Clinical Trial Tobacco Marijuana: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Clinical Trial Tobacco Marijuana (NCT02277665), a randomized na study. The primary outcome is Tobacco Quit Attempts.

2. Clinical Question

The trial addressed a treatment-development question: whether adding a tobacco-focused component to behavioral treatment for cannabis use disorder could be evaluated without adversely affecting cannabis outcomes. Its stated hypotheses included greater tobacco quit attempts and quit rates with the combined intervention, as well as no adverse effect on cannabis outcomes. These are hypotheses, not findings. The primary result described in the registry concerns tobacco quit attempts at Week 12; it does not by itself establish the effects on cannabis outcomes, tobacco cessation over a longer period, or the acceptability of the intervention.

The trial enrolled 67 participants and allocated them to one of the two behavioral-treatment groups. The registry record does not report detailed eligibility criteria, participant-level characteristics, or the specific way in which participants were assessed for a quit attempt. Those details would matter when judging how closely the enrolled group corresponds to the wider population the study intended to inform.

3. Trial Design

This was a randomized, parallel-group trial with no masking listed in the registry. It was marked completed and recorded an enrollment of 67 participants. The two arms differed by the addition of behavioral tobacco counseling and nicotine replacement therapy to the cannabis-use-disorder treatment in the combined-treatment arm. The registry lists the study phase as NA, consistent with a behavioral intervention rather than a drug-development phase.

Both arms received a 12 Week behavioral treatment involving weekly computer-assisted counseling and contingency management for cannabis use disorder. The combined arm also received behavioral tobacco counseling and nicotine replacement therapy. Thus, the comparison concerns the combined treatment package versus cannabis-use-disorder treatment alone; it does not isolate the separate contributions of tobacco counseling and nicotine replacement therapy.

The registry lists Dartmouth-Hitchcock Medical Center as the sponsor. The recorded study start was 2014-12-18 and primary completion was 2017-01-18. These dates describe the trial record and do not establish when results were analyzed or posted.

Random allocation is intended to make the treatment groups comparable on average, including with respect to factors that may affect outcomes. In a trial of this size, however, chance differences between groups remain possible. The absence of masking means participants and study staff knew which treatment was being delivered, according to the registry. That may be relevant when considering an outcome based on reporting whether a quit attempt occurred. The record does not describe how quit attempts were verified, the randomization procedure, allocation concealment, or whether outcome assessors were masked. Those aspects are not reported in the registry record.

The registry identifies six outcome measures and six analyses, but no secondary results are posted in the record represented by the available results. The primary outcome discussed in the result is the count of participants making at least one tobacco quit attempt at Week 12.

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.

Before assignment: All those who enrolled were assigned to a treatment condition

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 Tobacco Quit Attempts, measured over Week 12.

The primary endpoint is tobacco quit attempts at Week 12. The registry describes it as the percentage of participants who made at least one tobacco quit attempt. The arm results are reported as counts: 13.0 participants in the cannabis-use-disorder-only arm and 17.0 in the combined-treatment arm. The corresponding arm sizes are 33.0 and 34.0 participants.

A quit-attempt endpoint captures whether participants initiated an effort to stop tobacco use during the specified period. It is distinct from sustained abstinence: a participant can make an attempt without remaining tobacco-free, and the endpoint as described does not measure duration of abstinence or eventual cessation. It is therefore a behavioral outcome that can register an effort even when tobacco use continues.

This outcome was relevant to the intervention’s stated aim of addressing tobacco use during treatment for cannabis use disorder. It also fits the trial’s proof-of-concept focus on quit attempts and engagement with tobacco treatment. The registry’s description does not specify what counted as a quit attempt, whether reports were verified, or how missing Week 12 outcome information was handled. Those measurement details are not reported in the registry record. The endpoint should be interpreted as the recorded occurrence of at least one attempt by Week 12, rather than as evidence of successful or sustained tobacco cessation.

7. Results Posted in the Registry

• CUD and Tobacco Treatment: 50.0% (17/34) versus 39.4% (13/33) on control; risk difference 10.6 percentage points (95% CI -13.1 to 34.3), risk ratio 1.27 (0.74–2.18); unadjusted two-proportion z-test p = 0.383.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry presents a randomized comparison of two parallel groups. A parallel design assigns participants to one treatment group and follows the groups over the study period. Randomization aims to reduce systematic differences between groups at allocation, although it cannot prevent chance imbalances, especially in a modest-sized study. The record says the trial was randomized, but the specific sequence-generation method and allocation-concealment procedures are not reported in the registry record.

What does the primary comparison estimate?

The primary endpoint is binary at the participant level: each participant either made at least one tobacco quit attempt by Week 12 or did not. A risk or proportion describes the share of participants in an arm with the event. A risk difference compares those arm-specific proportions by subtraction. The posted unadjusted re-calculation uses this absolute scale, so its interpretation is the difference in the proportions with a recorded attempt. The estimate points toward a greater proportion in the combined-treatment arm, but the confidence interval spans zero, which represents no difference on the risk-difference scale.

An odds ratio compares the odds of an event between groups rather than the proportions themselves. A value of 1 is the null value: equal odds under the comparison and coding used. An odds ratio greater than 1 indicates higher odds in the numerator group, while a value below 1 indicates lower odds, assuming the event and reference group are defined consistently. The registry reports the sponsor’s result as an odds ratio of 0.56. Interpretation of its direction depends on the event and group coding, which the result entry does not fully explain. An odds ratio is also not the same quantity as a risk difference and should not be read directly as a difference in percentages.

A risk ratio compares the event proportions between groups. Its null value is also 1, but its scale differs from the odds ratio. These measures can convey different numerical magnitudes for the same data. The posted unadjusted summaries include alternative effect measures, but the main interpretation should remain anchored to the outcome and comparison actually described: participants with at least one attempt by Week 12 in the combined arm versus the cannabis-use-disorder-only arm.

What do the confidence interval and p-value add?

A confidence interval expresses uncertainty around an estimated effect under the statistical model. It gives a range of effect values compatible with the data and method at the stated confidence level; it is not a range that guarantees where the true effect lies. For the unadjusted risk-difference estimate, the interval crosses the no-difference value. That means the data do not distinguish clearly between a positive difference, no difference, and a negative difference at the stated interval level.

The re-calculated unadjusted p-value is 0.383. It measures the compatibility of the observed data with a no-difference null model, conditional on the method’s assumptions. It does not describe the size or clinical importance of an effect. A comparatively large p-value does not prove no effect, and a small p-value alone would not establish that an effect is important. The confidence interval and the estimate’s scale are needed to understand precision and plausible magnitude.

What is the sponsor’s test?

The registry lists a chi-squared analysis of an odds ratio, with an estimate of 0.56, a 95% confidence interval from .17 to 1.82, and p = 0.33. A chi-squared test assesses whether the observed group distribution is compatible with a null association under its assumptions. The odds-ratio interval includes 1, its null value. The re-calculation is an unadjusted analysis from posted arm summaries; it is not a reconstruction of every possible feature of the sponsor’s analysis. The registry does not report whether the sponsor’s analysis adjusted for other variables or used a different analysis population, nor does it document the event and reference-group coding in enough detail to reconcile the numerical estimates.

The endpoint is recorded as a count of participants with an event, but the underlying outcome is binary at the participant level. This differs from a count outcome in which one participant may contribute multiple events. For a binary participant-level result, the arm size and the number with the event define the basic comparison. The record does not state how participants with unavailable endpoint data were treated, so it is not possible to assess whether the posted counts include all randomized participants or only those with outcome information.

More generally, statistical inference depends on how outcomes, missing observations, and analysis populations are defined. Results can differ when studies use different coding, adjustment, or inclusion rules. Those distinctions should not be inferred from the estimates alone. In this trial, the registry supplies a sponsor analysis and arm-level counts, but it does not give the detail needed to reproduce the sponsor’s analysis exactly. The unadjusted re-calculation is therefore a complementary description of the posted summary data, not a replacement for the sponsor’s analysis plan or a claim that the two procedures are identical.

10. Confidence Intervals and Precision

The unadjusted re-calculated risk-difference interval spans zero, the value corresponding to no difference between the arms. This means the data are compatible with effects in either direction as well as with little or no difference. The interval is broad relative to the observed contrast, so the estimate is not precise enough to pin down a narrow range of plausible effects. The result should not be summarized only by the direction of the point estimate.

The sponsor’s reported odds-ratio interval runs from .17 to 1.82. Because it includes 1, the null value for an odds ratio, the sponsor’s analysis also leaves room for no difference. Its range includes values below and above the null, corresponding to different directions of association under a fixed event and group coding. The registry does not describe that coding in sufficient detail to give a fuller directional interpretation of the reported odds ratio.

Confidence intervals reflect sampling uncertainty under the analysis method, not every source of uncertainty. They do not account automatically for possible limitations in outcome measurement, missing data, selection of analyses, or generalizability. Those issues require information about the trial procedures and analysis plan. The registry does not report the handling of missing endpoint data, the prespecified analysis population, or the original analysis plan in enough detail to evaluate their impact.

The study’s enrollment of 67 participants and the arm counts indicate that the primary result rests on a limited dataset. The resulting uncertainty is consistent with the wide intervals. A confidence interval that includes the null is not proof of no effect, and it does not establish equivalence. Rather, the posted results leave considerable uncertainty about both the direction and magnitude of the difference in quit attempts.

11. 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 display gives planning illustrations as participants per arm for 80% power and participants per arm for 90% power. These figures are not total events. They use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set at the listed fraction of the observed effect. The original trial’s own sample-size assumptions and justification are not reported in the registry record.

In plain terms, the 80% power column gives an illustrative number of participants per arm for a design with an 80% chance of detecting the specified effect if that effect were the true one and the assumptions held. The 90% power column gives the corresponding illustration for a 90% chance. These are planning concepts for a future study, not probabilities that the observed trial result is correct. This is not observed power.

For planning at 100% of the observed effect, the illustration is 344 participants per arm for 80% power and 460 participants per arm for 90% power. Planning for 75% of the observed effect gives 609 participants per arm for 80% power and 814 participants per arm for 90% power. Planning for 50% of the observed effect gives 1,359 participants per arm for 80% power and 1,819 participants per arm for 90% power. Smaller effects require more participants because they are harder to distinguish from random variation.

These figures are conditional on the observed effect used as the planning basis. The observed estimate is imprecise, so it is not a guarantee of the effect a future trial would find. The illustrations should therefore be interpreted as consequences of the stated assumptions, not as a recommendation or as evidence that a particular future sample size is sufficient. The registry does not report the original design assumptions for this trial.

12. 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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.394; treatment proportion p₁ = 0.500; average p̄ = 0.447.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7031 = 1.3781.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6991 = 0.5884.
4. n = (1.3781 + 0.5884)² / (0.1061)² = 343.8, rounded up to 344 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
p0 <- 0.393939393939394                          # control proportion
p1 <- 0.5                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(2277665)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 344 (80% power), 460 (90% power)
 75% of observed effect: n per arm = 609 (80% power), 814 (90% power)
 50% of observed effect: n per arm = 1,359 (80% power), 1,819 (90% power)
power.prop.test: 343.8 
Simulated power at n = 344 per arm: 0.798 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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.

13. Analyses Reported by the Sponsor

The sponsor’s posted analysis uses a chi-squared method and reports an odds ratio of 0.56, a 95% confidence interval from .17 to 1.82, and p = 0.33. The unadjusted re-calculation from the arm-level summaries uses a risk difference and has a p-value of 0.383. These are not identical analyses: the effect measures differ, and the registry does not provide enough methodological detail to establish whether other features of the analysis also differ.

The sponsor’s confidence interval includes 1, the null value for an odds ratio, and the reported p-value does not indicate a clear departure from the null. The unadjusted risk-difference interval includes zero, its null value, and its point estimate is directed toward a higher event proportion in the combined-treatment arm. The distinct scales and unclear group coding mean the numerical estimates should not be treated as directly interchangeable.

The registry does not explain the sponsor’s event coding, reference-group orientation, analysis population, missing-data handling, or whether covariate adjustment was used. Without that information, the difference between the reported odds ratio and the unadjusted re-calculation cannot be fully reconciled. The sponsor analysis is the trial’s posted analysis; the re-calculation is a simple unadjusted analysis of the arm summaries and should be read as a complementary check, not as a substitute for a fully documented trial analysis.

14. Multiplicity, Missing Data and Analysis Populations

The registry record lists six outcome measures and six analyses, but no secondary results are posted. The primary result available for interpretation is tobacco quit attempts at Week 12. The absence of posted secondary findings means it is not possible to assess results for the other outcomes or how they relate to the trial’s stated aims concerning cannabis outcomes, quit rates, acceptability, and moderators.

When a study evaluates multiple outcomes or conducts multiple statistical tests, the chance of obtaining at least one apparently notable result can increase. Methods for addressing multiplicity may include designating a primary outcome in advance, adjusting inferential thresholds, or treating additional results as exploratory. The registry information does not report whether any multiplicity procedure was planned or used. It also does not give sufficient detail about the hierarchy or timing of the listed outcome measures to determine how multiplicity should affect interpretation.

The record does not report the original statistical analysis plan, prespecified analysis population, or the rules used to include participants in the primary analysis. Although the trial is described as randomized, that alone does not identify whether the analysis followed all randomized participants, only participants with observed Week 12 data, or another population. The handling of missing outcome data is also not reported in the registry record. If some participants lacked Week 12 information, the method used to handle those observations could affect the estimates; the record does not allow that possibility to be evaluated.

The outcome description does not specify the operational definition of a quit attempt or how it was ascertained. The record also does not state whether reports were verified or how repeated attempts were handled, although the endpoint is defined as having at least one attempt. These measurement details are relevant to reproducibility and interpretation. Without the analysis plan, missing-data rules, and fuller endpoint procedures, the posted result can be read as the reported comparison, but it cannot be audited against a complete prespecified statistical framework.

15. 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.

Most frequent non-serious adverse event terms

The registry reports 0 serious adverse events among 33 participants at risk in the cannabis-use-disorder-only arm and 0 among 34 participants at risk in the combined-treatment arm. It reports 0 deaths in each arm. These are the serious-event and death counts entered in the record.

For other adverse events, the registry lists 3 affected participants among 33 at risk in the cannabis-use-disorder-only arm and 5 among 34 at risk in the combined-treatment arm. These figures are counts of participants affected, not a measure of event severity, duration, or causal relationship to treatment. The registry record does not provide further event descriptions in the serious-adverse-event terms, and the information presented does not establish whether an event was caused by a study intervention.

A zero count of serious events in a trial does not establish that serious events cannot occur; it describes what was recorded among the participants at risk in this study. Likewise, the other-event counts should not be used alone to conclude that one treatment is safer. The registry does not provide details on ascertainment, timing, severity grading, or attribution sufficient for a fuller safety comparison.

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

The primary result describes the number of participants with at least one tobacco quit attempt by Week 12 in each randomized arm. The combined-treatment arm had 17.0 participants with a recorded attempt among 34.0 participants; the cannabis-use-disorder-only arm had 13.0 among 33.0. The unadjusted estimate points toward more participants with an attempt in the combined arm, but the confidence interval spans the no-difference value and the unadjusted p-value is 0.383. The sponsor’s reported odds-ratio interval also includes its null value, 1.

The result therefore does not provide a precise estimate of how much the groups differed, or establish a clear difference. It also does not demonstrate that the interventions are equivalent. The trial’s modest enrollment and the uncertainty around the estimates leave room for effects in either direction. Any interpretation should account for that imprecision rather than treating the raw count difference as a conclusive treatment effect.

The endpoint is a quit attempt, not sustained abstinence. A recorded attempt does not establish that tobacco use stopped, how long any cessation lasted, or whether the participant later resumed use. The result also does not answer whether cannabis outcomes were unaffected, whether the combined intervention was accepted by most eligible participants, or whether specific moderators predicted outcomes. Although these topics appear among the study’s stated hypotheses and aims, their findings are not reported in the registry record.

The comparison evaluates the combined tobacco-treatment package added to cannabis-use-disorder treatment. It cannot separate the effects of behavioral tobacco counseling from nicotine replacement therapy. Nor does a single Week 12 outcome establish the effects of the package in other populations, settings, or follow-up periods. The appropriate conclusion is limited to the posted comparison of recorded quit attempts at Week 12, with substantial uncertainty and without evidence in this record to answer the broader questions.

17. Why This Trial Matters Statistically

This trial is statistically informative as a small randomized comparison of an added tobacco-treatment package during behavioral treatment for cannabis use disorder. Its primary endpoint is concrete and participant-level: whether at least one quit attempt was made by Week 12. The comparison illustrates the distinction between observing different arm counts and estimating a treatment contrast with enough precision to support a clear conclusion.

Randomization provides a sound design framework for comparing interventions, but it does not make uncertainty disappear. With 67 participants enrolled, the primary estimate remains broad, and the posted confidence intervals include values compatible with different directions of effect. This is an important lesson in reading trial results: an observed point estimate describes the sample, while the interval helps show how uncertain the underlying contrast remains.

The record also illustrates why analysis details matter. The sponsor reports an odds ratio, while an unadjusted re-calculation based on arm summaries expresses the contrast as a risk difference. These measures answer related but distinct questions and use different null values. Without complete information about coding, analysis populations, and missing data, numerical differences between analyses cannot be fully explained.

Finally, the trial’s broader hypotheses concern more than the primary endpoint. The posted quit-attempt result cannot establish the effect on cannabis outcomes, successful tobacco cessation, acceptability, or treatment moderators. The registry lists several outcomes and analyses, but secondary results are not posted. The distinction between a study’s planned questions and the results available for assessment is essential when interpreting what a trial contributes.

18. 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 main limitation for interpreting this result is its imprecision. Enrollment was 67, and the confidence interval for the unadjusted primary contrast spans the no-difference value and effects in either direction. The point estimate should not be treated as a stable estimate of the effect expected in a larger or different population.

The endpoint is a reported quit attempt, not verified or sustained tobacco abstinence. The registry does not describe the operational definition, ascertainment method, or verification procedures. It also does not report follow-up beyond the Week 12 endpoint in the results. Consequently, the record cannot establish whether attempts led to cessation or persisted over time.

The study compared a combined package of behavioral tobacco counseling and nicotine replacement therapy with cannabis-use-disorder treatment alone. It cannot isolate the contribution of either tobacco-treatment component. The registry does not report enough detail about eligibility criteria or participant characteristics to judge the representativeness of the enrolled group. The absence of masking is also relevant to a participant-reported behavioral endpoint, while the method of outcome assessment is not described.

Important analytic details are not reported in the registry record: the original sample-size justification, analysis population, missing-data handling, full statistical plan, and any multiplicity approach. The sponsor’s odds-ratio result is not numerically identical to the unadjusted re-calculation, and the record does not give the coding and analysis details needed to reconcile them fully. Secondary results are not posted despite the registry listing six outcome measures and six analyses. These limitations constrain evaluation of the broader study aims and prevent a complete assessment of the planned analyses.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Clinical Trial Tobacco Marijuana — NCT02277665.
• Registry record last updated 2019-01-08; status last verified 2018-12.
• Results first posted 2019-01-08.
• 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.

22. Record Summary

NCT02277665 is a completed randomized, parallel-group trial sponsored by Dartmouth-Hitchcock Medical Center, with enrollment of 67. It compared a 12 Week behavioral treatment for cannabis use disorder alone with the same treatment combined with behavioral tobacco counseling and nicotine replacement therapy. The registry lists no masking and records the study start as 2014-12-18 and primary completion as 2017-01-18.

The primary endpoint was tobacco quit attempts at Week 12, described as the percentage of participants making at least one attempt. The posted counts were 13.0 of 33.0 in the cannabis-use-disorder-only arm and 17.0 of 34.0 in the combined-treatment arm. The unadjusted re-calculation is imprecise, and the sponsor’s reported odds-ratio confidence interval includes 1. The record does not report secondary outcome results or several details needed to assess the analysis plan and missing data.