practice, 43(3), 227–233. https://doi.org/
org/10.1016/j.beth.2020.07.002 10.1037/a0026235
Thorsell, J., Finnes, A., Dahl, J., Lundgren, T., Gybrant, M., Gordh, T., & Buhrman, M. Wolgast, M. (2014). What does the Acceptance and Action Questionnaire (AAQ-II) really
(2011). A comparative study of 2 manual-based self-help interventions, acceptance measure? Behavior Therapy, 45(6), 831–839. https://doi.org/10.1016/j.
and commitment therapy and applied relaxation, for persons with chronic pain. The beth.2014.07.002
Clinical Journal of Pain, 27(8), 716–723. https://doi.org/10.1097/
AJP.0b013e318219a933
17

ACT Techsupported Chronicconditions Meta BRAT 2022

Uploaded by

Copyright:

Available Formats

ACT Techsupported Chronicconditions Meta BRAT 2022

Uploaded by

Document Information

Original Description:

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

ACT Techsupported Chronicconditions Meta BRAT 2022

Uploaded by

Copyright:

Available Formats

Behaviour Research and Therapy 148 (2022) 103995

Contents lists available at ScienceDirect

Behaviour Research and Therapy

Technology-supported Acceptance and Commitment Therapy for chronic

Fig. 1. Flow diagram.

Fig. 2. Risk of bias domains.

Overweight and Obesity

Table 3 shows findings of exploratory subgroup analyses. Technol 4. Discussion

Therapist Contact Method

Function In-person/phone 8 − 0.31 (-0.65, 0.03) 54.3% Q (1) = 3.42, p = 0.06

Function Active 7 − 0.42 (-0.86, 0.01) 65.0% Q (1) = 0.30, p = 0.58

You might also like

ACT Techsupported Chronicconditions Meta BRAT 2022

Uploaded by

Copyright:

Available Formats

ACT Techsupported Chronicconditions Meta BRAT 2022

Uploaded by

Document Information

Original Description:

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

ACT Techsupported Chronicconditions Meta BRAT 2022

Uploaded by

Copyright:

Available Formats

Behaviour Research and Therapy 148 (2022) 103995

Contents lists available at ScienceDirect

Behaviour Research and Therapy

Technology-supported Acceptance and Commitment Therapy for chronic

Fig. 1. Flow diagram.

Fig. 2. Risk of bias domains.

Overweight and Obesity

Table 3 shows findings of exploratory subgroup analyses. Technol­ 4. Discussion

Therapist Contact Method

Function In-person/phone 8 − 0.31 (-0.65, 0.03) 54.3% Q (1) = 3.42, p = 0.06

Function Active 7 − 0.42 (-0.86, 0.01) 65.0% Q (1) = 0.30, p = 0.58

You might also like

Table 3 shows findings of exploratory subgroup analyses. Technol 4. Discussion