CoChat
Sign Up
Cited by 26Mindfulness and Compassion InterventionsSchizophrenia research and treatmentMental Health and Patient Involvement

Group acceptance and commitment therapy for patients and caregivers in psychosis services: Feasibility of training and a preliminary randomized controlled evaluation

Open access
Jolley, Suzanne, Johns, Louise C., O’Donoghue, Emma, Oliver, Joseph, Khondoker, Mizanur, Byrne, Majella, Butler, Lucy, De Rosa, Carmine +5 more·2020·British Journal of Clinical Psychology
DOI 10.1111/bjc.12265
Open this paper in CoChat — free accountOpen PDF

Abstract

Objective: Psychological interventions reduce the impact of psychosis, but widescale implementation is problematic. We tested the feasibility of group acceptance and commitment therapy for Psychosis (G‐ACTp), delivered by frontline staff, and co‐facilitated by service‐user experts‐by‐experience (SU‐EbyE), for service‐users and informal caregivers (ISRCTN: 68540929). We estimated recruitment/retention rates and outcome variability for future evaluation. Methods: Staff and SU‐EbyE facilitators completed 1‐day workshops, then delivered closely supervised G‐ACTp, comprising four sessions (weeks 1–4) and two boosters (10 and 12 weeks). Participants recruited from adult community psychosis services were randomized to receive G‐ACTp immediately or after 12 weeks, completing outcome assessments at 0, 4, and 12 weeks. Service‐use/month was calculated for 1‐year pre‐randomization, weeks 0–12, and 5‐year uncontrolled follow‐up. Results: Of 41 facilitators trained (29 staff, 12 SU‐EbyE), 29 (71%; 17 staff, 12 SU‐EbyE) delivered 18 G‐ACTp courses. Participant refusal rates were low (9% of service‐users [10/112]; 5% of caregivers [4/79]); 60% of those invited to participate attended ≥1 G‐ACTp session (64% of service‐users [39/61]; 56% of caregivers [35/63]). Randomization of facilitators and participants proved problematic and participant follow‐up was incomplete (78% [66/85]; 82% of service‐users [36/44]; 73% of caregivers [30/41]). Effect sizes ranged from very small to large mostly favouring treatment. Service‐use reductions require cautious interpretation, as very few participants incurred costs. Conclusions: Implementation appears feasible for service‐users; for caregivers, retention needs improving. Outcome variability indicated n = 100–300/arm followed up (α = 0.05, 90% power). Methodological limitations’ mean replication is needed: identified sources of potential bias may be reduced in a cluster randomized design with sessional outcome completion. Practitioner points: Group acceptance and commitment therapy can be successfully adapted for people with psychosis and their caregivers. Implementation (training and delivery) is possible in routine community mental health care settings. Clinical and economic outcomes are promising, but replication is needed. Recommendations are made for future studies.

Verified 2026-10-10 · 1 reference not checked yet

AI snapshot

Written from the paper’s full text.

TL;DR

This feasibility trial tested group Acceptance and Commitment Therapy for psychosis (G-ACTp), delivered by trained frontline staff and service-user experts-by-experience, for service-users and informal caregivers in community psychosis services. Training, delivery, and uptake were largely feasible, but recruitment/retention were only partially met, especially for caregivers, and preliminary randomised outcomes mostly favoured immediate treatment with small-to-large effect sizes.

Key findings

  • Of 41 facilitators trained (29 staff, 12 SU-EbyE), 29 (71%; 17 staff, 12 SU-EbyE) delivered 18 G-ACTp courses; ACT knowledge increased from pre-training mean 5.9 (SD 3.5) to post-training mean 7.4 (SD 3.2), t=-3.6, df=31, p=0.003.
  • Refusal rates were low (9% of service-users [10/112]; 5% of caregivers [4/79]); uptake of ≥1 G-ACTp session was 60% overall (74/124; 64% service-users [39/61]; 56% caregivers [35/63]); retention was 78% of randomised participants (66/85; 82% service-users [36/44]; 73% caregivers [30/41]).
  • Effect sizes ranged from very small to large, mostly favouring ACTnow; service-users improved on 12/13 outcomes at 12 weeks (average ES=0.3) and 11/12 at 4 weeks (average ES=0.4), while caregivers improved on 8/9 outcomes at 12 weeks (average ES=0.3) and 7/9 at 4 weeks (average ES=0.2).
  • The mixed linear model for wellbeing/distress gave an allocation coefficient of 9.6 (p=0.07, 95% CI -0.7 to 20.0); sample size estimation suggested n=100–300/arm for a future study.

Methods

  • Pragmatic randomised controlled feasibility trial in adult community psychosis services; participants were randomised 1:1 to immediate G-ACTp (ACTnow) or after 12 weeks (ACTlater), stratified by service-user/caregiver status, with outcomes at 0, 4, and 12 weeks and uncontrolled follow-up at 6 months and 5 years.
  • G-ACTp comprised four weekly 2-hour sessions plus boosters at 10 and 12 weeks (6 sessions, 12 hours over 12 weeks), delivered by trained frontline staff and SU-EbyE co-facilitators under supervision; usual care continued irrespective of allocation.
  • Missing data were handled under a missing-at-random assumption, with predictors of missingness identified by random intercept logistic regression and included in a linear mixed model for wellbeing/distress; effect sizes were calculated as group mean differences divided by pooled SD assuming a 0.5 correlation, and training data were analysed with paired t-tests.

Limitations

  • Stated in the paper: assessments were not blind at baseline or follow-up; external adherence ratings were not obtained because audio-recording consent was not given; medication status and other interventions were not examined; loss through non-adherence and follow-up elevated risk of bias and data were not missing at random; the study was small, had multiple outcomes, and was not designed to test efficacy; economic outcomes were based on very few individuals and were highly variable; recruitment was from a single service; there was no active control; and dissemination evaluation was observational.
  • Evident from the text: planned randomisation of facilitators proved infeasible, participant randomisation was problematic (11 randomisations unrealised, 8 participants invited unrandomised, 7 randomised in error post-intervention, and 8 ACTlater participants attended ACTnow and were excluded), and caregiver recruitment required a 6-month extension with poorer retention than service-users.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1111/bjc.12265

Cites

70 of 71 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

0 of 27 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    Attendance ranged from 1-6 sessions for both service-users (Mean=3.9,SD=1.9) and caregivers (Mean=4.4,SD=2.0).

    “Attendance ranged from 1-6 sessions for both service-users (Mean=3.9,SD=1.9) and caregivers (Mean=4.4,SD=2.0).”
    Quote in the paperSubstantial claim
  • Findingown data

    Estimated treatment effect for the proposed primary outcomes of wellbeing/distress using a linear mixed model was Coefficient=9.6, p=0.07, 95% CI -0.7 to 20.0.

    “Table 5 shows estimated treatment effects for the proposed primary outcomes of wellbeing/distress using a linear mixed model (Coefficient=9.6,p=0.07, 95% confidence interval -0.7, 20.0).”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: Missing data are missing at random (MAR).
  • Findingown data

    Pooled variances for wellbeing/distress (98.8 and 58.0 respectively) suggested n=100-300/arm to detect obtained treatment effects with 90% power and α=0.05.

    “Pooled variances for wellbeing/distress (98.8 and 58.0 respectively) suggested n=100-300/arm to detect obtained treatment effects with 90% power and α=0.05.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Knowledge of ACT, measured by the AKQ, increased significantly from pre-training mean 5.9 (SD 3.5) to post-training mean 7.4 (SD 3.2), t=-3.6, df=31, p=0.003.

    “Knowledge of ACT, measured by the AKQ, increased significantly (pre-training mean 5.9, SD 3.5; post-training mean 7.4, SD 3.2, t=-3.6,df=31, p=0.003), scores were similar between facilitator groups.”
    Quote in the paperSubstantial claim
  • Findingown data

    Half of all randomised participants completed uncontrolled 6-month follow-up.

    “Half of all randomised participants completed uncontrolled 6-month follow-up.”
    Quote in the paperSubstantial claim
  • May combine two claims. 50
  • Classified as probably not a substantial claim. 16

  • May combine two claims.

    Findingown data

    Refusals to participate were low (9% of service-users [10/112]; 5% of caregivers [4/79]), but 13 (4 service-users; 9 caregivers) did not complete baseline having consented.

    “Refusals to participate were low (9% of service-users [10/112]; 5% of caregivers [4/79]), but 13 (4 service-users; 9 caregivers) did not complete baseline having consented.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Caregivers improved on 8 of 9 outcomes measured at 12-weeks, average ES=0.3; and 7 of 9 outcomes at 4-weeks, average ES=0.2.

    “Caregivers improved on 8 of 9 outcomes measured at 12-weeks, average ES=0.3; and 7 of 9 outcomes at 4-weeks, average ES=0.2.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Service users improved on 12 of 13 outcomes measured at 12-weeks, average ES=0.3; and on 11 of 12 outcomes at 4-weeks, average ES=0.4.

    “Service users improved on 12 of 13 outcomes measured at 12-weeks, average ES=0.3; and on 11 of 12 outcomes at 4-weeks, average ES=0.4.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Satisfaction levels were high; qualitative feedback highlighted positive impact on recovery for service-users and improved resources to manage the caregiving relationship for caregivers.

    “Satisfaction levels were high; qualitative feedback highlighted positive impact on recovery for service-users and improved resources to manage the caregiving relationship for caregivers (Table 6).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The earliest of these deaths post-dated involvement by 6-months, and participation was not implicated in any incident.

    “The earliest of these post-dated involvement by 6-months, and participation was not implicated in any incident, by any party, including routine internal trust incident investigations.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Costs were incurred by very few individuals and outcomes were therefore highly variable and should be considered cautiously.

    “Costs were incurred by very few individuals and outcomes were therefore highly variable and should be considered cautiously.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Service-use data from medical records are complete and accurate.

  • May combine two claims.

    Findingown data

    All attendees were offered the opportunity to facilitate; 29 did so, all were judged competent to deliver/co-facilitate following training.

    “All attendees were offered the opportunity to facilitate; 29 did so, all were judged competent to deliver/co-facilitate following training.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Effect sizes across outcomes ranged from very small to large, mostly favouring ACTnow.

    “ESs across outcomes ranged from very small to large, mostly favouring ACTnow, signalling positive effects of intervention.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    During five-year follow-up, four participants died, all in the ACTnow group, at 12-, 14-, 29-, and 32-months post-randomisation.

    “During five-year follow-up, four participants sadly died, all in the ACTnow group, at 12-, 14-, 29-, and 32-months post-randomisation.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Follow-up data was available for 78% of those successfully randomised ([66/85]; 82% of service-users [36/44]; 73% of caregivers [30/41]).

    “Follow-up data was available for 78% of those successfully randomised ([66/85]; 82% of service-users [36/44]; 73% of caregivers [30/41]).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Over five years, uncontrolled costs across participants show small reductions and reduced variability compared to pre-baseline (£743/month, SD=£1930 pre-treatment; £217/person/month SD=£420 post-treatment), suggesting limited, but sustained, savings.

    “Over five years, uncontrolled costs across participants show small reductions and reduced variability compared to pre-baseline (£743/month, SD=£1930 pre-treatment; £217/person/month SD=£420 post-treatment), suggesting limited, but sustained, savings.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Service-use data from medical records are complete and accurate.

  • May combine two claims.

    Findingown data

    Staff training targets were met, but SU-EbyE targets (n=16) were not (achieved: n=12).

    “Staff training targets were met, but SU-EbyE targets (n=16) were not (achieved: n=12).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Service-user recruitment reached target (by October 2013); caregiver recruitment required a 6-month extension (to March 2014).

    “Service-user recruitment reached target (by October 2013); caregiver recruitment required a 6-month extension (to March 2014).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Significant predictors of missing data were Allocation, Time, and baseline distress, such that the likelihood of missing data was increased for waitlist participants, post-baseline assessments, and lower distress.

    “Significant predictors were Allocation (coeff.=-2.7,z=-2.7,p=0.006, 95% CI: -4.6 to -0.8);Time (coeff.=1.2, z=4.3; p<0.001; 95% CI: 0.6 to 1.7); and baseline distress, such that the likelihood of missing data was increased for waitlist participants, post-baseline assessments, and lower distress, but not by any other variable (|z| scores ≤ 1.5, p values all > 0.1).”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: Missing data are missing at random (MAR).

  • May combine two claims.

    Findingown data

    Wellbeing and distress outcomes at 6-month follow-up were comparable to post-treatment ACTnow scores, consistent with maintenance of change.

    “Wellbeing and distress outcomes were comparable to post-treatment ACTnow scores, consistent with maintenance of change (wellbeing mean: service-users=45.1 (SD=10.8,n=23), caregivers=50.7 (SD=8.2,n=21); distress mean: service-users=13.0(SD=7.4,n=23), caregivers=10.0 (SD=6.7,n=21)).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Accounting for treatment costs of £590.83/person, costs were lower for ACTnow during compared to before treatment, and compared to ACTlater, consistent with potential for cost-effective delivery.

    “Accounting for treatment costs of £590.83/person 2 costs were lower for ACTnow during compared to before treatment, and compared to ACTlater, consistent with potential for costeffective delivery (Table 4).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Service-use data from medical records are complete and accurate.

  • May combine two claims.

    Findingown data

    Uptake for ACTlater participants post-waitlist was 62% ([21/34]; 70% of service-users [14/20], mean attended=3.85,SD=1.92; 50% of caregivers [7/14], mean attended=4.43,SD=1.74).

    “Uptake for ACTlater participants post-waitlist was 62% ([21/34]; 70% of service-users [14/20], mean attended=3.85,SD=1.92; 50% of caregivers [7/14], mean attended=4.43,SD=1.74).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In practice, one additional ACTlater caregiver was followed up, but assessment data was corrupted, and one ACTnow service-user was excluded, thus 67/84 available participants completed follow-up (80%).

    “In practice, one additional ACTlater caregiver was followed up, but assessment data was corrupted, and one ACTnow service-user was excluded, thus 67/84 available participants completed follow-up (80%).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The exception to the overall positive trend favouring ACTnow for service-users was PSYRATS delusions; score increases were due to more participants reporting beliefs at follow-up, and allocation groups differed at baseline.

    “The exception to the overall positive trend favouring ACTnow for service-users was PSYRATS delusions; score increases were due to more participants reporting beliefs at follow-up, and allocation groups differed at baseline.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    For caregivers, the ES for mindfulness was 0 at 4-weeks, and having a full state of health on the EQ-5D appeared to be less likely following intervention, which may have been impacted by the high levels of physical health problems and disability in our caregivers.

    “For caregivers, the ES for mindfulness was 0 at 4-weeks, and having a full state of health on the EQ-5D appeared to be less likely following intervention, which may have been impacted by the high levels of physical health problems and disability in our caregivers.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingcitation

    Change in process measures was consistent with outcome change, and feedback highlights the accessibility of ACT constructs, as in previous qualitative studies (Bacon, Farhall, Fossey, 2014).

    “Change in process measures was consistent with outcome change, and feedback highlights the accessibility of ACT constructs, as in previous qualitative studies (Bacon, Farhall, Fossey, 2014).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Bacon, Farhall, Fossey, 2014

      Not checked in the free run.

  • May combine two claims.

    Findingnone given

    Implementation of the new group ACT intervention was feasible, with adjustment to allow for slower recruitment and poorer follow-up for caregivers.

    “Implementation of our new group ACT intervention was feasible, with adjustment to allow for slower recruitment and poorer follow-up for caregivers.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    Effect sizes were slightly larger in service-users compared to caregivers, and either comparable or slightly reduced from 4-weeks to 12-weeks, while caregiver outcomes improved from 4-weeks to 12-weeks.

    “Effect sizes were slightly larger in service-users compared to caregivers, and either comparable or slightly reduced from 4-weeks to 12-weeks, in line with previous research (Bach et al., 2012), while caregiver outcomes improved from 4-weeks to 12-weeks.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Bach et al., 2012

      Not checked in the free run.

  • May combine two claims.

    Findingnone given

    Satisfaction and feedback were good.

    “Satisfaction and feedback were good and preliminary treatment effects were consistent with positive impact of intervention.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Findingown data

    In the mixed linear model, the allocation x time interaction term was -7.2 (SE 6.2, p=0.2, 95% CI -19.3 to 4.8).

    “Allocation x Time interaction term 1 | -7.2 | 6.2 | -1.2 | 0.2 | -19.3, 4.8”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Brief courses of group ACT for psychosis (G-ACTp), comprising four sessions each of two hours, have shown promise in service evaluations and uncontrolled studies, improving pre-post outcomes for people with psychosis.

    “Importantly for implementation, brief courses of group ACT for psychosis (G-ACTp), comprising four sessions each of two hours, have shown promise in service evaluations and uncontrolled studies, improving pre-post outcomes for people with psychosis, suggesting potential for cost-effective, widescale provision (Morris et al., 2013;Johns et al., 2015;Butler et al., 2016;O'Donoghue et al., 2018).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Morris et al., 2013

      Not checked in the free run.

    • cites Johns et al., 2015 source

      Not checked in the free run.

    • cites Butler et al., 2016

      Not checked in the free run.

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    Acceptance and Commitment Therapy (ACT) aims to improve wellbeing, functioning, and quality of life by increasing psychological flexibility and reducing unhelpful self-regulation.

    “Acceptance and Commitment Therapy (ACT) is a process-oriented, contextual cognitive behavioural approach, aiming to improve wellbeing, functioning, and quality of life by increasing psychological flexibility and reducing unhelpful self-regulation (e.g.worry, rumination, suppression, avoidance; Hayes, Luoma, Bond, Masuda, & Lillis, 2006;Hayes, Strosahl, & Wilson, 2012).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Hayes, Luoma, Bond, Masuda, & Lillis, 2006

      Not checked in the free run.

    • cites Hayes, Strosahl, & Wilson, 2012

      Not checked in the free run.

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    Interventions promoting psychological flexibility are potentially helpful for many disorders including psychosis, improving symptoms, functioning, and service use, with indications of change through targeted mechanisms, and positive effects at follow-up, although findings are mixed and more rigorously controlled studies are needed.

    “Interventions promoting psychological flexibility are potentially helpful for many disorders (A-Tjak, Davis, Morina, Powers, Smits, & Emmelkamp, 2015) including psychosis, improving symptoms, functioning, and service use, with indications of change through targeted mechanisms, and positive effects at follow-up, although findings are mixed and more rigorously controlled studies are needed (Ost, 201”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites A-Tjak, Davis, Morina, Powers, Smits, & Emmelkamp, 2015

      Not checked in the free run.

    • cites Ost, 2014

      Not checked in the free run.

    • cites Khoury, Lecomte, Gaudiano, & Paquin, 2013

      Not checked in the free run.

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    Randomisation was carried out by a registered UK Clinical Trials Unit, employing blocks of randomly varying size, and was stratified by service-user or caregiver status.

    “Randomisation was carried out by a registered UK Clinical Trials Unit, employing blocks of randomly varying size, and was stratified by service-user or caregiver status.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductcitation

    Leads monitored adherence to protocol, rating ACT-consistent therapist behaviours, and ensured ACT-consistency off-manual.

    “Leads also monitored adherence to protocol, rating ACT-consistent therapist behaviours (Morris, 2013), and ensured ACT-consistency off-manual (e.g.responses to participant comments/questions).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Morris, 2013

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    Feasibility criteria were mostly met for training and delivery, and met for intervention uptake; recruitment and retention criteria were met for service-users but not for caregivers.

    “Feasibility criteria were mostly met for training and delivery, and met for intervention uptake; recruitment and retention criteria were met for service-users but not for caregivers.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationcitation

    Across outcomes, between-group ESs ranged from very small to large, but average ESs of 0.3-0.4 for service users are commensurate with meta-analyses of longer, individual cognitive behavioural therapies (e.g. NICE, 2014, ES=0.37).

    “Across outcomes, between-group ESs ranged from very small to large, but average ESs of 0.3-0.4 for service users are commensurate with meta-analyses of longer, individual cognitive behavioural therapies (e.g.NICE, 2014, ES=0.37).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites NICE, 2014

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    Randomisation initially occurred post-baseline, but difficulty contacting participants regarding allocation delayed early groups, disrupting other participants, room-bookings and facilitator schedules.

    “Randomisation initially occurred post-baseline, but difficulty contacting participants regarding allocation delayed early groups, disrupting other participants, room-bookings and facilitator schedules.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    Randomly occurring differences between allocation groups were found post-hoc only for PSYRATS delusions and ED attendances; the groups were otherwise comparable, although these analyses were not powered.

    “Randomly occurring differences between allocation groups were found post-hoc only for PSYRATS delusions and ED attendances, the groups were otherwise comparable, although these analyses were not powered.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    Adherence to randomised allocation was problematic; eight ACTlater participants attended ACTnow, so were excluded.

    “Adherence to randomised allocation was also problematic.Due to participant travel plans and other commitments, eight ACTlater participants (three service-users, five caregivers) attended ACTnow, so were excluded.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    Post-training competence to deliver/co-facilitate was judged by lead therapists, based on observation during groups and supervision, using sessional ACT adherence ratings.

    “Post-training competence to deliver/co-facilitate was judged by lead therapists, based on observation during groups and supervision, using sessional ACT adherence ratings (Morris, 2013).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Morris, 2013

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductown data

    Variables predicting missing data were included in the linear mixed model employed to estimate treatment effects for wellbeing/distress.

    “Variables predicting missing data were included in the linear mixed model employed to estimate treatment effects for wellbeing/distress, with outcomes at 4-weeks and 12-weeks as repeated measures, covarying for baseline wellbeing and distress, Measure (wellbeing/distress), Time (4 weeks, coded 0 and 12 weeks, coded 1), Status (serviceuser/caregiver), and an Intervention x Time interaction term, w”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: Missing data are missing at random (MAR).

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductown data

    Service use was calculated from medical records, blind to allocation.

    “Service use was calculated from medical records, blind to allocation, by counting emergency”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Service-use data from medical records are complete and accurate.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    Eight participants were invited unrandomised by the research team to ensure viable group numbers; seven of these were randomised in error post-intervention; all eight were excluded.

    “Eight participants were invited unrandomised by the research team to ensure viable group numbers; seven of these were randomised in error post-intervention: all eight were excluded.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Eleven randomisations remained unrealised through baseline non-attendance following consent.

    “However, eleven randomisations remained unrealised through baseline non-attendance following consent.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    Delivery was adherent according to lead therapist observation and rating, and was sufficient to deliver the intended number of groups.

    “Delivery was adherent according to lead therapist observation and rating, and was sufficient to deliver the intended number of groups.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    Indications were of a positive effect of intervention, with high levels of satisfaction and positive feedback, but with relatively small effects on the proposed primary outcomes that suggest sample sizes between n=100-300/arm for a future study.

    “Indications were of a positive effect of intervention, with high levels of satisfaction and positive feedback, but with relatively small effects on the proposed primary outcomes that suggest sample sizes between n=100-300/arm for a future study.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductnone given

    Recruitment took place in a single service, using service parameters for inclusion.

    “Recruitment took place in a single service, using service parameters for inclusion, and applicability in other contexts has not been tested.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationreasoning

    Sustained delivery in the host service and wider uptake are positive indications.

    “While sustained delivery in the host service and wider uptake are positive indications, formal evaluation of implementation, including uptake, competence and effectiveness of delivery, will be required, and selection criteria (e.g.permanent staff) should be applied.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    Our dissemination evaluation, beyond initial staff and SU-EbyE training, was observational, and participation voluntary, without selection criteria.

    “Our dissemination evaluation, beyond initial staff and SU-EbyE training, was observational, and participation voluntary, without selection criteria.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductreasoning

    The study was not powered to test baseline differences, which may therefore have gone undetected.

    “Only two randomly occurring baseline differences between allocation groups reached significance, but the study was not powered to test baseline differences, which may therefore have gone undetected.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The study had multiple outcomes.

    “The study was small, with multiple outcomes, and not designed to test efficacy.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The emerging evidence base for caregivers of people with a range of long-term conditions is similarly extended by the current findings, indicating that group ACT approaches have potential to improve wellbeing for caregivers of people with psychosis, and therefore comprise an intervention that could be delivered in routine services to support caregivers, in line with UK NICE recommendations.

    “The emerging evidence base for caregivers of people with a range of long-term conditions (e.g.Losada et al., 2015) is similarly extended by the current findings, indicating that group ACT approaches have potential to improve wellbeing for caregivers of people with psychosis, and therefore comprise an intervention that could be delivered in routine services to support caregivers, in line with UK N”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Losada et al., 2015

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    Uncontrolled follow-up outcomes suggest overall small reductions and reduced variability in longer-term service-use.

    “Uncontrolled follow-up outcomes suggest overall small reductions and reduced variability in longer-term service-use: future research should include a longer controlled evaluation of whether treatment costs are offset by reductions in follow-up service-use.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Service-use data from medical records are complete and accurate.

  • May combine two claims.

    Interpretationown data

    Disseminability is supported by sustained delivery, and wider uptake following training.

    “Disseminability is supported by sustained delivery, and wider uptake following training.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductnone given

    We did not examine effects of medication status and did not record changes, or receipt of other interventions.

    “We did not examine effects of medication status and did not record changes, or receipt of other interventions, and relied, without post hoc testing, upon randomisation to adequately control for any unintended differences between groups in these variables.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Randomisation adequately controls for unintended differences between groups in medication status and receipt of other interventions.

  • May combine two claims.

    Study design and conductown data

    Economic outcomes reflected service use by a small number of individuals, and were highly variable, requiring cautious interpretation.

    “Economic outcomes reflected service use by a small number of individuals, and were highly variable, requiring cautious interpretation.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductcitation

    Lead therapists were expert in the intervention, trained to competence in previous trials, followed a strict protocol, oversaw co-facilitators, and attended peer supervision sessions.

    “Lead therapists were, however, expert in the intervention, trained to competence in previous trials (Johns et al., 2015) and followed a strict protocol, overseeing the work of their co-facilitators, and attending peer supervision sessions.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Johns et al., 2015 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    Recruitment and retention of caregivers was problematic, primarily because of engagement difficulties, caregiving responsibilities, and other commitments, and our findings suggest that more time should be allowed for recruitment, and greater attrition expected, in caregiver compared to service-user participants.

    “However, recruitment and retention of caregivers was problematic, primarily because of engagement difficulties, caregiving responsibilities, and other commitments, and our findings suggest that more time should be allowed for recruitment, and greater attrition expected, in caregiver compared to service-user participants.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    Economic outcomes require cautious interpretation, but together with dissemination outcomes, are consistent with potential for cost-effective implementation.

    “Economic outcomes require cautious interpretation, but together with dissemination outcomes, are consistent with potential for cost-effective implementation.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Service-use data from medical records are complete and accurate.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationnone given

    Cluster randomisation and sessional outcome completion may reduce the risk of bias through difficulty adhering to allocation and loss to follow-up.

    “Cluster randomisation and sessional outcome completion may reduce the risk of bias through difficulty adhering to allocation and loss to follow-up: the high risk of bias in the current study limits interpretation of the size of effects, which should in any case be treated cautiously in a pilot, but should not impact variability of outcomes or other feasibility estimations.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationreasoning

    Because of the wide applicability of ACT approaches, training has greater potential to result in wider-scale implementation compared to condition-specific interventions with different competences according to presenting problems.

    “Because of the wide applicability of ACT approaches, training has greater potential to result in wider-scale implementation compared to condition-specific interventions with different competences according to presenting problems.”
    Quote in the paperNot a substantial claim
57 of 57

  • The authors state this as an assumption.

    Missing data are missing at random (MAR).

    “Loss to follow-up and missing data were handled under the missing at random (MAR) assumption, whereby predictors of missingness were identified using a series of random intercept logistic regression analyses (White et al., 2011;Stata version 15, Statacorp, 2017) and adjusted for in the analysis model.”

  • The paper relies on this without support.

    Outcomes of G-ACTp have not been evaluated in a randomised controlled trial and the feasibility of frontline, recovery-focused delivery is unknown.

    “However, outcomes have yet to be evaluated in a randomised controlled trial (RCT) and the feasibility of frontline, recovery-focused delivery is unknown.”

  • The paper relies on this without support.

    Service-use data from medical records are complete and accurate.

    “Service use was calculated from medical records, blind to allocation, by counting emergency”

  • The paper relies on this without support.

    Randomisation adequately controls for unintended differences between groups in medication status and receipt of other interventions.

    “We did not examine effects of medication status and did not record changes, or receipt of other interventions, and relied, without post hoc testing, upon randomisation to adequately control for any unintended differences between groups in these variables.”

  • The paper relies on this without support.

    Data were not missing at random.

    “The loss of participants, through non-adherence to allocation and loss to followup elevates the risk of bias in the study, and, while predictors of missing data were controlled in the analyses, data were not missing at random: replication will be required.”
5 of 5

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Consistentz = -2.7reported p = .006 · recomputed p = .007

    “,z=-2.7,p = .006”

1 of 1

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

  • ACTnow antipsychotic percentage does not match its denominator

    Minor

    The table states that medication data were collected for 17 ACTnow participants. Eleven of 17 is about 65%, not the reported 76%.

    Reported 76 · recomputed 64.705882

  • ACTlater antipsychotic percentage does not match its denominator

    Minor

    The table states that medication data were collected for 17 ACTlater participants. Ten of 17 is about 59%, not the reported 76%.

    Reported 76 · recomputed 58.823529

  • ACTnow caregiver ethnicity percentages are reversed

    Minor

    The table reports 10 BME and 11 non-BME caregivers among the 21 who stated their ethnicity. The BME percentage is 48%, not 52%; 52% corresponds to the 11 non-BME caregivers.

    Reported 52 · recomputed 47.619048

3 of 3

Cites

Flagged first, then by citations

5 of the 76 references couldn’t be matched to a known paper.

Research works printed but not linked: 6

  1. Connell J., 2007, CORE‐10 User Manual, Version 1.1
  2. Curtis L., 2016, Unit costs of health and social care 2016
  3. Curtis L., 2018, Unit costs of health and social care 2018
  4. National Institute for Health & Care Excellence, 2014, Psychosis and Schizophrenia in adults
  5. StataCorp, 2017, Stata Statistical Software: Release 15
  6. The Schizophrenia Commission, 2012, The abandoned illness: A report from the Schizophrenia Commission

Cited by

Sorted by citations

Related

Authors
Jolley, S., Johns, L. C., O’Donoghue, E., Oliver, J., Khondoker, M., Byrne, M., Butler, L., De Rosa, C., Leal, D., McGovern, J., Rasiukeviciute, B., Sim, F., & Morris, E.
Affiliations
  • King's College London (GB)
  • University College London (GB)
  • University of East Anglia (GB)
  • South London and Maudsley NHS Foundation Trust London UK
  • South London and Maudsley NHS Foundation Trust (GB)
  • La Trobe University (AU)
Corresponding author
Jolley, Suzanne
ORCID
Venue
British Journal of Clinical Psychology
Publisher
Wiley
Published
2020-09-17
Volume
59
Issue
4
Pages
524-551
ISSN
  • 0144-6657 (ISSN-L)
  • 2044-8260
DOI
10.1111/bjc.12265
OpenAlex
W3087326605
Semantic Scholar
193037646fafe748123afa966a66f9141ddba499
Cited by
26
References
76
Citations by year
2026: 5 · 2025: 4 · 2024: 4 · 2023: 6 · 2022: 4 · 2021: 3
FWCI
1.30
Type
journal article
Language
en
Open access
Licence
Locations
  • British Journal of Clinical Psychology (publishedVersion)
  • PubMed (publishedVersion)
  • Research Portal (King's College London) (submittedVersion, other-oa)
  • UEA Digital Repository (University of East Anglia) (submittedVersion)
  • UCL Discovery (University College London) (submittedVersion)
  • Figshare (submittedVersion)
  • La Trobe University
Funders
Maudsley Charity (764)
Topics
  • Social Sciences › Psychology › Clinical Psychology › Mindfulness and Compassion Interventions
  • Health Sciences › Medicine › Psychiatry and Mental health › Schizophrenia research and treatment
  • Health Sciences › Health Professions › General Health Professions › Mental Health and Patient Involvement
Keywords
psychosis, informal caregivers, community mental health services, feasibility, implementation, employee training, co-facilitation, participant recruitment and retention, treatment effect size, service utilization, outcome variability, cluster randomized designs
MeSH terms
Adult, Feasibility Studies, Female, Humans, Male, Middle Aged, Psychotherapy, Group, Psychotic Disorders, Treatment Outcome, Caregivers, Acceptance and Commitment Therapy

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-10.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-10.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review