Development/Testing of SUMMIT: a Tool to Help Patients Manage Pain While Tapering Opioids: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Development/Testing of SUMMIT: a Tool to Help Patients Manage Pain While Tapering Opioids (NCT04746833), a randomized na study. The primary outcome is Daily Morphine Milligram Equivalents (MME).

2. Clinical Question

This question concerns a measured outcome, not a direct test of every component of the intervention or of the safety of opioid tapering. The primary measure is daily morphine milligram equivalents based on participant self report. The registry record does not report the eligibility criteria, baseline opioid use, tapering schedules, or other participant characteristics needed to describe the enrolled population in detail. Nor does it report an equivalence margin, so the sponsor analysis label alone does not establish what range of differences was considered acceptably equivalent.

3. Trial Design

The registry record identifies NCT04746833 as a completed study sponsored by the VA Office of Research and Development. It lists chronic pain as the condition, enrollment of 44, and dates from 2021-03-24 to primary completion on 2023-03-23. The phase is listed as NA.

Allocation is listed as randomized, masking as none, and the study model as single-group. At the same time, the posted primary outcome contains results for SUMMIT and Control, and the sponsor analysis compares the two groups. Those entries do not fully clarify how the design-model label relates to the reported two-group results. Additional design details are not reported in the registry record.

The intervention listings identify SUMMIT and control as behavioral interventions. The SUMMIT arm description refers to a Motivational Interviewing session with a study psychologist and a multicomponent tool intended to support opioid tapering while managing pain. The control description identifies My Pain Diary: Chronic Pain Management, an eHealth comparator described as primarily a pain-monitoring app. The registry record does not provide the complete intervention descriptions, the allocation procedure, or details of follow-up and adherence.

The primary outcome is daily morphine milligram equivalents at 9-months. The posted arm summaries include 20 participants in SUMMIT and 19 in Control for this outcome. Enrollment and outcome counts need not represent the same analysis population, and the reasons for any difference are not reported in the registry record. The record also posts a secondary Behavior Intervention Rating Scale outcome at six months. The protocol's original sample-size assumptions, analysis populations, and detailed statistical plan are not reported in the registry record.

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.

Overall Study

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.

6. Primary Endpoint

The registry lists the primary outcome as Daily Morphine Milligram Equivalents (MME), measured over 9-months.

The primary endpoint is Daily Morphine Milligram Equivalents (MME), reported as a mean in morphine milligram equivalents per day at 9-months. The registry description says the measure is based on participant self report. MME expresses opioid dosage on a common scale, allowing group-level daily opioid amounts to be summarized using a mean. The registry record does not specify the exact self-report instrument, recall period, or handling of missing reports.

In a study focused on support for opioid tapering, a measure of daily opioid dosage is relevant because it captures a medication-use outcome alongside the stated goal of helping participants manage pain. A mean describes the center of the observed group values, but it does not show how values are distributed among participants. The posted standard deviations are large relative to the group means, indicating substantial variation in the observed values. This makes the mean difference and its confidence interval important to consider together rather than interpreting group averages alone.

The endpoint is not, by itself, a full account of pain, function, participant experience, or intervention acceptability. The registry record lists the Behavior Intervention Rating Scale as a secondary outcome, but other details about the range of outcomes relevant to participants are not reported in the registry record. The 9-month measure should therefore be understood as a particular self-reported dosage outcome at a stated follow-up time, rather than a complete assessment of the intervention's effects or of the tapering process.

7. Results Posted in the Registry

• SUMMIT: mean 25.10 versus 21.90 on control; difference 3.20 (95% CI -20.23 to 26.63), standardized effect d = 0.08; z-test p = 0.789.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Behavior Intervention Rating Scale (BIRS)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • SUMMIT: mean 4.40 versus 4.50 on control; difference -0.10 (95% CI -0.36 to 0.16), standardized effect d = -0.25; z-test p = 0.454.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between SUMMIT and the control. The interval gives the range of effects that remain compatible with the data.

The registry record lists one secondary outcome, the Behavior Intervention Rating Scale, measured at six months. Its posted group summaries and sponsor analysis represent a measure distinct from daily MME, so it adds a different perspective on the study rather than serving as a substitute for the primary dosage outcome.

Across the posted primary and secondary outcomes, the registry record does not present a coherent basis for claiming a definitive group difference. The secondary result should be considered alongside, but not as confirmation of, the primary comparison. The two outcomes address different constructs and timeframes, and the registry record does not report a joint analysis that combines them.

The secondary analysis is reported with an unadjusted p-value. When multiple outcomes or analyses are examined, the chance of observing a nominally notable result by chance can increase. Statistical methods can account for multiplicity, but the registry record does not report whether a multiplicity strategy was prespecified or applied. Consequently, an unadjusted secondary result generally carries less evidential weight than a result from a prespecified analysis that accounts for the full set of tests.

The record also does not give a detailed rationale for the secondary endpoint, its role in the study's hierarchy of outcomes, or whether it was intended to support a formal confirmatory claim. The secondary result is best treated as additional descriptive evidence from this study, with interpretation limited by the posted information. It does not resolve the uncertainty around the primary comparison or establish a broader pattern beyond the outcomes actually recorded.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Behavior Intervention Rating Scale (BIRS)

9. Statistical Interpretation

10. Statistical Methods Explained

The posted summaries describe a continuous outcome using group means and standard deviations. The comparison is a mean difference, with the SUMMIT mean compared against the Control mean. A mean difference retains the outcome's original units: morphine milligram equivalents per day. The recalculated comparison is unadjusted, meaning it uses the posted summary statistics without accounting for other participant characteristics or design features.

What does a mean and standard deviation convey?

A mean gives the arithmetic average of the observed values in a group. A standard deviation describes how spread out individual values are around that mean. It is not the standard error of the mean and is not a confidence interval. In this study, the standard deviations are large relative to the means, so the group averages should not be read as if participants had similar daily MME values. The posted summaries do not describe the distribution, such as whether values are skewed or include unusually high observations.

What is the mean difference?

The posted mean difference is 3.2, calculated for SUMMIT relative to Control. A positive value indicates that the SUMMIT group mean is higher in the posted summaries; a negative value would indicate a lower mean. The sign describes direction for this specified ordering only. It does not by itself establish the size, importance, or cause of a group difference.

How should the p-value be understood?

The recalculated unadjusted p-value is 0.789. In general, a p-value describes how unusual a result at least as extreme as the observed result would be if the statistical model's null hypothesis were true. It depends on the test and its assumptions. It is not the probability that the null hypothesis is true, and it is not a measure of the probability that the result will replicate. A larger p-value does not prove equivalence or prove there is no difference.

The sponsor-reported analysis is identified as a two-sided t-test and as an equivalence analysis. The registry record gives p = .91 but does not post the analysis estimate, confidence interval, or equivalence margin. Equivalence testing generally asks whether the uncertainty around a treatment contrast falls within prespecified bounds for differences regarded as acceptably small. Without the margin and corresponding estimate or interval, the sponsor analysis label cannot be fully interpreted as a test of equivalence from the posted record alone.

Why is the confidence interval central?

A confidence interval gives a range of values compatible with the observed data under the method used to calculate it. In general, a 95% confidence procedure is one that would cover the underlying parameter in 95% of repeated samples under its assumptions. The interval for the unadjusted mean difference extends from -20.23 to 26.63. It includes zero, as well as differences in both directions. Its width signals limited precision, so the point estimate should not be treated as a precise estimate of the group contrast.

What assumptions matter?

A standard t-test compares means while accounting for sampling variability. Interpretation depends on the suitability of the test for the data and analysis structure. The registry record does not report diagnostic checks, transformations, robust methods, covariate adjustment, or details about how repeated or missing observations were handled. The summary statistics alone do not allow those choices to be assessed.

Over-dispersion, in general, means that observations vary more than a simple model expects. Excess variability can make standard errors larger than a model that assumes less variation would suggest, and an analysis that does not account for it can overstate precision. The posted standard deviations show appreciable spread, but the registry record does not report a formal over-dispersion assessment. It is not appropriate to infer a particular distribution or model from the summaries alone.

What is the role of recalculation?

The unadjusted recalculation uses the posted means, standard deviations, and group counts to describe the mean contrast and its uncertainty. It is a simple summary-statistic comparison, not an individual-participant analysis and not an adjusted model. The registry's own analysis is reported separately. Since the posted record does not provide enough information to reproduce that analysis in full, the recalculation should be read as a complementary statistical description rather than as a reconstruction of the complete sponsor analysis.

11. Confidence Intervals and Precision

The 95% confidence interval for the unadjusted mean difference is -20.23 to 26.63 morphine milligram equivalents per day. The interval is centered around the posted estimate of 3.2, but spans negative and positive values. In practical statistical terms, the result is compatible with the SUMMIT mean being lower than the Control mean, with little difference, or with the SUMMIT mean being higher. The interval does not identify one direction with precision.

A confidence interval is not a range containing a fixed proportion of individual participants' values. It is an interval estimate for the group mean difference. In general, a 95% confidence procedure is designed to cover the true parameter in 95% of repeated samples if the method's assumptions hold. That repeated-sampling interpretation does not mean there is a 95% probability that this particular fixed parameter lies in this particular interval.

The size of the interval reflects the uncertainty in the estimate, which is influenced by the variability and number of observations as well as the statistical method. The posted standard deviations are 44.7 for SUMMIT and 28.6 for Control, and the primary outcome summaries include 20 and 19 participants, respectively. The registry record does not report a baseline-adjusted estimate or additional analysis details that could change the interval.

The interval includes zero, but this is not evidence that the groups are identical. Nor does the interval alone establish equivalence: an equivalence conclusion requires a prespecified margin and an analysis that evaluates the estimate against that margin. The margin is not reported in the registry record. The width of the interval is therefore especially important: it shows that the unadjusted estimate leaves considerable uncertainty about the magnitude and direction of the group contrast.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total participants. The accompanying curve should be read as a planning illustration based on the same assumptions as the table, not as a record of the trial's observed enrollment or its original design calculation. The registry record's original sample-size assumptions and justification are not reported in the registry record.

The illustration uses a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set as a stated fraction of the observed effect. For each assumed effect, the 80% power column gives the estimated participants per arm for an 80% chance of detecting that effect under the planning assumptions if it is present. The 90% power column gives the corresponding number for a 90% chance. These are design-planning quantities, not guarantees for a future study.

The rows vary the effect used for planning: 100%, 75%, or 50% of the observed effect. The figures rise as the assumed effect becomes smaller, because a smaller difference is harder to distinguish from random variation and therefore requires more participants under the same assumptions. For the primary outcome, the per-arm figures are 2,183 and 2,923 for 100% of the observed effect, 3,881 and 5,196 for 75%, and 8,732 and 11,690 for 50%, for 80% and 90% power respectively.

These sample-size values are not observed power. They are planning illustrations computed from the observed effect and stated assumptions. They do not establish what power the completed study had, and they should not be interpreted as the study's original target or justification.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 3.200; pooled standard deviation σ = 37.736.
2. n = 2 × (1.96 + 0.842)² × 37.736² / 3.200² = 2,182.9, rounded up to 2,183 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 3.2                      # observed difference in means
sigma <- 37.7355628380272                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(4746833)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 2,183 (80% power), 2,923 (90% power)
 75% of observed effect: n per arm = 3,881 (80% power), 5,196 (90% power)
 50% of observed effect: n per arm = 8,732 (80% power), 11,690 (90% power)
power.t.test: 2183.9 
Simulated power at n = 2,183 per arm: 0.803 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry record reports the sponsor's analysis as a two-sided t-test labeled EQUIVALENCE, with p = .91. It does not post an estimate, confidence interval, or equivalence margin for that analysis. The label and p-value therefore provide limited information about the size of the contrast and about how the analysis evaluated equivalence.

The recalculated unadjusted comparison from the posted group summaries gives a mean difference of 3.2, a 95% confidence interval from -20.23 to 26.63, and an unadjusted p-value of 0.789. These results are based on the posted group means, standard deviations, and counts. The recalculation is not an adjusted analysis and may not reproduce the sponsor's full analysis, whose detailed specifications are not reported in the registry record.

The two p-values are not identical, and they should not be treated as interchangeable. The sponsor's result describes the analysis entered in the registry, while the recalculation describes a straightforward unadjusted contrast from the posted summaries. The registry record does not explain why the results differ or provide enough detail to determine the exact analytical choices behind the sponsor's reported value. Neither p-value alone establishes equivalence or demonstrates that the groups are identical.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study assesses several outcomes, timepoints, groups, or statistical contrasts. Testing more than one hypothesis can increase the chance that at least one result appears notable by chance. The registry record lists three outcome measures overall and posts one secondary outcome, but it does not report a multiplicity adjustment, a testing hierarchy, or a prespecified plan for interpreting the collection of analyses. The unadjusted secondary result should therefore be treated cautiously, and the record does not support a claim that family-wise error or another multiplicity concern was controlled.

The registry record does not report how missing primary or secondary outcome data were handled. It does not describe whether analyses used complete cases, imputation, a likelihood-based approach, or another method. It also does not report the reasons for missingness or compare participants with and without observed outcomes. The posted outcome counts are smaller than the enrollment entry, but the record does not explain the difference or establish which analysis population each set of counts represents.

Analysis populations are not defined in the posted record. There is no reported statement identifying an intention-to-treat population, a per-protocol population, or another population for the sponsor's test. Although allocation is listed as randomized, the record does not explain whether participants were analyzed according to assigned group or whether any participants were excluded from the outcome analysis.

These omissions matter because missing-data choices, analysis population, and multiplicity procedures can affect both estimates and uncertainty. Without those details, the posted mean comparison remains a useful description of the available summaries but cannot be assumed to reflect a particular prespecified analysis strategy. The original statistical analysis plan and the protocol's detailed handling of missingness are not reported in the registry record.

16. Safety Results

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

Deaths

Other (non-serious) adverse events

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

The registry adverse-event entries report serious adverse events for 0 of 24 participants at risk in SUMMIT and 0 of 20 participants at risk in Control. The same entries report 0 other adverse events for each group and 0 deaths in each group. These are the event counts recorded for the stated numbers at risk.

A zero count in the registry entry means no events of that listed category were recorded among the participants included in that entry. It does not establish that an event could not occur, that all possible harms were assessed, or that the interventions have identical safety profiles. Event counts can be interpreted only in the context of the defined observation period, event collection procedures, and participants included.

The registry record does not report detailed adverse-event definitions, collection methods, follow-up duration for safety, or a breakdown of serious-event terms. The numbers at risk also differ from the primary outcome counts, and the record does not explain the relevant analysis populations. These entries should therefore be read as registry-reported counts, not as a complete assessment of risk or as a comparative safety test.

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

The posted primary outcome describes average self-reported daily MME at 9-months in the two reported groups. The unadjusted mean difference is 3.2, with a 95% confidence interval from -20.23 to 26.63. The interval is broad and includes values in both directions, so the estimate does not pin down the direction or magnitude of the underlying group contrast. The unadjusted p-value of 0.789 does not establish that there is no difference.

The result does not show whether participants reduced opioid use from their own baseline. Baseline MME summaries and a change-from-baseline comparison are not reported in the registry record. Nor does the result establish whether participants experienced changes in pain, function, quality of life, or other outcomes beyond those listed in the record. Daily MME is one self-reported measure at a specified follow-up, not a complete account of the tapering experience.

The registry describes a randomized allocation but also labels the model single-group, while posting results for SUMMIT and Control. The record does not clarify this design labeling or provide detailed information about allocation, analysis populations, and missing data. Those gaps limit how fully the posted contrast can be interpreted as an estimate of an intervention effect.

The sponsor's analysis is labeled as an equivalence analysis, but the equivalence margin and corresponding estimate and interval are not reported. A p-value alone, including the sponsor-reported p = .91, is not enough to establish that two interventions are equivalent. The available results should therefore be read as imprecise group summaries and a reported sponsor test, rather than as proof of equivalence, a definitive treatment effect, or evidence of no effect.

The study's stated aim is to develop and pilot test a multicomponent tool intended to support opioid tapering and pain management. The posted results do not by themselves establish how the tool performs in broader settings or whether it achieves the intended goals. Those questions require interpretation in light of the study's design, participant selection, intervention delivery, and complete analysis plan, details that are not reported in the registry record.

18. Why This Trial Matters Statistically

This study raises a statistical question that is common in early intervention research: how to interpret an uncertain estimate from a small pilot study without turning a non-significant test into a claim of no difference. The posted mean difference is accompanied by a wide confidence interval. That interval makes uncertainty visible and shows why a point estimate or p-value alone is not an adequate summary.

The comparison also illustrates the distinction between a conventional difference test and an equivalence question. A nonsignificant difference test does not demonstrate equivalence. Equivalence requires a defined range of differences considered acceptably small and an analysis that evaluates uncertainty against that range. Although the sponsor analysis is labeled as equivalence, the margin and interval are not reported in the registry record. The label therefore cannot be assessed fully from the posted entry.

High variability is another important feature. Means can conceal substantial participant-to-participant variation, and imprecise estimates may remain compatible with materially different contrasts. Clear reporting of distributions, baseline values, analysis populations, and missing-data methods would help readers understand that variability and the scope of the comparison.

Finally, planning illustrations based on an observed effect can be useful for thinking about future study scale, but they are not evidence that the original trial was powered for its purpose. The record's original design assumptions are not reported. Keeping planning calculations separate from observed results helps prevent an exploratory or pilot comparison from being mistaken for a definitive test.

19. 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.

The registry record does not report detailed eligibility criteria or participant characteristics beyond chronic pain and the study's context of long-term opioid therapy. It therefore does not allow readers to assess how closely the analyzed participants match the broader population of interest. The brief description emphasizes veterans, but the posted record does not provide a full account of the enrolled sample's characteristics.

The design entries are difficult to reconcile: allocation is listed as randomized, the model as single-group, and results are posted for SUMMIT and Control. Further explanation of the study structure and analysis is not reported in the registry record. Masking is listed as none, but the record does not provide details about outcome collection or whether any outcome assessment procedures reduced potential bias.

The primary endpoint is self-reported daily MME, and details of the instrument and recall period are not reported. The posted group standard deviations indicate substantial variability, while the confidence interval is wide. The record does not provide baseline outcome summaries or a baseline-adjusted result, making it impossible to assess change from baseline using the posted results.

Further limitations include missing information on analysis populations, missing-data handling, adherence, intervention fidelity, and the original sample-size rationale. The sponsor-reported equivalence result lacks a posted margin and confidence interval. The registry record also does not report a multiplicity plan. These omissions constrain interpretation of the primary and secondary results and prevent a full assessment of the analysis choices behind them.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Development/Testing of SUMMIT: a Tool to Help Patients Manage Pain While Tapering Opioids — NCT04746833.
• Registry record last updated 2025-09-12; status last verified 2025-08.
• Results first posted 2025-09-12.
• 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.

23. Record Summary

NCT04746833 is a completed study sponsored by the VA Office of Research and Development, with enrollment listed as 44 and chronic pain as the condition. Its stated goal is to develop and test a multicomponent tool intended to support opioid tapering while managing pain. The record lists randomized allocation and no masking, while also labeling the model single-group and posting results for SUMMIT and Control.

The primary outcome is self-reported daily MME at 9-months. The posted unadjusted comparison gives a mean difference of 3.2 and a 95% confidence interval from -20.23 to 26.63. The sponsor's analysis is labeled as an equivalence test, but its estimate, confidence interval, and equivalence margin are not reported. A secondary Behavior Intervention Rating Scale outcome is also posted.

Interpretation is limited by uncertainty in the primary estimate and by missing details on eligibility, analysis populations, missing data, multiplicity, and the sponsor's analysis specifications. The registry results describe the posted outcomes but do not establish equivalence or a definitive effect.