New Technologies to Help Manage ADHD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for New Technologies to Help Manage ADHD (NCT02390791), a randomized na study. The primary outcome is Medication Continuity: The Number of Days Covered With Medicine.

2. Clinical Question

3. Trial Design

This was a completed, randomized, parallel-group, single-masked, cluster-randomized controlled trial sponsored by Children's Hospital Medical Center, Cincinnati. The registry reports enrollment of 85. The study start is listed as 2015-09 and primary completion as 2017-08. The phase is recorded as NA.

The comparison was between enhanced myADHDportal.com and the treatment as usual standard portal. The enhanced software included family-management support; the standard portal was the standard version of the same web software. The registry describes the study as evaluating the general effectiveness of the intervention compared with control on medication continuity and more proximal outcomes during the first 6 months of treatment.

Cluster randomization means assignment occurred at the cluster level rather than by treating every participant as an independently randomized unit. The cluster unit, number of clusters, allocation procedures, and the number of participants in each primary-outcome analysis group are not reported in the registry record. The identity of the masked party is also not reported in the registry record. These details matter for understanding the design and assessing how the analysis handled dependencies among participants within clusters.

The registry lists 10 outcome measures and 10 analyses. Its secondary outcomes include parent-reported beliefs about medication, decisional conflict, trust in the provider, working alliance, perceived social support, parent distress, parent-rated side effects, and parent log-ins to myADHDportal.com. The record does not give enough information to determine the analysis population or the cluster structure used for those outcomes.

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 Medication Continuity: The Number of Days Covered With Medicine, measured over first 6 months of treatment.

The primary endpoint is medication continuity, measured as the number of days covered with medicine during the first 6 months of treatment. The registry says this measure is calculated from an audit of prescriptions written. It is a continuous outcome, with the mean identified as the parameter of interest and the standard deviation as the dispersion measure.

A prescription-based measure can describe whether prescriptions provide coverage over time. It is not, by itself, a direct measure of whether medicine was taken on each covered day. The endpoint therefore represents continuity as operationalized by prescription records, rather than every aspect of medication use or the child's broader clinical status. This distinction is important when interpreting what a group comparison can address.

Medication continuity is relevant to the intervention's stated purpose because the enhanced portal was designed to support family involvement in maintaining continuity. The registry's description does not specify the detailed rules for translating prescription records into covered days, how overlapping prescriptions or gaps were handled, or how records unavailable for some participants were treated. Those details are not reported in the registry record. The posted analysis also does not include a numerical mean, group difference, or confidence interval for the endpoint, so the size and precision of any difference cannot be assessed from the reported primary result alone.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary result comes from a generalized linear model, as identified in the registry. The model family, link function, included predictors, variance assumptions, and handling of the cluster structure are not reported in the registry record. Those choices can affect how a model relates the observed data to an estimated group contrast, so the model label alone does not fully describe the analysis.

Why does cluster randomization matter?

In a cluster-randomized study, groups of participants are assigned together. Outcomes among people in the same cluster can be correlated because they share settings, providers, or other influences. General statistical reasoning therefore treats the cluster structure as part of the design, not as a minor detail. If a model treats all participants as independent when their outcomes are correlated, its uncertainty estimates may be too small. Methods that account for clustering can incorporate that dependence when estimating uncertainty. The registry does not report the cluster unit, cluster count, or how the sponsor's model addressed within-cluster correlation, so it is not possible to assess those details from the record.

What does the generalized linear model label tell readers?

In general, a generalized linear model is a flexible modeling framework for relating an outcome to explanatory variables. The particular form depends on the outcome and the model specification. The registry identifies the primary outcome as a mean of a continuous measure, but does not name the distributional family or link function. It also does not report covariate adjustment or the exact contrast used. Those elements should not be guessed from the method label.

How should p=0.04 be read?

A p-value is calculated under a statistical model and a null hypothesis. It indicates how unusual data at least as incompatible with that null would be if the model and null were correct. It does not quantify the size of the observed difference, the probability that the null is true, or the chance that the findings will recur. Here, the registry reports p=0.04 for a superiority analysis, but no effect estimate or confidence interval accompanies it. That limits interpretation of both the direction and precision of the result.

A confidence interval, in general, gives a range of values compatible with the data and the assumptions of the analysis, using a stated confidence procedure. It can communicate uncertainty about an effect estimate in a way that a p-value alone cannot. No confidence interval is reported for this primary result. Consequently, readers cannot use the registry result to judge the range of group differences reasonably compatible with the analysis.

What can planning illustrations show?

Sample-size planning illustrations are conditional calculations, not outcomes observed in the trial and not a measure of the trial's observed power. The planning assumptions described for these illustrations include a two-sided significance level of 0.05, equal allocation between arms, a normal approximation, and an effect to detect set as a chosen fraction of the observed effect. For time-to-event outcomes, the stated convention is a total number of events; other outcomes are expressed as participants per arm. These assumptions describe the illustrations rather than establishing the trial's original design rationale. The registry does not report the trial's sample-size justification or the assumptions used when planning the trial.

When planning uses an effect estimated from a completed study, the illustration depends on that estimate and on the chosen fraction of it. It does not demonstrate that the trial had a specified probability of detecting an effect, and it should not be described as observed power. Here, because the primary effect estimate is not reported in the registry record, the numerical basis for such an illustration cannot be evaluated from the reported primary result.

Why are re-calculations not presented?

A simple unadjusted re-calculation from summary means and standard deviations would not reflect the cluster-randomized structure. The registry explicitly notes that an individual-level re-analysis would be misleading and reports only the sponsor's own analysis. In general, a re-calculation from summary statistics is useful as a transparent check only when the required statistics and design assumptions support it. It cannot replace a cluster-aware analysis where dependence within clusters matters. In this record, the primary group summaries, effect estimate, and uncertainty interval are not reported, so there is no basis for an independent unadjusted calculation.

10. Confidence Intervals for Each Arm

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

R code and output

arm <- c("Enhanced myADHDportal.Com", "Treatment as Usual Standard Portal")
n   <- c(40, 45)
m   <- c(108.8, 133.3)
sp  <- c(45.8, 50.9)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Enhanced myADHDportal.Com  40  108.800  45.800  94.152  123.448
CARM  Treatment as Usual Standard Portal  45  133.300  50.900  118.008  148.592

11. Analyses Reported by the Sponsor

The registry reports the sponsor's own generalized linear model analysis for the primary medication-continuity comparison, with p=0.04 and a superiority classification. It does not report the estimated effect, confidence interval, or the model details needed to reproduce that analysis. Therefore, the result can be described, but its magnitude and precision cannot be independently checked from the primary outcome summaries in the record.

No simple individual-level re-calculation is presented. The registry identifies the trial as cluster randomized and states that an individual-level re-analysis would be misleading. That caution is important: a re-calculation that ignores within-cluster dependence could produce an inappropriate uncertainty estimate. The sponsor's analysis is thus the only inferential result reported for this comparison, rather than one estimate alongside an independent recalculation.

The posted information does not establish whether the sponsor's model fully accounted for the clustering or which other modeling choices were used. These details are not reported in the registry record. There is no reported numerical discrepancy between the sponsor's result and a recalculation to resolve. The result should consequently be read as the registry's sponsor-analysis p-value, not as a fully reproducible estimate of the group difference.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists 10 outcome measures and 10 analyses. Alongside medication continuity, the listed secondary outcomes cover parent-reported medication beliefs, decisional conflict, trust in the provider, working alliance, perceived social support, parent distress, parent-rated side effects, and parent log-ins. Considering multiple outcomes can increase the chance that at least one result appears notable by chance if many separate tests are conducted. General approaches include prespecifying a primary outcome, adjusting tests, or interpreting secondary findings as exploratory. The registry does not report whether a multiplicity adjustment was planned or used, or how the listed analyses relate to the listed outcomes.

The record identifies medication continuity as the primary outcome, but does not provide a complete statistical analysis plan or indicate whether other outcomes were designated as confirmatory. The p=0.04 result should therefore be interpreted in the context of the information reported, without assuming an adjustment or a particular testing hierarchy. Results for the listed secondary outcomes are not reported in the registry record.

Missing-data procedures are not reported in the registry record. The record does not describe the amount or pattern of missing outcome data, how unavailable prescription audits were handled, or whether any imputation or sensitivity analysis was performed. These details matter because the analysis can depend on which participants have observed outcome information and on assumptions about missing observations.

The analysis population is also not reported in the registry record. It is not possible to determine from the stated result whether all randomized participants were included, whether participants were analyzed according to assignment, or whether eligibility for the analysis depended on having particular data. The number of clusters and the way participants were represented in the model are not reported either. These omissions limit assessment of both the handling of missing data and the relationship between the randomized design and the primary analysis.

13. 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.

For the enhanced myADHDportal.com group, the registry records 0 serious adverse events among 40 participants at risk and 0 other adverse events among 40 at risk; deaths are recorded as 0. For the treatment as usual standard portal group, it records 0 serious adverse events among 45 participants at risk and 0 other adverse events among 45 at risk; deaths are also recorded as 0.

These are descriptive counts in the registry. They indicate that no events in the listed categories were recorded for the participants at risk in either group. They do not show that the probability of an event is zero, and they do not establish that the groups have equivalent safety profiles. With no events recorded, the registry counts alone do not provide a precise estimate of between-group differences. The record does not report event definitions beyond the serious and other categories, follow-up details for these counts, or a statistical comparison of adverse events. Safety conclusions should therefore remain limited to the counts recorded.

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

The result establishes that the registry reports a sponsor-conducted superiority analysis of medication continuity with p=0.04. The primary endpoint was days covered with medicine, based on an audit of prescriptions written during the first 6 months of treatment. The trial compared an enhanced portal with a standard portal and used cluster randomization.

The result does not establish the size or direction of the group difference. No group means, estimated contrast, or confidence interval are reported in the registry record. Without those quantities, the finding cannot be translated into a number of days, and its precision cannot be judged. The reported p-value does not indicate how important any difference might be in practice.

The endpoint also has a specific scope. Prescription-based coverage is not identical to taking medicine on each covered day, and the registry does not describe additional measures of medication-taking behavior for the primary endpoint. The result therefore concerns continuity as measured through prescription records, not every dimension of medication use or outcomes beyond that measure.

The primary outcome is distinct from the listed parent-reported and portal-use secondary outcomes. Their results are not reported in the registry record, so the primary p-value cannot stand in for findings on family beliefs, distress, provider relationships, side effects, or portal use. Similarly, the posted adverse-event counts are not a complete comparative safety analysis.

Finally, cluster randomization and the unreported details of the statistical model limit independent assessment. The registry cautions against a simple individual-level re-analysis. In the absence of information about clustering, missing data, multiplicity, and analysis populations, the result should be read narrowly as the reported model-based p-value, not as a complete account of the evidence or a conclusion about outcomes not measured by the primary endpoint.

15. Why This Trial Matters Statistically

This trial is statistically instructive because its allocation was cluster randomized. The design affects how outcome data should be analyzed: participants within a cluster may share influences, so their observations may not provide as much independent information as observations from separate clusters. Recognizing this dependence is central to obtaining appropriate uncertainty estimates. The registry's warning against an individual-level re-analysis makes the design issue explicit.

The record also illustrates the difference between a test result and an effect estimate. It reports p=0.04, but does not report the group difference or confidence interval. A p-value alone cannot communicate how many days the groups differed or how precisely that difference was estimated. Reporting the effect and its uncertainty would allow readers to assess the result's magnitude as well as its compatibility with the null model.

Measurement choices matter too. The endpoint uses prescription audits to define days covered with medicine. That operational definition is relevant to the stated goal of medication continuity, while remaining distinct from direct evidence about whether medicine was taken. Clear descriptions of how prescription records become covered days would help readers understand what the measure captures.

Finally, the registry lists multiple outcomes but does not report multiplicity procedures, missing-data methods, or the analysis population. Those omissions do not erase the reported result, but they constrain how fully it can be evaluated. Together, the design and reporting details show why statistical interpretation requires attention to the estimand, measurement, model, and uncertainty—not only a p-value.

16. 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 primary result is difficult to assess quantitatively because the registry does not report the estimated group difference, group-level means, standard deviations, or confidence interval. The reported p=0.04 gives no direct information about direction or magnitude. The model is identified as a generalized linear model, but its family, link function, covariates, and contrast are not reported in the registry record.

Several design details needed to appraise a cluster-randomized analysis are not reported in the registry record: the cluster unit, number of clusters, allocation procedure, and handling of within-cluster dependence. The identity of the masked party is not reported in the registry record. Although the registry warns that a simple individual-level re-analysis would be misleading, the model details needed to evaluate the sponsor's approach are not available.

Information about missing outcomes, analysis populations, multiplicity, and the statistical plan is also not reported in the registry record. It is not possible to assess the extent of missingness, how missing data were handled, whether all randomized participants contributed to the primary analysis, or whether the many listed outcomes were subject to an adjustment or testing hierarchy.

The primary endpoint is based on prescription records, and the record does not detail the rules used to derive covered days. The stated primary timeframe is the first 6 months of treatment; other follow-up for this outcome is not described. Finally, the posted safety counts are small descriptive summaries without a reported comparative analysis. These limitations constrain interpretation of the primary result and do not support extrapolation to outcomes or settings not described in the registry.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: New Technologies to Help Manage ADHD — NCT02390791.
• Registry record last updated 2019-09-13; status last verified 2019-08.
• Results first posted 2019-09-13.
• 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.

20. Record Summary

NCT02390791, titled New Technologies to Help Manage ADHD, is recorded as a completed, randomized, parallel, single-masked, cluster-randomized trial sponsored by Children's Hospital Medical Center, Cincinnati. Enrollment was 85. The study start is listed as 2015-09, with primary completion in 2017-08.

The trial compared enhanced myADHDportal.com with the treatment as usual standard portal. Its primary outcome was medication continuity, defined as days covered with medicine from prescription audits during the first 6 months of treatment. The registry reports a generalized linear model superiority analysis with p=0.04, but no effect estimate or confidence interval. It also records 0 serious adverse events and 0 other adverse events in each group, with 0 deaths in each group.