Tailored Non-Pharmacotherapy Services for Chronic Pain: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Tailored Non-Pharmacotherapy Services for Chronic Pain (NCT04523714), a randomized na study. The primary outcome is Minimal Clinically Important Difference (MCID) in Pain Severity at 3 Months (Yes / No).

2. Clinical Question

3. Trial Design

The RESOLVE study is described in the registry record as a multicenter comparative effectiveness trial of two telehealth cognitive behavioral therapy-based chronic pain programs. The trial used randomized allocation and a parallel-group model. The registry lists single masking, but does not report who was masked or how masking was maintained. The study is marked completed.

The three groups were an online CBT-CP-based program, a virtual health coach-led CBT-CP-based program, and usual care plus information. The online program was self-completed; the coach-led program used telephone or videoconference sessions. Both interventions consisted of eight interactive sessions, approximately one per week. The control group received the specified chronic pain resource guide and had no restriction on access to other available pharmacologic or nonpharmacologic treatments.

Total enrollment was 2333. The primary outcome summaries list 776 participants in the online group, 778 in the health coach-led group, and 777 in the usual care plus information group. The sponsor was Kaiser Permanente. The recorded study start was 2021-01-14, and primary completion was 2024-04-30.

The registry identifies the trial as not applicable for phase. It does not report the trial's original sample-size justification, allocation procedures beyond randomized allocation, the identity of the masked parties, or additional protocol details. The interventions are behavioral rather than drug treatments, and the comparison is between two delivery approaches and a usual-care-plus-information group.

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

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: The 2 individuals randomized in error are not reported/included in the Baseline Analysis Population as they were not eligible (i.e., 779 in the Usual Care Plus in the Participant Flow vs. 777 in Baseline Analysis Population). The 2 individuals who were randomized in error (i.e., totaling 2,333 in the Participant Flow) were withdrawn immediately after randomization when the error was identified. That is why the Baseline Analytic Population = 2,331.

6. Primary Endpoint

The registry lists the primary outcome as Minimal Clinically Important Difference (MCID) in Pain Severity at 3 Months (Yes / No), measured over Baseline to 3 months.

The primary endpoint was the adjusted percentage of participants with a minimal clinically important difference in pain severity from baseline to 3 months. The registry defines this outcome as a 30% or greater improvement in the pain severity score. The outcome is binary: a participant either meets the improvement criterion or does not.

Pain severity was assessed using a modified 11-item version of the Brief Pain Inventory - Short Form, consistent with IMMPACT guidelines. The registry describes the BPI-SF range as 0 to 10, with higher scores indicating worse pain severity. The endpoint therefore translates change on a pain-severity measure into an easily interpretable threshold: the proportion of participants reaching the prespecified level of improvement.

As a general statistical point, a binary threshold can aid communication about the share of participants meeting a defined criterion. It also simplifies a continuous score into two categories, so it does not show the full distribution of score changes or how far participants who met the threshold improved. The registry record does not report why this threshold was selected beyond defining the MCID, nor does it provide additional details about the adjustment used to estimate the percentages.

7. Results Posted in the Registry

• Online CBT-CP Based Program: 26.5% (206/776) versus 20.8% (162/777) on control; risk difference 5.7 percentage points (95% CI 1.5 to 9.9), risk ratio 1.27 (1.06–1.52); unadjusted two-proportion z-test p = 0.008.
• Health Coach-led CBT-CP Based Program: 32.0% (249/778) versus 20.8% (162/777) on control; risk difference 11.2 percentage points (95% CI 6.8 to 15.5), risk ratio 1.54 (1.29–1.82); unadjusted two-proportion z-test p = <0.001.

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.

MCID in Pain Severity at 6 Months (Yes / No)

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Online CBT-CP Based Program: 32.9% (255/776) versus 22.9% (178/777) on control; risk difference 10.0 percentage points (95% CI 5.5 to 14.4), risk ratio 1.43 (1.22–1.69); unadjusted two-proportion z-test p = <0.001.
  • Health Coach-led CBT-CP Based Program: 37.1% (289/778) versus 22.9% (178/777) on control; risk difference 14.2 percentage points (95% CI 9.7 to 18.7), risk ratio 1.62 (1.38–1.90); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Online CBT-CP Based Program than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Health Coach-led CBT-CP Based Program than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

MCID in Pain Severity at 12 Months (Yes / No)

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Online CBT-CP Based Program: 36.0% (279/776) versus 27.2% (211/777) on control; risk difference 8.8 percentage points (95% CI 4.2 to 13.4), risk ratio 1.32 (1.14–1.54); unadjusted two-proportion z-test p = <0.001.
  • Health Coach-led CBT-CP Based Program: 38.3% (298/778) versus 27.2% (211/777) on control; risk difference 11.1 percentage points (95% CI 6.5 to 15.8), risk ratio 1.41 (1.22–1.63); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Online CBT-CP Based Program than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Health Coach-led CBT-CP Based Program than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry lists secondary outcomes covering MCID in pain severity at 6 months and 12 months, pain severity and intensity scores, pain-related interference, social role functioning, physical functioning, and patient global impression of change for pain. The detailed numeric results available for the two later MCID time points show a broad pattern aligned with the primary outcome: the adjusted percentages meeting the MCID threshold are higher in each CBT-CP group than in usual care plus information, and the reported relative risks for both intervention groups versus usual care are above 1. The record also reports pairwise comparisons for the online and health coach-led programs at those later time points.

Those later observations concern the same threshold but different follow-up times, so they provide context about whether the primary pattern is also present beyond 3 months. They are not independent replications of the primary endpoint: they involve the same trial and related outcome definitions. The registry also names other secondary measures, but detailed numerical estimates for pain intensity, interference, social role functioning, physical functioning, and patient global impression are not reported in the registry record.

Secondary outcomes are generally interpreted with more caution when many outcomes or comparisons are examined. Unless multiplicity is addressed, the chance of at least one apparently notable result can increase as the number of statistical tests grows. The registry record lists multiple secondary measures and analyses but does not report a multiplicity procedure. These results can therefore be viewed as supportive or exploratory context, but they carry less confirmatory weight than a prespecified primary outcome analyzed under a clearly described testing plan. The record does not provide enough information to determine the extent to which the secondary analyses were prespecified.

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

• MCID in Pain Severity at 6 Months (Yes / No)
• MCID in Pain Severity at 12 Months (Yes / No)
• Pain Severity Score
• Pain Intensity Score
• Pain-related Interference Score
• Social Role Functioning
• Physical Functioning
• Patient Global Impression of Change (PGIC) - Pain

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary, because the registry classifies participants according to whether they reached the defined pain-severity improvement threshold. For this kind of outcome, results can be expressed as a percentage in each group and as a contrast between groups. The registry reports adjusted percentages for the three groups and uses modified Poisson regression for its pairwise comparisons.

What does a relative risk express?

A relative risk compares the probability of an outcome in one group with its probability in another group. A relative risk of 1 represents equal probabilities under the comparison; a value above 1 means the outcome is estimated to be more frequent in the first-listed group, while a value below 1 means it is estimated to be less frequent. For the primary comparisons, the registry reported relative risks of 1.28 for the online program versus usual care plus information, and 1.54 for the health coach-led program versus usual care plus information. It also reported 1.20 for the online program versus the health coach-led program.

The registry identifies modified Poisson regression as the method used to estimate relative risks. As a general statistical concept, Poisson regression is commonly associated with modeling event counts, while a modified approach can be used to estimate relative risks for binary outcomes. The registry record does not report the model's covariates, variance estimator, or other implementation details, so those aspects cannot be characterized from the record.

How should confidence intervals and p-values be read?

A confidence interval gives a range of estimates compatible with the data and the statistical model under repeated use of the same procedure. A narrower interval indicates greater precision than a wider interval, all else equal. The registry reports 95% confidence intervals for its relative risks. The interval for online versus usual care plus information is 1.06 to 1.55; for health coach-led versus usual care plus information it is 1.30 to 1.82; and for online versus health coach-led it is 1.03 to 1.40.

A p-value describes how unusual data at least as incompatible with a specified null hypothesis would be if that null hypothesis and the model assumptions held. It is not the probability that the null hypothesis is true, and it does not quantify the size or practical importance of an effect. The reported primary comparison p-values are each p < 0.001. The registry record does not state whether or how these comparisons were adjusted for multiple testing.

What do the recalculated contrasts add?

The unadjusted recalculations use the posted group summary statistics and express comparisons as risk differences. A risk difference is the difference between two group probabilities, unlike a relative risk, which is a ratio. The recalculations are not adjusted for covariates and are not substitutes for the sponsor's regression analysis. They provide an additional description of the group summaries using a different effect scale. The registry's reported adjusted relative risks remain the relevant results for describing the sponsor's own analysis.

The posted adjusted percentages and the observed group counts are related but are not interchangeable: an adjusted percentage is a model-based estimate, whereas a raw count or proportion is a direct summary of recorded outcomes and denominators. Adjustments can account for modeled variables, but the particular variables and procedures used for these adjusted percentages are not reported in the registry record. The record also does not provide enough detail to establish whether the adjustment model for the percentages was the same as the modified Poisson model.

Because the endpoint turns a score change into a yes-or-no result, the analysis focuses on the proportion crossing the MCID threshold. As a general point, this method may be easier to interpret than a model coefficient on a score scale, but it loses information about participants' exact score changes. The registry record describes other pain-related outcomes as well, which provide distinct measures of the trial's questions; their interpretation should not be collapsed into the primary binary endpoint.

11. Confidence Intervals and Precision

The primary adjusted percentages have 95% confidence intervals that convey uncertainty around each group's estimated proportion meeting the MCID criterion. The online program estimate is 26.6%, with an interval from 23.4% to 30.2%. The health coach-led estimate is 32.0%, with an interval from 29.3% to 35.0%. The usual care plus information estimate is 20.8%, with an interval from 18.0% to 24.0%. These intervals describe uncertainty in the estimates, not the range of individual participants' outcomes.

The registry's relative-risk intervals also indicate the precision of the pairwise comparisons. For the online program versus usual care plus information, the estimate is 1.28 and the interval is 1.06 to 1.55. For the health coach-led program versus usual care plus information, the estimate is 1.54 and the interval is 1.30 to 1.82. For online versus health coach-led, the estimate is 1.20 and the interval is 1.03 to 1.40. Each interval excludes 1, the null value for a relative risk, while the interval for the online versus health coach-led comparison comes relatively close to that value.

As a general statistical concept, an interval should be read as a range of plausible effect estimates under the analysis, not as a guarantee that the true effect lies within it. Its width reflects sampling variability and depends on the model and data. Confidence intervals do not address all sources of uncertainty, including missing outcome data, measurement limitations, model specification, or multiplicity. Those details are not fully reported in the registry record, so the intervals should not be treated as resolving those uncertainties.

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 illustrations based on the observed effect, not a report of the trial's original design calculations. For these binary outcomes, the stated quantities are participants per arm, not total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set to a stated fraction of the observed effect. The registry record does not report the original sample-size assumptions or justification.

The 80% power column gives the estimated participants per arm for a hypothetical study to have an 80% chance of detecting the specified effect if that effect and the planning assumptions hold. The 90% power column gives the corresponding estimate for a 90% chance. These are design-planning concepts, not probabilities that the observed finding is true. For example, for the health coach-led program versus usual care plus information at the primary endpoint, planning for 100% of the observed effect gives estimates of 245 participants per arm for 80% power and 327 participants per arm for 90% power. Planning for 75% of the observed effect gives estimates of 420 and 562 participants per arm, respectively.

Smaller effects are harder to distinguish from random variation, so the planning figures require more participants per arm when the target effect is a smaller fraction of the observed effect. The figures are conditional on the stated assumptions and should not be read as a recommendation or as a description of the actual trial's design. This is not observed power.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.208; treatment proportion p₁ = 0.265; average p̄ = 0.237.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6014 = 1.1787.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6000 = 0.5050.
4. n = (1.1787 + 0.5050)² / (0.0570)² = 873.4, rounded up to 874 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
p0 <- 0.208494208494208                          # control proportion
p1 <- 0.265463917525773                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(4523714)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 874 (80% power), 1,169 (90% power)
 75% of observed effect: n per arm = 1,521 (80% power), 2,036 (90% power)
 50% of observed effect: n per arm = 3,349 (80% power), 4,482 (90% power)
power.prop.test: 873.4 
Simulated power at n = 874 per arm: 0.803 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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's own primary comparisons use modified Poisson regression and report adjusted relative risks. They are the sponsor's analysis results: 1.28 for the online program versus usual care plus information, 1.54 for the health coach-led program versus usual care plus information, and 1.20 for online versus health coach-led. Each comparison has a reported 95% confidence interval and p-value.

The recalculated contrasts are simple, unadjusted comparisons derived from posted summary statistics and are expressed as risk differences. They answer a related but different statistical question from an adjusted relative risk. They do not reproduce the sponsor's modeling choices, covariate adjustment, or analysis population, details of which are not fully reported in the registry record. A difference in effect scale should not be treated as a disagreement in results.

No discrepancy is flagged between the posted summaries and the sponsor's reported analyses. The sponsor's reported model-based results should therefore be kept distinct from the unadjusted recalculations rather than blended into a single estimate. The adjusted percentages are useful for understanding the reported group-level outcome, while the relative risks describe the sponsor's pairwise comparisons. Neither form of estimate alone establishes causality or clinical importance.

15. Multiplicity, Missing Data and Analysis Populations

The registry record reports a primary endpoint and multiple secondary outcomes, along with pairwise comparisons among trial groups. It does not report whether the statistical plan controlled the family-wise error rate, used another multiplicity adjustment, or treated secondary analyses as exploratory. When multiple hypotheses are tested, the chance of finding at least one result that appears statistically notable can rise, even when no true group differences exist. This is a general statistical concern, not evidence that any particular reported result is false.

The record also does not report how missing outcome data were handled, how many participants contributed outcome data at each follow-up, or whether sensitivity analyses were conducted. Missingness can affect interpretation if the probability of having an observed outcome is related to the outcome itself or differs between groups. Without details on the amount, reasons, and handling of missing data, the possible influence of missingness cannot be assessed from the registry record.

Enrollment and outcome-summary group sizes are reported, but those figures alone do not establish the analysis population or explain whether all randomized participants were included in each analysis. The registry record does not describe an intention-to-treat analysis, a per-protocol analysis, or another population definition for the reported regression results. It also does not provide enough information to determine whether the adjusted percentages and relative risks used the same participants or adjustment variables.

These omissions limit evaluation of the analysis plan and robustness, but they do not change what the posted estimates state. They mean that readers cannot verify from the registry record whether multiplicity, missing data, or population selection affected the reported uncertainty and comparisons. Additional protocol or analysis-plan information is 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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hospitalizations (8.4%).

The registry lists serious adverse events affecting 55 participants in the online CBT-CP group, 65 in the health coach-led group, and 63 in the usual care plus information group. The listed numbers at risk are 776, 778, and 777, respectively. The registry also lists other adverse events affecting 1, 16, and 0 participants in those groups, respectively. Deaths are listed as 3 in the online group, 7 in the health coach-led group, and 7 in usual care plus information.

Hospitalizations are the serious adverse event term identified as the top serious term, with 8.4% reported. The record does not state the denominator for this term in the information, nor does it provide enough detail to interpret the event definition or timing. These counts and the reported term describe what the registry lists; they do not establish whether an event was caused by an intervention. Differences in these counts should not be interpreted as comparative safety effects without the relevant definitions, follow-up, and analysis.

As a general point, adverse-event summaries are easier to interpret when the event definitions, observation period, participant-level denominators, and attribution procedures are clear. Those details are not fully reported in the registry record. The figures should therefore be read as descriptive safety reporting rather than a complete comparative safety analysis.

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

The primary finding describes the proportion of participants who met a defined pain-severity improvement threshold during baseline to 3 months. The registry reports higher adjusted percentages in both CBT-CP groups than in the usual care plus information group, and its modified Poisson comparisons report relative risks above 1 for each intervention compared with that group. The interval estimates convey uncertainty around these group-level comparisons.

This result does not mean every participant in either program met the threshold, and it does not describe the magnitude of change for participants who did or did not reach it. The binary endpoint records whether improvement crossed the 30% criterion; it does not preserve the full range of pain-severity score changes. It also does not, by itself, answer questions about pain-related interference, physical functioning, social role functioning, or other outcomes listed in the registry.

The trial compares telehealth CBT-CP delivery approaches with usual care plus information, but the registry record does not report enough details about implementation, adherence, missing data, or adjustment procedures to resolve all interpretive questions. The usual care group could access pharmacologic and nonpharmacologic treatments without restriction, so the comparison is against that care context rather than against no care.

Statistical evidence and practical importance are different considerations. A confidence interval and p-value characterize statistical uncertainty under an analysis; they do not determine whether the size of a difference matters to patients or clinicians. The registry defines the MCID threshold, but the record does not establish how the reported group-level comparisons translate into individual decisions or broader clinical value. No medical recommendation follows from these statistics alone.

18. Why This Trial Matters Statistically

Statistically, this trial illustrates the difference between reporting group-level outcome percentages and reporting comparative effects. The adjusted percentages describe the share meeting the threshold in each group, while relative risks compare those shares on a ratio scale. An unadjusted recalculation can use another scale, such as a risk difference, but should not be mistaken for the sponsor's adjusted model-based analysis.

The three-arm design also makes the choice of comparison important. The trial included an online program, a health coach-led program, and usual care plus information. Comparisons against usual care address intervention-versus-control questions; comparing the two intervention arms addresses a delivery-format question. Those contrasts have separate estimates and uncertainty, and should not be collapsed into a single claim about the interventions.

Use of a thresholded binary endpoint provides an interpretable measure of how many participants reached a defined level of improvement. At the same time, thresholding discards information about the amount of score change. Reading the binary endpoint alongside the other pain-related outcomes would provide a fuller account, but detailed numeric results for several named secondary measures are not reported in the registry record.

Finally, the record offers an opportunity to distinguish inferential results from design planning. Planning illustrations based on observed effects can show how required sample sizes change when a smaller effect is targeted, but they do not document the original design rationale and are not evidence of the study's observed power. Clear separation of these concepts helps prevent post hoc planning calculations from being mistaken for trial design assumptions.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The registry record leaves important design and analysis details unspecified. It does not report the original sample-size justification, the identities of those masked, the adjustment variables for the reported percentages or regression comparisons, or the variance-estimation details for the modified Poisson regression. The record also does not state how multiplicity was handled, how missing outcome data were treated, or which analysis population was used.

For the primary endpoint, the thresholded yes-or-no outcome does not describe the full distribution of pain-severity changes. The record names several other secondary outcomes, but detailed numeric estimates for several of those measures are not reported in the registry record. The later MCID outcomes add follow-up context, but they remain related threshold-based measures from the same trial rather than independent evidence.

The comparator allowed participants unrestricted access to available pharmacologic and nonpharmacologic treatment. That reflects the described usual-care context, but the registry record does not quantify the treatments participants received outside the trial programs. It also does not provide adherence, completion, or engagement details for the online or coach-led sessions. These omissions make it difficult to distinguish the effects of program assignment from differences in participation or background care.

The study is identified as multicenter, but additional information about sites, participant characteristics, and generalizability is not reported in the registry record. The record provides descriptive adverse-event counts but not enough detail to assess event timing, attribution, or comparative safety. These limitations constrain interpretation of mechanisms, implementation, and external applicability; they do not erase the reported outcome estimates, but they qualify the conclusions that can be drawn from them.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Tailored Non-Pharmacotherapy Services for Chronic Pain — NCT04523714.
• Registry record last updated 2026-01-22; status last verified 2026-01.
• Results first posted 2026-01-22.
• 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

The completed RESOLVE randomized, parallel-group trial enrolled 2333 participants with chronic pain. It compared an online CBT-CP-based program and a virtual health coach-led CBT-CP-based program with usual care plus information. The primary endpoint was the adjusted percentage meeting the defined pain-severity MCID from baseline to 3 months.

The registry reports adjusted percentages of 26.6% for the online group, 32.0% for the health coach-led group, and 20.8% for usual care plus information. Modified Poisson regression comparisons reported relative risks above 1 for both intervention groups versus usual care plus information, and for online versus health coach-led. Confidence intervals and p-values are reported for these comparisons.

Several details needed to assess multiplicity, missing data, analysis populations, and model implementation are not reported in the registry record. The findings describe the posted analyses and should be interpreted with those reporting limits in mind.