Reducing Cannabis Use for Sleep Among Adults Using Medical Cannabis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Reducing Cannabis Use for Sleep Among Adults Using Medical Cannabis (NCT03964974), a randomized na study. The primary outcome is Change From Baseline Insomnia Severity Index Score at Study Completion.

2. Clinical Question

The study description places the question in the context of adults who use medical cannabis and insomnia-related concerns, including cannabis use for sleep. Its stated purpose was to adapt and gather pilot data on a CBTi-CB intervention and sleep- and cannabis-related outcomes. The registered primary endpoint focuses specifically on perceived insomnia severity, not on cannabis consumption or every outcome described in the study rationale.

The primary question is comparative: the outcome in the CBTi-CB arm needs to be considered against the outcome in the SHE arm. A change within one arm alone would not establish how the interventions compare. The registry identifies a primary outcome and a timeframe, but does not report the numerical results needed to assess the direction, size, or precision of that comparison. Detailed eligibility criteria and the exact analysis population are also not reported in the registry record.

3. Trial Design

The registry describes a randomized, single-masked, parallel-group behavioral study. Participants were assigned to CBTi-CB or Sleep Hygiene Education. The record lists enrollment as 57. It does not describe the randomization sequence, how allocation was concealed, who was masked, or the practical meaning of masking for these interventions.

The study is listed as completed, with a start date of 2020-02-10 and a primary completion date of 2021-07-13. Its phase is listed as NA. The sponsor is the University of Michigan. These registry entries describe the study record; they do not establish how closely the conduct followed any unreported protocol or whether all planned procedures were completed.

The intervention descriptions identify both arms as behavioral, but the arm descriptions contain no further detail. The registry does not report session frequency, intervention duration, provider qualifications, adherence, treatment fidelity, or what participants in each arm were instructed to do. Those details would help readers judge how the approaches were implemented and how similar the comparison was in other respects.

The registered primary outcome is change from baseline in Insomnia Severity Index score at study completion, with a timeframe of 16 Weeks. The outcome is described as a least-squares mean with standard error. Although the record lists one outcome measure and one analysis, it also indicates that the primary outcome has several categories or time points and was not reduced to one comparison. The specific result estimates and the analysis details needed to resolve that issue are not reported in the registry record.

The study is described as a pilot-data effort in its summary. A formal sample-size justification, target enrollment rationale, interim-analysis plan, and protocol-level statistical plan are not reported in the registry record. The enrollment figure alone does not show whether the study was designed to detect a particular difference or how precisely it could estimate one.

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

Reasons for not completing this period:

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 From Baseline Insomnia Severity Index Score at Study Completion, measured over 16 Weeks.

The primary endpoint is change from baseline in the Insomnia Severity Index (ISI) score at study completion. The registry gives a timeframe of 16 Weeks and describes the outcome as a least-squares mean, with standard error as the measure of dispersion. The ISI is a self-report instrument intended to measure a participant's perception of insomnia symptoms and their effects.

The instrument has seven items covering perceived difficulty initiating sleep, staying asleep, and early morning awakening; satisfaction with the current sleep pattern; interference with daily functioning; noticeability of impairment attributed to sleep problems; and distress or concern about those problems. Its range is 0 to 28, with 28 corresponding to maximum severity. A change score is intended to express how the measured score differs from baseline at the specified assessment; interpretation requires knowing the direction used for that change and the actual estimates, which are not reported in the registry record.

Using a defined symptom measure gives the study a consistent way to quantify perceived insomnia severity across participants. It is relevant to the stated focus on sleep, but it does not by itself measure every sleep-related experience or cannabis-related outcome mentioned in the study summary. The record does not report an estimate of change in either arm, a between-arm contrast, or an uncertainty interval. Therefore, the endpoint definition is available, but the size and precision of the observed comparative result cannot be assessed from the posted results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registered design is randomized and parallel-group. In general, random assignment is intended to make treatment groups comparable at the point of assignment, on average, and supports a comparison of outcomes between assigned groups. It does not guarantee that every baseline characteristic will match, nor does it by itself resolve problems such as missing outcome data or departures from the assigned intervention. The registry does not report the randomization procedure, allocation concealment, or analysis population.

What does a change score represent?

In general, a change score compares a participant's measurement at follow-up with that participant's baseline measurement. A group summary of these changes describes the average pattern within an arm, while a comparison between arms addresses how the average changes differ. The registry names change from baseline in ISI score at study completion as the primary endpoint and gives 16 Weeks as the timeframe. It does not report the actual change estimates or the precise definition of the contrast, including the sign convention for change.

What does a least-squares mean convey?

In general, a least-squares mean is an estimated mean from a statistical model. Its value can depend on the model structure and on whether variables such as baseline score are included. The registry labels its primary outcome as a least-squares mean, but the model specification, included variables, and estimation details are not reported in the registry record. The label alone is not enough to infer that a particular adjustment was made.

How should the standard error be read?

In general, a standard error describes the precision of an estimated quantity under the analysis used. It is distinct from the standard deviation, which describes variability among individual observations. A smaller standard error generally indicates a more precise estimate, while a larger one indicates greater uncertainty. Neither can be interpreted in isolation from the estimate and the analysis that produced it. The registry lists standard error as the dispersion measure but does not post the relevant primary outcome estimates or standard errors.

A confidence interval is a general way to express the uncertainty around an estimate by giving a range of values compatible with the data and method under stated assumptions. Its width reflects precision as well as the chosen confidence level. The registry record does not report a confidence interval for the primary comparison, so no interval-based interpretation is possible. A p-value, if reported, would address compatibility with a specified null model; it would not measure the size or practical importance of an effect. No primary-comparison p-value is reported.

Why does analysis detail matter?

General statistical interpretation depends on knowing what observations entered the analysis, what model was fitted, how repeated or multiple assessments were handled, and how missing values were treated. These choices can affect both estimates and their uncertainty. The registry notes that the primary outcome has several categories or time points and was not reduced to one comparison, but it does not identify those categories or explain the analysis choice. The statistical plan, covariate adjustment, and missing-data methods are not reported in the registry record.

Over-dispersion is a general concern when count data vary more than an assumed model allows; it can make model-based uncertainty too small if ignored. It is not a reason to reinterpret this registered continuous ISI outcome as a count outcome. For a continuous measure, the relevant questions include the distribution of scores or changes, the adequacy of the model, and the uncertainty around the estimated comparison. The registry does not report diagnostics or assumption checks.

How do planning illustrations differ from trial results?

Sample-size calculations are general planning tools. A calculation based on an observed effect asks how many participants per arm might be needed under specified assumptions to detect an effect of that size at a chosen power and significance level. Such a calculation depends on assumptions, including the effect used and the statistical method. It does not show the power actually achieved in the completed study, and it cannot repair uncertainty in an unreported effect estimate. Any planning illustration is not the trial's own design justification unless the record explicitly says so; the registry does not report that justification.

These concepts clarify what would be needed for a complete reading: arm-specific outcome estimates, a defined between-arm contrast, an uncertainty measure, and enough analysis information to understand how the estimates were produced. Those elements are not reported for the primary outcome in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when researchers examine multiple outcomes, time points, groups, or contrasts. In general, making many statistical comparisons can increase the chance that at least one appears notable by chance. A prespecified primary comparison and an appropriate multiplicity strategy can help clarify which result is confirmatory and how other results should be interpreted. The registry identifies a primary outcome but also notes that it has several categories or time points and was not reduced to one comparison. The categories, time points, hierarchy of comparisons, and any adjustment for multiplicity are not reported in the registry record.

The registry lists one outcome measure and one analysis, but those labels do not resolve the note about multiple categories or time points. The exact analysis target and the relationship between it and the 16 Weeks timeframe are unclear from the posted record. The protocol, analysis plan, and any rationale for selecting a particular contrast are not reported.

Missing data can affect interpretation if participants lack follow-up ISI scores, especially when the likelihood of missingness differs between groups or relates to outcomes. General methods may use assumptions or models to address incomplete data, but those assumptions need to be understood. The registry does not report the amount or pattern of missing primary outcome data, reasons for missingness, imputation methods, or sensitivity analyses.

The analysis population is also not reported in the registry record. It is therefore unclear whether the primary analysis included all randomized participants, only participants with observed outcomes, or another defined group. Without these details, readers cannot assess how exclusions or incomplete follow-up may have affected the result. The record does not report a prespecified statistical plan against which to compare the posted analysis.

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 5% reporting threshold.

The registry lists serious adverse events, other adverse events, and deaths separately for the two arms. For CBTi-CB, the record lists 0 serious adverse events among 30 participants at risk, 0 other adverse events among 30 participants at risk, and 0 deaths. For Sleep Education, it lists 0 serious adverse events among 27 participants at risk, 0 other adverse events among 27 participants at risk, and 0 deaths. The serious adverse event term list is empty.

These are the recorded counts and risk-group denominators; they should not be converted into a claim that an intervention is risk-free. A zero count in a registry report does not establish that no event could occur, and it does not show the precision of an estimated event rate. The record does not provide event-specific follow-up duration, definitions beyond the serious/other categories, or further detail about how adverse events were collected.

The figures are descriptive safety reporting, not a statistical test comparing the arms. With no reported events in these categories, the registry does not offer an estimate of a between-arm safety difference. The listed counts also do not describe symptom changes or other outcomes that were not included in these adverse-event fields. Readers should distinguish the recorded event counts from broader conclusions about safety.

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

The record supports a description of the study question and design: a randomized parallel comparison of CBTi-CB with Sleep Hygiene Education among adults using medical cannabis, with change in ISI score at study completion as the registered primary outcome. It also provides a timeframe of 16 Weeks and describes the outcome as a least-squares mean with standard error. These details help define what the study was intended to assess.

The posted record does not report the numerical primary outcome results. It therefore does not establish the direction or size of change in either arm, the difference between arms, or the uncertainty around that difference. It cannot support a conclusion that one approach performed better, that the approaches were similar, or that there was no effect. An unreported result should not be treated as a null result.

The outcome itself is a self-report measure of perceived insomnia severity. It does not, by itself, answer every question about sleep or cannabis use in the study rationale. The registry's study summary mentions sleep- and cannabis-related outcomes, but the primary endpoint described in the record is ISI change. Other outcome results are not available in the record.

The safety entries report zero serious adverse events, zero other adverse events, and zero deaths in each arm, with the listed numbers at risk. Those entries are descriptive registry counts and do not establish the absence of all possible harms. Likewise, the study's completed status does not mean that the registry contains all details needed to judge analysis quality or outcome interpretation.

Overall, the registry record identifies a clinically relevant comparative question but does not provide the primary estimates needed to answer it. Interpretation would require the outcome values, a clearly defined between-arm contrast, uncertainty estimates, and information about analysis populations and missing data. The record does not report those details, so conclusions about comparative efficacy or the precision of any effect are not warranted from these entries alone.

13. Why This Trial Matters Statistically

This trial is statistically informative as an example of the gap between a registered question and an interpretable result. The record specifies a randomized comparison, two behavioral approaches, a self-report endpoint, and an assessment timeframe. Those features frame a comparative analysis of change in insomnia severity. Yet without posted estimates and their uncertainty, the central statistical question remains unanswered by the registry record.

The outcome is a continuous measure with a stated range and a defined direction of maximum severity. That makes clear what construct was intended to be measured, while also highlighting why the estimate and its scale matter. Statistical interpretation needs more than a label such as least-squares mean: readers need the actual arm results, the contrast of interest, and the standard error or confidence interval. The model specification is also needed to understand how a model-based mean was obtained.

The registry's note about several categories or time points is important because the analysis target affects interpretation and multiplicity. Without knowing which assessment or comparison is primary, readers cannot tell whether a reported result would correspond to the stated endpoint or how other analyses might relate to it. Clear reporting of the prespecified contrast and analysis population would strengthen the connection between design and results.

The pilot framing also makes transparent separation of estimation and hypothesis testing valuable. Pilot data may help describe feasibility or inform planning, but the record does not report a design justification or primary effect estimate. Sample-size illustrations based on observed effects are planning aids, not evidence of achieved power. The available information therefore supports careful discussion of what was registered and what remains unknown, rather than an efficacy conclusion.

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.

Several trial-specific details needed to evaluate the comparison are not reported in the registry record. The arm descriptions are blank, so the content and delivery of CBTi-CB and Sleep Hygiene Education cannot be characterized beyond their names and behavioral classification. Session structure, intervention exposure, adherence, fidelity, and any differences in contact between arms are not described.

The record identifies single masking but does not say who was masked or how masking was implemented. For behavioral interventions, this detail can be important to understand what expectations or assessment procedures might influence self-reported outcomes. Randomization is stated, but the sequence-generation and allocation-concealment procedures are not reported.

The primary outcome definition is available, but the primary estimates, the between-arm contrast, and an uncertainty interval are not. The note about multiple categories or time points leaves the exact analysis target unclear. Model specification, covariate adjustment, assumption checks, multiplicity procedures, and the analysis population are not reported in the registry record.

Follow-up completeness and missing-data handling are also not described. The record does not report how many participants had an ISI assessment at the relevant time, reasons for missing measurements, or sensitivity analyses. The enrollment figure does not substitute for those details or show the precision of the endpoint estimate.

Finally, the study summary describes an effort to gather pilot data, but the registry does not report a sample-size justification or a target effect for design purposes. The registry record therefore offers limited information for judging how the trial was planned, how it was conducted, and how robust any primary finding might be. These limitations concern what can be interpreted from the registry entry; they do not establish that the underlying study lacked procedures or data that were not reported there.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Reducing Cannabis Use for Sleep Among Adults Using Medical Cannabis — NCT03964974.
• Registry record last updated 2022-09-14; status last verified 2022-08.
• Results first posted 2022-09-14.
• 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

NCT03964974 is a completed randomized, single-masked, parallel behavioral study sponsored by the University of Michigan. Its listed enrollment is 57, and its phase is NA. The record gives a start date of 2020-02-10 and a primary completion date of 2021-07-13.

The comparison is Cognitive Behavioral Therapy for Insomnia in Cannabis Users (CBTi-CB) versus Sleep Hygiene Education (SHE). The registered primary endpoint is change from baseline in ISI score at study completion, with a timeframe of 16 Weeks; it is described as a least-squares mean with standard error.

The registry lists zero serious adverse events, zero other adverse events, and zero deaths in each arm. It does not report numerical primary outcome estimates, the between-arm effect, uncertainty intervals, detailed analysis methods, or missing-data handling. The comparative outcome therefore cannot be interpreted from the posted registry results.