This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Cognitive Behavioral Therapy and Real-Time Pain Management Intervention for Sickle Cell Via Mobile Applications (NCT04419168), a randomized na study. The primary outcome is 6-month Change in PROMIS Pain Interference.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT04419168 |
| Phase | NA |
| Condition | Chronic Pain, Depression, Anxiety, Opioid Use |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Single |
| Enrollment | 359 |
| Arms in the primary outcome | 2 |
| Lead sponsor | University of Pittsburgh |
| Results posted | Yes |
2. Clinical Question
Population
Adults with sickle cell disease who report chronic pain. The registry lists chronic pain, depression and anxiety, and opioid use among the trial conditions. The brief summary describes a comparative effectiveness trial in this population.
Intervention
Computerized cognitive behavioral therapy delivered by mobile phone. The program is described as teaching users to recognize negative thoughts and emotions, use cognitive skills and problem-solving, and apply coping behaviors such as distraction, activity scheduling, and relaxation. The intervention emphasizes skills.
Comparator
Mobile-delivered education about pain and sickle cell disease, including chronic pain, healthy lifestyle tips, and details about sickle cell disease. Both groups were also to receive weekly health-coach follow-up, more frequent if requested or needed, for at least 3 months.
Primary question
At 6 months, how did PROMIS Pain Interference change from baseline in the cCBT and m-Education groups, and what between-group difference did the registry report? The scale reflects the effect of pain on daily activities and other aspects of life; higher T-scores indicate greater pain interference.
3. Trial Design
The registry describes a randomized, parallel-group trial with single masking and lists its phase as NA. It records an enrollment of 359 and a completed status. The brief summary describes 350 adults with sickle cell disease and chronic pain, with planned assignment of 175 to mobile computerized cognitive behavioral therapy and 175 to mobile education. For the primary outcome, the posted arm records list n=181 for cCBT and n=178 for m-Education. The registry does not explain how these enrollment and outcome counts relate to one another.
The interventions were delivered by mobile application. The cCBT program focused on recognizing thoughts and emotions, cognitive skills, problem-solving, and coping behaviors. The education program addressed chronic pain, lifestyle tips, and sickle cell disease. Both arms also had health-coach follow-up for at least 3 months, with more frequent contact if requested or needed. Participants were to track pain, mood, and medication use during two-week periods at baseline and at 3, 6, and 12 months. They were also to have access to a study-associated online support group. Routine care, including opioid pain management and novel therapies, was to continue.
The listed primary endpoint is change in PROMIS Pain Interference from baseline to 6 months. The registry also lists secondary assessments, including pain intensity, depression, anxiety, quality of life, self-efficacy, pain interference, and healthcare use. The primary analysis is described as a mixed models analysis of mean difference in final values. The registry record does not report the identities of the people masked, allocation concealment procedures, or additional details of the randomization process.
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
| Stage | cCBT | m-Education |
|---|---|---|
| STARTED | 181 | 178 |
| 3-month Follow up Visit | 134 | 135 |
| COMPLETED | 125 | 125 |
| NOT COMPLETED | 56 | 53 |
Reasons for not completing this period:
| Reason | cCBT | m-Education |
|---|---|---|
| Protocol Violation | 1 | 1 |
| Lost to Follow-up | 7 | 7 |
| Missed 6-month Visit | 48 | 45 |
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.
| Characteristic | Unit | cCBT | m-Education | Total |
|---|---|---|---|---|
| Participants analysed | 181 | 178 | 359 | |
| Age, Continuous | years | 36.4 (SD 9.9) | 36.3 (SD 11.1) | 36.3 (SD 10.5) |
| Sex/Gender, Customized: Biological Sex – Male | Participants | 69 | 49 | 118 |
| Sex/Gender, Customized: Biological Sex – Female | Participants | 111 | 128 | 239 |
| Sex/Gender, Customized: Biological Sex – Prefer not to answer | Participants | 1 | 1 | 2 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 6 | 9 | 15 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 156 | 152 | 308 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 19 | 17 | 36 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 | 1 | 1 |
| Race (NIH/OMB): Asian | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Black or African American | Participants | 169 | 163 | 332 |
| Race (NIH/OMB): White | Participants | 0 | 2 | 2 |
| Race (NIH/OMB): More than one race | Participants | 7 | 8 | 15 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 5 | 4 | 9 |
| PROMIS Pain Interference | T-score | 62.2 (SD 7.3) | 62.9 (SD 7) | 62.6 (SD 7.1) |
| Average pain intensity | units on a scale | 4.2 (SD 2.5) | 4.5 (SD 2.5) | 4.4 (SD 2.5) |
| Patient Health Questionnaire (PHQ) | units on a scale | 11 (SD 5) | 10 (SD 5) | 10 (SD 5) |
| Generalized Anxiety Disorder Scale | units on a scale | 9 (SD 5) | 9 (SD 5) | 9 (SD 5) |
| Adult Sickle Cell Quality of Life Measurement Information System (ASCQ-ME): Social Functioning | T-score | 47.4 (SD 8.6) | 47.1 (SD 7.1) | 47.3 (SD 7.9) |
| Adult Sickle Cell Quality of Life Measurement Information System (ASCQ-ME): Emotional Impact | T-score | 47.2 (SD 8.8) | 48.8 (SD 8.8) | 48.0 (SD 8.8) |
| Sickle Cell Self-Efficacy Scale (SCSES) | units on a scale | 28 (SD 6) | 29 (SD 7) | 29 (SD 7) |
| Total Body Area Shaded (Painimation) | mean percentage of total body area shade | 13 (SD 14.7) | 14 (SD 15.4) | 13.5 (SD 15.0) |
| Current Opioid Misuse Measure (COMM) | units on a scale | 9 (SD 6) | 8 (SD 5) | 9 (SD 5) |
6. Primary Endpoint
The registry lists the primary outcome as 6-month Change in PROMIS Pain Interference, measured over Baseline and 6 months.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | T-score |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is change in PROMIS Pain Interference from baseline to 6 months. The measure uses the 8-item PROMIS-8a instrument to assess how pain affects relevant parts of a person’s life, including social, cognitive, emotional, physical, and recreational activities during the past 7 days. Each item uses a 1–5 Likert scale. The raw total ranges from 8 to 40 and is transformed to a T-score using a reference-population mean of 50 and standard deviation of 10. Higher T-scores indicate greater pain interference.
A change score describes movement from baseline rather than the absolute level at follow-up. In this trial, negative arm-level mean changes indicate lower scores at 6 months on average than at baseline; because higher scores represent greater interference, this direction is consistent with reduced measured interference. The arm-level estimates are reported with 95% confidence intervals. Those intervals convey uncertainty around each estimated mean change, rather than the range of individual participants’ changes.
The comparison between arms is the main evidence for whether the average change differed between the intervention groups. An improvement within each arm does not, by itself, establish that one intervention led to a greater change than the other. The registry reports the between-group analysis using a mixed models approach.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
For PROMIS Pain Interference, the posted mean change was -2.13 T-score points in the cCBT group, with a 95% confidence interval from -3.42 to -0.84. In the m-Education group, the mean change was -2.66, with a 95% confidence interval from -3.97 to -1.36. Both estimated changes are negative, corresponding to lower average pain-interference scores at 6 months than at baseline. The individual arm estimates do not answer whether the average change differed between the groups.
The registry’s mixed models analysis reports a mean difference in final values of .54, with a 95% confidence interval from -1.30 to 2.37 and p=.57. The interval includes zero, so the posted comparison does not establish a clear between-group difference in the primary endpoint. Its breadth also leaves room for differences in either direction; the estimate should not be read as proof that the interventions are equivalent or that any possible difference is absent.
The secondary results include rates of opioid prescriptions, emergency department visits, and hospitalizations over 12 months. The registry’s negative binomial regression analyses report uncertainty intervals that include the rate-ratio null value of 1 for these comparisons, and the reported p-values do not provide clear evidence of a difference. These secondary analyses should be interpreted cautiously, particularly because the registry record does not report whether multiplicity was addressed. They do not overturn or sharpen the primary comparison.
Arm-level estimates and between-group estimates answer different questions. The arm-level confidence intervals describe estimated mean changes within each group. The mixed-model estimate addresses the comparison between groups, with its own uncertainty. Overlap or non-overlap of separate arm intervals is not a substitute for the reported between-group analysis.
Overall, the posted primary result is compatible with a range of between-group effects, and it does not show clear evidence that mean pain interference changed differently between cCBT and m-Education at 6 months. That conclusion is specific to the registered primary measure and analysis. It does not establish that the programs are interchangeable, and it does not answer every clinical or behavioral question that might be relevant to people with sickle cell disease and chronic pain.
9. Statistical Methods Explained
The primary outcome is continuous: it is a T-score, and the registry reports mean change by arm. Continuous outcomes are commonly described with a mean and a measure of uncertainty. Here, the arm-level confidence intervals quantify uncertainty in the estimated mean changes. They are not prediction intervals for individual participants, and they do not show where most individual scores fall.
What does the between-group estimate represent?
The primary comparison is reported as a mean difference in final values from a mixed models analysis. Mixed models are a general approach for analyzing measurements that may be related within a participant, such as observations collected at different times. A model can use information across observations to estimate the group comparison while accounting for within-person correlation. The registry identifies the method but does not report the exact model specification, covariance structure, covariates, or handling of incomplete measurements. Those details are not reported in the registry record.
The reported 95% confidence interval for the primary comparison runs from -1.30 to 2.37, and the p-value is .57. A confidence interval gives a range of values compatible with the data under the assumptions of the analysis; it is not a guarantee that the true effect lies within the interval. For a difference, zero is the null value of no between-group difference. Because the interval crosses zero, the result does not provide clear statistical evidence of a difference. A non-significant p-value does not establish equality: the interval still includes effects in either direction.
How should the count outcomes be understood?
The secondary healthcare-use outcomes are counts expressed as rates per person-year. A rate accounts for the amount of time represented by the observation, which can matter when participants contribute different amounts of follow-up. The registry reports sponsor analyses using negative binomial regression and expresses the comparison as an incidence rate ratio. For a rate ratio, 1 is the null value: values on either side correspond to different estimated rates, while uncertainty intervals that include 1 do not clearly distinguish the groups.
Count data often have variability that exceeds what a simple Poisson model expects. This is called over-dispersion. A negative binomial model can accommodate extra variation and may be more suitable when counts vary more than a Poisson model allows. The registry names negative binomial regression for its reported analyses. The simple recalculations associated with the posted secondary rate summaries use a Poisson-based approach; those are unadjusted comparisons and are not replacements for the sponsor’s reported model. Differences in modeling assumptions can produce estimates and uncertainty intervals that are not identical.
The recalculated comparisons are useful as an independent descriptive check on the posted arm summaries, but they do not reproduce the full sponsor analysis. The sponsor’s methods and estimates remain the reported inferential results. The registry record does not provide further detail about model diagnostics, covariate adjustment, or whether the negative binomial model’s assumptions were assessed.
What do the planning illustrations mean?
Planning illustrations use the observed effect as a reference and consider detecting a stated fraction of that effect. The posted assumptions specify a two-sided significance level of 0.05, equal allocation, and a normal approximation. These are illustrative calculations, not the trial’s original design justification and not observed power. A sample-size estimate based on an observed effect is especially uncertain because the estimated effect itself can be imprecise. Smaller assumed effects require more participants per arm in the illustration. These calculations do not establish that the trial was adequately or inadequately powered for its actual design.
The registry does not report the original sample-size justification, whether there were interim analyses, or the planned statistical analysis protocol. The planning illustrations therefore should not be used to infer what effect the investigators originally intended to detect, what power the completed trial achieved, or whether the study met a prespecified target.
What information is missing for interpretation?
Although the registry reports the primary model type and several secondary model results, it does not report enough detail to reconstruct the complete analysis. The exact definition of the analysis population, the approach to missing outcome data, and the treatment of departures from the assigned intervention are not reported in the registry record. Nor does the record state whether adjustments were made for multiple comparisons. These gaps limit assessment of robustness and mean that the reported estimates should be read as posted, rather than as a fully reproducible account of every analytic decision.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Mixed Models Analysis | Mean Difference (Final Values) .54 | -1.30 to 2.37 | 0.570 |
The sponsor’s reported primary analysis is a mixed models analysis of the mean difference in final values. It gives an estimate of .54, a 95% confidence interval from -1.30 to 2.37, and p=.57. That is the main reported comparison for the primary endpoint. The arm-level mean changes and their confidence intervals provide context, but they are not a substitute for the model-based comparison.
For the secondary rate outcomes, the sponsor reports negative binomial regression estimates as incidence rate ratios. The accompanying simple recalculations are unadjusted comparisons derived from the posted summary statistics and use a Poisson-based approach. They can help readers see how the arm-level rates relate to a basic rate comparison, but they do not account for all features of the sponsor’s model and should not be treated as the sponsor’s analysis.
The registry record does not flag a discrepancy between the sponsor analyses and the recalculations. Even so, the model-based analyses are the sponsor’s reported inferential results; the recalculations are supplementary statistical checks. Details such as model specification, covariate adjustment, and missing-data handling are not reported in the registry record, so the analysis cannot be fully reconstructed from these summaries alone.
11. Multiplicity, Missing Data and Analysis Populations
The registry lists a primary outcome and multiple secondary outcomes, with 20 analyses recorded. Testing multiple outcomes can increase the chance that at least one result appears statistically notable even when there is no underlying effect. Common approaches include adjusting significance thresholds, controlling a family-wise error rate, controlling a false discovery rate, or specifying a hierarchy in which some outcomes are tested only after earlier criteria are met. Whether any such approach was planned or used is not reported in the registry record.
The primary comparison should therefore remain the central result, while secondary findings warrant cautious interpretation. The registry lists 17 secondary outcomes, including patient-reported measures and healthcare-use outcomes. Their different constructs and assessment timeframes make it important not to treat them as repeated measurements of a single endpoint. The record does not explain which secondary outcomes, if any, were considered confirmatory.
Missing-data handling is also not reported in the registry record. The record does not describe the extent or pattern of missing primary or secondary outcome observations, the assumptions used to handle missingness, or whether sensitivity analyses were conducted. These details matter because analyses based on incomplete follow-up can be affected if missingness is related to outcomes or group assignment.
The analysis populations are not fully described. The record does not state whether the primary model followed an intention-to-treat principle, how participants who did not adhere to their assigned program were handled, or how eligibility for each outcome analysis was determined. It reports arm-level primary outcome sample sizes of n=181 for cCBT and n=178 for m-Education, while the brief summary describes a planned sample of N = 350 and 175 per group and the registry lists enrollment of 359. The relationship among these counts is not explained in the registry record. They should not be combined or treated as interchangeable counts without further information.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| cCBT | 3 | 181 | 1.7% |
| m-Education | 1 | 178 | 0.6% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| cCBT | 3 of 181 |
| m-Education | 1 of 178 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| cCBT | 0 of 181 |
| m-Education | 0 of 178 |
Most frequent serious adverse event terms
| Event term | Organ system | cCBT | m-Education |
|---|---|---|---|
| General disorders and administration site conditions | General disorders | 3 of 181 | 0 of 178 |
| Cardiac Arrest | Cardiac disorders | 0 of 181 | 1 of 178 |
Serious events with the highest arm-level rate: general disorders and administration site conditions (1.7%), cardiac arrest (0.6%).
The registry reports serious adverse events for both groups. For cCBT, 3 serious adverse events are recorded among 181 at risk; for m-Education, 1 is recorded among 178 at risk. The record also reports 0 other adverse events in each group. These are the posted event counts, not a complete description of event timing, severity beyond the serious classification, or causal attribution.
Deaths are recorded as 3 for cCBT and 1 for m-Education. The registry summary does not state whether the deaths overlap with the serious adverse event counts, provide causes, or establish a relationship to an intervention. It would therefore be inappropriate to assume that these are separate sets of events or to infer causation from the counts alone.
The listed serious-event terms include general disorders and administration site conditions at 1.7% and cardiac arrest at 0.6%. These categories and percentages should be read as registry safety reporting, not as evidence that an intervention caused an event. The record does not provide enough detail to assess individual cases, comparative risk with precision, or the completeness of safety follow-up. Safety findings require context on event definitions, ascertainment, and attribution, which is not reported in the registry record.
13. What the Result Does, and Does Not, Mean
The primary result describes the average change in reported pain interference from baseline to 6 months and the estimated difference between the randomized groups. Both groups had negative mean changes on the PROMIS Pain Interference measure, while the mixed models comparison had a confidence interval that included zero and a p-value of .57. The posted data do not establish a clear difference between cCBT and mobile education on this primary endpoint.
This is not the same as proving that the interventions have identical effects. The confidence interval permits a range of possible between-group differences, and the registry does not report an equivalence or non-inferiority design with a prespecified margin. A result that does not provide clear evidence of superiority cannot be recast as proof of equivalence.
The finding also does not show that neither group changed. Each arm’s estimated mean change was negative, and each arm-level interval is below zero. Those within-group estimates describe change over time in the participants represented in the analysis; they do not isolate the contribution of the assigned program from other influences. Both groups continued routine care, and both received additional study supports, including health-coach follow-up and access to an online support group. The comparison is between these packages as described, not between receiving an intervention and receiving no care.
The results are specific to the registered population, endpoint, assessment time, and analyses. They do not establish effects on every aspect of pain, mental health, quality of life, medication use, or healthcare use. Nor do the posted summaries establish longer-term effects beyond the reported outcome timeframes. The registry’s secondary comparisons do not supply clear evidence of differences in the listed healthcare-use rates, but they should be interpreted with caution because multiplicity and analytic details are not reported.
Clinical interpretation requires more than a p-value. The size and precision of the estimated effect, the measure’s meaning, the population studied, and the analysis methods all matter. The registry record gives useful outcome summaries but leaves important details about missing data, analysis populations, and the full statistical plan unreported.
14. Why This Trial Matters Statistically
This trial is statistically informative because it compares two active mobile approaches in adults with sickle cell disease and chronic pain. The comparator was not an untreated group: participants in both groups were to receive health-coach follow-up, use mobile tracking, have access to an online support group, and continue routine care. Consequently, the primary contrast addresses the difference between the cCBT and education programs within that broader study context.
The primary endpoint uses a patient-reported measure tied to the impact of pain on daily life. Reporting both arm-level changes and a between-group analysis helps distinguish observed change within each arm from evidence about a difference between assigned programs. The mixed models estimate and its confidence interval are more relevant to the comparative question than a comparison of whether each group changed from baseline in isolation.
The registry also illustrates why precision matters. The primary interval includes effects in either direction, so the absence of clear statistical evidence is not evidence that all meaningful differences have been ruled out. Secondary healthcare-use outcomes add another perspective, but their count-based nature and multiple testing create separate interpretive issues. The registry’s negative binomial analyses and the simpler rate recalculations also show why a statistical method’s assumptions matter when analyzing event counts.
Finally, transparency about analysis choices is essential. The registry names several methods but does not report the full statistical plan, missing-data procedures, multiplicity handling, or analysis populations. Those omissions do not erase the posted results, but they limit how completely readers can assess their robustness. Clear reporting of those details would support stronger interpretation of a trial that addresses a consequential chronic-pain question.
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.
The registry record does not provide enough detail to evaluate the full analysis plan. It does not report the exact mixed-model specification, covariate adjustments, assumptions about correlation across repeated measures, or model diagnostics. The negative binomial model is named for the secondary count outcomes, but the record does not describe its specification or assessment of assumptions. This makes it difficult to reproduce the sponsor analyses from the summaries alone.
Information about missing observations and analysis populations is also absent. The record does not explain how incomplete follow-up was handled, whether sensitivity analyses were performed, or whether participants were analyzed according to randomized assignment. The primary outcome arm counts and enrollment descriptions are not reconciled in the registry record, so their differing contexts cannot be determined from the summaries.
The masking label is single, but the identity of the masked parties is not reported. The record also does not describe allocation concealment, intervention adherence, intervention exposure, or whether participants used the programs as intended. These details could help readers judge possible sources of bias and the practical contrast between study arms.
Both groups received support beyond their assigned mobile program, and routine care continued. This is relevant to interpretation: the comparison reflects the programs within the described study context, not an isolated effect of cCBT against no intervention. In addition, the registry does not report a multiplicity strategy despite recording numerous outcomes and analyses. Secondary findings therefore need cautious interpretation.
The posted safety summary offers event counts and selected terms, but does not explain overlap between serious events and deaths, causes, attribution, or case-level details. These limitations are specific to the available registry record; they do not establish that the information was absent from other trial documents.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Cognitive Behavioral Therapy and Real-Time Pain Management Intervention for Sickle Cell Via Mobile Applications — NCT04419168.
- Registry record last updated 2025-02-19; status last verified 2025-01.
- Results first posted 2025-02-19.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
The registry records a completed randomized, parallel-group trial of mobile computerized cognitive behavioral therapy versus mobile education for adults with sickle cell disease and chronic pain. The primary endpoint was change in PROMIS Pain Interference from baseline to 6 months. The sponsor’s mixed models analysis reported a mean difference of .54, with a 95% confidence interval from -1.30 to 2.37 and p=.57; the result does not show clear evidence of a between-group difference.
Posted secondary analyses address opioid prescriptions, emergency department visits, and hospitalizations over 12 months. Their sponsor-reported negative binomial comparisons do not provide clear evidence of group differences. The registry record does not report the multiplicity approach, missing-data methods, or complete analysis-population definitions, limiting assessment of those results and their robustness.