Impact of Pillboxes on Medical Adherence: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Impact of Pillboxes on Medical Adherence (NCT03861845), a randomized na study. The primary outcome is Medication Adherence at 1 Month.

2. Clinical Question

3. Trial Design

This was a randomized, double-masked, parallel-group study. The registry record lists the study as completed, with an enrollment of 15 and phase listed as NA. The study dates are listed as 2019-02-10 for the start and 2019-05-01 for primary completion. The record describes three pillbox intervention arms rather than a no-intervention arm.

All three intervention descriptions include reflection on medication routines and education and training on how to use the pillbox. The arms differ in the pillbox approach: standard off-the-shelf, custom off-the-shelf, or custom designed and manufactured. The study summary describes three visits with the research team for baseline data collection, receipt of the pillbox, and follow-up data collection. Further details about visit timing beyond the primary outcome timeframe are not reported in the registry record.

The posted primary outcome is medication adherence at 1 month. The three arm summaries report a mean score, standard deviation, and number of participants: the standard off-the-shelf group has n=6, the custom off-the-shelf group has n=5, and the custom designed and manufactured group has n=4. The registry record also lists “Number of Participants Who Completed the Study” as a secondary outcome, but does not post its result.

Randomization and masking are recorded, but the method used to generate the allocation sequence, conceal assignment, or maintain masking is not reported in the registry record. The specific sample-size justification and statistical analysis plan are also not reported in the registry record. These omissions limit how fully the design and the intended analysis can be assessed.

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.

Population: All participants enrolled in the study were analyzed. All participants completed the study.

6. Primary Endpoint

The registry lists the primary outcome as Medication Adherence at 1 Month, measured over 1 month.

The primary endpoint is medication adherence at 1 month, measured with the Adherence to Refills and Medications Scale (ARMS). The record describes ARMS as a 7-item assessment. Participants indicate the extent of their adherence on a 4-point Likert-like scale, and the total score ranges from 7 - 28. Higher scores indicate worse medication adherence, so lower scores correspond to better reported adherence on this measure.

Medication adherence is relevant to the study because the interventions are intended to support medication routines through pillbox use, reflection, and education and training. A structured scale gives the study a consistent way to summarize participants’ reported adherence across the intervention groups. The registry record posts each group’s mean and standard deviation, rather than individual participant scores or a model-based estimate.

The endpoint is a continuous score, and the posted summary uses the mean as its measure of central tendency and the standard deviation as its measure of spread. These summaries describe the recorded scores in each arm; they do not, on their own, quantify the uncertainty around a difference between arms. The registry record does not report further details about how ARMS responses were collected, handled if incomplete, or incorporated into an inferential analysis.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is continuous: it is a numerical score, and the registry record summarizes it with a mean and standard deviation. The mean gives the arithmetic average of the observed scores in an arm. It is useful for describing the group’s central tendency, but can be affected by unusually high or low observations. The standard deviation describes how dispersed the observed scores are around the mean. It is not a confidence interval and does not show how precisely the mean or a difference between arms has been estimated.

What do the group summaries tell us?

The posted values describe the participants whose results contributed to each arm summary. The group means are 11.83, 10.2, and 9.33; their standard deviations are 2.71, 2.17, and 2.08, respectively. The corresponding group sizes are 6, 5, and 4. Since higher ARMS scores indicate worse adherence, the ordering of these means can be described in terms of the scale’s direction. But the summaries do not include a formal estimate of uncertainty for contrasts between groups. They also do not reveal the distribution of individual scores, which could matter when assessing whether a mean is an adequate summary.

Why are confidence intervals useful?

In general, a confidence interval expresses the uncertainty around an estimate under a specified statistical method. For a comparison of group means, it can show which values of a difference are compatible with the data and analysis assumptions. A narrow interval generally indicates greater precision than a wide interval. The registry record does not report confidence intervals for these arm means or comparisons, so the precision of a difference cannot be assessed from posted inferential results.

What would a comparison require?

A comparative analysis would need to specify the contrast of interest, such as comparing one pillbox arm with another, and the statistical model or test used. It would also need to address whether adjustment for baseline values or other factors was intended. Those analysis details are not reported in the registry record. The record lists a continuous endpoint but does not post any analysis estimates or tests. No unadjusted comparison has been recalculated for this interpretation, and there is no sponsor analysis to compare against.

In general, an analysis of continuous outcomes may use a model that estimates differences in means, with uncertainty expressed through a confidence interval and a test statistic or p-value. The validity and interpretation of such an analysis depend on its assumptions and the study design. With small groups, estimates can be particularly uncertain, and distributional assumptions may be difficult to assess. The registry record does not describe a planned model, its assumptions, or diagnostics, so those choices cannot be evaluated.

How should the ARMS score be understood?

The ARMS description identifies a 7-item assessment using a 4-point Likert-like response scale. Its total score ranges from 7 - 28, with higher scores indicating worse medication adherence. A mean summarizes the total scores as though the scale can be represented numerically at the group level; that is the summary approach posted in the record. The meaning of a score change depends on the measurement properties and context of the instrument. The registry record does not report a threshold for a meaningful difference, validation details for this study population, or a minimal important difference. A clinical interpretation of any hypothetical contrast would therefore require information beyond the posted summaries.

General statistical reasoning also distinguishes statistical significance from practical importance. A small p-value, if one were reported, would not by itself establish that a difference is important to participants. Conversely, absence of a small p-value would not prove that groups are equivalent. Equivalence or non-inferiority requires a prespecified margin and suitable design and analysis; these are not reported in the registry record. No such conclusion can be drawn from means and standard deviations alone.

What is the role of sample-size planning?

The registry record describes sample-size illustrations as based on the observed effect, a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. It does not report resulting participant counts for those illustrations. Such planning calculations are not observed power and do not alter the interpretation of the participants actually enrolled. The trial enrolled 15 participants, divided across the three posted groups as n=6, n=5, and n=4. The record does not report the trial’s own sample-size justification.

More generally, power is a design-stage probability tied to assumptions about the true effect, variability, analysis, and sample size. It is not a property that can establish a result after the study has been completed. Planning illustrations based on an observed effect can be especially dependent on the observed data. They cannot substitute for reporting an effect estimate with its uncertainty, and they do not provide evidence that the observed group means represent a stable population pattern.

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("Standard Off-the-shelf Pillbox", "Custom Off-the-shelf", "Custom Designed and Manufactured")
n   <- c(6, 5, 4)
m   <- c(11.83, 10.2, 9.33)
sp  <- c(2.71, 2.17, 2.08)
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  Standard Off-the-shelf Pillbox  6  11.830  2.710  8.986  14.674
CARM  Custom Off-the-shelf  5  10.200  2.170  7.506  12.894
CARM  Custom Designed and Manufactured  4  9.330  2.080  6.020  12.640

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists three outcome measures and identifies one secondary outcome: the number of participants who completed the study. It does not post a result for that secondary outcome. The record reports zero analyses, and it does not provide an inferential analysis of the primary endpoint. A multiplicity strategy—how the analysis would handle potential false-positive findings across multiple outcomes or comparisons—is not reported in the registry record.

Multiplicity is a general concern when several outcomes or contrasts are examined. Testing many hypotheses can increase the chance that at least one result appears unusual by chance. Possible approaches depend on which outcomes and comparisons were prespecified and what conclusions are intended. Without a reported analysis plan or results for the listed outcomes, it is not possible to assess how multiplicity was addressed in this trial. No adjustment should be assumed.

Missing-data handling is not reported in the registry record. The record does not state whether all enrolled participants contributed a one-month ARMS score, how incomplete questionnaires were handled, or whether any missing outcome values were imputed. The arm counts associated with the posted summaries are 6, 5, and 4; they do not by themselves explain whether every randomized participant completed follow-up or was included in the analysis.

The analysis population is also not reported in the registry record. It is not stated whether the posted results follow an intention-to-treat approach, include only participants with observed outcome data, or use another definition. These details matter because excluding participants or handling incomplete outcomes differently can affect the interpretation of group summaries. The record also does not report protocol deviations or the number of participants assessed at each visit. These gaps prevent a full assessment of missingness, attrition, and the population represented by the endpoint results.

12. 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 record posts serious adverse event figures separately for the three pillbox groups. The standard off-the-shelf group had 0 serious adverse events among 6 participants at risk; the custom off-the-shelf group had 0 among 5 at risk; and the custom designed and manufactured group had 0 among 4 at risk. The record also reports 0 other adverse events in each arm and 0 deaths in each arm.

These figures describe the events recorded in this trial during the period covered by the registry results. Zero reported events does not demonstrate that an intervention has no risks. With small groups, the absence of observed events offers limited information about how frequently uncommon events might occur in a broader population. The study summary identifies potential study risks including loss of time, boredom, invasion of privacy, and medication errors, but the posted adverse-event counts do not provide further detail about those risks or their monitoring.

The registry record does not list serious adverse event terms. It also does not report event timing, severity grading beyond the serious versus other categorization, relatedness assessments, or additional safety analyses. The figures should therefore be read as counts reported for the study, not as a complete estimate of the safety profile of the pillbox approaches.

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

The posted results establish that the registry record contains one-month ARMS group summaries for three pillbox approaches. The mean scores are 11.83 for the standard off-the-shelf pillbox, 10.2 for the custom off-the-shelf pillbox, and 9.33 for the custom designed and manufactured pillbox. Because higher scores indicate worse adherence, the lower means are in a direction associated with better adherence on this scale. The results also report standard deviations and group sizes, which help describe the observed data.

Those observations do not establish that the differences between group means are statistically reliable, clinically important, or caused by the pillbox approach. The trial is randomized, but the registry record does not post inferential comparisons, confidence intervals, p-values, or an analysis plan. The numerical summaries alone cannot account for uncertainty, and they do not establish equivalence among the groups either. A finding of no reported analysis is not evidence of no difference.

The endpoint is a self-reported scale score, not a direct measure of every medication-taking behavior or a clinical outcome. It is measured at 1 month, so the posted endpoint does not describe adherence over a longer period. The record does not report an important-difference threshold for ARMS, or additional information that would connect the observed score patterns to health outcomes.

The record also does not report the analysis population, missing-data approach, or a result for the listed completion outcome. The small group sizes and absence of inferential results restrict the conclusions that can be drawn. The trial’s results are best treated as descriptive observations from the participants whose outcomes contributed to each posted summary. They can motivate further study, but they are not a basis for a treatment recommendation or a definitive comparative claim.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of the difference between descriptive summaries and comparative evidence. The registry record posts means and standard deviations for a continuous outcome across three randomized arms. These statistics characterize the observed scores, but a reader also needs an estimate of uncertainty around between-group contrasts to assess how precisely the groups differ. That inferential step is absent from the posted results.

The group sizes are small: n=6, n=5, and n=4. In small groups, estimates can be sensitive to individual observations, and standard deviations may themselves be uncertain. The posted means have a clear ordering, but ordering alone does not establish a stable effect or rule out chance variation. This distinction is important in clinical research, where a numerical pattern can be mistaken for a conclusive result if uncertainty is not reported.

The study also illustrates how endpoint interpretation depends on the measurement scale. The ARMS description specifies that higher scores indicate worse adherence. Keeping that direction explicit is necessary when describing the means. At the same time, a score is not automatically a direct measure of adherence behavior or clinical benefit, and the registry record does not provide a threshold for an important score difference.

Finally, the study record leaves important analysis details unreported: the statistical plan, missing-data approach, analysis population, multiplicity strategy, and result for the secondary completion outcome. Reporting those elements alongside effect estimates and uncertainty would make the evidence easier to evaluate. The posted summaries remain useful as a transparent description of the recorded outcome, while the missing inferential information limits comparative conclusions.

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 primary limitation for interpreting the posted outcome is the small number of participants in each arm: 6, 5, and 4. Such group sizes provide limited information about the variability of scores and make the arm summaries sensitive to individual observations. The registry record does not post individual scores, confidence intervals, or comparative estimates that would help readers assess uncertainty.

The registry record does not report a statistical analysis plan, the method used to analyze the primary outcome, or an inferential result. It also does not report the trial’s own sample-size justification. Although the record describes sample-size illustrations based on the observed effect and specified assumptions, it does not give resulting participant counts. Those illustrations cannot replace an analysis of the observed trial data.

Further limitations are the absence of reported detail about the analysis population, missing-data handling, and participant completion by arm. The record names a secondary completion outcome but does not post its result. It also does not report the method of random sequence generation, allocation concealment, or how masking was maintained. These details would help assess the conduct and interpretation of the trial.

The endpoint is a 1-month ARMS score. The record does not report how incomplete scale responses were handled, an important-difference threshold, or additional measurement properties for this study. Follow-up beyond the stated timeframe and longer-term adherence results are not reported in the registry record. The posted safety counts also have limited ability to characterize uncommon events given the small group sizes, and further safety details are not reported.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Impact of Pillboxes on Medical Adherence — NCT03861845.
• Registry record last updated 2020-08-26; status last verified 2020-08.
• Results first posted 2020-08-26.
• 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.

19. Record Summary

The registry record describes a completed, randomized, double-masked, parallel-group study of pillbox approaches for medication adherence. It lists enrollment of 15 and three arms: standard off-the-shelf, custom off-the-shelf, and custom designed and manufactured pillboxes. The primary outcome is the ARMS score at 1 month, for which higher scores indicate worse adherence.

Posted mean scores are 11.83, 10.2, and 9.33, with standard deviations of 2.71, 2.17, and 2.08, and group sizes of 6, 5, and 4, respectively. The record reports no inferential analyses or comparative estimates. It lists one secondary outcome, study completion, without a posted result. Serious adverse events, other adverse events, and deaths are reported as 0 in each arm.

Important analysis details—including the statistical plan, analysis population, and missing-data approach—are not reported in the registry record. The posted results therefore support a descriptive reading of the recorded scores, not a definitive conclusion about differences among the pillbox approaches.