Improving Medication Adherence in the Alabama Black Belt: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Improving Medication Adherence in the Alabama Black Belt (NCT02274844), a randomized na study. The primary outcome is Change in Self Reported Medication Adherence.

2. Clinical Question

The study was motivated by the importance of medication adherence in rural Alabama and by prior community-engaged research described in the registry summary. Its stated hypothesis was that an intervention developed within the Corbin and Strauss framework could improve adherence and health outcomes compared with usual care. That is the hypothesis being tested, not a result established by the record.

The intervention bundled educational storytelling and individual telephone coaching, so the clinical question concerns the program as a whole. The record does not specify a comparison that isolates the contribution of either component. It also lists quality of life, medication-use self-efficacy, health-care visits, hospital stays, emergency visits, and diabetes medication counts as secondary outcomes. Those measures broaden the study questions, but do not replace the stated primary endpoint.

3. Trial Design

This was a randomized, parallel-group study with no masking reported. The registry lists the phase as NA and the status as completed. Enrollment was 473. The recorded study start was 2016-04, and primary completion was 2019-01-31. The sponsor was Weill Medical College of Cornell University.

Participants were assigned to Peer Coaching or Usual Care. The peer-coaching intervention was the Living Well with Diabetes Program: educational DVDs integrating community-member storytelling and one-on-one telephone peer coaching. The comparator received an educational DVD on general health and wellness topics, without diabetes-related peer storytelling. Thus, both arms involved educational material, while the diabetes-focused stories and coaching distinguished the intervention described in the registry.

Randomization is intended to support a fair comparison by distributing participant characteristics across study arms on average. In any particular trial, however, random assignment does not guarantee that every characteristic is balanced, and the registry record does not report baseline characteristics or the randomization procedure. It also does not report allocation concealment or a planned allocation ratio. The record’s designation of randomized allocation does not by itself answer those implementation questions.

The primary outcome was assessed at baseline and 6 months. The registry describes it as change in self-reported medication adherence, using a score from 0 to 3, with higher scores indicating worse adherence. It reports exclusions from the primary-outcome analysis in both arms because participants could not be contacted, had health reasons, or did not have time. The record does not report the number analyzed at each assessment or an analysis population definition.

The registry lists several secondary outcomes, including quality-of-life measures, self-efficacy measures, health-care utilization, and medication counts. Further detail about their results and statistical handling is not reported in the registry record. The record does not state whether assessors were masked, how the intervention was delivered consistently, or whether participants in different arms could have shared intervention materials.

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

Analysis population for the primary outcome, as defined in the registry: 39 participants in the peer coaching arm and 31 participants in the usual care arm were not analyzed because the participants were unable to be contacted, had health reasons, or did not have time.

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 Change in Self Reported Medication Adherence, measured over Baseline, 6 months.

The primary endpoint is change in self-reported medication adherence, assessed from baseline to 6 months. The measure is described as a score on a scale from 0 to 3; a higher score means worse adherence. The endpoint therefore represents participants’ reports about their medication-taking behavior, rather than a direct count of doses taken or an objective measure of medication availability.

Change over time is useful because it relates follow-up status to the participant’s starting point. Comparing change between randomized arms can help assess whether the intervention group’s score moved differently from the usual-care group’s score over the assessment period. A between-arm comparison is important: a change within one arm alone does not establish that the intervention caused a difference, because changes can occur over time in either arm.

Self-report can capture behavior that may be difficult to observe directly, but it also depends on recall and how participants interpret and answer questions. The registry record does not report further details about the instrument, its validation, scoring procedures beyond the stated range and direction, or how missing items were handled. It also does not provide the primary-outcome means, standard deviations, estimated change, confidence interval, or p-value.

The record indicates that some participants were not analyzed for this outcome, with inability to contact them, health reasons, or lack of time given as reasons. Because the numerical primary comparison is not reported in the registry record, the magnitude and precision of any difference in change cannot be described from this record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What a change-score comparison means. The primary outcome is a continuous score measured at baseline and 6 months. In general, an analysis of change compares how much the score moves over time in one arm with how much it moves in the other. Because higher scores indicate worse adherence, the direction of a change must be interpreted using that scoring rule. The registry record does not report the actual changes or a single between-arm estimate.

A change within the peer-coaching arm by itself would not answer the randomized trial question. Participants’ scores might change over time for reasons unrelated to the intervention. The comparison arm provides a reference for those shared influences. The contrast between the arms is the part relevant to the intervention comparison, subject to the assumptions and limitations of the analysis.

What a confidence interval would describe

In general, a confidence interval expresses uncertainty around an estimated effect. A narrow interval indicates greater precision than a wide interval, while an interval spanning effects in different directions can leave the direction uncertain. It does not give the probability that a particular treatment effect is true. No primary-outcome confidence interval is reported in the registry record, so the precision of the estimated comparison cannot be assessed from the record.

What a p-value can and cannot answer

In general, a p-value measures how surprising data at least as incompatible with a specified null hypothesis would be if that hypothesis and the statistical model were correct. It is not the probability that the null hypothesis is true, nor a measure of clinical importance. The registry record does not report a primary-outcome p-value or identify a single primary comparison to which one could apply.

Why standard deviations matter

For a continuous outcome, the standard deviation describes how spread out participant scores are around the mean. It is not the same as uncertainty in the estimated mean or the between-arm contrast. Greater variability can make it harder to estimate a difference precisely, all else being equal. The primary outcome is described as a mean with standard deviation, but the numerical arm-level values are not reported in the registry record.

Planning illustrations and observed power

In general, sample-size planning asks how many participants would be needed to detect a specified effect under stated assumptions. The registry record describes planning assumptions using a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. Such counts are planning illustrations, not observed power and not a measure of whether the trial’s result is reliable. The numerical observed effect and the resulting planning counts are not reported in the registry record, so no numerical planning interpretation can be made from the available details.

Planning calculations based on an observed effect also depend on that effect estimate. In general, an estimate from a study can be uncertain; using it for planning does not remove that uncertainty. A sample-size illustration does not establish that the trial was adequately powered, and a completed study’s observed result should be interpreted using its estimated effect and uncertainty rather than a retrospective power calculation. The registry record does not report a sample-size justification or the assumptions used for the original trial design.

Over-dispersion and model fit

In general, over-dispersion means that outcome variability is greater than a statistical model expects. If a model assumes less variability than the data contain, its uncertainty estimates can be too small and its tests can appear more decisive than warranted. This concept is especially relevant when the outcome distribution or model assumptions differ from a simple continuous-outcome model. The registry record describes a continuous mean outcome but does not report a fitted model, diagnostics, or any assessment of over-dispersion. There is no basis in the record to say whether it was present or relevant to the primary analysis.

What randomization contributes

In general, random assignment helps make the comparison groups similar in expectation, including on factors that were not measured. It does not prevent missing data, guarantee exact balance in a particular sample, or eliminate problems caused by deviations after assignment. The registry identifies the allocation as randomized, but details such as sequence generation, concealment, adherence to assignment, and analysis by assigned group are not reported.

The record reports one analysis and 16 outcome measures, but does not set out the statistical model, covariate adjustment, handling of repeated observations, or analysis population for the primary endpoint. Without those details and the actual estimate, the statistical methods and precision of the primary comparison cannot be independently characterized.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 16 outcome measures, including the primary medication-adherence measure and secondary measures of quality of life, self-efficacy, health-care visits, hospital stays, emergency visits, and medication counts. When multiple outcomes or comparisons are assessed, the chance of at least one apparently unusual result can increase. In general, a prespecified hierarchy or multiplicity adjustment can help address that issue. The registry record does not report whether multiplicity was considered, which outcomes were included in any adjustment, or how the primary outcome’s repeated time points or categories were handled.

The primary outcome is identified as change from baseline to 6 months, but the record does not reduce it to one reported comparison. The record also lists one analysis, without explaining its relationship to the multiple outcome measures. Details of any prespecified analysis plan, outcome hierarchy, interim analyses, or statistical adjustment are not reported in the registry record.

Missing outcome data are a material interpretive issue. The registry states that 39 participants in the peer-coaching arm and 31 in the usual-care arm were not analyzed for the primary outcome because they could not be contacted, had health reasons, or did not have time. It does not report the number with complete data at each assessment, whether these reasons overlapped, or whether the excluded participants differed from those analyzed.

In general, analysis of only participants with observed outcomes can be biased if missingness is related to outcome values or other factors that differ between arms. Methods such as imputation or sensitivity analyses can examine the effect of assumptions about missing outcomes, but the registry record does not report whether these or other approaches were used. It also does not define an intention-to-treat or per-protocol analysis population. Therefore, the effect of missing data and the population represented by any primary estimate cannot be determined from the record.

11. 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 safety counts for Peer Coaching and Usual Care. For Peer Coaching, the listed at-risk count is 203, with zero serious adverse events, zero other adverse events, and zero deaths. For Usual Care, the listed at-risk count is 270, also with zero serious adverse events, zero other adverse events, and zero deaths.

These are the counts recorded in the registry’s safety fields. They do not establish that no adverse experiences occurred under every definition or throughout every period of follow-up. The registry record does not report the adverse-event definitions, the safety observation period, collection procedures, or further details about event monitoring. Zero recorded events should therefore be interpreted as a registry report, not as proof of absence of risk.

The at-risk counts are safety denominators and should not be assumed to be the number of participants included in the primary-outcome analysis. The record separately reports participants not analyzed for the primary endpoint. Safety reporting and outcome analysis can involve different populations and follow-up completeness.

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

The registry record establishes the study question, intervention components, comparator, endpoint definition, and some information about enrollment and missing primary-outcome data. It does not report a single numerical primary-outcome comparison. As a result, the record cannot establish the direction or magnitude of a difference between peer coaching and usual care, or the precision of such a difference.

The intended comparison is between a program combining diabetes-focused educational storytelling and telephone peer coaching and a general-health DVD comparator. Because those intervention components were delivered together, a comparison would concern the full program package; it would not isolate the separate effect of storytelling or coaching. The record does not report enough intervention-fidelity detail to assess how consistently the program was delivered.

The endpoint is self-reported, and the scale’s direction is important: higher scores indicate worse adherence. A score is not itself a direct measure of medication ingestion. The registry does not report detailed measurement procedures, psychometric information, or the numerical results needed to interpret score changes.

The randomized design is a strength for comparing groups, but it does not resolve incomplete outcome data or establish that every analytic decision preserved the intended randomized comparison. The reported non-analysis of participants for contact, health, or time reasons may affect interpretation, and the registry does not describe how missing data were handled. The record also does not report adjustment for multiple outcomes or provide a prespecified statistical analysis plan.

Accordingly, the documented details support a description of what the trial was designed to compare, not a conclusion that the program did or did not change adherence. No claim about clinical importance, statistical significance, or effectiveness can be drawn from a primary result that is not reported in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically notable for asking a community-based behavioral question using a patient-reported continuous outcome. The intervention was developed around peer stories and individual coaching, while the comparator also received educational material. That design makes the comparison more specific than intervention versus no contact, but it still evaluates a bundled program rather than separate components.

The outcome’s direction needs careful attention: higher adherence scores represent worse adherence. Any interpretation of an estimated change would need to preserve that scoring direction and distinguish within-arm change from the between-arm contrast. The registry describes assessment at baseline and 6 months but does not report the numerical comparison, so neither its direction nor precision can be evaluated.

Missing primary-outcome data are also important. The registry gives reasons for non-analysis in both arms, but does not state how missing outcomes were handled or whether those excluded differed from participants analyzed. Statistical conclusions can depend on those details, particularly when outcome availability may be related to health or participation circumstances.

The record lists a range of secondary measures alongside the primary endpoint. This raises general questions about outcome prioritization and multiplicity, but the registry does not report the relevant analysis plan or any adjustment. The record therefore illustrates why study design information and an outcome definition are not substitutes for estimates, uncertainty intervals, and transparent analysis details.

For biostatistical interpretation, the key distinction is between a stated hypothesis and evidence bearing on that hypothesis. The trial’s registry entry documents the question and broad design, but it does not provide the numerical primary-result evidence required to quantify the comparison.

14. 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 most consequential limitation for interpreting the primary question is that a single numerical comparison is not reported in the registry record. Arm-specific outcome summaries, a between-arm change estimate, uncertainty interval, and p-value are not available in the record. Therefore, the direction, magnitude, and precision of the primary result cannot be assessed.

The record reports that 39 participants in the Peer Coaching arm and 31 in the Usual Care arm were not analyzed for the primary outcome. Reasons included inability to contact participants, health reasons, or lack of time. It does not report the number assessed at each time point, the characteristics of those not analyzed, or the method used to address missingness.

The endpoint is self-reported medication adherence. The record gives the score range and direction but not further instrument details, validation information, or procedures for handling incomplete responses. Self-report and missing follow-up can affect the interpretation of an adherence measure, although their impact cannot be quantified from the registry information.

Additional limitations of the available record include absent detail on the randomization procedure, allocation concealment, baseline characteristics, intervention fidelity, and the primary statistical model. The registry lists multiple outcome measures but does not report a multiplicity strategy or a statistical analysis plan. It also does not report whether analyses followed randomized assignment or whether sensitivity analyses were conducted.

The safety fields list zero serious adverse events, other adverse events, and deaths in both arms, but event definitions and observation procedures are not reported. These omissions limit the conclusions that can be drawn about safety reporting. The limitations describe gaps in the registry record; they do not establish that the trial itself did not collect or analyze additional information.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Improving Medication Adherence in the Alabama Black Belt — NCT02274844.
• Registry record last updated 2020-10-29; status last verified 2020-10.
• Results first posted 2020-10-29.
• 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.

18. Record Summary

NCT02274844, titled “Improving Medication Adherence in the Alabama Black Belt,” is a completed randomized, parallel-group study sponsored by Weill Medical College of Cornell University. The registry lists enrollment of 473, phase NA, a start of 2016-04, and primary completion on 2019-01-31.

The study compared the Living Well with Diabetes Program, including storytelling DVDs and one-on-one telephone peer coaching, with usual care consisting of a general-health educational DVD. The primary outcome was change in self-reported medication-adherence score from baseline to 6 months; higher scores on the 0-to-3 scale indicate worse adherence.

The registry record does not report a single numerical primary-outcome comparison. It notes participants not analyzed for that outcome in both arms, and lists secondary outcomes spanning quality of life, self-efficacy, health-care use, and medication counts. Safety fields report zero serious adverse events, other adverse events, and deaths in both arms.