Acceptance and Commitment Therapy for the Inpatient Treatment of Psychosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Acceptance and Commitment Therapy for the Inpatient Treatment of Psychosis (NCT01981356), a randomized early/phase 1 study. The primary outcome is Brief Psychiatric Rating Scale (Overall & Gorham, 1962).

2. Clinical Question

The study was intended to explore whether ACT could be delivered as an inpatient adjunct and whether it appeared feasible, acceptable, safe, and potentially useful. The posted primary results describe symptom-score changes in each arm; they do not establish a treatment effect. The outcome groups are small, and the record does not report an inferential estimate comparing the groups. Results should therefore be read as descriptive observations from a pilot rather than as a definitive estimate of comparative effectiveness.

3. Trial Design

This was a randomized, parallel-group trial with no masking reported. The registry record lists the study as early phase 1 and completed, with enrollment of 18. The study start was 2014-09 and primary completion was 2015-04. It was sponsored by the VA Office of Research and Development.

The two assigned conditions were Acceptance and Commitment Therapy (ACT) and Treatment as Usual (TAU). Participants assigned to ACT were offered 4 ACT sessions. TAU included psychopharmacology, case management, and psychotherapy. The TAU description also specifies brief contacts with ACT facilitators every other day for support and questions, while avoiding discussion or recommendation of therapy techniques. These contacts are part of the comparator as described, so the contrast is ACT sessions versus TAU with facilitator contact, not ACT versus no additional contact.

The registry record describes the study population as Veterans with psychosis receiving inpatient care at a single VHA site. The primary outcome was assessed over the duration of the hospital stay. The reported mean stay was 24.0 days, with SD 15.8 days; individuals’ follow-up duration therefore related to their hospital stay rather than to a single fixed follow-up duration reported for everyone.

Although enrollment was 18, the primary outcome results list 5 participants in ACT and 4 in TAU. The record does not explain the difference between enrollment and the numbers contributing to these reported outcome summaries. Random assignment supports a planned comparison, but the posted outcome summaries alone do not show whether all randomized participants were included in an analysis or how any unavailable measurements were handled.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Although 18 participants were randomized to treatment condition, the consent of two participants was invalid. Thus, their data could not be utilized and is not reported other than for identification of participant flow.

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Although 18 participants were randomized to treatment condition, the consent of two participants was invalid. Thus, their data could not be utilized and is not reported other than for identification of participant flow.

6. Primary Endpoint

The registry lists the primary outcome as Brief Psychiatric Rating Scale (Overall & Gorham, 1962), measured over Participants were followed for the duration of hospital stay (Mean = 24.0 days, SD = 15.8)..

The primary endpoint was change on the Brief Psychiatric Rating Scale, which assesses broad symptom domains including affect disturbance, positive and negative symptoms, resistance or hostility, and activation, as well as specific symptoms such as delusions. Items were averaged to produce a total scale score.

Results are expressed as percentage of total possible change. The record defines this as follow-up score minus baseline score, divided by the total points in the scale. The possible range is described as -100% change to +100% change. A -100% change corresponds to a decrease from the highest possible scale value at baseline to the lowest at follow-up; +100% corresponds to the reverse. Under this scoring convention, decreases in percentage change are considered better because they represent reduced symptoms.

A broad symptom measure can capture change across several clinically relevant domains rather than focusing on only one symptom. In this trial, the reported mean change was -8.0 for ACT and -11.6 for TAU. These are arm-specific descriptive summaries; the registry record does not provide a formal between-arm estimate or confidence interval. The measure was collected over the hospital stay, whose reported mean duration was 24.0 days (SD 15.8). Variation in stay duration is relevant when interpreting change because the follow-up interval was tied to hospital stay.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

For a continuous outcome, an arm mean describes the average observed score among participants included in that arm’s summary. Here the outcome is percentage change in Brief Psychiatric Rating Scale score, and the registry record reports means, standard deviations, and participant counts. The mean is a compact description of the group, but it does not show the full distribution of individual scores. This matters especially when group sizes are small: a few observations can influence the mean, and the available summaries provide limited information about the shape or range of the data.

The standard deviation (SD) describes how dispersed individual observations are around their group mean. It is not the same as a confidence interval and does not directly express the uncertainty of the mean or of a difference between arms. ACT has SD 13.8 and TAU has SD 5.8. The reported group sizes are 5 and 4, respectively. The SDs therefore help describe observed variability, but they do not establish whether the groups differ beyond chance variation. Confidence intervals for the means or a contrast are not reported in the registry record.

A treatment comparison in a randomized trial is usually expressed as a contrast between assigned groups, accompanied by uncertainty. Such an estimate would make clear both the direction and magnitude of the observed difference under a specified analysis. The registry record gives arm-specific means but no formal between-arm estimate, test statistic, or p-value. The posted summaries should not be turned into a claim of statistical significance or a precise treatment effect. A simple re-calculation from summary statistics, if shown elsewhere, would be an unadjusted calculation and would not replace a sponsor’s own analysis; no sponsor analysis is reported in the registry record.

The endpoint’s sign convention is essential. The scale is expressed as percentage of total possible change, calculated from follow-up and baseline scores relative to the scale’s total points. Negative changes represent decreases from baseline, and the registry record defines decreases as better outcomes because they indicate reduced symptoms. Interpretation should use that definition rather than assume that positive change is preferable. The reported arm means are both negative, but their descriptive values alone do not show whether the difference between assigned conditions is statistically reliable or attributable to treatment.

The outcome was assessed for the duration of hospital stay, not at a common calendar follow-up time specified for every participant. The reported mean stay was 24.0 days and SD was 15.8. In general, when observation periods vary, the amount of time over which change can occur may also vary. That is a consideration in interpreting the score summaries, not evidence that duration caused any particular result. The registry record does not describe an analysis accounting for differences in follow-up duration.

General statistical explanation: confidence intervals give a range of values compatible with the data under the assumptions of the analysis. They help communicate precision, whereas a p-value addresses the compatibility of data with a specified null model. Neither can be recovered from the posted arm means without making additional assumptions and calculations, and neither is reported for this primary comparison. With very small groups, uncertainty is particularly important; a numerical mean should not be mistaken for a precise population effect.

General statistical explanation: over-dispersion means that observed variability is greater than a model expects. It can make standard errors and significance tests unreliable if the model assumes less variation than is present. The registry record reports SDs but does not report an analysis model or an over-dispersion assessment. There is therefore no basis for saying that an over-dispersion adjustment was used or needed in this trial.

The record describes randomized allocation, a parallel model, and no masking. Randomization is intended to support a fair comparison by assigning participants to conditions, while lack of masking can matter when outcomes or care involve subjective judgments. The record does not state whether outcome assessors were masked, so that detail is not reported in the registry record. Nor does it report the statistical model, covariate adjustment, handling of repeated measurements, or a prespecified analysis population. These details should not be inferred from the fact of randomization or from the reported summaries.

The registry record includes a planning-illustration approach based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. Such calculations are illustrations for future planning, not observed power and not evidence that the trial had a particular chance of detecting an effect. The record’s description says that these are planning calculations, and they should not be mistaken for trial results or a design justification. A sample-size justification for this trial is not reported in the registry record.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Acceptance and Commitment Therapy (ACT)", "Treatment as Usual (TAU)")
n   <- c(5, 4)
m   <- c(-8, -11.6)
sp  <- c(13.8, 5.8)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Acceptance and Commitment Therapy (ACT)  5  -8.000  13.800  -25.135  9.135
CARM  Treatment as Usual (TAU)  4  -11.600  5.800  -20.829  -2.371

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 11 outcome measures and 10 posted secondary outcomes, in addition to the primary endpoint. Multiple outcomes can raise the chance of at least one apparently notable result when many statistical tests are conducted. In general, a multiplicity plan may specify which outcomes are confirmatory, how tests are ordered, or whether error rates are adjusted. A multiplicity strategy is not reported in the registry record, and the record lists no analyses. It is therefore not possible to determine whether any adjustment or prespecified testing hierarchy was used.

For the primary endpoint, the posted arm summaries include 5 participants in ACT and 4 in TAU, while total enrollment was 18. The registry record does not explain this difference, identify the number with missing measurements, or state how missing data were handled. It does not report whether participants were analyzed in their assigned groups, whether a per-protocol population was used, or what rules defined inclusion in the outcome analysis. These analysis-population details are not reported in the registry record.

Missing outcome data can affect estimates if the participants with recorded measurements differ systematically from those without them. This is a general concern, not a claim about what occurred in this study. Without reasons for the differing counts or a description of missing-data procedures, the extent and direction of any possible effect cannot be assessed. The posted means should consequently be understood as summaries for the stated outcome group sizes, not automatically as summaries of all randomized participants.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry record lists 12 participants at risk for serious adverse events in the ACT group and 12 in the TAU group. It reports 0 serious adverse events affecting participants in either group. For other adverse events, it likewise lists 12 at risk in each group and 0 affected in either group. These are the safety figures recorded for the groups and denominators stated in the record.

Zero recorded events means that none were reported among the listed participants at risk; it does not prove that an intervention is risk-free. The safety denominators do not match the participant counts in the primary outcome summaries, and the registry record does not explain the difference. The deaths field is not reported in the registry record. The serious adverse event terms are also not reported in the registry record. These results should be read as limited descriptive safety reporting from a small pilot, not as a precise estimate of event risk or as a comparison capable of ruling out uncommon events.

13. What the Result Does, and Does Not, Mean

The posted primary results show arm-specific average changes in Brief Psychiatric Rating Scale scores during the hospital stay: -8.0 (SD 13.8; n=5) for ACT and -11.6 (SD 5.8; n=4) for TAU. Because lower percentage change is defined as better, the signs and scoring convention should be considered together. The registry record does not report a formal statistical comparison, confidence interval, or p-value. The means therefore describe observed group summaries without establishing a reliable difference between assignments.

The study was a randomized pilot at a single VHA site, with 18 enrolled participants. The primary outcome summaries cover fewer participants than total enrollment, and the reason is not reported in the registry record. The record’s intended focus included feasibility, acceptability, safety, implementation barriers and facilitators, and potential effectiveness. The primary outcome figures alone cannot answer all of those questions, nor do they establish whether the intervention can be implemented broadly.

In general, a pilot can provide descriptive information and help identify practical questions for later research, but small group summaries are not a substitute for a sufficiently precise comparative estimate. This is a general statistical distinction, not a judgment about treatment. No clinical recommendation follows from these results. The figures also do not show whether any observed change was caused by the assigned condition, whether it persisted after discharge, or how outcomes would differ in other settings; those findings are not reported in the registry record.

What can be stated is limited to the posted measures, participant counts, and safety counts. What cannot be stated from these results is whether ACT is superior, equivalent, or inferior to TAU, or whether the recorded differences reflect treatment rather than chance, missing data, or other factors. The absence of a reported inferential analysis leaves the comparative uncertainty unresolved.

14. Why This Trial Matters Statistically

This trial is statistically notable because it combines randomized assignment with very small outcome groups and descriptive reporting. Randomization creates a framework for comparing interventions, but a useful comparison also needs an estimate of the contrast and its uncertainty. The registry record reports means and standard deviations for each arm, yet no formal between-arm analysis. This illustrates why arm-by-arm summaries should not be treated as a completed test of comparative effectiveness.

The difference between total enrollment and the primary outcome group counts is another important issue. Enrollment was 18, whereas the outcome summaries include 5 ACT participants and 4 TAU participants. Without an account of missingness and the analysis population, readers cannot tell how representative those summaries are of randomized participants. Transparent reporting of participant flow and outcome availability is important for interpreting both precision and potential bias.

The endpoint also depends on careful attention to scale direction. The registry defines decreases in percentage change as better, so a reader must not apply a generic assumption that a higher score is always preferable. The measure spans broad symptom domains, while follow-up is tied to hospital stay and therefore has a reported mean duration of 24.0 days (SD 15.8). Both the scoring rule and timing are necessary context for understanding the posted values.

Finally, this record distinguishes pilot aims from confirmatory evidence. Feasibility, acceptability, safety, and implementation considerations are relevant study questions, but the available primary summaries do not establish a precise comparative effect. The case emphasizes the value of reporting analysis plans, missing-data handling, uncertainty intervals, and results for the intended analysis population alongside descriptive statistics.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several limitations are specific to the information in this registry record. The study enrolled 18 participants at a single VHA site, and the primary outcome summaries include 5 participants in ACT and 4 in TAU. The record does not explain why the outcome counts are smaller than enrollment, so completeness of follow-up and the potential effect of missing observations cannot be assessed.

The record reports a mean and standard deviation for each arm but no formal between-arm estimate, confidence interval, p-value, or analysis. It does not describe the statistical model, covariate adjustment, handling of repeated measurements, or analysis population. These details are not reported in the registry record, limiting assessment of how the posted summaries were generated and how uncertainty should be quantified.

The study was unmasked according to the design record. Whether outcome assessors were masked is not reported in the registry record. Follow-up lasted for the duration of hospital stay, with mean 24.0 days and SD 15.8; the record does not describe an approach for accounting for variation in stay duration. The primary outcome counts also differ from the safety denominators, and the record does not explain why.

Safety reporting lists zero serious and zero other adverse events among the stated at-risk groups, but the death counts are not reported in the registry record. The small groups and limited reporting cannot establish that uncommon events were absent. A sample-size justification, missing-data plan, multiplicity strategy, and detailed analysis plan are not reported in the registry record. These limitations constrain interpretation of the descriptive findings and preclude conclusions about comparative treatment effects.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Acceptance and Commitment Therapy for the Inpatient Treatment of Psychosis — NCT01981356.
• Registry record last updated 2016-04-14; status last verified 2016-04.
• Results first posted 2016-04-14.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry record describes a completed, randomized, parallel-group, unmasked early-phase 1 pilot sponsored by the VA Office of Research and Development. Enrollment was 18, and the study took place from 2014-09 to primary completion in 2015-04. It compared ACT, offered as 4 sessions, with TAU consisting of psychopharmacology, case management, psychotherapy, and brief facilitator contacts.

The primary outcome was percentage change on the Brief Psychiatric Rating Scale over the hospital stay. The reported mean (SD) was -8.0 (13.8) for ACT, n=5, and -11.6 (5.8) for TAU, n=4. Decreases are defined as better outcomes. A formal comparison and confidence intervals are not reported in the registry record. Safety summaries list 0 serious and 0 other adverse events in each group among 12 participants at risk per group; deaths are not reported in the registry record.