Tailored Self-Management of Temporomandibular Disorders (TMD) Using Health Information Technology: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Tailored Self-Management of Temporomandibular Disorders (TMD) Using Health Information Technology (NCT03036215), a randomized na study. The primary outcome is Assessing Graded Chronic Pain Severity Evaluation Feasibility of the PACT Program.

2. Clinical Question

3. Trial Design

This was a completed, randomized, parallel-group pilot registered as NCT03036215. The registry lists an enrollment of 80 and identifies the study as a behavioral intervention; phase is listed as NA. The trial compared PACT with traditional self-care. Masking is listed as none, so the record does not describe a masked comparison. The sponsor was HealthPartners Institute.

The registry describes the project as a feasibility and acceptability study intended to inform a subsequent trial of tailored self-management. PACT is described as a personalized web-based program of exercise and behavioral changes, supported by a health coach. Its care description calls it an 8-week progressive training program. Traditional self-care involved changes in daily behaviors and usual dental care, including possible use of a splint or anti-inflammatory medications.

The study start date is 2017-05-08, and primary completion is listed as 2017-11-09. The status is completed. The primary outcome timeframe is change from baseline to 16 weeks post intervention. For the posted primary outcome, the registry lists 39 participants in the control arm and 36 in the PACT arm; these are the group sizes associated with the reported outcome, not a statement that every enrolled participant contributed outcome data.

The registry record does not report the original sample-size justification, the randomization procedure, allocation concealment, or detailed eligibility criteria. It also does not report whether the study had interim analyses, the original statistical analysis plan, or the analysis population definitions. Those absences limit what can be inferred about the design and conduct beyond the randomized, parallel, non-masked pilot description.

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 Assessing Graded Chronic Pain Severity Evaluation Feasibility of the PACT Program, measured over Change from baseline to 16 weeks post intervention.

The primary endpoint is change from baseline to 16 weeks post intervention in Graded Chronic Pain Scale pain severity. The record describes the measure as mean intensity ratings for current, worst, and average pain, with three items scored on a 0–100 scale, where 100 is most severe. The reported unit is units on a scale, and the outcome is summarized as a mean with a standard deviation.

Pain severity is relevant to the stated aim because the PACT program was designed to support self-management of TMD pain. Measuring change from baseline allows the reported outcome to describe how scores differed over the intervention follow-up, rather than only reporting a single follow-up score. Comparing mean changes between randomized arms estimates the observed difference in average change between the groups.

That contrast is a continuous-outcome summary: it describes the average, not how every participant's pain changed. A mean can conceal variation among individuals, which is why the group standard deviations and uncertainty around the arm difference are important. The registry does not describe how the three pain ratings were combined into the endpoint, or the exact subtraction convention used to define change. Accordingly, the negative reported values can be stated as negative changes, but their coding details are not reported in the registry record.

The endpoint is also described in the context of feasibility and acceptability. The posted pain severity result is an outcome measure, but it does not itself establish whether recruitment, adherence, acceptability, or other feasibility targets were met. Those targets and their results are not reported in the registry record.

7. Results Posted in the Registry

• PACT Experimental Arm: mean -19.00 versus -12.00 on control; difference -7.00 (95% CI -14.42 to 0.42), standardized effect d = -0.43; z-test p = 0.065.

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.

Graded Chronic Pain Interference

• 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:
  • PACT Experimental Arm: mean -11.00 versus -5.80 on control; difference -5.20 (95% CI -13.79 to 3.39), standardized effect d = -0.27; z-test p = 0.235.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between PACT Experimental Arm and the control. The interval gives the range of effects that remain compatible with the data.

The registry identifies Graded Chronic Pain Interference as a secondary outcome and posts results for that measure. Its recalculated mean difference is -5.2 units, with a confidence interval from -13.79 to 3.389 and an unadjusted p-value of 0.235. The interval includes zero, so the estimate does not precisely distinguish a difference between arms from no difference.

At the level of point estimates, the secondary result has the same sign as the primary result when expressed as PACT minus traditional self-care. This broad directional similarity does not establish that the outcomes agree statistically or that the intervention has a consistent effect across pain domains. Both estimates are uncertain, and their intervals include zero. The endpoint measures different aspects of the reported pain experience, so they should not be treated as interchangeable outcomes.

Unadjusted secondary findings carry less weight when several outcomes or analyses are examined. Each additional test creates another opportunity for a result to appear notable by chance, even if no underlying difference exists. A multiplicity adjustment or a prespecified hierarchy can help address that issue, but neither is described in the registry record. The unadjusted p-value therefore should not be read as a definitive secondary-outcome conclusion.

The registry lists three outcome measures and reports three analyses, while one secondary outcome is identified in the record. The additional measures and their results are not described in the available outcome information. There is also no reported protocol hierarchy or account of which outcomes were intended to be confirmatory. Without that context, the secondary result is best treated as descriptive and supportive rather than as independent confirmation of the primary finding.

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

• Graded Chronic Pain Interference

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous and is reported as a mean with a standard deviation. A mean difference compares the two arm-level average changes. In the recalculation, the contrast is expressed as the PACT mean minus the traditional self-care mean: -19.0 minus -12.0 is reported as -7.0. This is an unadjusted contrast based on the posted arm summaries, rather than a covariate-adjusted estimate from individual participant data.

The standard deviation describes how spread out individual outcome values are around the arm mean. It is not the same quantity as the standard error. The primary arm standard deviations are 16.8 for traditional self-care and 16.0 for PACT. The recalculation gives a pooled standard deviation of 16.42. Pooling combines the two arm spreads into a common scale for describing the outcome variation; it does not mean that every participant had similar scores. The record does not report the sponsor's method for estimating the treatment contrast, so the pooled-standard-deviation method should be understood as part of the posted recalculation, not as a claim about the sponsor's analysis.

A standard error describes the sampling uncertainty of an estimate, rather than the variation among individual participants. The reported standard errors for the primary arm means are 2.69 for control and 2.667 for PACT. Standard errors help determine how much uncertainty surrounds an estimated mean or contrast. They generally become smaller when data are more informative, such as when there are more observations or less variability. In this pilot, the arm-level standard deviations are appreciably larger than the arm-level standard errors, which is expected because these quantities describe different sources of variation.

A confidence interval gives a range of values compatible with the estimate under the statistical procedure and its assumptions. For the recalculated primary difference, the interval runs from -14.42 to an upper bound of 0.424. Because zero lies in that interval, the result does not clearly separate a between-arm difference from no difference. A confidence interval is not a guarantee that the true effect lies within its limits, nor does it give the probability that a particular value is true. Its width conveys precision: wider intervals leave more uncertainty about the size and direction of the underlying difference.

The unadjusted p-value for the primary recalculated contrast is 0.065. A p-value describes how unusual data at least as incompatible with a specified null model would be, if that model and the test assumptions held. It is not the probability that the null hypothesis is true, and it does not measure the importance of an effect. The p-value is unadjusted, so it does not account for testing multiple outcomes or other analyses. The registry does not report whether the sponsor used the same test, any covariate adjustments, or a prespecified decision threshold for interpreting results.

Randomization supports a comparison between assigned groups, but does not by itself resolve all potential problems in a pilot analysis. Group sizes associated with the primary outcome are 39 and 36, while the registry enrollment is 80. The record does not explain the difference between these counts, report participant flow, or specify the analysis population. It is not reported in the registry record whether all randomized participants were analyzed as assigned, whether missing outcome values were imputed, or whether any participants were excluded from the outcome analysis.

Over-dispersion, in a general statistical sense, means that the observed variability is greater than a model assumes. When a method underestimates variability, standard errors can be too small and confidence intervals can be too narrow. The primary outcome is continuous, and the posted arm standard deviations describe its observed spread. The registry does not report a model-based dispersion check or any correction for over-dispersion; the term is therefore a general consideration, not a reported issue in this trial.

Planning calculations are distinct from the trial's observed analysis. The registry record includes a planning illustration based on the observed effect, using a two-sided significance level of 0.05, equal allocation, and a normal approximation. It considers effects set to stated fractions of the observed effect. These assumptions describe that illustration only. The original trial's sample-size assumptions and justification are not reported in the registry record, and the planning figures do not establish the power of the completed pilot.

The record does not report a detailed statistical analysis plan, whether outcomes were prespecified in a protocol, handling of multiplicity, or interim analyses. These omissions matter because analytic choices can affect estimates and uncertainty. The posted recalculations offer a transparent unadjusted reading of the summary statistics, but they cannot substitute for participant-level analyses or recover methods that were not registered in the available record.

11. Confidence Intervals and Precision

Precision concerns how tightly an estimate identifies the range of plausible values. The primary recalculated mean difference is -7.0 units, and its confidence interval extends from -14.42 to 0.424 as displayed to three decimal places for the upper bound. The interval includes zero, and spans values consistent with a negative difference as well as a small positive difference. The estimate therefore leaves uncertainty about both whether the arms differ and the size of any difference.

The interval's width should be considered alongside the point estimate. A point estimate alone can appear decisive even when the data permit a broad range of underlying effects. Here, the confidence interval indicates limited precision for distinguishing the observed contrast from no difference. The p-value of 0.065 is another summary of the data relative to a null model, but it does not replace the interval's direct account of uncertainty in the size and direction of the estimate.

For the secondary interference outcome, the recalculated mean difference is -5.2 units, with an interval from -13.79 to 3.389. This interval also includes zero and allows values in either direction. The record reports 39 control participants and 36 PACT participants for both posted outcomes. The standard deviations and standard errors indicate meaningful variation around the arm means, which contributes to uncertainty in the between-arm contrasts.

These intervals are recalculated from summary statistics and are unadjusted. They do not account for multiplicity, missing-data uncertainty, or any model adjustments that may have been used in an analysis not described in the registry. Confidence intervals quantify sampling uncertainty under their calculation assumptions; they do not address every source of uncertainty in a trial. The sponsor's posted analysis does not include confidence limits in the registry record, so the recalculated intervals should not be mistaken for sponsor-reported intervals.

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 sample-size figures are planning illustrations, not observed trial results. For these continuous outcomes, the figures are participants per arm, not total events. The calculation assumptions listed in the registry are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is set to the stated fraction of the observed effect. The original trial's design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the number of participants per arm that the planning calculation estimates would be needed to have an 80% chance of detecting the specified effect under those assumptions, if that effect were the true one. The 90% power column uses a higher chance of detecting that same specified effect, and therefore generally calls for more participants per arm. These are planning concepts, not guarantees: actual power depends on the true effect, variability, analysis choices, and the quality and completeness of the data.

Each planning row uses a different fraction of the observed effect. Planning for 75% of the observed effect requires more participants than planning for 100% of the observed effect; planning for 50% requires still more. This reflects a general statistical principle: smaller differences are harder to distinguish from random variation, so more information is needed to detect them at the same power and significance level. The same pattern appears in the secondary-outcome planning illustration.

This is not observed power. The planning figures do not measure what the completed pilot's analysis achieved, and they do not show that the trial was designed with these assumptions. The registry record does not report the trial's own original sample-size rationale.

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 = -7.000; pooled standard deviation σ = 16.421.
2. n = 2 × (1.96 + 0.842)² × 16.421² / -7.000² = 86.4, rounded up to 87 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 <- -7                      # observed difference in means
sigma <- 16.4213025695951                     # 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(3036215)
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 = 87 (80% power), 116 (90% power)
 75% of observed effect: n per arm = 154 (80% power), 206 (90% power)
 50% of observed effect: n per arm = 346 (80% power), 463 (90% power)
power.t.test: 87.4 
Simulated power at n = 87 per arm: 0.798 (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 reports sponsor analyses as net mean differences of -7 for the primary outcome and -5.2 for the posted secondary outcome. It does not include a method, confidence interval, or p-value for either sponsor analysis. The recalculated primary contrast from the posted arm means is -7.0, and the recalculated secondary contrast is -5.2. Thus, the point estimates agree at the displayed precision.

The recalculated estimates are simple, unadjusted comparisons of the posted arm summaries. They do not establish which analysis method the sponsor used, and they should not be presented as though they are the sponsor's complete analysis. The registry does not describe any adjustment for baseline values, covariates, clustering, or missing data in the sponsor results. Whether any such procedures were used is not reported in the registry record.

The record indicates no discrepancy between the posted sponsor point estimates and the recalculated point estimates. The sponsor's own analysis therefore remains the relevant reported analysis, while the recalculations add an unadjusted estimate of uncertainty from the available summary statistics. Since the sponsor results contain no inferential statistics, the recalculated confidence intervals and p-values should be clearly distinguished from the sponsor's reported values.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study examines more than one outcome, time point, subgroup, or analysis. With repeated testing, the chance of at least one apparently unusual result can increase even if there is no true effect. This is a general statistical concern. The registry lists three outcome measures and three analyses, and identifies Graded Chronic Pain Interference as a secondary outcome. It does not report a multiplicity adjustment, testing hierarchy, or other strategy for controlling the interpretation of multiple results.

The primary and secondary p-values discussed in the posted results are unadjusted. They should be interpreted as individual comparisons rather than as results corrected for the full set of analyses. The registry record does not clarify whether all planned outcomes and analyses were posted, whether additional exploratory analyses were conducted, or how outcome selection was handled. Those details are not reported in the registry record.

Missing-data handling is also not described. The registry gives an enrollment of 80, while the primary and secondary outcome summaries list 39 participants in control and 36 in PACT. The record does not explain the participant counts, state how many participants were randomized, give reasons for missing outcomes, or describe imputation or other missing-data methods. Without this information, it is difficult to assess whether the reported outcome groups represent all randomized participants or whether missingness could have influenced the estimates.

Analysis populations are not specified. The record does not say whether the outcomes were analyzed according to assigned group, restricted to participants who completed the intervention, or based on another population. It also does not provide a protocol-defined analysis plan or identify how deviations from such a plan were handled. These omissions do not invalidate the posted summary statistics, but they limit assessment of bias and reproducibility.

Interim analyses and any stopping rules are not reported. Baseline adjustment, covariate selection, and sensitivity analyses are also not described. These unknowns should not be filled in by assumption: the registry record supports an unadjusted reading of the posted summaries, but not a detailed reconstruction of the sponsor's analysis strategy.

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 reports zero serious adverse events among 41 participants at risk in traditional self-care and zero among 39 participants at risk in PACT. It also reports zero other adverse events in both groups: 0 among 41 at risk in control and 0 among 39 at risk in PACT. Deaths are reported as 0 in each arm. No serious adverse-event terms are listed.

These figures mean that no events were recorded in the registry's adverse-event categories for the stated numbers at risk. They do not establish that no symptoms or other events occurred outside the reported categories, and they do not prove that the risk of an event is zero. A pilot with limited participant numbers can provide only limited information about uncommon events. The registry record does not give definitions, event-collection procedures, severity classifications beyond the categories shown, or duration of safety follow-up. The counts should therefore be read as the adverse-event reporting available in the record, not as a comprehensive conclusion about safety.

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

The posted results describe the average changes recorded in the two randomized groups and an unadjusted estimate of their difference. For pain severity, the recalculated difference is -7.0 units, with a confidence interval that includes zero. For pain interference, the recalculated difference is -5.2 units, and its interval also includes zero. Both estimates are uncertain, so neither provides a precise separation of the groups in these posted summaries.

These results do not show that PACT did or did not reduce pain compared with traditional self-care. The study is described as a feasibility and acceptability pilot for a subsequent trial, and its outcome summaries alone do not answer whether the program was feasible or acceptable. Those feasibility and acceptability outcomes are not reported in the registry record. The record also does not report a confirmatory efficacy analysis plan or original sample-size justification.

The mean differences represent group averages. They do not reveal how individual participants responded, whether outcomes varied by baseline characteristics, or whether some participants improved while others did not. The registry does not provide individual-level data or subgroup results. The endpoint change convention is also not described, so the sign of the mean differences should not be converted into a claim about the direction of clinical benefit without that detail.

Confidence intervals and unadjusted p-values describe statistical uncertainty under particular assumptions; they do not determine clinical importance. The record does not provide a threshold for a clinically meaningful difference on the reported scale. It also lacks information about missing data, analysis populations, multiplicity, and detailed statistical methods, which limits the strength of conclusions that can be drawn from the posted results.

The most defensible reading is descriptive: the registry reports negative mean changes in both groups, a negative PACT-minus-control contrast for each posted outcome, and uncertainty that includes no between-arm difference. These data can inform the design and questions of a larger study, but they do not by themselves settle whether the program has an effect.

18. Why This Trial Matters Statistically

This trial is statistically informative in part because it illustrates the distinction between a point estimate and its precision. The primary contrast is -7.0 units, but its confidence interval includes zero and permits effects on either side of that value. A single estimate can therefore give an incomplete account of what the data support. Interval width and the underlying spread of the outcome matter when interpreting a pilot comparison.

The pilot's stated purpose was to examine feasibility and acceptability of methods for a subsequent clinical trial. That role affects how the pain estimates should be read. A pilot can generate estimates of variability and possible effect size that may help with planning, but those estimates can be unstable, especially when participant numbers are limited. The registry's planning illustration makes this relationship explicit by showing that targeting smaller fractions of the observed effect requires more participants per arm.

The record also demonstrates why statistical transparency matters. Sponsor point estimates are posted, but the sponsor's methods, confidence intervals, and p-values are not. Recalculations from arm summaries can add context, yet they cannot replace a prespecified analysis or explain missingness, multiplicity, or the analysis population. Those details determine how reproducible and interpretable an estimate is.

Finally, the primary and secondary outcomes address related but distinct aspects of pain, and both recalculated intervals include zero. Treating the outcomes as multiple pieces of descriptive evidence, rather than as isolated tests for a definitive answer, is important. Clear reporting of outcome definitions, analysis choices, participant flow, and feasibility targets would strengthen interpretation of a later trial.

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 the exact change-score convention or explain how the three GCPS pain ratings were combined. This limits interpretation of the sign and construction of the primary endpoint. It also does not report a threshold for clinical importance, so the numerical size of the difference cannot be judged against a stated meaningful-change criterion.

Outcome group sizes are 39 for traditional self-care and 36 for PACT, while the registry enrollment is 80. Participant flow and reasons for the difference are not reported. Missing-data methods and analysis populations are not reported either. These gaps make it difficult to assess completeness of the outcome data or whether the reported groups reflect the intended randomized comparison.

The sponsor's reported analysis contains point estimates but no confidence intervals, p-values, or method description. The recalculated estimates are unadjusted and rely on posted summary statistics, so they cannot account for individual-level covariates or clarify any analysis decisions. The registry also does not report the original sample-size rationale, a detailed statistical plan, multiplicity handling, interim analyses, or the treatment of protocol deviations.

The study is described as a pilot, and the posted pain outcomes do not include the feasibility or acceptability findings named in the study purpose. One secondary outcome is posted, while the registry lists three outcome measures. The other outcome details are not available in the record. The relatively limited outcome group sizes and wide intervals leave substantial uncertainty about the effect estimates, and the adverse-event counts alone cannot characterize uncommon events or unreported safety information.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Tailored Self-Management of Temporomandibular Disorders (TMD) Using Health Information Technology — NCT03036215.
• Registry record last updated 2018-06-15; status last verified 2017-01.
• Results first posted 2018-06-15.
• 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

NCT03036215 is a completed randomized, parallel-group, non-masked pilot sponsored by HealthPartners Institute. It compared an 8-week tailored, web-based PACT self-management program supported by a health coach with traditional self-care for adults with TMD pain. The registry lists enrollment of 80 and a primary outcome of change in GCPS pain severity from baseline to 16 weeks post intervention.

The recalculated primary mean difference is -7.0 units, with an interval that includes zero. The posted secondary outcome, pain interference, also has an interval including zero. Sponsor analyses report net mean differences of -7 and -5.2, respectively, without inferential statistics. Feasibility and acceptability results, detailed analysis methods, missing-data handling, and the original sample-size justification are not reported in the registry record.